diff options
Diffstat (limited to 'libraries')
89 files changed, 16353 insertions, 16343 deletions
diff --git a/libraries/LocalPeer/CMakeLists.txt b/libraries/LocalPeer/CMakeLists.txt index 99e3fe4d..f476da38 100644 --- a/libraries/LocalPeer/CMakeLists.txt +++ b/libraries/LocalPeer/CMakeLists.txt @@ -12,18 +12,18 @@ include/LocalPeer.h ) if(UNIX) - list(APPEND SINGLE_SOURCES - src/LockedFile_unix.cpp - ) + list(APPEND SINGLE_SOURCES + src/LockedFile_unix.cpp + ) endif() if(WIN32) - list(APPEND SINGLE_SOURCES - src/LockedFile_win.cpp - ) + list(APPEND SINGLE_SOURCES + src/LockedFile_win.cpp + ) endif() add_library(LocalPeer STATIC ${SINGLE_SOURCES}) target_include_directories(LocalPeer PUBLIC include) -qt5_use_modules(LocalPeer Core Network) +target_link_libraries(LocalPeer Qt5::Core Qt5::Network) diff --git a/libraries/LocalPeer/include/LocalPeer.h b/libraries/LocalPeer/include/LocalPeer.h index 7558f18e..a24e4775 100644 --- a/libraries/LocalPeer/include/LocalPeer.h +++ b/libraries/LocalPeer/include/LocalPeer.h @@ -50,51 +50,51 @@ class LockedFile; class ApplicationId { public: /* methods */ - // traditional app = installed system wide and used in a multi-user environment - static ApplicationId fromTraditionalApp(); - // ID based on a path with all the application data (no two instances with the same data path should run) - static ApplicationId fromPathAndVersion(const QString & dataPath, const QString & version); - // custom ID - static ApplicationId fromCustomId(const QString & id); - // custom ID, based on a raw string previously acquired from 'toString' - static ApplicationId fromRawString(const QString & id); + // traditional app = installed system wide and used in a multi-user environment + static ApplicationId fromTraditionalApp(); + // ID based on a path with all the application data (no two instances with the same data path should run) + static ApplicationId fromPathAndVersion(const QString & dataPath, const QString & version); + // custom ID + static ApplicationId fromCustomId(const QString & id); + // custom ID, based on a raw string previously acquired from 'toString' + static ApplicationId fromRawString(const QString & id); - QString toString() - { - return m_id; - } + QString toString() + { + return m_id; + } private: /* methods */ - ApplicationId(const QString & value) - { - m_id = value; - } + ApplicationId(const QString & value) + { + m_id = value; + } private: /* data */ - QString m_id; + QString m_id; }; class LocalPeer : public QObject { - Q_OBJECT + Q_OBJECT public: - LocalPeer(QObject *parent, const ApplicationId &appId); - ~LocalPeer(); - bool isClient(); - bool sendMessage(const QString &message, int timeout); - ApplicationId applicationId() const; + LocalPeer(QObject *parent, const ApplicationId &appId); + ~LocalPeer(); + bool isClient(); + bool sendMessage(const QString &message, int timeout); + ApplicationId applicationId() const; Q_SIGNALS: - void messageReceived(const QString &message); + void messageReceived(const QString &message); protected Q_SLOTS: - void receiveConnection(); + void receiveConnection(); protected: - ApplicationId id; - QString socketName; - std::unique_ptr<QLocalServer> server; - std::unique_ptr<LockedFile> lockFile; + ApplicationId id; + QString socketName; + std::unique_ptr<QLocalServer> server; + std::unique_ptr<LockedFile> lockFile; }; diff --git a/libraries/LocalPeer/src/LocalPeer.cpp b/libraries/LocalPeer/src/LocalPeer.cpp index f71c62d8..129f3abc 100644 --- a/libraries/LocalPeer/src/LocalPeer.cpp +++ b/libraries/LocalPeer/src/LocalPeer.cpp @@ -67,60 +67,60 @@ static const char* ack = "ack"; ApplicationId ApplicationId::fromTraditionalApp() { - QString protoId = QCoreApplication::applicationFilePath(); + QString protoId = QCoreApplication::applicationFilePath(); #if defined(Q_OS_WIN) - protoId = protoId.toLower(); + protoId = protoId.toLower(); #endif - auto prefix = protoId.section(QLatin1Char('/'), -1); - prefix.remove(QRegExp("[^a-zA-Z]")); - prefix.truncate(6); - QByteArray idc = protoId.toUtf8(); - quint16 idNum = qChecksum(idc.constData(), idc.size()); - auto socketName = QLatin1String("qtsingleapp-") + prefix + QLatin1Char('-') + QString::number(idNum, 16); + auto prefix = protoId.section(QLatin1Char('/'), -1); + prefix.remove(QRegExp("[^a-zA-Z]")); + prefix.truncate(6); + QByteArray idc = protoId.toUtf8(); + quint16 idNum = qChecksum(idc.constData(), idc.size()); + auto socketName = QLatin1String("qtsingleapp-") + prefix + QLatin1Char('-') + QString::number(idNum, 16); #if defined(Q_OS_WIN) - if (!pProcessIdToSessionId) - { - QLibrary lib("kernel32"); - pProcessIdToSessionId = (PProcessIdToSessionId)lib.resolve("ProcessIdToSessionId"); - } - if (pProcessIdToSessionId) - { - DWORD sessionId = 0; - pProcessIdToSessionId(GetCurrentProcessId(), &sessionId); - socketName += QLatin1Char('-') + QString::number(sessionId, 16); - } + if (!pProcessIdToSessionId) + { + QLibrary lib("kernel32"); + pProcessIdToSessionId = (PProcessIdToSessionId)lib.resolve("ProcessIdToSessionId"); + } + if (pProcessIdToSessionId) + { + DWORD sessionId = 0; + pProcessIdToSessionId(GetCurrentProcessId(), &sessionId); + socketName += QLatin1Char('-') + QString::number(sessionId, 16); + } #else - socketName += QLatin1Char('-') + QString::number(::getuid(), 16); + socketName += QLatin1Char('-') + QString::number(::getuid(), 16); #endif - return ApplicationId(socketName); + return ApplicationId(socketName); } ApplicationId ApplicationId::fromPathAndVersion(const QString& dataPath, const QString& version) { - QCryptographicHash shasum(QCryptographicHash::Algorithm::Sha1); - QString result = dataPath + QLatin1Char('-') + version; - shasum.addData(result.toUtf8()); - return ApplicationId(QLatin1String("qtsingleapp-") + QString::fromLatin1(shasum.result().toHex())); + QCryptographicHash shasum(QCryptographicHash::Algorithm::Sha1); + QString result = dataPath + QLatin1Char('-') + version; + shasum.addData(result.toUtf8()); + return ApplicationId(QLatin1String("qtsingleapp-") + QString::fromLatin1(shasum.result().toHex())); } ApplicationId ApplicationId::fromCustomId(const QString& id) { - return ApplicationId(QLatin1String("qtsingleapp-") + id); + return ApplicationId(QLatin1String("qtsingleapp-") + id); } ApplicationId ApplicationId::fromRawString(const QString& id) { - return ApplicationId(id); + return ApplicationId(id); } LocalPeer::LocalPeer(QObject * parent, const ApplicationId &appId) - : QObject(parent), id(appId) + : QObject(parent), id(appId) { - socketName = id.toString(); - server.reset(new QLocalServer()); - QString lockName = QDir(QDir::tempPath()).absolutePath() + QLatin1Char('/') + socketName + QLatin1String("-lockfile"); - lockFile.reset(new LockedFile(lockName)); - lockFile->open(QIODevice::ReadWrite); + socketName = id.toString(); + server.reset(new QLocalServer()); + QString lockName = QDir(QDir::tempPath()).absolutePath() + QLatin1Char('/') + socketName + QLatin1String("-lockfile"); + lockFile.reset(new LockedFile(lockName)); + lockFile->open(QIODevice::ReadWrite); } LocalPeer::~LocalPeer() @@ -129,113 +129,113 @@ LocalPeer::~LocalPeer() ApplicationId LocalPeer::applicationId() const { - return id; + return id; } bool LocalPeer::isClient() { - if (lockFile->isLocked()) - return false; + if (lockFile->isLocked()) + return false; - if (!lockFile->lock(LockedFile::WriteLock, false)) - return true; + if (!lockFile->lock(LockedFile::WriteLock, false)) + return true; - bool res = server->listen(socketName); + bool res = server->listen(socketName); #if defined(Q_OS_UNIX) - // ### Workaround - if (!res && server->serverError() == QAbstractSocket::AddressInUseError) { - QFile::remove(QDir::cleanPath(QDir::tempPath())+QLatin1Char('/')+socketName); - res = server->listen(socketName); - } + // ### Workaround + if (!res && server->serverError() == QAbstractSocket::AddressInUseError) { + QFile::remove(QDir::cleanPath(QDir::tempPath())+QLatin1Char('/')+socketName); + res = server->listen(socketName); + } #endif - if (!res) - qWarning("QtSingleCoreApplication: listen on local socket failed, %s", qPrintable(server->errorString())); - QObject::connect(server.get(), SIGNAL(newConnection()), SLOT(receiveConnection())); - return false; + if (!res) + qWarning("QtSingleCoreApplication: listen on local socket failed, %s", qPrintable(server->errorString())); + QObject::connect(server.get(), SIGNAL(newConnection()), SLOT(receiveConnection())); + return false; } bool LocalPeer::sendMessage(const QString &message, int timeout) { - if (!isClient()) - return false; - - QLocalSocket socket; - bool connOk = false; - for(int i = 0; i < 2; i++) { - // Try twice, in case the other instance is just starting up - socket.connectToServer(socketName); - connOk = socket.waitForConnected(timeout/2); - if (connOk || i) - { - break; - } - std::this_thread::sleep_for(std::chrono::milliseconds(250)); - } - if (!connOk) - { - return false; - } - - QByteArray uMsg(message.toUtf8()); - QDataStream ds(&socket); - - ds.writeBytes(uMsg.constData(), uMsg.size()); - if(!socket.waitForBytesWritten(timeout)) - { - return false; - } - - // wait for 'ack' - if(!socket.waitForReadyRead(timeout)) - { - return false; - } - - // make sure we got 'ack' - if(!(socket.read(qstrlen(ack)) == ack)) - { - return false; - } - return true; + if (!isClient()) + return false; + + QLocalSocket socket; + bool connOk = false; + for(int i = 0; i < 2; i++) { + // Try twice, in case the other instance is just starting up + socket.connectToServer(socketName); + connOk = socket.waitForConnected(timeout/2); + if (connOk || i) + { + break; + } + std::this_thread::sleep_for(std::chrono::milliseconds(250)); + } + if (!connOk) + { + return false; + } + + QByteArray uMsg(message.toUtf8()); + QDataStream ds(&socket); + + ds.writeBytes(uMsg.constData(), uMsg.size()); + if(!socket.waitForBytesWritten(timeout)) + { + return false; + } + + // wait for 'ack' + if(!socket.waitForReadyRead(timeout)) + { + return false; + } + + // make sure we got 'ack' + if(!(socket.read(qstrlen(ack)) == ack)) + { + return false; + } + return true; } void LocalPeer::receiveConnection() { - QLocalSocket* socket = server->nextPendingConnection(); - if (!socket) - { - return; - } - - while (socket->bytesAvailable() < (int)sizeof(quint32)) - { - socket->waitForReadyRead(); - } - QDataStream ds(socket); - QByteArray uMsg; - quint32 remaining; - ds >> remaining; - uMsg.resize(remaining); - int got = 0; - char* uMsgBuf = uMsg.data(); - do - { - got = ds.readRawData(uMsgBuf, remaining); - remaining -= got; - uMsgBuf += got; - } while (remaining && got >= 0 && socket->waitForReadyRead(2000)); - if (got < 0) - { - qWarning("QtLocalPeer: Message reception failed %s", socket->errorString().toLatin1().constData()); - delete socket; - return; - } - QString message(QString::fromUtf8(uMsg)); - socket->write(ack, qstrlen(ack)); - socket->waitForBytesWritten(1000); - socket->waitForDisconnected(1000); // make sure client reads ack - delete socket; - emit messageReceived(message); //### (might take a long time to return) + QLocalSocket* socket = server->nextPendingConnection(); + if (!socket) + { + return; + } + + while (socket->bytesAvailable() < (int)sizeof(quint32)) + { + socket->waitForReadyRead(); + } + QDataStream ds(socket); + QByteArray uMsg; + quint32 remaining; + ds >> remaining; + uMsg.resize(remaining); + int got = 0; + char* uMsgBuf = uMsg.data(); + do + { + got = ds.readRawData(uMsgBuf, remaining); + remaining -= got; + uMsgBuf += got; + } while (remaining && got >= 0 && socket->waitForReadyRead(2000)); + if (got < 0) + { + qWarning("QtLocalPeer: Message reception failed %s", socket->errorString().toLatin1().constData()); + delete socket; + return; + } + QString message(QString::fromUtf8(uMsg)); + socket->write(ack, qstrlen(ack)); + socket->waitForBytesWritten(1000); + socket->waitForDisconnected(1000); // make sure client reads ack + delete socket; + emit messageReceived(message); //### (might take a long time to return) } diff --git a/libraries/LocalPeer/src/LockedFile.cpp b/libraries/LocalPeer/src/LockedFile.cpp index a4951bfe..73294a16 100644 --- a/libraries/LocalPeer/src/LockedFile.cpp +++ b/libraries/LocalPeer/src/LockedFile.cpp @@ -41,70 +41,70 @@ #include "LockedFile.h" /*! - \class QtLockedFile - - \brief The QtLockedFile class extends QFile with advisory locking - functions. - - A file may be locked in read or write mode. Multiple instances of - \e QtLockedFile, created in multiple processes running on the same - machine, may have a file locked in read mode. Exactly one instance - may have it locked in write mode. A read and a write lock cannot - exist simultaneously on the same file. - - The file locks are advisory. This means that nothing prevents - another process from manipulating a locked file using QFile or - file system functions offered by the OS. Serialization is only - guaranteed if all processes that access the file use - QLockedFile. Also, while holding a lock on a file, a process - must not open the same file again (through any API), or locks - can be unexpectedly lost. - - The lock provided by an instance of \e QtLockedFile is released - whenever the program terminates. This is true even when the - program crashes and no destructors are called. + \class QtLockedFile + + \brief The QtLockedFile class extends QFile with advisory locking + functions. + + A file may be locked in read or write mode. Multiple instances of + \e QtLockedFile, created in multiple processes running on the same + machine, may have a file locked in read mode. Exactly one instance + may have it locked in write mode. A read and a write lock cannot + exist simultaneously on the same file. + + The file locks are advisory. This means that nothing prevents + another process from manipulating a locked file using QFile or + file system functions offered by the OS. Serialization is only + guaranteed if all processes that access the file use + QLockedFile. Also, while holding a lock on a file, a process + must not open the same file again (through any API), or locks + can be unexpectedly lost. + + The lock provided by an instance of \e QtLockedFile is released + whenever the program terminates. This is true even when the + program crashes and no destructors are called. */ /*! \enum QtLockedFile::LockMode - This enum describes the available lock modes. + This enum describes the available lock modes. - \value ReadLock A read lock. - \value WriteLock A write lock. - \value NoLock Neither a read lock nor a write lock. + \value ReadLock A read lock. + \value WriteLock A write lock. + \value NoLock Neither a read lock nor a write lock. */ /*! - Constructs an unlocked \e QtLockedFile object. This constructor - behaves in the same way as \e QFile::QFile(). + Constructs an unlocked \e QtLockedFile object. This constructor + behaves in the same way as \e QFile::QFile(). - \sa QFile::QFile() + \sa QFile::QFile() */ LockedFile::LockedFile() - : QFile() + : QFile() { #ifdef Q_OS_WIN - wmutex = 0; - rmutex = 0; + wmutex = 0; + rmutex = 0; #endif - m_lock_mode = NoLock; + m_lock_mode = NoLock; } /*! - Constructs an unlocked QtLockedFile object with file \a name. This - constructor behaves in the same way as \e QFile::QFile(const - QString&). + Constructs an unlocked QtLockedFile object with file \a name. This + constructor behaves in the same way as \e QFile::QFile(const + QString&). - \sa QFile::QFile() + \sa QFile::QFile() */ LockedFile::LockedFile(const QString &name) - : QFile(name) + : QFile(name) { #ifdef Q_OS_WIN - wmutex = 0; - rmutex = 0; + wmutex = 0; + rmutex = 0; #endif - m_lock_mode = NoLock; + m_lock_mode = NoLock; } /*! @@ -122,72 +122,72 @@ Returns true if successful; otherwise false. */ bool LockedFile::open(OpenMode mode) { - if (mode & QIODevice::Truncate) { - qWarning("QtLockedFile::open(): Truncate mode not allowed."); - return false; - } - return QFile::open(mode); + if (mode & QIODevice::Truncate) { + qWarning("QtLockedFile::open(): Truncate mode not allowed."); + return false; + } + return QFile::open(mode); } /*! - Returns \e true if this object has a in read or write lock; - otherwise returns \e false. + Returns \e true if this object has a in read or write lock; + otherwise returns \e false. - \sa lockMode() + \sa lockMode() */ bool LockedFile::isLocked() const { - return m_lock_mode != NoLock; + return m_lock_mode != NoLock; } /*! - Returns the type of lock currently held by this object, or \e - QtLockedFile::NoLock. + Returns the type of lock currently held by this object, or \e + QtLockedFile::NoLock. - \sa isLocked() + \sa isLocked() */ LockedFile::LockMode LockedFile::lockMode() const { - return m_lock_mode; + return m_lock_mode; } /*! - \fn bool QtLockedFile::lock(LockMode mode, bool block = true) + \fn bool QtLockedFile::lock(LockMode mode, bool block = true) - Obtains a lock of type \a mode. The file must be opened before it - can be locked. + Obtains a lock of type \a mode. The file must be opened before it + can be locked. - If \a block is true, this function will block until the lock is - aquired. If \a block is false, this function returns \e false - immediately if the lock cannot be aquired. + If \a block is true, this function will block until the lock is + aquired. If \a block is false, this function returns \e false + immediately if the lock cannot be aquired. - If this object already has a lock of type \a mode, this function - returns \e true immediately. If this object has a lock of a - different type than \a mode, the lock is first released and then a - new lock is obtained. + If this object already has a lock of type \a mode, this function + returns \e true immediately. If this object has a lock of a + different type than \a mode, the lock is first released and then a + new lock is obtained. - This function returns \e true if, after it executes, the file is - locked by this object, and \e false otherwise. + This function returns \e true if, after it executes, the file is + locked by this object, and \e false otherwise. - \sa unlock(), isLocked(), lockMode() + \sa unlock(), isLocked(), lockMode() */ /*! - \fn bool QtLockedFile::unlock() + \fn bool QtLockedFile::unlock() - Releases a lock. + Releases a lock. - If the object has no lock, this function returns immediately. + If the object has no lock, this function returns immediately. - This function returns \e true if, after it executes, the file is - not locked by this object, and \e false otherwise. + This function returns \e true if, after it executes, the file is + not locked by this object, and \e false otherwise. - \sa lock(), isLocked(), lockMode() + \sa lock(), isLocked(), lockMode() */ /*! - \fn QtLockedFile::~QtLockedFile() + \fn QtLockedFile::~QtLockedFile() - Destroys the \e QtLockedFile object. If any locks were held, they - are released. + Destroys the \e QtLockedFile object. If any locks were held, they + are released. */ diff --git a/libraries/LocalPeer/src/LockedFile.h b/libraries/LocalPeer/src/LockedFile.h index 8c178250..2f29ee20 100644 --- a/libraries/LocalPeer/src/LockedFile.h +++ b/libraries/LocalPeer/src/LockedFile.h @@ -48,30 +48,30 @@ class LockedFile : public QFile { public: - enum LockMode { NoLock = 0, ReadLock, WriteLock }; + enum LockMode { NoLock = 0, ReadLock, WriteLock }; - LockedFile(); - LockedFile(const QString &name); - ~LockedFile(); + LockedFile(); + LockedFile(const QString &name); + ~LockedFile(); - bool open(OpenMode mode); + bool open(OpenMode mode); - bool lock(LockMode mode, bool block = true); - bool unlock(); - bool isLocked() const; - LockMode lockMode() const; + bool lock(LockMode mode, bool block = true); + bool unlock(); + bool isLocked() const; + LockMode lockMode() const; - private: + private: #ifdef Q_OS_WIN - Qt::HANDLE wmutex; - Qt::HANDLE rmutex; - QVector<Qt::HANDLE> rmutexes; - QString mutexname; + Qt::HANDLE wmutex; + Qt::HANDLE rmutex; + QVector<Qt::HANDLE> rmutexes; + QString mutexname; - Qt::HANDLE getMutexHandle(int idx, bool doCreate); - bool waitMutex(Qt::HANDLE mutex, bool doBlock); + Qt::HANDLE getMutexHandle(int idx, bool doCreate); + bool waitMutex(Qt::HANDLE mutex, bool doBlock); #endif - LockMode m_lock_mode; + LockMode m_lock_mode; }; diff --git a/libraries/LocalPeer/src/LockedFile_unix.cpp b/libraries/LocalPeer/src/LockedFile_unix.cpp index 4b68916c..6becc89e 100644 --- a/libraries/LocalPeer/src/LockedFile_unix.cpp +++ b/libraries/LocalPeer/src/LockedFile_unix.cpp @@ -47,68 +47,68 @@ bool LockedFile::lock(LockMode mode, bool block) { - if (!isOpen()) { - qWarning("QtLockedFile::lock(): file is not opened"); - return false; - } + if (!isOpen()) { + qWarning("QtLockedFile::lock(): file is not opened"); + return false; + } - if (mode == NoLock) - return unlock(); + if (mode == NoLock) + return unlock(); - if (mode == m_lock_mode) - return true; + if (mode == m_lock_mode) + return true; - if (m_lock_mode != NoLock) - unlock(); + if (m_lock_mode != NoLock) + unlock(); - struct flock fl; - fl.l_whence = SEEK_SET; - fl.l_start = 0; - fl.l_len = 0; - fl.l_type = (mode == ReadLock) ? F_RDLCK : F_WRLCK; - int cmd = block ? F_SETLKW : F_SETLK; - int ret = fcntl(handle(), cmd, &fl); + struct flock fl; + fl.l_whence = SEEK_SET; + fl.l_start = 0; + fl.l_len = 0; + fl.l_type = (mode == ReadLock) ? F_RDLCK : F_WRLCK; + int cmd = block ? F_SETLKW : F_SETLK; + int ret = fcntl(handle(), cmd, &fl); - if (ret == -1) { - if (errno != EINTR && errno != EAGAIN) - qWarning("QtLockedFile::lock(): fcntl: %s", strerror(errno)); - return false; - } + if (ret == -1) { + if (errno != EINTR && errno != EAGAIN) + qWarning("QtLockedFile::lock(): fcntl: %s", strerror(errno)); + return false; + } - m_lock_mode = mode; - return true; + m_lock_mode = mode; + return true; } bool LockedFile::unlock() { - if (!isOpen()) { - qWarning("QtLockedFile::unlock(): file is not opened"); - return false; - } - - if (!isLocked()) - return true; - - struct flock fl; - fl.l_whence = SEEK_SET; - fl.l_start = 0; - fl.l_len = 0; - fl.l_type = F_UNLCK; - int ret = fcntl(handle(), F_SETLKW, &fl); - - if (ret == -1) { - qWarning("QtLockedFile::lock(): fcntl: %s", strerror(errno)); - return false; - } - - m_lock_mode = NoLock; - return true; + if (!isOpen()) { + qWarning("QtLockedFile::unlock(): file is not opened"); + return false; + } + + if (!isLocked()) + return true; + + struct flock fl; + fl.l_whence = SEEK_SET; + fl.l_start = 0; + fl.l_len = 0; + fl.l_type = F_UNLCK; + int ret = fcntl(handle(), F_SETLKW, &fl); + + if (ret == -1) { + qWarning("QtLockedFile::lock(): fcntl: %s", strerror(errno)); + return false; + } + + m_lock_mode = NoLock; + return true; } LockedFile::~LockedFile() { - if (isOpen()) - unlock(); + if (isOpen()) + unlock(); } diff --git a/libraries/LocalPeer/src/LockedFile_win.cpp b/libraries/LocalPeer/src/LockedFile_win.cpp index f93ef27f..93d2c73b 100644 --- a/libraries/LocalPeer/src/LockedFile_win.cpp +++ b/libraries/LocalPeer/src/LockedFile_win.cpp @@ -48,158 +48,158 @@ Qt::HANDLE LockedFile::getMutexHandle(int idx, bool doCreate) { - if (mutexname.isEmpty()) { - QFileInfo fi(*this); - mutexname = QString::fromLatin1(MUTEX_PREFIX) - + fi.absoluteFilePath().toLower(); - } - QString mname(mutexname); - if (idx >= 0) - mname += QString::number(idx); - - Qt::HANDLE mutex; - if (doCreate) { - mutex = CreateMutexW(NULL, FALSE, (LPCWSTR)mname.utf16()); - if (!mutex) { - qErrnoWarning("QtLockedFile::lock(): CreateMutex failed"); - return 0; - } - } - else { - OpenMutexW(SYNCHRONIZE | MUTEX_MODIFY_STATE, FALSE, (LPCWSTR)mname.utf16()); - if (!mutex) { - if (GetLastError() != ERROR_FILE_NOT_FOUND) - qErrnoWarning("QtLockedFile::lock(): OpenMutex failed"); - return 0; - } - } - return mutex; + if (mutexname.isEmpty()) { + QFileInfo fi(*this); + mutexname = QString::fromLatin1(MUTEX_PREFIX) + + fi.absoluteFilePath().toLower(); + } + QString mname(mutexname); + if (idx >= 0) + mname += QString::number(idx); + + Qt::HANDLE mutex; + if (doCreate) { + mutex = CreateMutexW(NULL, FALSE, (LPCWSTR)mname.utf16()); + if (!mutex) { + qErrnoWarning("QtLockedFile::lock(): CreateMutex failed"); + return 0; + } + } + else { + mutex = OpenMutexW(SYNCHRONIZE | MUTEX_MODIFY_STATE, FALSE, (LPCWSTR)mname.utf16()); + if (!mutex) { + if (GetLastError() != ERROR_FILE_NOT_FOUND) + qErrnoWarning("QtLockedFile::lock(): OpenMutex failed"); + return 0; + } + } + return mutex; } bool LockedFile::waitMutex(Qt::HANDLE mutex, bool doBlock) { - Q_ASSERT(mutex); - DWORD res = WaitForSingleObject(mutex, doBlock ? INFINITE : 0); - switch (res) { - case WAIT_OBJECT_0: - case WAIT_ABANDONED: - return true; - break; - case WAIT_TIMEOUT: - break; - default: - qErrnoWarning("QtLockedFile::lock(): WaitForSingleObject failed"); - } - return false; + Q_ASSERT(mutex); + DWORD res = WaitForSingleObject(mutex, doBlock ? INFINITE : 0); + switch (res) { + case WAIT_OBJECT_0: + case WAIT_ABANDONED: + return true; + break; + case WAIT_TIMEOUT: + break; + default: + qErrnoWarning("QtLockedFile::lock(): WaitForSingleObject failed"); + } + return false; } bool LockedFile::lock(LockMode mode, bool block) { - if (!isOpen()) { - qWarning("QtLockedFile::lock(): file is not opened"); - return false; - } - - if (mode == NoLock) - return unlock(); - - if (mode == m_lock_mode) - return true; - - if (m_lock_mode != NoLock) - unlock(); - - if (!wmutex && !(wmutex = getMutexHandle(-1, true))) - return false; - - if (!waitMutex(wmutex, block)) - return false; - - if (mode == ReadLock) { - int idx = 0; - for (; idx < MAX_READERS; idx++) { - rmutex = getMutexHandle(idx, false); - if (!rmutex || waitMutex(rmutex, false)) - break; - CloseHandle(rmutex); - } - bool ok = true; - if (idx >= MAX_READERS) { - qWarning("QtLockedFile::lock(): too many readers"); - rmutex = 0; - ok = false; - } - else if (!rmutex) { - rmutex = getMutexHandle(idx, true); - if (!rmutex || !waitMutex(rmutex, false)) - ok = false; - } - if (!ok && rmutex) { - CloseHandle(rmutex); - rmutex = 0; - } - ReleaseMutex(wmutex); - if (!ok) - return false; - } - else { - Q_ASSERT(rmutexes.isEmpty()); - for (int i = 0; i < MAX_READERS; i++) { - Qt::HANDLE mutex = getMutexHandle(i, false); - if (mutex) - rmutexes.append(mutex); - } - if (rmutexes.size()) { - DWORD res = WaitForMultipleObjects(rmutexes.size(), rmutexes.constData(), - TRUE, block ? INFINITE : 0); - if (res != WAIT_OBJECT_0 && res != WAIT_ABANDONED) { - if (res != WAIT_TIMEOUT) - qErrnoWarning("QtLockedFile::lock(): WaitForMultipleObjects failed"); - m_lock_mode = WriteLock; // trick unlock() to clean up - semiyucky - unlock(); - return false; - } - } - } - - m_lock_mode = mode; - return true; + if (!isOpen()) { + qWarning("QtLockedFile::lock(): file is not opened"); + return false; + } + + if (mode == NoLock) + return unlock(); + + if (mode == m_lock_mode) + return true; + + if (m_lock_mode != NoLock) + unlock(); + + if (!wmutex && !(wmutex = getMutexHandle(-1, true))) + return false; + + if (!waitMutex(wmutex, block)) + return false; + + if (mode == ReadLock) { + int idx = 0; + for (; idx < MAX_READERS; idx++) { + rmutex = getMutexHandle(idx, false); + if (!rmutex || waitMutex(rmutex, false)) + break; + CloseHandle(rmutex); + } + bool ok = true; + if (idx >= MAX_READERS) { + qWarning("QtLockedFile::lock(): too many readers"); + rmutex = 0; + ok = false; + } + else if (!rmutex) { + rmutex = getMutexHandle(idx, true); + if (!rmutex || !waitMutex(rmutex, false)) + ok = false; + } + if (!ok && rmutex) { + CloseHandle(rmutex); + rmutex = 0; + } + ReleaseMutex(wmutex); + if (!ok) + return false; + } + else { + Q_ASSERT(rmutexes.isEmpty()); + for (int i = 0; i < MAX_READERS; i++) { + Qt::HANDLE mutex = getMutexHandle(i, false); + if (mutex) + rmutexes.append(mutex); + } + if (rmutexes.size()) { + DWORD res = WaitForMultipleObjects(rmutexes.size(), rmutexes.constData(), + TRUE, block ? INFINITE : 0); + if (res != WAIT_OBJECT_0 && res != WAIT_ABANDONED) { + if (res != WAIT_TIMEOUT) + qErrnoWarning("QtLockedFile::lock(): WaitForMultipleObjects failed"); + m_lock_mode = WriteLock; // trick unlock() to clean up - semiyucky + unlock(); + return false; + } + } + } + + m_lock_mode = mode; + return true; } bool LockedFile::unlock() { - if (!isOpen()) { - qWarning("QtLockedFile::unlock(): file is not opened"); - return false; - } - - if (!isLocked()) - return true; - - if (m_lock_mode == ReadLock) { - ReleaseMutex(rmutex); - CloseHandle(rmutex); - rmutex = 0; - } - else { - foreach(Qt::HANDLE mutex, rmutexes) { - ReleaseMutex(mutex); - CloseHandle(mutex); - } - rmutexes.clear(); - ReleaseMutex(wmutex); - } - - m_lock_mode = LockedFile::NoLock; - return true; + if (!isOpen()) { + qWarning("QtLockedFile::unlock(): file is not opened"); + return false; + } + + if (!isLocked()) + return true; + + if (m_lock_mode == ReadLock) { + ReleaseMutex(rmutex); + CloseHandle(rmutex); + rmutex = 0; + } + else { + foreach(Qt::HANDLE mutex, rmutexes) { + ReleaseMutex(mutex); + CloseHandle(mutex); + } + rmutexes.clear(); + ReleaseMutex(wmutex); + } + + m_lock_mode = LockedFile::NoLock; + return true; } LockedFile::~LockedFile() { - if (isOpen()) - unlock(); - if (wmutex) - CloseHandle(wmutex); + if (isOpen()) + unlock(); + if (wmutex) + CloseHandle(wmutex); } diff --git a/libraries/classparser/CMakeLists.txt b/libraries/classparser/CMakeLists.txt index db266f53..3fe7591d 100644 --- a/libraries/classparser/CMakeLists.txt +++ b/libraries/classparser/CMakeLists.txt @@ -6,7 +6,7 @@ set(CMAKE_AUTOMOC ON) include(TestBigEndian) test_big_endian(BIGENDIAN) if(${BIGENDIAN}) - add_definitions(-DMULTIMC_BIG_ENDIAN) + add_definitions(-DMULTIMC_BIG_ENDIAN) endif(${BIGENDIAN}) # Find Qt @@ -38,5 +38,4 @@ add_definitions(-DCLASSPARSER_LIBRARY) add_library(MultiMC_classparser STATIC ${CLASSPARSER_SOURCES} ${CLASSPARSER_HEADERS}) target_include_directories(MultiMC_classparser PUBLIC "${CMAKE_CURRENT_SOURCE_DIR}/include") -target_link_libraries(MultiMC_classparser MultiMC_quazip) -qt5_use_modules(MultiMC_classparser Core) +target_link_libraries(MultiMC_classparser MultiMC_quazip Qt5::Core) diff --git a/libraries/classparser/include/classparser.h b/libraries/classparser/include/classparser.h index 23a65589..6c619e47 100644 --- a/libraries/classparser/include/classparser.h +++ b/libraries/classparser/include/classparser.h @@ -1,4 +1,4 @@ -/* Copyright 2013-2018 MultiMC Contributors +/* Copyright 2013-2019 MultiMC Contributors * * Authors: Orochimarufan <orochimarufan.x3@gmail.com> * diff --git a/libraries/classparser/include/classparser_config.h b/libraries/classparser/include/classparser_config.h index db8f40a3..ee053de6 100644 --- a/libraries/classparser/include/classparser_config.h +++ b/libraries/classparser/include/classparser_config.h @@ -1,4 +1,4 @@ -/* Copyright 2013-2018 MultiMC Contributors +/* Copyright 2013-2019 MultiMC Contributors * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. diff --git a/libraries/classparser/src/annotations.cpp b/libraries/classparser/src/annotations.cpp index d1a7c046..18a9e880 100644 --- a/libraries/classparser/src/annotations.cpp +++ b/libraries/classparser/src/annotations.cpp @@ -6,80 +6,80 @@ namespace java { std::string annotation::toString() { - std::ostringstream ss; - ss << "Annotation type : " << type_index << " - " << pool[type_index].str_data << std::endl; - ss << "Contains " << name_val_pairs.size() << " pairs:" << std::endl; - for (unsigned i = 0; i < name_val_pairs.size(); i++) - { - std::pair<uint16_t, element_value *> &val = name_val_pairs[i]; - auto name_idx = val.first; - ss << pool[name_idx].str_data << "(" << name_idx << ")" - << " = " << val.second->toString() << std::endl; - } - return ss.str(); + std::ostringstream ss; + ss << "Annotation type : " << type_index << " - " << pool[type_index].str_data << std::endl; + ss << "Contains " << name_val_pairs.size() << " pairs:" << std::endl; + for (unsigned i = 0; i < name_val_pairs.size(); i++) + { + std::pair<uint16_t, element_value *> &val = name_val_pairs[i]; + auto name_idx = val.first; + ss << pool[name_idx].str_data << "(" << name_idx << ")" + << " = " << val.second->toString() << std::endl; + } + return ss.str(); } annotation *annotation::read(util::membuffer &input, constant_pool &pool) { - uint16_t type_index = 0; - input.read_be(type_index); - annotation *ann = new annotation(type_index, pool); + uint16_t type_index = 0; + input.read_be(type_index); + annotation *ann = new annotation(type_index, pool); - uint16_t num_pairs = 0; - input.read_be(num_pairs); - while (num_pairs) - { - uint16_t name_idx = 0; - // read name index - input.read_be(name_idx); - auto elem = element_value::readElementValue(input, pool); - // read value - ann->add_pair(name_idx, elem); - num_pairs--; - } - return ann; + uint16_t num_pairs = 0; + input.read_be(num_pairs); + while (num_pairs) + { + uint16_t name_idx = 0; + // read name index + input.read_be(name_idx); + auto elem = element_value::readElementValue(input, pool); + // read value + ann->add_pair(name_idx, elem); + num_pairs--; + } + return ann; } element_value *element_value::readElementValue(util::membuffer &input, - java::constant_pool &pool) + java::constant_pool &pool) { - element_value_type type = INVALID; - input.read(type); - uint16_t index = 0; - uint16_t index2 = 0; - std::vector<element_value *> vals; - switch (type) - { - case PRIMITIVE_BYTE: - case PRIMITIVE_CHAR: - case PRIMITIVE_DOUBLE: - case PRIMITIVE_FLOAT: - case PRIMITIVE_INT: - case PRIMITIVE_LONG: - case PRIMITIVE_SHORT: - case PRIMITIVE_BOOLEAN: - case STRING: - input.read_be(index); - return new element_value_simple(type, index, pool); - case ENUM_CONSTANT: - input.read_be(index); - input.read_be(index2); - return new element_value_enum(type, index, index2, pool); - case CLASS: // Class - input.read_be(index); - return new element_value_class(type, index, pool); - case ANNOTATION: // Annotation - // FIXME: runtime visibility info needs to be passed from parent - return new element_value_annotation(ANNOTATION, annotation::read(input, pool), pool); - case ARRAY: // Array - input.read_be(index); - for (int i = 0; i < index; i++) - { - vals.push_back(element_value::readElementValue(input, pool)); - } - return new element_value_array(ARRAY, vals, pool); - default: - throw new java::classfile_exception(); - } + element_value_type type = INVALID; + input.read(type); + uint16_t index = 0; + uint16_t index2 = 0; + std::vector<element_value *> vals; + switch (type) + { + case PRIMITIVE_BYTE: + case PRIMITIVE_CHAR: + case PRIMITIVE_DOUBLE: + case PRIMITIVE_FLOAT: + case PRIMITIVE_INT: + case PRIMITIVE_LONG: + case PRIMITIVE_SHORT: + case PRIMITIVE_BOOLEAN: + case STRING: + input.read_be(index); + return new element_value_simple(type, index, pool); + case ENUM_CONSTANT: + input.read_be(index); + input.read_be(index2); + return new element_value_enum(type, index, index2, pool); + case CLASS: // Class + input.read_be(index); + return new element_value_class(type, index, pool); + case ANNOTATION: // Annotation + // FIXME: runtime visibility info needs to be passed from parent + return new element_value_annotation(ANNOTATION, annotation::read(input, pool), pool); + case ARRAY: // Array + input.read_be(index); + for (int i = 0; i < index; i++) + { + vals.push_back(element_value::readElementValue(input, pool)); + } + return new element_value_array(ARRAY, vals, pool); + default: + throw new java::classfile_exception(); + } } }
\ No newline at end of file diff --git a/libraries/classparser/src/annotations.h b/libraries/classparser/src/annotations.h index dd603af3..15bf05a4 100644 --- a/libraries/classparser/src/annotations.h +++ b/libraries/classparser/src/annotations.h @@ -7,21 +7,21 @@ namespace java { enum element_value_type : uint8_t { - INVALID = 0, - STRING = 's', - ENUM_CONSTANT = 'e', - CLASS = 'c', - ANNOTATION = '@', - ARRAY = '[', // one array dimension - PRIMITIVE_INT = 'I', // integer - PRIMITIVE_BYTE = 'B', // signed byte - PRIMITIVE_CHAR = 'C', // Unicode character code point in the Basic Multilingual Plane, - // encoded with UTF-16 - PRIMITIVE_DOUBLE = 'D', // double-precision floating-point value - PRIMITIVE_FLOAT = 'F', // single-precision floating-point value - PRIMITIVE_LONG = 'J', // long integer - PRIMITIVE_SHORT = 'S', // signed short - PRIMITIVE_BOOLEAN = 'Z' // true or false + INVALID = 0, + STRING = 's', + ENUM_CONSTANT = 'e', + CLASS = 'c', + ANNOTATION = '@', + ARRAY = '[', // one array dimension + PRIMITIVE_INT = 'I', // integer + PRIMITIVE_BYTE = 'B', // signed byte + PRIMITIVE_CHAR = 'C', // Unicode character code point in the Basic Multilingual Plane, + // encoded with UTF-16 + PRIMITIVE_DOUBLE = 'D', // double-precision floating-point value + PRIMITIVE_FLOAT = 'F', // single-precision floating-point value + PRIMITIVE_LONG = 'J', // long integer + PRIMITIVE_SHORT = 'S', // signed short + PRIMITIVE_BOOLEAN = 'Z' // true or false }; /** * The element_value structure is a discriminated union representing the value of an @@ -37,21 +37,21 @@ enum element_value_type : uint8_t class element_value { protected: - element_value_type type; - constant_pool &pool; + element_value_type type; + constant_pool &pool; public: - element_value(element_value_type type, constant_pool &pool) : type(type), pool(pool) {}; - virtual ~element_value() {} + element_value(element_value_type type, constant_pool &pool) : type(type), pool(pool) {}; + virtual ~element_value() {} - element_value_type getElementValueType() - { - return type; - } + element_value_type getElementValueType() + { + return type; + } - virtual std::string toString() = 0; + virtual std::string toString() = 0; - static element_value *readElementValue(util::membuffer &input, constant_pool &pool); + static element_value *readElementValue(util::membuffer &input, constant_pool &pool); }; /** @@ -62,58 +62,58 @@ public: class annotation { public: - typedef std::vector<std::pair<uint16_t, element_value *>> value_list; + typedef std::vector<std::pair<uint16_t, element_value *>> value_list; protected: - /** - * The value of the type_index item must be a valid index into the constant_pool table. - * The constant_pool entry at that index must be a CONSTANT_Utf8_info (§4.4.7) structure - * representing a field descriptor representing the annotation type corresponding - * to the annotation represented by this annotation structure. - */ - uint16_t type_index; - /** - * map between element_name_index and value. - * - * The value of the element_name_index item must be a valid index into the constant_pool - *table. - * The constant_pool entry at that index must be a CONSTANT_Utf8_info (§4.4.7) structure - *representing - * a valid field descriptor (§4.3.2) that denotes the name of the annotation type element - *represented - * by this element_value_pairs entry. - */ - value_list name_val_pairs; - /** - * Reference to the parent constant pool - */ - constant_pool &pool; + /** + * The value of the type_index item must be a valid index into the constant_pool table. + * The constant_pool entry at that index must be a CONSTANT_Utf8_info (§4.4.7) structure + * representing a field descriptor representing the annotation type corresponding + * to the annotation represented by this annotation structure. + */ + uint16_t type_index; + /** + * map between element_name_index and value. + * + * The value of the element_name_index item must be a valid index into the constant_pool + *table. + * The constant_pool entry at that index must be a CONSTANT_Utf8_info (§4.4.7) structure + *representing + * a valid field descriptor (§4.3.2) that denotes the name of the annotation type element + *represented + * by this element_value_pairs entry. + */ + value_list name_val_pairs; + /** + * Reference to the parent constant pool + */ + constant_pool &pool; public: - annotation(uint16_t type_index, constant_pool &pool) - : type_index(type_index), pool(pool) {}; - ~annotation() - { - for (unsigned i = 0; i < name_val_pairs.size(); i++) - { - delete name_val_pairs[i].second; - } - } - void add_pair(uint16_t key, element_value *value) - { - name_val_pairs.push_back(std::make_pair(key, value)); - } - ; - value_list::const_iterator begin() - { - return name_val_pairs.cbegin(); - } - value_list::const_iterator end() - { - return name_val_pairs.cend(); - } - std::string toString(); - static annotation *read(util::membuffer &input, constant_pool &pool); + annotation(uint16_t type_index, constant_pool &pool) + : type_index(type_index), pool(pool) {}; + ~annotation() + { + for (unsigned i = 0; i < name_val_pairs.size(); i++) + { + delete name_val_pairs[i].second; + } + } + void add_pair(uint16_t key, element_value *value) + { + name_val_pairs.push_back(std::make_pair(key, value)); + } + ; + value_list::const_iterator begin() + { + return name_val_pairs.cbegin(); + } + value_list::const_iterator end() + { + return name_val_pairs.cend(); + } + std::string toString(); + static annotation *read(util::membuffer &input, constant_pool &pool); }; typedef std::vector<annotation *> annotation_table; @@ -121,158 +121,158 @@ typedef std::vector<annotation *> annotation_table; class element_value_simple : public element_value { protected: - /// index of the constant in the constant pool - uint16_t index; + /// index of the constant in the constant pool + uint16_t index; public: - element_value_simple(element_value_type type, uint16_t index, constant_pool &pool) - : element_value(type, pool), index(index) { - // TODO: verify consistency - }; - uint16_t getIndex() - { - return index; - } - virtual std::string toString() - { - return pool[index].toString(); - } - ; + element_value_simple(element_value_type type, uint16_t index, constant_pool &pool) + : element_value(type, pool), index(index) { + // TODO: verify consistency + }; + uint16_t getIndex() + { + return index; + } + virtual std::string toString() + { + return pool[index].toString(); + } + ; }; /// The enum_const_value item is used if the tag item is 'e'. class element_value_enum : public element_value { protected: - /** - * The value of the type_name_index item must be a valid index into the constant_pool table. - * The constant_pool entry at that index must be a CONSTANT_Utf8_info (§4.4.7) structure - * representing a valid field descriptor (§4.3.2) that denotes the internal form of the - * binary - * name (§4.2.1) of the type of the enum constant represented by this element_value - * structure. - */ - uint16_t typeIndex; - /** - * The value of the const_name_index item must be a valid index into the constant_pool - * table. - * The constant_pool entry at that index must be a CONSTANT_Utf8_info (§4.4.7) structure - * representing the simple name of the enum constant represented by this element_value - * structure. - */ - uint16_t valueIndex; + /** + * The value of the type_name_index item must be a valid index into the constant_pool table. + * The constant_pool entry at that index must be a CONSTANT_Utf8_info (§4.4.7) structure + * representing a valid field descriptor (§4.3.2) that denotes the internal form of the + * binary + * name (§4.2.1) of the type of the enum constant represented by this element_value + * structure. + */ + uint16_t typeIndex; + /** + * The value of the const_name_index item must be a valid index into the constant_pool + * table. + * The constant_pool entry at that index must be a CONSTANT_Utf8_info (§4.4.7) structure + * representing the simple name of the enum constant represented by this element_value + * structure. + */ + uint16_t valueIndex; public: - element_value_enum(element_value_type type, uint16_t typeIndex, uint16_t valueIndex, - constant_pool &pool) - : element_value(type, pool), typeIndex(typeIndex), valueIndex(valueIndex) - { - // TODO: verify consistency - } - uint16_t getValueIndex() - { - return valueIndex; - } - uint16_t getTypeIndex() - { - return typeIndex; - } - virtual std::string toString() - { - return "enum value"; - } - ; + element_value_enum(element_value_type type, uint16_t typeIndex, uint16_t valueIndex, + constant_pool &pool) + : element_value(type, pool), typeIndex(typeIndex), valueIndex(valueIndex) + { + // TODO: verify consistency + } + uint16_t getValueIndex() + { + return valueIndex; + } + uint16_t getTypeIndex() + { + return typeIndex; + } + virtual std::string toString() + { + return "enum value"; + } + ; }; class element_value_class : public element_value { protected: - /** - * The class_info_index item must be a valid index into the constant_pool table. - * The constant_pool entry at that index must be a CONSTANT_Utf8_info (§4.4.7) structure - * representing the return descriptor (§4.3.3) of the type that is reified by the class - * represented by this element_value structure. - * - * For example, 'V' for Void.class, 'Ljava/lang/Object;' for Object, etc. - * - * Or in plain english, you can store type information in annotations. Yay. - */ - uint16_t classIndex; + /** + * The class_info_index item must be a valid index into the constant_pool table. + * The constant_pool entry at that index must be a CONSTANT_Utf8_info (§4.4.7) structure + * representing the return descriptor (§4.3.3) of the type that is reified by the class + * represented by this element_value structure. + * + * For example, 'V' for Void.class, 'Ljava/lang/Object;' for Object, etc. + * + * Or in plain english, you can store type information in annotations. Yay. + */ + uint16_t classIndex; public: - element_value_class(element_value_type type, uint16_t classIndex, constant_pool &pool) - : element_value(type, pool), classIndex(classIndex) - { - // TODO: verify consistency - } - uint16_t getIndex() - { - return classIndex; - } - virtual std::string toString() - { - return "class"; - } - ; + element_value_class(element_value_type type, uint16_t classIndex, constant_pool &pool) + : element_value(type, pool), classIndex(classIndex) + { + // TODO: verify consistency + } + uint16_t getIndex() + { + return classIndex; + } + virtual std::string toString() + { + return "class"; + } + ; }; /// nested annotations... yay class element_value_annotation : public element_value { private: - annotation *nestedAnnotation; + annotation *nestedAnnotation; public: - element_value_annotation(element_value_type type, annotation *nestedAnnotation, - constant_pool &pool) - : element_value(type, pool), nestedAnnotation(nestedAnnotation) {}; - ~element_value_annotation() - { - if (nestedAnnotation) - { - delete nestedAnnotation; - nestedAnnotation = nullptr; - } - } - virtual std::string toString() - { - return "nested annotation"; - } - ; + element_value_annotation(element_value_type type, annotation *nestedAnnotation, + constant_pool &pool) + : element_value(type, pool), nestedAnnotation(nestedAnnotation) {}; + ~element_value_annotation() + { + if (nestedAnnotation) + { + delete nestedAnnotation; + nestedAnnotation = nullptr; + } + } + virtual std::string toString() + { + return "nested annotation"; + } + ; }; /// and arrays! class element_value_array : public element_value { public: - typedef std::vector<element_value *> elem_vec; + typedef std::vector<element_value *> elem_vec; protected: - elem_vec values; + elem_vec values; public: - element_value_array(element_value_type type, std::vector<element_value *> &values, - constant_pool &pool) - : element_value(type, pool), values(values) {}; - ~element_value_array() - { - for (unsigned i = 0; i < values.size(); i++) - { - delete values[i]; - } - } - ; - elem_vec::const_iterator begin() - { - return values.cbegin(); - } - elem_vec::const_iterator end() - { - return values.cend(); - } - virtual std::string toString() - { - return "array"; - } - ; + element_value_array(element_value_type type, std::vector<element_value *> &values, + constant_pool &pool) + : element_value(type, pool), values(values) {}; + ~element_value_array() + { + for (unsigned i = 0; i < values.size(); i++) + { + delete values[i]; + } + } + ; + elem_vec::const_iterator begin() + { + return values.cbegin(); + } + elem_vec::const_iterator end() + { + return values.cend(); + } + virtual std::string toString() + { + return "array"; + } + ; }; }
\ No newline at end of file diff --git a/libraries/classparser/src/classfile.h b/libraries/classparser/src/classfile.h index a5e7ee50..1616a828 100644 --- a/libraries/classparser/src/classfile.h +++ b/libraries/classparser/src/classfile.h @@ -11,146 +11,146 @@ namespace java class classfile : public util::membuffer { public: - classfile(char *data, std::size_t size) : membuffer(data, size) - { - valid = false; - is_synthetic = false; - read_be(magic); - if (magic != 0xCAFEBABE) - throw new classfile_exception(); - read_be(minor_version); - read_be(major_version); - constants.load(*this); - read_be(access_flags); - read_be(this_class); - read_be(super_class); + classfile(char *data, std::size_t size) : membuffer(data, size) + { + valid = false; + is_synthetic = false; + read_be(magic); + if (magic != 0xCAFEBABE) + throw new classfile_exception(); + read_be(minor_version); + read_be(major_version); + constants.load(*this); + read_be(access_flags); + read_be(this_class); + read_be(super_class); - // Interfaces - uint16_t iface_count = 0; - read_be(iface_count); - while (iface_count) - { - uint16_t iface; - read_be(iface); - interfaces.push_back(iface); - iface_count--; - } + // Interfaces + uint16_t iface_count = 0; + read_be(iface_count); + while (iface_count) + { + uint16_t iface; + read_be(iface); + interfaces.push_back(iface); + iface_count--; + } - // Fields - // read fields (and attributes from inside fields) (and possible inner classes. yay for - // recursion!) - // for now though, we will ignore all attributes - /* - * field_info - * { - * u2 access_flags; - * u2 name_index; - * u2 descriptor_index; - * u2 attributes_count; - * attribute_info attributes[attributes_count]; - * } - */ - uint16_t field_count = 0; - read_be(field_count); - while (field_count) - { - // skip field stuff - skip(6); - // and skip field attributes - uint16_t attr_count = 0; - read_be(attr_count); - while (attr_count) - { - skip(2); - uint32_t attr_length = 0; - read_be(attr_length); - skip(attr_length); - attr_count--; - } - field_count--; - } + // Fields + // read fields (and attributes from inside fields) (and possible inner classes. yay for + // recursion!) + // for now though, we will ignore all attributes + /* + * field_info + * { + * u2 access_flags; + * u2 name_index; + * u2 descriptor_index; + * u2 attributes_count; + * attribute_info attributes[attributes_count]; + * } + */ + uint16_t field_count = 0; + read_be(field_count); + while (field_count) + { + // skip field stuff + skip(6); + // and skip field attributes + uint16_t attr_count = 0; + read_be(attr_count); + while (attr_count) + { + skip(2); + uint32_t attr_length = 0; + read_be(attr_length); + skip(attr_length); + attr_count--; + } + field_count--; + } - // class methods - /* - * method_info - * { - * u2 access_flags; - * u2 name_index; - * u2 descriptor_index; - * u2 attributes_count; - * attribute_info attributes[attributes_count]; - * } - */ - uint16_t method_count = 0; - read_be(method_count); - while (method_count) - { - skip(6); - // and skip method attributes - uint16_t attr_count = 0; - read_be(attr_count); - while (attr_count) - { - skip(2); - uint32_t attr_length = 0; - read_be(attr_length); - skip(attr_length); - attr_count--; - } - method_count--; - } + // class methods + /* + * method_info + * { + * u2 access_flags; + * u2 name_index; + * u2 descriptor_index; + * u2 attributes_count; + * attribute_info attributes[attributes_count]; + * } + */ + uint16_t method_count = 0; + read_be(method_count); + while (method_count) + { + skip(6); + // and skip method attributes + uint16_t attr_count = 0; + read_be(attr_count); + while (attr_count) + { + skip(2); + uint32_t attr_length = 0; + read_be(attr_length); + skip(attr_length); + attr_count--; + } + method_count--; + } - // class attributes - // there are many kinds of attributes. this is just the generic wrapper structure. - // type is decided by attribute name. extensions to the standard are *possible* - // class annotations are one kind of a attribute (one per class) - /* - * attribute_info - * { - * u2 attribute_name_index; - * u4 attribute_length; - * u1 info[attribute_length]; - * } - */ - uint16_t class_attr_count = 0; - read_be(class_attr_count); - while (class_attr_count) - { - uint16_t name_idx = 0; - read_be(name_idx); - uint32_t attr_length = 0; - read_be(attr_length); + // class attributes + // there are many kinds of attributes. this is just the generic wrapper structure. + // type is decided by attribute name. extensions to the standard are *possible* + // class annotations are one kind of a attribute (one per class) + /* + * attribute_info + * { + * u2 attribute_name_index; + * u4 attribute_length; + * u1 info[attribute_length]; + * } + */ + uint16_t class_attr_count = 0; + read_be(class_attr_count); + while (class_attr_count) + { + uint16_t name_idx = 0; + read_be(name_idx); + uint32_t attr_length = 0; + read_be(attr_length); - auto name = constants[name_idx]; - if (name.str_data == "RuntimeVisibleAnnotations") - { - uint16_t num_annotations = 0; - read_be(num_annotations); - while (num_annotations) - { - visible_class_annotations.push_back(annotation::read(*this, constants)); - num_annotations--; - } - } - else - skip(attr_length); - class_attr_count--; - } - valid = true; - } - ; - bool valid; - bool is_synthetic; - uint32_t magic; - uint16_t minor_version; - uint16_t major_version; - constant_pool constants; - uint16_t access_flags; - uint16_t this_class; - uint16_t super_class; - // interfaces this class implements ? must be. investigate. - std::vector<uint16_t> interfaces; - // FIXME: doesn't free up memory on delete - java::annotation_table visible_class_annotations; + auto name = constants[name_idx]; + if (name.str_data == "RuntimeVisibleAnnotations") + { + uint16_t num_annotations = 0; + read_be(num_annotations); + while (num_annotations) + { + visible_class_annotations.push_back(annotation::read(*this, constants)); + num_annotations--; + } + } + else + skip(attr_length); + class_attr_count--; + } + valid = true; + } + ; + bool valid; + bool is_synthetic; + uint32_t magic; + uint16_t minor_version; + uint16_t major_version; + constant_pool constants; + uint16_t access_flags; + uint16_t this_class; + uint16_t super_class; + // interfaces this class implements ? must be. investigate. + std::vector<uint16_t> interfaces; + // FIXME: doesn't free up memory on delete + java::annotation_table visible_class_annotations; }; }
\ No newline at end of file diff --git a/libraries/classparser/src/classparser.cpp b/libraries/classparser/src/classparser.cpp index 8837781f..e0b6728d 100644 --- a/libraries/classparser/src/classparser.cpp +++ b/libraries/classparser/src/classparser.cpp @@ -1,4 +1,4 @@ -/* Copyright 2013-2018 MultiMC Contributors +/* Copyright 2013-2019 MultiMC Contributors * * Authors: Orochimarufan <orochimarufan.x3@gmail.com> * @@ -26,60 +26,58 @@ namespace classparser QString GetMinecraftJarVersion(QString jarName) { - QString version; + QString version; - // check if minecraft.jar exists - QFile jar(jarName); - if (!jar.exists()) - return version; + // check if minecraft.jar exists + QFile jar(jarName); + if (!jar.exists()) + return version; - // open minecraft.jar - QuaZip zip(&jar); - if (!zip.open(QuaZip::mdUnzip)) - return version; + // open minecraft.jar + QuaZip zip(&jar); + if (!zip.open(QuaZip::mdUnzip)) + return version; - // open Minecraft.class - zip.setCurrentFile("net/minecraft/client/Minecraft.class", QuaZip::csSensitive); - QuaZipFile Minecraft(&zip); - if (!Minecraft.open(QuaZipFile::ReadOnly)) - return version; + // open Minecraft.class + zip.setCurrentFile("net/minecraft/client/Minecraft.class", QuaZip::csSensitive); + QuaZipFile Minecraft(&zip); + if (!Minecraft.open(QuaZipFile::ReadOnly)) + return version; - // read Minecraft.class - qint64 size = Minecraft.size(); - char *classfile = new char[size]; - Minecraft.read(classfile, size); + // read Minecraft.class + qint64 size = Minecraft.size(); + char *classfile = new char[size]; + Minecraft.read(classfile, size); - // parse Minecraft.class - try - { - char *temp = classfile; - java::classfile MinecraftClass(temp, size); - java::constant_pool constants = MinecraftClass.constants; - for (java::constant_pool::container_type::const_iterator iter = constants.begin(); - iter != constants.end(); iter++) - { - const java::constant &constant = *iter; - if (constant.type != java::constant::j_string_data) - continue; - const std::string &str = constant.str_data; - qDebug() << QString::fromStdString(str); - if (str.compare(0, 20, "Minecraft Minecraft ") == 0) - { - version = str.substr(20).data(); - break; - } - } - } - catch (java::classfile_exception &) - { - } + // parse Minecraft.class + try + { + char *temp = classfile; + java::classfile MinecraftClass(temp, size); + java::constant_pool constants = MinecraftClass.constants; + for (java::constant_pool::container_type::const_iterator iter = constants.begin(); + iter != constants.end(); iter++) + { + const java::constant &constant = *iter; + if (constant.type != java::constant_type_t::j_string_data) + continue; + const std::string &str = constant.str_data; + qDebug() << QString::fromStdString(str); + if (str.compare(0, 20, "Minecraft Minecraft ") == 0) + { + version = str.substr(20).data(); + break; + } + } + } + catch (const java::classfile_exception &) { } - // clean up - delete[] classfile; - Minecraft.close(); - zip.close(); - jar.close(); + // clean up + delete[] classfile; + Minecraft.close(); + zip.close(); + jar.close(); - return version; + return version; } } diff --git a/libraries/classparser/src/constants.h b/libraries/classparser/src/constants.h index 9c74ab20..3b6c3b7a 100644 --- a/libraries/classparser/src/constants.h +++ b/libraries/classparser/src/constants.h @@ -4,152 +4,161 @@ namespace java { +enum class constant_type_t : uint8_t +{ + j_hole = 0, // HACK: this is a hole in the array, because java is crazy + j_string_data = 1, + j_int = 3, + j_float = 4, + j_long = 5, + j_double = 6, + j_class = 7, + j_string = 8, + j_fieldref = 9, + j_methodref = 10, + j_interface_methodref = 11, + j_nameandtype = 12 + // FIXME: missing some constant types, see https://docs.oracle.com/javase/specs/jvms/se7/html/jvms-4.html#jvms-4.4 +}; + +struct ref_type_t +{ + /** + * Class reference: + * an index within the constant pool to a UTF-8 string containing + * the fully qualified class name (in internal format) + * Used for j_class, j_fieldref, j_methodref and j_interface_methodref + */ + uint16_t class_idx; + // used for j_fieldref, j_methodref and j_interface_methodref + uint16_t name_and_type_idx; +}; + +struct name_and_type_t +{ + uint16_t name_index; + uint16_t descriptor_index; +}; + class constant { public: - enum type_t : uint8_t - { - j_hole = 0, // HACK: this is a hole in the array, because java is crazy - j_string_data = 1, - j_int = 3, - j_float = 4, - j_long = 5, - j_double = 6, - j_class = 7, - j_string = 8, - j_fieldref = 9, - j_methodref = 10, - j_interface_methodref = 11, - j_nameandtype = 12 - // FIXME: missing some constant types, see https://docs.oracle.com/javase/specs/jvms/se7/html/jvms-4.html#jvms-4.4 - } type; + constant_type_t type = constant_type_t::j_hole; - constant(util::membuffer &buf) - { - buf.read(type); + constant(util::membuffer &buf) + { + buf.read(type); - // load data depending on type - switch (type) - { - case j_float: - case j_int: - buf.read_be(int_data); // same as float data really - break; - case j_double: - case j_long: - buf.read_be(long_data); // same as double - break; - case j_class: - buf.read_be(ref_type.class_idx); - break; - case j_fieldref: - case j_methodref: - case j_interface_methodref: - buf.read_be(ref_type.class_idx); - buf.read_be(ref_type.name_and_type_idx); - break; - case j_string: - buf.read_be(index); - break; - case j_string_data: - // HACK HACK: for now, we call these UTF-8 and do no further processing. - // Later, we should do some decoding. It's really modified UTF-8 - // * U+0000 is represented as 0xC0,0x80 invalid character - // * any single zero byte ends the string - // * characters above U+10000 are encoded like in CESU-8 - buf.read_jstr(str_data); - break; - case j_nameandtype: - buf.read_be(name_and_type.name_index); - buf.read_be(name_and_type.descriptor_index); - break; - default: - // invalid constant type! - throw new classfile_exception(); - } - } + // load data depending on type + switch (type) + { + case constant_type_t::j_float: + buf.read_be(data.int_data); + break; + case constant_type_t::j_int: + buf.read_be(data.int_data); // same as float data really + break; + case constant_type_t::j_double: + buf.read_be(data.long_data); + break; + case constant_type_t::j_long: + buf.read_be(data.long_data); // same as double + break; + case constant_type_t::j_class: + buf.read_be(data.ref_type.class_idx); + break; + case constant_type_t::j_fieldref: + case constant_type_t::j_methodref: + case constant_type_t::j_interface_methodref: + buf.read_be(data.ref_type.class_idx); + buf.read_be(data.ref_type.name_and_type_idx); + break; + case constant_type_t::j_string: + buf.read_be(data.index); + break; + case constant_type_t::j_string_data: + // HACK HACK: for now, we call these UTF-8 and do no further processing. + // Later, we should do some decoding. It's really modified UTF-8 + // * U+0000 is represented as 0xC0,0x80 invalid character + // * any single zero byte ends the string + // * characters above U+10000 are encoded like in CESU-8 + buf.read_jstr(str_data); + break; + case constant_type_t::j_nameandtype: + buf.read_be(data.name_and_type.name_index); + buf.read_be(data.name_and_type.descriptor_index); + break; + default: + // invalid constant type! + throw new classfile_exception(); + } + } + constant(int) + { + } - constant(int) - { - type = j_hole; - } + std::string toString() + { + std::ostringstream ss; + switch (type) + { + case constant_type_t::j_hole: + ss << "Fake legacy entry"; + break; + case constant_type_t::j_float: + ss << "Float: " << data.float_data; + break; + case constant_type_t::j_double: + ss << "Double: " << data.double_data; + break; + case constant_type_t::j_int: + ss << "Int: " << data.int_data; + break; + case constant_type_t::j_long: + ss << "Long: " << data.long_data; + break; + case constant_type_t::j_string_data: + ss << "StrData: " << str_data; + break; + case constant_type_t::j_string: + ss << "Str: " << data.index; + break; + case constant_type_t::j_fieldref: + ss << "FieldRef: " << data.ref_type.class_idx << " " << data.ref_type.name_and_type_idx; + break; + case constant_type_t::j_methodref: + ss << "MethodRef: " << data.ref_type.class_idx << " " << data.ref_type.name_and_type_idx; + break; + case constant_type_t::j_interface_methodref: + ss << "IfMethodRef: " << data.ref_type.class_idx << " " << data.ref_type.name_and_type_idx; + break; + case constant_type_t::j_class: + ss << "Class: " << data.ref_type.class_idx; + break; + case constant_type_t::j_nameandtype: + ss << "NameAndType: " << data.name_and_type.name_index << " " + << data.name_and_type.descriptor_index; + break; + default: + ss << "Invalid entry (" << int(type) << ")"; + break; + } + return ss.str(); + } - std::string toString() - { - std::ostringstream ss; - switch (type) - { - case j_hole: - ss << "Fake legacy entry"; - break; - case j_float: - ss << "Float: " << float_data; - break; - case j_double: - ss << "Double: " << double_data; - break; - case j_int: - ss << "Int: " << int_data; - break; - case j_long: - ss << "Long: " << long_data; - break; - case j_string_data: - ss << "StrData: " << str_data; - break; - case j_string: - ss << "Str: " << index; - break; - case j_fieldref: - ss << "FieldRef: " << ref_type.class_idx << " " << ref_type.name_and_type_idx; - break; - case j_methodref: - ss << "MethodRef: " << ref_type.class_idx << " " << ref_type.name_and_type_idx; - break; - case j_interface_methodref: - ss << "IfMethodRef: " << ref_type.class_idx << " " << ref_type.name_and_type_idx; - break; - case j_class: - ss << "Class: " << ref_type.class_idx; - break; - case j_nameandtype: - ss << "NameAndType: " << name_and_type.name_index << " " - << name_and_type.descriptor_index; - break; - default: - ss << "Invalid entry (" << int(type) << ")"; - break; - } - return ss.str(); - } + std::string str_data; /** String data in 'modified utf-8'.*/ - std::string str_data; /** String data in 'modified utf-8'.*/ - // store everything here. - union - { - int32_t int_data; - int64_t long_data; - float float_data; - double double_data; - uint16_t index; - struct - { - /** - * Class reference: - * an index within the constant pool to a UTF-8 string containing - * the fully qualified class name (in internal format) - * Used for j_class, j_fieldref, j_methodref and j_interface_methodref - */ - uint16_t class_idx; - // used for j_fieldref, j_methodref and j_interface_methodref - uint16_t name_and_type_idx; - } ref_type; - struct - { - uint16_t name_index; - uint16_t descriptor_index; - } name_and_type; - }; + // store everything here. + union + { + int32_t int_data; + int64_t long_data; + float float_data; + double double_data; + uint16_t index; + ref_type_t ref_type; + name_and_type_t name_and_type; + } data = {0}; }; /** @@ -159,64 +168,64 @@ public: class constant_pool { public: - /** - * Create a pool of constants - */ - constant_pool() - { - } - /** - * Load a java constant pool - */ - void load(util::membuffer &buf) - { - // FIXME: @SANITY this should check for the end of buffer. - uint16_t length = 0; - buf.read_be(length); - length--; - const constant *last_constant = nullptr; - while (length) - { - const constant &cnst = constant(buf); - constants.push_back(cnst); - last_constant = &constants[constants.size() - 1]; - if (last_constant->type == constant::j_double || - last_constant->type == constant::j_long) - { - // push in a fake constant to preserve indexing - constants.push_back(constant(0)); - length -= 2; - } - else - { - length--; - } - } - } - typedef std::vector<java::constant> container_type; - /** - * Access constants based on jar file index numbers (index of the first element is 1) - */ - java::constant &operator[](std::size_t constant_index) - { - if (constant_index == 0 || constant_index > constants.size()) - { - throw new classfile_exception(); - } - return constants[constant_index - 1]; - } - ; - container_type::const_iterator begin() const - { - return constants.begin(); - } - ; - container_type::const_iterator end() const - { - return constants.end(); - } + /** + * Create a pool of constants + */ + constant_pool() + { + } + /** + * Load a java constant pool + */ + void load(util::membuffer &buf) + { + // FIXME: @SANITY this should check for the end of buffer. + uint16_t length = 0; + buf.read_be(length); + length--; + const constant *last_constant = nullptr; + while (length) + { + const constant &cnst = constant(buf); + constants.push_back(cnst); + last_constant = &constants[constants.size() - 1]; + if (last_constant->type == constant_type_t::j_double || + last_constant->type == constant_type_t::j_long) + { + // push in a fake constant to preserve indexing + constants.push_back(constant(0)); + length -= 2; + } + else + { + length--; + } + } + } + typedef std::vector<java::constant> container_type; + /** + * Access constants based on jar file index numbers (index of the first element is 1) + */ + java::constant &operator[](std::size_t constant_index) + { + if (constant_index == 0 || constant_index > constants.size()) + { + throw new classfile_exception(); + } + return constants[constant_index - 1]; + } + ; + container_type::const_iterator begin() const + { + return constants.begin(); + } + ; + container_type::const_iterator end() const + { + return constants.end(); + } private: - container_type constants; + container_type constants; }; } diff --git a/libraries/classparser/src/javaendian.h b/libraries/classparser/src/javaendian.h index d488b382..076bff5e 100644 --- a/libraries/classparser/src/javaendian.h +++ b/libraries/classparser/src/javaendian.h @@ -9,68 +9,68 @@ namespace util #ifdef MULTIMC_BIG_ENDIAN inline uint64_t bigswap(uint64_t x) { - return x; + return x; } ; inline uint32_t bigswap(uint32_t x) { - return x; + return x; } ; inline uint16_t bigswap(uint16_t x) { - return x; + return x; } ; inline int64_t bigswap(int64_t x) { - return x; + return x; } ; inline int32_t bigswap(int32_t x) { - return x; + return x; } ; inline int16_t bigswap(int16_t x) { - return x; + return x; } ; #else inline uint64_t bigswap(uint64_t x) { - return (x >> 56) | ((x << 40) & 0x00FF000000000000) | ((x << 24) & 0x0000FF0000000000) | - ((x << 8) & 0x000000FF00000000) | ((x >> 8) & 0x00000000FF000000) | - ((x >> 24) & 0x0000000000FF0000) | ((x >> 40) & 0x000000000000FF00) | (x << 56); + return (x >> 56) | ((x << 40) & 0x00FF000000000000) | ((x << 24) & 0x0000FF0000000000) | + ((x << 8) & 0x000000FF00000000) | ((x >> 8) & 0x00000000FF000000) | + ((x >> 24) & 0x0000000000FF0000) | ((x >> 40) & 0x000000000000FF00) | (x << 56); } -; + inline uint32_t bigswap(uint32_t x) { - return (x >> 24) | ((x << 8) & 0x00FF0000) | ((x >> 8) & 0x0000FF00) | (x << 24); + return (x >> 24) | ((x << 8) & 0x00FF0000) | ((x >> 8) & 0x0000FF00) | (x << 24); } -; + inline uint16_t bigswap(uint16_t x) { - return (x >> 8) | (x << 8); + return (x >> 8) | (x << 8); } -; + inline int64_t bigswap(int64_t x) { - return (x >> 56) | ((x << 40) & 0x00FF000000000000) | ((x << 24) & 0x0000FF0000000000) | - ((x << 8) & 0x000000FF00000000) | ((x >> 8) & 0x00000000FF000000) | - ((x >> 24) & 0x0000000000FF0000) | ((x >> 40) & 0x000000000000FF00) | (x << 56); + return (x >> 56) | ((x << 40) & 0x00FF000000000000) | ((x << 24) & 0x0000FF0000000000) | + ((x << 8) & 0x000000FF00000000) | ((x >> 8) & 0x00000000FF000000) | + ((x >> 24) & 0x0000000000FF0000) | ((x >> 40) & 0x000000000000FF00) | (x << 56); } -; + inline int32_t bigswap(int32_t x) { - return (x >> 24) | ((x << 8) & 0x00FF0000) | ((x >> 8) & 0x0000FF00) | (x << 24); + return (x >> 24) | ((x << 8) & 0x00FF0000) | ((x >> 8) & 0x0000FF00) | (x << 24); } -; + inline int16_t bigswap(int16_t x) { - return (x >> 8) | (x << 8); + return (x >> 8) | (x << 8); } -; + #endif } diff --git a/libraries/classparser/src/membuffer.h b/libraries/classparser/src/membuffer.h index ab83412a..f81c9705 100644 --- a/libraries/classparser/src/membuffer.h +++ b/libraries/classparser/src/membuffer.h @@ -10,54 +10,54 @@ namespace util class membuffer { public: - membuffer(char *buffer, std::size_t size) - { - current = start = buffer; - end = start + size; - } - ~membuffer() - { - // maybe? possibly? left out to avoid confusion. for now. - // delete start; - } - /** - * Read some value. That's all ;) - */ - template <class T> void read(T &val) - { - val = *(T *)current; - current += sizeof(T); - } - /** - * Read a big-endian number - * valid for 2-byte, 4-byte and 8-byte variables - */ - template <class T> void read_be(T &val) - { - val = util::bigswap(*(T *)current); - current += sizeof(T); - } - /** - * Read a string in the format: - * 2B length (big endian, unsigned) - * length bytes data - */ - void read_jstr(std::string &str) - { - uint16_t length = 0; - read_be(length); - str.append(current, length); - current += length; - } - /** - * Skip N bytes - */ - void skip(std::size_t N) - { - current += N; - } + membuffer(char *buffer, std::size_t size) + { + current = start = buffer; + end = start + size; + } + ~membuffer() + { + // maybe? possibly? left out to avoid confusion. for now. + // delete start; + } + /** + * Read some value. That's all ;) + */ + template <class T> void read(T &val) + { + val = *(T *)current; + current += sizeof(T); + } + /** + * Read a big-endian number + * valid for 2-byte, 4-byte and 8-byte variables + */ + template <class T> void read_be(T &val) + { + val = util::bigswap(*(T *)current); + current += sizeof(T); + } + /** + * Read a string in the format: + * 2B length (big endian, unsigned) + * length bytes data + */ + void read_jstr(std::string &str) + { + uint16_t length = 0; + read_be(length); + str.append(current, length); + current += length; + } + /** + * Skip N bytes + */ + void skip(std::size_t N) + { + current += N; + } private: - char *start, *end, *current; + char *start, *end, *current; }; } diff --git a/libraries/hoedown/include/hoedown/autolink.h b/libraries/hoedown/include/hoedown/autolink.h index 528885c9..953e7807 100644 --- a/libraries/hoedown/include/hoedown/autolink.h +++ b/libraries/hoedown/include/hoedown/autolink.h @@ -15,7 +15,7 @@ extern "C" { *************/ typedef enum hoedown_autolink_flags { - HOEDOWN_AUTOLINK_SHORT_DOMAINS = (1 << 0) + HOEDOWN_AUTOLINK_SHORT_DOMAINS = (1 << 0) } hoedown_autolink_flags; @@ -28,15 +28,15 @@ int hoedown_autolink_is_safe(const uint8_t *data, size_t size); /* hoedown_autolink__www: search for the next www link in data */ size_t hoedown_autolink__www(size_t *rewind_p, hoedown_buffer *link, - uint8_t *data, size_t offset, size_t size, hoedown_autolink_flags flags); + uint8_t *data, size_t offset, size_t size, hoedown_autolink_flags flags); /* hoedown_autolink__email: search for the next email in data */ size_t hoedown_autolink__email(size_t *rewind_p, hoedown_buffer *link, - uint8_t *data, size_t offset, size_t size, hoedown_autolink_flags flags); + uint8_t *data, size_t offset, size_t size, hoedown_autolink_flags flags); /* hoedown_autolink__url: search for the next URL in data */ size_t hoedown_autolink__url(size_t *rewind_p, hoedown_buffer *link, - uint8_t *data, size_t offset, size_t size, hoedown_autolink_flags flags); + uint8_t *data, size_t offset, size_t size, hoedown_autolink_flags flags); #ifdef __cplusplus diff --git a/libraries/hoedown/include/hoedown/buffer.h b/libraries/hoedown/include/hoedown/buffer.h index d7703f8d..062d86ce 100644 --- a/libraries/hoedown/include/hoedown/buffer.h +++ b/libraries/hoedown/include/hoedown/buffer.h @@ -28,14 +28,14 @@ typedef void *(*hoedown_realloc_callback)(void *, size_t); typedef void (*hoedown_free_callback)(void *); struct hoedown_buffer { - uint8_t *data; /* actual character data */ - size_t size; /* size of the string */ - size_t asize; /* allocated size (0 = volatile buffer) */ - size_t unit; /* reallocation unit size (0 = read-only buffer) */ - - hoedown_realloc_callback data_realloc; - hoedown_free_callback data_free; - hoedown_free_callback buffer_free; + uint8_t *data; /* actual character data */ + size_t size; /* size of the string */ + size_t asize; /* allocated size (0 = volatile buffer) */ + size_t unit; /* reallocation unit size (0 = read-only buffer) */ + + hoedown_realloc_callback data_realloc; + hoedown_free_callback data_free; + hoedown_free_callback buffer_free; }; typedef struct hoedown_buffer hoedown_buffer; @@ -52,11 +52,11 @@ void *hoedown_realloc(void *ptr, size_t size) __attribute__ ((malloc)); /* hoedown_buffer_init: initialize a buffer with custom allocators */ void hoedown_buffer_init( - hoedown_buffer *buffer, - size_t unit, - hoedown_realloc_callback data_realloc, - hoedown_free_callback data_free, - hoedown_free_callback buffer_free + hoedown_buffer *buffer, + size_t unit, + hoedown_realloc_callback data_realloc, + hoedown_free_callback data_free, + hoedown_free_callback buffer_free ); /* hoedown_buffer_uninit: uninitialize an existing buffer */ @@ -116,15 +116,15 @@ void hoedown_buffer_free(hoedown_buffer *buf); /* HOEDOWN_BUFPUTSL: optimized hoedown_buffer_puts of a string literal */ #define HOEDOWN_BUFPUTSL(output, literal) \ - hoedown_buffer_put(output, (const uint8_t *)literal, sizeof(literal) - 1) + hoedown_buffer_put(output, (const uint8_t *)literal, sizeof(literal) - 1) /* HOEDOWN_BUFSETSL: optimized hoedown_buffer_sets of a string literal */ #define HOEDOWN_BUFSETSL(output, literal) \ - hoedown_buffer_set(output, (const uint8_t *)literal, sizeof(literal) - 1) + hoedown_buffer_set(output, (const uint8_t *)literal, sizeof(literal) - 1) /* HOEDOWN_BUFEQSL: optimized hoedown_buffer_eqs of a string literal */ #define HOEDOWN_BUFEQSL(output, literal) \ - hoedown_buffer_eq(output, (const uint8_t *)literal, sizeof(literal) - 1) + hoedown_buffer_eq(output, (const uint8_t *)literal, sizeof(literal) - 1) #ifdef __cplusplus diff --git a/libraries/hoedown/include/hoedown/document.h b/libraries/hoedown/include/hoedown/document.h index a8178fec..210c565e 100644 --- a/libraries/hoedown/include/hoedown/document.h +++ b/libraries/hoedown/include/hoedown/document.h @@ -16,68 +16,68 @@ extern "C" { *************/ typedef enum hoedown_extensions { - /* block-level extensions */ - HOEDOWN_EXT_TABLES = (1 << 0), - HOEDOWN_EXT_FENCED_CODE = (1 << 1), - HOEDOWN_EXT_FOOTNOTES = (1 << 2), - - /* span-level extensions */ - HOEDOWN_EXT_AUTOLINK = (1 << 3), - HOEDOWN_EXT_STRIKETHROUGH = (1 << 4), - HOEDOWN_EXT_UNDERLINE = (1 << 5), - HOEDOWN_EXT_HIGHLIGHT = (1 << 6), - HOEDOWN_EXT_QUOTE = (1 << 7), - HOEDOWN_EXT_SUPERSCRIPT = (1 << 8), - HOEDOWN_EXT_MATH = (1 << 9), - - /* other flags */ - HOEDOWN_EXT_NO_INTRA_EMPHASIS = (1 << 11), - HOEDOWN_EXT_SPACE_HEADERS = (1 << 12), - HOEDOWN_EXT_MATH_EXPLICIT = (1 << 13), - - /* negative flags */ - HOEDOWN_EXT_DISABLE_INDENTED_CODE = (1 << 14) + /* block-level extensions */ + HOEDOWN_EXT_TABLES = (1 << 0), + HOEDOWN_EXT_FENCED_CODE = (1 << 1), + HOEDOWN_EXT_FOOTNOTES = (1 << 2), + + /* span-level extensions */ + HOEDOWN_EXT_AUTOLINK = (1 << 3), + HOEDOWN_EXT_STRIKETHROUGH = (1 << 4), + HOEDOWN_EXT_UNDERLINE = (1 << 5), + HOEDOWN_EXT_HIGHLIGHT = (1 << 6), + HOEDOWN_EXT_QUOTE = (1 << 7), + HOEDOWN_EXT_SUPERSCRIPT = (1 << 8), + HOEDOWN_EXT_MATH = (1 << 9), + + /* other flags */ + HOEDOWN_EXT_NO_INTRA_EMPHASIS = (1 << 11), + HOEDOWN_EXT_SPACE_HEADERS = (1 << 12), + HOEDOWN_EXT_MATH_EXPLICIT = (1 << 13), + + /* negative flags */ + HOEDOWN_EXT_DISABLE_INDENTED_CODE = (1 << 14) } hoedown_extensions; #define HOEDOWN_EXT_BLOCK (\ - HOEDOWN_EXT_TABLES |\ - HOEDOWN_EXT_FENCED_CODE |\ - HOEDOWN_EXT_FOOTNOTES ) + HOEDOWN_EXT_TABLES |\ + HOEDOWN_EXT_FENCED_CODE |\ + HOEDOWN_EXT_FOOTNOTES ) #define HOEDOWN_EXT_SPAN (\ - HOEDOWN_EXT_AUTOLINK |\ - HOEDOWN_EXT_STRIKETHROUGH |\ - HOEDOWN_EXT_UNDERLINE |\ - HOEDOWN_EXT_HIGHLIGHT |\ - HOEDOWN_EXT_QUOTE |\ - HOEDOWN_EXT_SUPERSCRIPT |\ - HOEDOWN_EXT_MATH ) + HOEDOWN_EXT_AUTOLINK |\ + HOEDOWN_EXT_STRIKETHROUGH |\ + HOEDOWN_EXT_UNDERLINE |\ + HOEDOWN_EXT_HIGHLIGHT |\ + HOEDOWN_EXT_QUOTE |\ + HOEDOWN_EXT_SUPERSCRIPT |\ + HOEDOWN_EXT_MATH ) #define HOEDOWN_EXT_FLAGS (\ - HOEDOWN_EXT_NO_INTRA_EMPHASIS |\ - HOEDOWN_EXT_SPACE_HEADERS |\ - HOEDOWN_EXT_MATH_EXPLICIT ) + HOEDOWN_EXT_NO_INTRA_EMPHASIS |\ + HOEDOWN_EXT_SPACE_HEADERS |\ + HOEDOWN_EXT_MATH_EXPLICIT ) #define HOEDOWN_EXT_NEGATIVE (\ - HOEDOWN_EXT_DISABLE_INDENTED_CODE ) + HOEDOWN_EXT_DISABLE_INDENTED_CODE ) typedef enum hoedown_list_flags { - HOEDOWN_LIST_ORDERED = (1 << 0), - HOEDOWN_LI_BLOCK = (1 << 1) /* <li> containing block data */ + HOEDOWN_LIST_ORDERED = (1 << 0), + HOEDOWN_LI_BLOCK = (1 << 1) /* <li> containing block data */ } hoedown_list_flags; typedef enum hoedown_table_flags { - HOEDOWN_TABLE_ALIGN_LEFT = 1, - HOEDOWN_TABLE_ALIGN_RIGHT = 2, - HOEDOWN_TABLE_ALIGN_CENTER = 3, - HOEDOWN_TABLE_ALIGNMASK = 3, - HOEDOWN_TABLE_HEADER = 4 + HOEDOWN_TABLE_ALIGN_LEFT = 1, + HOEDOWN_TABLE_ALIGN_RIGHT = 2, + HOEDOWN_TABLE_ALIGN_CENTER = 3, + HOEDOWN_TABLE_ALIGNMASK = 3, + HOEDOWN_TABLE_HEADER = 4 } hoedown_table_flags; typedef enum hoedown_autolink_type { - HOEDOWN_AUTOLINK_NONE, /* used internally when it is not an autolink*/ - HOEDOWN_AUTOLINK_NORMAL, /* normal http/http/ftp/mailto/etc link */ - HOEDOWN_AUTOLINK_EMAIL /* e-mail link without explit mailto: */ + HOEDOWN_AUTOLINK_NONE, /* used internally when it is not an autolink*/ + HOEDOWN_AUTOLINK_NORMAL, /* normal http/http/ftp/mailto/etc link */ + HOEDOWN_AUTOLINK_EMAIL /* e-mail link without explit mailto: */ } hoedown_autolink_type; @@ -89,57 +89,57 @@ struct hoedown_document; typedef struct hoedown_document hoedown_document; struct hoedown_renderer_data { - void *opaque; + void *opaque; }; typedef struct hoedown_renderer_data hoedown_renderer_data; /* hoedown_renderer - functions for rendering parsed data */ struct hoedown_renderer { - /* state object */ - void *opaque; - - /* block level callbacks - NULL skips the block */ - void (*blockcode)(hoedown_buffer *ob, const hoedown_buffer *text, const hoedown_buffer *lang, const hoedown_renderer_data *data); - void (*blockquote)(hoedown_buffer *ob, const hoedown_buffer *content, const hoedown_renderer_data *data); - void (*header)(hoedown_buffer *ob, const hoedown_buffer *content, int level, const hoedown_renderer_data *data); - void (*hrule)(hoedown_buffer *ob, const hoedown_renderer_data *data); - void (*list)(hoedown_buffer *ob, const hoedown_buffer *content, hoedown_list_flags flags, const hoedown_renderer_data *data); - void (*listitem)(hoedown_buffer *ob, const hoedown_buffer *content, hoedown_list_flags flags, const hoedown_renderer_data *data); - void (*paragraph)(hoedown_buffer *ob, const hoedown_buffer *content, const hoedown_renderer_data *data); - void (*table)(hoedown_buffer *ob, const hoedown_buffer *content, const hoedown_renderer_data *data); - void (*table_header)(hoedown_buffer *ob, const hoedown_buffer *content, const hoedown_renderer_data *data); - void (*table_body)(hoedown_buffer *ob, const hoedown_buffer *content, const hoedown_renderer_data *data); - void (*table_row)(hoedown_buffer *ob, const hoedown_buffer *content, const hoedown_renderer_data *data); - void (*table_cell)(hoedown_buffer *ob, const hoedown_buffer *content, hoedown_table_flags flags, const hoedown_renderer_data *data); - void (*footnotes)(hoedown_buffer *ob, const hoedown_buffer *content, const hoedown_renderer_data *data); - void (*footnote_def)(hoedown_buffer *ob, const hoedown_buffer *content, unsigned int num, const hoedown_renderer_data *data); - void (*blockhtml)(hoedown_buffer *ob, const hoedown_buffer *text, const hoedown_renderer_data *data); - - /* span level callbacks - NULL or return 0 prints the span verbatim */ - int (*autolink)(hoedown_buffer *ob, const hoedown_buffer *link, hoedown_autolink_type type, const hoedown_renderer_data *data); - int (*codespan)(hoedown_buffer *ob, const hoedown_buffer *text, const hoedown_renderer_data *data); - int (*double_emphasis)(hoedown_buffer *ob, const hoedown_buffer *content, const hoedown_renderer_data *data); - int (*emphasis)(hoedown_buffer *ob, const hoedown_buffer *content, const hoedown_renderer_data *data); - int (*underline)(hoedown_buffer *ob, const hoedown_buffer *content, const hoedown_renderer_data *data); - int (*highlight)(hoedown_buffer *ob, const hoedown_buffer *content, const hoedown_renderer_data *data); - int (*quote)(hoedown_buffer *ob, const hoedown_buffer *content, const hoedown_renderer_data *data); - int (*image)(hoedown_buffer *ob, const hoedown_buffer *link, const hoedown_buffer *title, const hoedown_buffer *alt, const hoedown_renderer_data *data); - int (*linebreak)(hoedown_buffer *ob, const hoedown_renderer_data *data); - int (*link)(hoedown_buffer *ob, const hoedown_buffer *content, const hoedown_buffer *link, const hoedown_buffer *title, const hoedown_renderer_data *data); - int (*triple_emphasis)(hoedown_buffer *ob, const hoedown_buffer *content, const hoedown_renderer_data *data); - int (*strikethrough)(hoedown_buffer *ob, const hoedown_buffer *content, const hoedown_renderer_data *data); - int (*superscript)(hoedown_buffer *ob, const hoedown_buffer *content, const hoedown_renderer_data *data); - int (*footnote_ref)(hoedown_buffer *ob, unsigned int num, const hoedown_renderer_data *data); - int (*math)(hoedown_buffer *ob, const hoedown_buffer *text, int displaymode, const hoedown_renderer_data *data); - int (*raw_html)(hoedown_buffer *ob, const hoedown_buffer *text, const hoedown_renderer_data *data); - - /* low level callbacks - NULL copies input directly into the output */ - void (*entity)(hoedown_buffer *ob, const hoedown_buffer *text, const hoedown_renderer_data *data); - void (*normal_text)(hoedown_buffer *ob, const hoedown_buffer *text, const hoedown_renderer_data *data); - - /* miscellaneous callbacks */ - void (*doc_header)(hoedown_buffer *ob, int inline_render, const hoedown_renderer_data *data); - void (*doc_footer)(hoedown_buffer *ob, int inline_render, const hoedown_renderer_data *data); + /* state object */ + void *opaque; + + /* block level callbacks - NULL skips the block */ + void (*blockcode)(hoedown_buffer *ob, const hoedown_buffer *text, const hoedown_buffer *lang, const hoedown_renderer_data *data); + void (*blockquote)(hoedown_buffer *ob, const hoedown_buffer *content, const hoedown_renderer_data *data); + void (*header)(hoedown_buffer *ob, const hoedown_buffer *content, int level, const hoedown_renderer_data *data); + void (*hrule)(hoedown_buffer *ob, const hoedown_renderer_data *data); + void (*list)(hoedown_buffer *ob, const hoedown_buffer *content, hoedown_list_flags flags, const hoedown_renderer_data *data); + void (*listitem)(hoedown_buffer *ob, const hoedown_buffer *content, hoedown_list_flags flags, const hoedown_renderer_data *data); + void (*paragraph)(hoedown_buffer *ob, const hoedown_buffer *content, const hoedown_renderer_data *data); + void (*table)(hoedown_buffer *ob, const hoedown_buffer *content, const hoedown_renderer_data *data); + void (*table_header)(hoedown_buffer *ob, const hoedown_buffer *content, const hoedown_renderer_data *data); + void (*table_body)(hoedown_buffer *ob, const hoedown_buffer *content, const hoedown_renderer_data *data); + void (*table_row)(hoedown_buffer *ob, const hoedown_buffer *content, const hoedown_renderer_data *data); + void (*table_cell)(hoedown_buffer *ob, const hoedown_buffer *content, hoedown_table_flags flags, const hoedown_renderer_data *data); + void (*footnotes)(hoedown_buffer *ob, const hoedown_buffer *content, const hoedown_renderer_data *data); + void (*footnote_def)(hoedown_buffer *ob, const hoedown_buffer *content, unsigned int num, const hoedown_renderer_data *data); + void (*blockhtml)(hoedown_buffer *ob, const hoedown_buffer *text, const hoedown_renderer_data *data); + + /* span level callbacks - NULL or return 0 prints the span verbatim */ + int (*autolink)(hoedown_buffer *ob, const hoedown_buffer *link, hoedown_autolink_type type, const hoedown_renderer_data *data); + int (*codespan)(hoedown_buffer *ob, const hoedown_buffer *text, const hoedown_renderer_data *data); + int (*double_emphasis)(hoedown_buffer *ob, const hoedown_buffer *content, const hoedown_renderer_data *data); + int (*emphasis)(hoedown_buffer *ob, const hoedown_buffer *content, const hoedown_renderer_data *data); + int (*underline)(hoedown_buffer *ob, const hoedown_buffer *content, const hoedown_renderer_data *data); + int (*highlight)(hoedown_buffer *ob, const hoedown_buffer *content, const hoedown_renderer_data *data); + int (*quote)(hoedown_buffer *ob, const hoedown_buffer *content, const hoedown_renderer_data *data); + int (*image)(hoedown_buffer *ob, const hoedown_buffer *link, const hoedown_buffer *title, const hoedown_buffer *alt, const hoedown_renderer_data *data); + int (*linebreak)(hoedown_buffer *ob, const hoedown_renderer_data *data); + int (*link)(hoedown_buffer *ob, const hoedown_buffer *content, const hoedown_buffer *link, const hoedown_buffer *title, const hoedown_renderer_data *data); + int (*triple_emphasis)(hoedown_buffer *ob, const hoedown_buffer *content, const hoedown_renderer_data *data); + int (*strikethrough)(hoedown_buffer *ob, const hoedown_buffer *content, const hoedown_renderer_data *data); + int (*superscript)(hoedown_buffer *ob, const hoedown_buffer *content, const hoedown_renderer_data *data); + int (*footnote_ref)(hoedown_buffer *ob, unsigned int num, const hoedown_renderer_data *data); + int (*math)(hoedown_buffer *ob, const hoedown_buffer *text, int displaymode, const hoedown_renderer_data *data); + int (*raw_html)(hoedown_buffer *ob, const hoedown_buffer *text, const hoedown_renderer_data *data); + + /* low level callbacks - NULL copies input directly into the output */ + void (*entity)(hoedown_buffer *ob, const hoedown_buffer *text, const hoedown_renderer_data *data); + void (*normal_text)(hoedown_buffer *ob, const hoedown_buffer *text, const hoedown_renderer_data *data); + + /* miscellaneous callbacks */ + void (*doc_header)(hoedown_buffer *ob, int inline_render, const hoedown_renderer_data *data); + void (*doc_footer)(hoedown_buffer *ob, int inline_render, const hoedown_renderer_data *data); }; typedef struct hoedown_renderer hoedown_renderer; @@ -150,9 +150,9 @@ typedef struct hoedown_renderer hoedown_renderer; /* hoedown_document_new: allocate a new document processor instance */ hoedown_document *hoedown_document_new( - const hoedown_renderer *renderer, - hoedown_extensions extensions, - size_t max_nesting + const hoedown_renderer *renderer, + hoedown_extensions extensions, + size_t max_nesting ) __attribute__ ((malloc)); /* hoedown_document_render: render regular Markdown using the document processor */ diff --git a/libraries/hoedown/include/hoedown/html.h b/libraries/hoedown/include/hoedown/html.h index e46e7fd6..7c68809a 100644 --- a/libraries/hoedown/include/hoedown/html.h +++ b/libraries/hoedown/include/hoedown/html.h @@ -16,16 +16,16 @@ extern "C" { *************/ typedef enum hoedown_html_flags { - HOEDOWN_HTML_SKIP_HTML = (1 << 0), - HOEDOWN_HTML_ESCAPE = (1 << 1), - HOEDOWN_HTML_HARD_WRAP = (1 << 2), - HOEDOWN_HTML_USE_XHTML = (1 << 3) + HOEDOWN_HTML_SKIP_HTML = (1 << 0), + HOEDOWN_HTML_ESCAPE = (1 << 1), + HOEDOWN_HTML_HARD_WRAP = (1 << 2), + HOEDOWN_HTML_USE_XHTML = (1 << 3) } hoedown_html_flags; typedef enum hoedown_html_tag { - HOEDOWN_HTML_TAG_NONE = 0, - HOEDOWN_HTML_TAG_OPEN, - HOEDOWN_HTML_TAG_CLOSE + HOEDOWN_HTML_TAG_NONE = 0, + HOEDOWN_HTML_TAG_OPEN, + HOEDOWN_HTML_TAG_CLOSE } hoedown_html_tag; @@ -34,19 +34,19 @@ typedef enum hoedown_html_tag { *********/ struct hoedown_html_renderer_state { - void *opaque; + void *opaque; - struct { - int header_count; - int current_level; - int level_offset; - int nesting_level; - } toc_data; + struct { + int header_count; + int current_level; + int level_offset; + int nesting_level; + } toc_data; - hoedown_html_flags flags; + hoedown_html_flags flags; - /* extra callbacks */ - void (*link_attributes)(hoedown_buffer *ob, const hoedown_buffer *url, const hoedown_renderer_data *data); + /* extra callbacks */ + void (*link_attributes)(hoedown_buffer *ob, const hoedown_buffer *url, const hoedown_renderer_data *data); }; typedef struct hoedown_html_renderer_state hoedown_html_renderer_state; @@ -64,13 +64,13 @@ hoedown_html_tag hoedown_html_is_tag(const uint8_t *data, size_t size, const cha /* hoedown_html_renderer_new: allocates a regular HTML renderer */ hoedown_renderer *hoedown_html_renderer_new( - hoedown_html_flags render_flags, - int nesting_level + hoedown_html_flags render_flags, + int nesting_level ) __attribute__ ((malloc)); /* hoedown_html_toc_renderer_new: like hoedown_html_renderer_new, but the returned renderer produces the Table of Contents */ hoedown_renderer *hoedown_html_toc_renderer_new( - int nesting_level + int nesting_level ) __attribute__ ((malloc)); /* hoedown_html_renderer_free: deallocate an HTML renderer */ diff --git a/libraries/hoedown/include/hoedown/stack.h b/libraries/hoedown/include/hoedown/stack.h index bf9b439b..d1855f4f 100644 --- a/libraries/hoedown/include/hoedown/stack.h +++ b/libraries/hoedown/include/hoedown/stack.h @@ -15,9 +15,9 @@ extern "C" { *********/ struct hoedown_stack { - void **item; - size_t size; - size_t asize; + void **item; + size_t size; + size_t asize; }; typedef struct hoedown_stack hoedown_stack; diff --git a/libraries/hoedown/src/autolink.c b/libraries/hoedown/src/autolink.c index 9bc7fad5..3063b1a0 100644 --- a/libraries/hoedown/src/autolink.c +++ b/libraries/hoedown/src/autolink.c @@ -8,274 +8,274 @@ #ifndef _MSC_VER #include <strings.h> #else -#define strncasecmp _strnicmp +#define strncasecmp _strnicmp #endif int hoedown_autolink_is_safe(const uint8_t *data, size_t size) { - static const size_t valid_uris_count = 6; - static const char *valid_uris[] = { - "http://", "https://", "/", "#", "ftp://", "mailto:" - }; - static const size_t valid_uris_size[] = { 7, 8, 1, 1, 6, 7 }; - size_t i; - - for (i = 0; i < valid_uris_count; ++i) { - size_t len = valid_uris_size[i]; - - if (size > len && - strncasecmp((char *)data, valid_uris[i], len) == 0 && - isalnum(data[len])) - return 1; - } - - return 0; + static const size_t valid_uris_count = 6; + static const char *valid_uris[] = { + "http://", "https://", "/", "#", "ftp://", "mailto:" + }; + static const size_t valid_uris_size[] = { 7, 8, 1, 1, 6, 7 }; + size_t i; + + for (i = 0; i < valid_uris_count; ++i) { + size_t len = valid_uris_size[i]; + + if (size > len && + strncasecmp((char *)data, valid_uris[i], len) == 0 && + isalnum(data[len])) + return 1; + } + + return 0; } static size_t autolink_delim(uint8_t *data, size_t link_end, size_t max_rewind, size_t size) { - uint8_t cclose, copen = 0; - size_t i; - - for (i = 0; i < link_end; ++i) - if (data[i] == '<') { - link_end = i; - break; - } - - while (link_end > 0) { - if (strchr("?!.,:", data[link_end - 1]) != NULL) - link_end--; - - else if (data[link_end - 1] == ';') { - size_t new_end = link_end - 2; - - while (new_end > 0 && isalpha(data[new_end])) - new_end--; - - if (new_end < link_end - 2 && data[new_end] == '&') - link_end = new_end; - else - link_end--; - } - else break; - } - - if (link_end == 0) - return 0; - - cclose = data[link_end - 1]; - - switch (cclose) { - case '"': copen = '"'; break; - case '\'': copen = '\''; break; - case ')': copen = '('; break; - case ']': copen = '['; break; - case '}': copen = '{'; break; - } - - if (copen != 0) { - size_t closing = 0; - size_t opening = 0; - size_t i = 0; - - /* Try to close the final punctuation sign in this same line; - * if we managed to close it outside of the URL, that means that it's - * not part of the URL. If it closes inside the URL, that means it - * is part of the URL. - * - * Examples: - * - * foo http://www.pokemon.com/Pikachu_(Electric) bar - * => http://www.pokemon.com/Pikachu_(Electric) - * - * foo (http://www.pokemon.com/Pikachu_(Electric)) bar - * => http://www.pokemon.com/Pikachu_(Electric) - * - * foo http://www.pokemon.com/Pikachu_(Electric)) bar - * => http://www.pokemon.com/Pikachu_(Electric)) - * - * (foo http://www.pokemon.com/Pikachu_(Electric)) bar - * => foo http://www.pokemon.com/Pikachu_(Electric) - */ - - while (i < link_end) { - if (data[i] == copen) - opening++; - else if (data[i] == cclose) - closing++; - - i++; - } - - if (closing != opening) - link_end--; - } - - return link_end; + uint8_t cclose, copen = 0; + size_t i; + + for (i = 0; i < link_end; ++i) + if (data[i] == '<') { + link_end = i; + break; + } + + while (link_end > 0) { + if (strchr("?!.,:", data[link_end - 1]) != NULL) + link_end--; + + else if (data[link_end - 1] == ';') { + size_t new_end = link_end - 2; + + while (new_end > 0 && isalpha(data[new_end])) + new_end--; + + if (new_end < link_end - 2 && data[new_end] == '&') + link_end = new_end; + else + link_end--; + } + else break; + } + + if (link_end == 0) + return 0; + + cclose = data[link_end - 1]; + + switch (cclose) { + case '"': copen = '"'; break; + case '\'': copen = '\''; break; + case ')': copen = '('; break; + case ']': copen = '['; break; + case '}': copen = '{'; break; + } + + if (copen != 0) { + size_t closing = 0; + size_t opening = 0; + size_t i = 0; + + /* Try to close the final punctuation sign in this same line; + * if we managed to close it outside of the URL, that means that it's + * not part of the URL. If it closes inside the URL, that means it + * is part of the URL. + * + * Examples: + * + * foo http://www.pokemon.com/Pikachu_(Electric) bar + * => http://www.pokemon.com/Pikachu_(Electric) + * + * foo (http://www.pokemon.com/Pikachu_(Electric)) bar + * => http://www.pokemon.com/Pikachu_(Electric) + * + * foo http://www.pokemon.com/Pikachu_(Electric)) bar + * => http://www.pokemon.com/Pikachu_(Electric)) + * + * (foo http://www.pokemon.com/Pikachu_(Electric)) bar + * => foo http://www.pokemon.com/Pikachu_(Electric) + */ + + while (i < link_end) { + if (data[i] == copen) + opening++; + else if (data[i] == cclose) + closing++; + + i++; + } + + if (closing != opening) + link_end--; + } + + return link_end; } static size_t check_domain(uint8_t *data, size_t size, int allow_short) { - size_t i, np = 0; - - if (!isalnum(data[0])) - return 0; - - for (i = 1; i < size - 1; ++i) { - if (strchr(".:", data[i]) != NULL) np++; - else if (!isalnum(data[i]) && data[i] != '-') break; - } - - if (allow_short) { - /* We don't need a valid domain in the strict sense (with - * least one dot; so just make sure it's composed of valid - * domain characters and return the length of the the valid - * sequence. */ - return i; - } else { - /* a valid domain needs to have at least a dot. - * that's as far as we get */ - return np ? i : 0; - } + size_t i, np = 0; + + if (!isalnum(data[0])) + return 0; + + for (i = 1; i < size - 1; ++i) { + if (strchr(".:", data[i]) != NULL) np++; + else if (!isalnum(data[i]) && data[i] != '-') break; + } + + if (allow_short) { + /* We don't need a valid domain in the strict sense (with + * least one dot; so just make sure it's composed of valid + * domain characters and return the length of the the valid + * sequence. */ + return i; + } else { + /* a valid domain needs to have at least a dot. + * that's as far as we get */ + return np ? i : 0; + } } size_t hoedown_autolink__www( - size_t *rewind_p, - hoedown_buffer *link, - uint8_t *data, - size_t max_rewind, - size_t size, - unsigned int flags) + size_t *rewind_p, + hoedown_buffer *link, + uint8_t *data, + size_t max_rewind, + size_t size, + unsigned int flags) { - size_t link_end; + size_t link_end; - if (max_rewind > 0 && !ispunct(data[-1]) && !isspace(data[-1])) - return 0; + if (max_rewind > 0 && !ispunct(data[-1]) && !isspace(data[-1])) + return 0; - if (size < 4 || memcmp(data, "www.", strlen("www.")) != 0) - return 0; + if (size < 4 || memcmp(data, "www.", strlen("www.")) != 0) + return 0; - link_end = check_domain(data, size, 0); + link_end = check_domain(data, size, 0); - if (link_end == 0) - return 0; + if (link_end == 0) + return 0; - while (link_end < size && !isspace(data[link_end])) - link_end++; + while (link_end < size && !isspace(data[link_end])) + link_end++; - link_end = autolink_delim(data, link_end, max_rewind, size); + link_end = autolink_delim(data, link_end, max_rewind, size); - if (link_end == 0) - return 0; + if (link_end == 0) + return 0; - hoedown_buffer_put(link, data, link_end); - *rewind_p = 0; + hoedown_buffer_put(link, data, link_end); + *rewind_p = 0; - return (int)link_end; + return (int)link_end; } size_t hoedown_autolink__email( - size_t *rewind_p, - hoedown_buffer *link, - uint8_t *data, - size_t max_rewind, - size_t size, - unsigned int flags) + size_t *rewind_p, + hoedown_buffer *link, + uint8_t *data, + size_t max_rewind, + size_t size, + unsigned int flags) { - size_t link_end, rewind; - int nb = 0, np = 0; + size_t link_end, rewind; + int nb = 0, np = 0; - for (rewind = 0; rewind < max_rewind; ++rewind) { - uint8_t c = data[-1 - rewind]; + for (rewind = 0; rewind < max_rewind; ++rewind) { + uint8_t c = data[-1 - rewind]; - if (isalnum(c)) - continue; + if (isalnum(c)) + continue; - if (strchr(".+-_", c) != NULL) - continue; + if (strchr(".+-_", c) != NULL) + continue; - break; - } + break; + } - if (rewind == 0) - return 0; + if (rewind == 0) + return 0; - for (link_end = 0; link_end < size; ++link_end) { - uint8_t c = data[link_end]; + for (link_end = 0; link_end < size; ++link_end) { + uint8_t c = data[link_end]; - if (isalnum(c)) - continue; + if (isalnum(c)) + continue; - if (c == '@') - nb++; - else if (c == '.' && link_end < size - 1) - np++; - else if (c != '-' && c != '_') - break; - } + if (c == '@') + nb++; + else if (c == '.' && link_end < size - 1) + np++; + else if (c != '-' && c != '_') + break; + } - if (link_end < 2 || nb != 1 || np == 0 || - !isalpha(data[link_end - 1])) - return 0; + if (link_end < 2 || nb != 1 || np == 0 || + !isalpha(data[link_end - 1])) + return 0; - link_end = autolink_delim(data, link_end, max_rewind, size); + link_end = autolink_delim(data, link_end, max_rewind, size); - if (link_end == 0) - return 0; + if (link_end == 0) + return 0; - hoedown_buffer_put(link, data - rewind, link_end + rewind); - *rewind_p = rewind; + hoedown_buffer_put(link, data - rewind, link_end + rewind); + *rewind_p = rewind; - return link_end; + return link_end; } size_t hoedown_autolink__url( - size_t *rewind_p, - hoedown_buffer *link, - uint8_t *data, - size_t max_rewind, - size_t size, - unsigned int flags) + size_t *rewind_p, + hoedown_buffer *link, + uint8_t *data, + size_t max_rewind, + size_t size, + unsigned int flags) { - size_t link_end, rewind = 0, domain_len; + size_t link_end, rewind = 0, domain_len; - if (size < 4 || data[1] != '/' || data[2] != '/') - return 0; + if (size < 4 || data[1] != '/' || data[2] != '/') + return 0; - while (rewind < max_rewind && isalpha(data[-1 - rewind])) - rewind++; + while (rewind < max_rewind && isalpha(data[-1 - rewind])) + rewind++; - if (!hoedown_autolink_is_safe(data - rewind, size + rewind)) - return 0; + if (!hoedown_autolink_is_safe(data - rewind, size + rewind)) + return 0; - link_end = strlen("://"); + link_end = strlen("://"); - domain_len = check_domain( - data + link_end, - size - link_end, - flags & HOEDOWN_AUTOLINK_SHORT_DOMAINS); + domain_len = check_domain( + data + link_end, + size - link_end, + flags & HOEDOWN_AUTOLINK_SHORT_DOMAINS); - if (domain_len == 0) - return 0; + if (domain_len == 0) + return 0; - link_end += domain_len; - while (link_end < size && !isspace(data[link_end])) - link_end++; + link_end += domain_len; + while (link_end < size && !isspace(data[link_end])) + link_end++; - link_end = autolink_delim(data, link_end, max_rewind, size); + link_end = autolink_delim(data, link_end, max_rewind, size); - if (link_end == 0) - return 0; + if (link_end == 0) + return 0; - hoedown_buffer_put(link, data - rewind, link_end + rewind); - *rewind_p = rewind; + hoedown_buffer_put(link, data - rewind, link_end + rewind); + *rewind_p = rewind; - return link_end; + return link_end; } diff --git a/libraries/hoedown/src/buffer.c b/libraries/hoedown/src/buffer.c index 1c7ba55a..024a8bcc 100644 --- a/libraries/hoedown/src/buffer.c +++ b/libraries/hoedown/src/buffer.c @@ -8,301 +8,301 @@ void * hoedown_malloc(size_t size) { - void *ret = malloc(size); + void *ret = malloc(size); - if (!ret) { - fprintf(stderr, "Allocation failed.\n"); - abort(); - } + if (!ret) { + fprintf(stderr, "Allocation failed.\n"); + abort(); + } - return ret; + return ret; } void * hoedown_calloc(size_t nmemb, size_t size) { - void *ret = calloc(nmemb, size); + void *ret = calloc(nmemb, size); - if (!ret) { - fprintf(stderr, "Allocation failed.\n"); - abort(); - } + if (!ret) { + fprintf(stderr, "Allocation failed.\n"); + abort(); + } - return ret; + return ret; } void * hoedown_realloc(void *ptr, size_t size) { - void *ret = realloc(ptr, size); + void *ret = realloc(ptr, size); - if (!ret) { - fprintf(stderr, "Allocation failed.\n"); - abort(); - } + if (!ret) { + fprintf(stderr, "Allocation failed.\n"); + abort(); + } - return ret; + return ret; } void hoedown_buffer_init( - hoedown_buffer *buf, - size_t unit, - hoedown_realloc_callback data_realloc, - hoedown_free_callback data_free, - hoedown_free_callback buffer_free) + hoedown_buffer *buf, + size_t unit, + hoedown_realloc_callback data_realloc, + hoedown_free_callback data_free, + hoedown_free_callback buffer_free) { - assert(buf); - - buf->data = NULL; - buf->size = buf->asize = 0; - buf->unit = unit; - buf->data_realloc = data_realloc; - buf->data_free = data_free; - buf->buffer_free = buffer_free; + assert(buf); + + buf->data = NULL; + buf->size = buf->asize = 0; + buf->unit = unit; + buf->data_realloc = data_realloc; + buf->data_free = data_free; + buf->buffer_free = buffer_free; } void hoedown_buffer_uninit(hoedown_buffer *buf) { - assert(buf && buf->unit); - buf->data_free(buf->data); + assert(buf && buf->unit); + buf->data_free(buf->data); } hoedown_buffer * hoedown_buffer_new(size_t unit) { - hoedown_buffer *ret = hoedown_malloc(sizeof (hoedown_buffer)); - hoedown_buffer_init(ret, unit, hoedown_realloc, free, free); - return ret; + hoedown_buffer *ret = hoedown_malloc(sizeof (hoedown_buffer)); + hoedown_buffer_init(ret, unit, hoedown_realloc, free, free); + return ret; } void hoedown_buffer_free(hoedown_buffer *buf) { - if (!buf) return; - assert(buf && buf->unit); + if (!buf) return; + assert(buf && buf->unit); - buf->data_free(buf->data); + buf->data_free(buf->data); - if (buf->buffer_free) - buf->buffer_free(buf); + if (buf->buffer_free) + buf->buffer_free(buf); } void hoedown_buffer_reset(hoedown_buffer *buf) { - assert(buf && buf->unit); + assert(buf && buf->unit); - buf->data_free(buf->data); - buf->data = NULL; - buf->size = buf->asize = 0; + buf->data_free(buf->data); + buf->data = NULL; + buf->size = buf->asize = 0; } void hoedown_buffer_grow(hoedown_buffer *buf, size_t neosz) { - size_t neoasz; - assert(buf && buf->unit); + size_t neoasz; + assert(buf && buf->unit); - if (buf->asize >= neosz) - return; + if (buf->asize >= neosz) + return; - neoasz = buf->asize + buf->unit; - while (neoasz < neosz) - neoasz += buf->unit; + neoasz = buf->asize + buf->unit; + while (neoasz < neosz) + neoasz += buf->unit; - buf->data = (uint8_t *) buf->data_realloc(buf->data, neoasz); - buf->asize = neoasz; + buf->data = (uint8_t *) buf->data_realloc(buf->data, neoasz); + buf->asize = neoasz; } void hoedown_buffer_put(hoedown_buffer *buf, const uint8_t *data, size_t size) { - assert(buf && buf->unit); + assert(buf && buf->unit); - if (buf->size + size > buf->asize) - hoedown_buffer_grow(buf, buf->size + size); + if (buf->size + size > buf->asize) + hoedown_buffer_grow(buf, buf->size + size); - memcpy(buf->data + buf->size, data, size); - buf->size += size; + memcpy(buf->data + buf->size, data, size); + buf->size += size; } void hoedown_buffer_puts(hoedown_buffer *buf, const char *str) { - hoedown_buffer_put(buf, (const uint8_t *)str, strlen(str)); + hoedown_buffer_put(buf, (const uint8_t *)str, strlen(str)); } void hoedown_buffer_putc(hoedown_buffer *buf, uint8_t c) { - assert(buf && buf->unit); + assert(buf && buf->unit); - if (buf->size >= buf->asize) - hoedown_buffer_grow(buf, buf->size + 1); + if (buf->size >= buf->asize) + hoedown_buffer_grow(buf, buf->size + 1); - buf->data[buf->size] = c; - buf->size += 1; + buf->data[buf->size] = c; + buf->size += 1; } int hoedown_buffer_putf(hoedown_buffer *buf, FILE *file) { - assert(buf && buf->unit); + assert(buf && buf->unit); - while (!(feof(file) || ferror(file))) { - hoedown_buffer_grow(buf, buf->size + buf->unit); - buf->size += fread(buf->data + buf->size, 1, buf->unit, file); - } + while (!(feof(file) || ferror(file))) { + hoedown_buffer_grow(buf, buf->size + buf->unit); + buf->size += fread(buf->data + buf->size, 1, buf->unit, file); + } - return ferror(file); + return ferror(file); } void hoedown_buffer_set(hoedown_buffer *buf, const uint8_t *data, size_t size) { - assert(buf && buf->unit); + assert(buf && buf->unit); - if (size > buf->asize) - hoedown_buffer_grow(buf, size); + if (size > buf->asize) + hoedown_buffer_grow(buf, size); - memcpy(buf->data, data, size); - buf->size = size; + memcpy(buf->data, data, size); + buf->size = size; } void hoedown_buffer_sets(hoedown_buffer *buf, const char *str) { - hoedown_buffer_set(buf, (const uint8_t *)str, strlen(str)); + hoedown_buffer_set(buf, (const uint8_t *)str, strlen(str)); } int hoedown_buffer_eq(const hoedown_buffer *buf, const uint8_t *data, size_t size) { - if (buf->size != size) return 0; - return memcmp(buf->data, data, size) == 0; + if (buf->size != size) return 0; + return memcmp(buf->data, data, size) == 0; } int hoedown_buffer_eqs(const hoedown_buffer *buf, const char *str) { - return hoedown_buffer_eq(buf, (const uint8_t *)str, strlen(str)); + return hoedown_buffer_eq(buf, (const uint8_t *)str, strlen(str)); } int hoedown_buffer_prefix(const hoedown_buffer *buf, const char *prefix) { - size_t i; + size_t i; - for (i = 0; i < buf->size; ++i) { - if (prefix[i] == 0) - return 0; + for (i = 0; i < buf->size; ++i) { + if (prefix[i] == 0) + return 0; - if (buf->data[i] != prefix[i]) - return buf->data[i] - prefix[i]; - } + if (buf->data[i] != prefix[i]) + return buf->data[i] - prefix[i]; + } - return 0; + return 0; } void hoedown_buffer_slurp(hoedown_buffer *buf, size_t size) { - assert(buf && buf->unit); + assert(buf && buf->unit); - if (size >= buf->size) { - buf->size = 0; - return; - } + if (size >= buf->size) { + buf->size = 0; + return; + } - buf->size -= size; - memmove(buf->data, buf->data + size, buf->size); + buf->size -= size; + memmove(buf->data, buf->data + size, buf->size); } const char * hoedown_buffer_cstr(hoedown_buffer *buf) { - assert(buf && buf->unit); + assert(buf && buf->unit); - if (buf->size < buf->asize && buf->data[buf->size] == 0) - return (char *)buf->data; + if (buf->size < buf->asize && buf->data[buf->size] == 0) + return (char *)buf->data; - hoedown_buffer_grow(buf, buf->size + 1); - buf->data[buf->size] = 0; + hoedown_buffer_grow(buf, buf->size + 1); + buf->data[buf->size] = 0; - return (char *)buf->data; + return (char *)buf->data; } void hoedown_buffer_printf(hoedown_buffer *buf, const char *fmt, ...) { - va_list ap; - int n; + va_list ap; + int n; - assert(buf && buf->unit); + assert(buf && buf->unit); - if (buf->size >= buf->asize) - hoedown_buffer_grow(buf, buf->size + 1); + if (buf->size >= buf->asize) + hoedown_buffer_grow(buf, buf->size + 1); - va_start(ap, fmt); - n = vsnprintf((char *)buf->data + buf->size, buf->asize - buf->size, fmt, ap); - va_end(ap); + va_start(ap, fmt); + n = vsnprintf((char *)buf->data + buf->size, buf->asize - buf->size, fmt, ap); + va_end(ap); - if (n < 0) { + if (n < 0) { #ifndef _MSC_VER - return; + return; #else - va_start(ap, fmt); - n = _vscprintf(fmt, ap); - va_end(ap); + va_start(ap, fmt); + n = _vscprintf(fmt, ap); + va_end(ap); #endif - } + } - if ((size_t)n >= buf->asize - buf->size) { - hoedown_buffer_grow(buf, buf->size + n + 1); + if ((size_t)n >= buf->asize - buf->size) { + hoedown_buffer_grow(buf, buf->size + n + 1); - va_start(ap, fmt); - n = vsnprintf((char *)buf->data + buf->size, buf->asize - buf->size, fmt, ap); - va_end(ap); - } + va_start(ap, fmt); + n = vsnprintf((char *)buf->data + buf->size, buf->asize - buf->size, fmt, ap); + va_end(ap); + } - if (n < 0) - return; + if (n < 0) + return; - buf->size += n; + buf->size += n; } void hoedown_buffer_put_utf8(hoedown_buffer *buf, unsigned int c) { - unsigned char unichar[4]; - - assert(buf && buf->unit); - - if (c < 0x80) { - hoedown_buffer_putc(buf, c); - } - else if (c < 0x800) { - unichar[0] = 192 + (c / 64); - unichar[1] = 128 + (c % 64); - hoedown_buffer_put(buf, unichar, 2); - } - else if (c - 0xd800u < 0x800) { - HOEDOWN_BUFPUTSL(buf, "\xef\xbf\xbd"); - } - else if (c < 0x10000) { - unichar[0] = 224 + (c / 4096); - unichar[1] = 128 + (c / 64) % 64; - unichar[2] = 128 + (c % 64); - hoedown_buffer_put(buf, unichar, 3); - } - else if (c < 0x110000) { - unichar[0] = 240 + (c / 262144); - unichar[1] = 128 + (c / 4096) % 64; - unichar[2] = 128 + (c / 64) % 64; - unichar[3] = 128 + (c % 64); - hoedown_buffer_put(buf, unichar, 4); - } - else { - HOEDOWN_BUFPUTSL(buf, "\xef\xbf\xbd"); - } + unsigned char unichar[4]; + + assert(buf && buf->unit); + + if (c < 0x80) { + hoedown_buffer_putc(buf, c); + } + else if (c < 0x800) { + unichar[0] = 192 + (c / 64); + unichar[1] = 128 + (c % 64); + hoedown_buffer_put(buf, unichar, 2); + } + else if (c - 0xd800u < 0x800) { + HOEDOWN_BUFPUTSL(buf, "\xef\xbf\xbd"); + } + else if (c < 0x10000) { + unichar[0] = 224 + (c / 4096); + unichar[1] = 128 + (c / 64) % 64; + unichar[2] = 128 + (c % 64); + hoedown_buffer_put(buf, unichar, 3); + } + else if (c < 0x110000) { + unichar[0] = 240 + (c / 262144); + unichar[1] = 128 + (c / 4096) % 64; + unichar[2] = 128 + (c / 64) % 64; + unichar[3] = 128 + (c % 64); + hoedown_buffer_put(buf, unichar, 4); + } + else { + HOEDOWN_BUFPUTSL(buf, "\xef\xbf\xbd"); + } } diff --git a/libraries/hoedown/src/document.c b/libraries/hoedown/src/document.c index 8ba82e47..e9e2ab11 100644 --- a/libraries/hoedown/src/document.c +++ b/libraries/hoedown/src/document.c @@ -10,7 +10,7 @@ #ifndef _MSC_VER #include <strings.h> #else -#define strncasecmp _strnicmp +#define strncasecmp _strnicmp #endif #define REF_TABLE_SIZE 8 @@ -18,7 +18,7 @@ #define BUFFER_BLOCK 0 #define BUFFER_SPAN 1 -#define HOEDOWN_LI_END 8 /* internal list flag */ +#define HOEDOWN_LI_END 8 /* internal list flag */ const char *hoedown_find_block_tag(const char *str, unsigned int len); @@ -28,35 +28,35 @@ const char *hoedown_find_block_tag(const char *str, unsigned int len); /* link_ref: reference to a link */ struct link_ref { - unsigned int id; + unsigned int id; - hoedown_buffer *link; - hoedown_buffer *title; + hoedown_buffer *link; + hoedown_buffer *title; - struct link_ref *next; + struct link_ref *next; }; /* footnote_ref: reference to a footnote */ struct footnote_ref { - unsigned int id; + unsigned int id; - int is_used; - unsigned int num; + int is_used; + unsigned int num; - hoedown_buffer *contents; + hoedown_buffer *contents; }; /* footnote_item: an item in a footnote_list */ struct footnote_item { - struct footnote_ref *ref; - struct footnote_item *next; + struct footnote_ref *ref; + struct footnote_item *next; }; /* footnote_list: linked list of footnote_item */ struct footnote_list { - unsigned int count; - struct footnote_item *head; - struct footnote_item *tail; + unsigned int count; + struct footnote_item *head; + struct footnote_item *tail; }; /* char_trigger: function pointer to render active chars */ @@ -81,51 +81,51 @@ static size_t char_superscript(hoedown_buffer *ob, hoedown_document *doc, uint8_ static size_t char_math(hoedown_buffer *ob, hoedown_document *doc, uint8_t *data, size_t offset, size_t size); enum markdown_char_t { - MD_CHAR_NONE = 0, - MD_CHAR_EMPHASIS, - MD_CHAR_CODESPAN, - MD_CHAR_LINEBREAK, - MD_CHAR_LINK, - MD_CHAR_LANGLE, - MD_CHAR_ESCAPE, - MD_CHAR_ENTITY, - MD_CHAR_AUTOLINK_URL, - MD_CHAR_AUTOLINK_EMAIL, - MD_CHAR_AUTOLINK_WWW, - MD_CHAR_SUPERSCRIPT, - MD_CHAR_QUOTE, - MD_CHAR_MATH + MD_CHAR_NONE = 0, + MD_CHAR_EMPHASIS, + MD_CHAR_CODESPAN, + MD_CHAR_LINEBREAK, + MD_CHAR_LINK, + MD_CHAR_LANGLE, + MD_CHAR_ESCAPE, + MD_CHAR_ENTITY, + MD_CHAR_AUTOLINK_URL, + MD_CHAR_AUTOLINK_EMAIL, + MD_CHAR_AUTOLINK_WWW, + MD_CHAR_SUPERSCRIPT, + MD_CHAR_QUOTE, + MD_CHAR_MATH }; static char_trigger markdown_char_ptrs[] = { - NULL, - &char_emphasis, - &char_codespan, - &char_linebreak, - &char_link, - &char_langle_tag, - &char_escape, - &char_entity, - &char_autolink_url, - &char_autolink_email, - &char_autolink_www, - &char_superscript, - &char_quote, - &char_math + NULL, + &char_emphasis, + &char_codespan, + &char_linebreak, + &char_link, + &char_langle_tag, + &char_escape, + &char_entity, + &char_autolink_url, + &char_autolink_email, + &char_autolink_www, + &char_superscript, + &char_quote, + &char_math }; struct hoedown_document { - hoedown_renderer md; - hoedown_renderer_data data; - - struct link_ref *refs[REF_TABLE_SIZE]; - struct footnote_list footnotes_found; - struct footnote_list footnotes_used; - uint8_t active_char[256]; - hoedown_stack work_bufs[2]; - hoedown_extensions ext_flags; - size_t max_nesting; - int in_link_body; + hoedown_renderer md; + hoedown_renderer_data data; + + struct link_ref *refs[REF_TABLE_SIZE]; + struct footnote_list footnotes_found; + struct footnote_list footnotes_used; + uint8_t active_char[256]; + hoedown_stack work_bufs[2]; + hoedown_extensions ext_flags; + size_t max_nesting; + int in_link_body; }; /*************************** @@ -135,177 +135,177 @@ struct hoedown_document { static hoedown_buffer * newbuf(hoedown_document *doc, int type) { - static const size_t buf_size[2] = {256, 64}; - hoedown_buffer *work = NULL; - hoedown_stack *pool = &doc->work_bufs[type]; + static const size_t buf_size[2] = {256, 64}; + hoedown_buffer *work = NULL; + hoedown_stack *pool = &doc->work_bufs[type]; - if (pool->size < pool->asize && - pool->item[pool->size] != NULL) { - work = pool->item[pool->size++]; - work->size = 0; - } else { - work = hoedown_buffer_new(buf_size[type]); - hoedown_stack_push(pool, work); - } + if (pool->size < pool->asize && + pool->item[pool->size] != NULL) { + work = pool->item[pool->size++]; + work->size = 0; + } else { + work = hoedown_buffer_new(buf_size[type]); + hoedown_stack_push(pool, work); + } - return work; + return work; } static void popbuf(hoedown_document *doc, int type) { - doc->work_bufs[type].size--; + doc->work_bufs[type].size--; } static void unscape_text(hoedown_buffer *ob, hoedown_buffer *src) { - size_t i = 0, org; - while (i < src->size) { - org = i; - while (i < src->size && src->data[i] != '\\') - i++; + size_t i = 0, org; + while (i < src->size) { + org = i; + while (i < src->size && src->data[i] != '\\') + i++; - if (i > org) - hoedown_buffer_put(ob, src->data + org, i - org); + if (i > org) + hoedown_buffer_put(ob, src->data + org, i - org); - if (i + 1 >= src->size) - break; + if (i + 1 >= src->size) + break; - hoedown_buffer_putc(ob, src->data[i + 1]); - i += 2; - } + hoedown_buffer_putc(ob, src->data[i + 1]); + i += 2; + } } static unsigned int hash_link_ref(const uint8_t *link_ref, size_t length) { - size_t i; - unsigned int hash = 0; + size_t i; + unsigned int hash = 0; - for (i = 0; i < length; ++i) - hash = tolower(link_ref[i]) + (hash << 6) + (hash << 16) - hash; + for (i = 0; i < length; ++i) + hash = tolower(link_ref[i]) + (hash << 6) + (hash << 16) - hash; - return hash; + return hash; } static struct link_ref * add_link_ref( - struct link_ref **references, - const uint8_t *name, size_t name_size) + struct link_ref **references, + const uint8_t *name, size_t name_size) { - struct link_ref *ref = hoedown_calloc(1, sizeof(struct link_ref)); + struct link_ref *ref = hoedown_calloc(1, sizeof(struct link_ref)); - ref->id = hash_link_ref(name, name_size); - ref->next = references[ref->id % REF_TABLE_SIZE]; + ref->id = hash_link_ref(name, name_size); + ref->next = references[ref->id % REF_TABLE_SIZE]; - references[ref->id % REF_TABLE_SIZE] = ref; - return ref; + references[ref->id % REF_TABLE_SIZE] = ref; + return ref; } static struct link_ref * find_link_ref(struct link_ref **references, uint8_t *name, size_t length) { - unsigned int hash = hash_link_ref(name, length); - struct link_ref *ref = NULL; + unsigned int hash = hash_link_ref(name, length); + struct link_ref *ref = NULL; - ref = references[hash % REF_TABLE_SIZE]; + ref = references[hash % REF_TABLE_SIZE]; - while (ref != NULL) { - if (ref->id == hash) - return ref; + while (ref != NULL) { + if (ref->id == hash) + return ref; - ref = ref->next; - } + ref = ref->next; + } - return NULL; + return NULL; } static void free_link_refs(struct link_ref **references) { - size_t i; + size_t i; - for (i = 0; i < REF_TABLE_SIZE; ++i) { - struct link_ref *r = references[i]; - struct link_ref *next; + for (i = 0; i < REF_TABLE_SIZE; ++i) { + struct link_ref *r = references[i]; + struct link_ref *next; - while (r) { - next = r->next; - hoedown_buffer_free(r->link); - hoedown_buffer_free(r->title); - free(r); - r = next; - } - } + while (r) { + next = r->next; + hoedown_buffer_free(r->link); + hoedown_buffer_free(r->title); + free(r); + r = next; + } + } } static struct footnote_ref * create_footnote_ref(struct footnote_list *list, const uint8_t *name, size_t name_size) { - struct footnote_ref *ref = hoedown_calloc(1, sizeof(struct footnote_ref)); + struct footnote_ref *ref = hoedown_calloc(1, sizeof(struct footnote_ref)); - ref->id = hash_link_ref(name, name_size); + ref->id = hash_link_ref(name, name_size); - return ref; + return ref; } static int add_footnote_ref(struct footnote_list *list, struct footnote_ref *ref) { - struct footnote_item *item = hoedown_calloc(1, sizeof(struct footnote_item)); - if (!item) - return 0; - item->ref = ref; + struct footnote_item *item = hoedown_calloc(1, sizeof(struct footnote_item)); + if (!item) + return 0; + item->ref = ref; - if (list->head == NULL) { - list->head = list->tail = item; - } else { - list->tail->next = item; - list->tail = item; - } - list->count++; + if (list->head == NULL) { + list->head = list->tail = item; + } else { + list->tail->next = item; + list->tail = item; + } + list->count++; - return 1; + return 1; } static struct footnote_ref * find_footnote_ref(struct footnote_list *list, uint8_t *name, size_t length) { - unsigned int hash = hash_link_ref(name, length); - struct footnote_item *item = NULL; + unsigned int hash = hash_link_ref(name, length); + struct footnote_item *item = NULL; - item = list->head; + item = list->head; - while (item != NULL) { - if (item->ref->id == hash) - return item->ref; - item = item->next; - } + while (item != NULL) { + if (item->ref->id == hash) + return item->ref; + item = item->next; + } - return NULL; + return NULL; } static void free_footnote_ref(struct footnote_ref *ref) { - hoedown_buffer_free(ref->contents); - free(ref); + hoedown_buffer_free(ref->contents); + free(ref); } static void free_footnote_list(struct footnote_list *list, int free_refs) { - struct footnote_item *item = list->head; - struct footnote_item *next; + struct footnote_item *item = list->head; + struct footnote_item *next; - while (item) { - next = item->next; - if (free_refs) - free_footnote_ref(item->ref); - free(item); - item = next; - } + while (item) { + next = item->next; + if (free_refs) + free_footnote_ref(item->ref); + free(item); + item = next; + } } @@ -323,16 +323,16 @@ free_footnote_list(struct footnote_list *list, int free_refs) static int _isspace(int c) { - return c == ' ' || c == '\n'; + return c == ' ' || c == '\n'; } /* is_empty_all: verify that all the data is spacing */ static int is_empty_all(const uint8_t *data, size_t size) { - size_t i = 0; - while (i < size && _isspace(data[i])) i++; - return i == size; + size_t i = 0; + while (i < size && _isspace(data[i])) i++; + return i == size; } /* @@ -342,19 +342,19 @@ is_empty_all(const uint8_t *data, size_t size) static void replace_spacing(hoedown_buffer *ob, const uint8_t *data, size_t size) { - size_t i = 0, mark; - hoedown_buffer_grow(ob, size); - while (1) { - mark = i; - while (i < size && data[i] != '\n') i++; - hoedown_buffer_put(ob, data + mark, i - mark); + size_t i = 0, mark; + hoedown_buffer_grow(ob, size); + while (1) { + mark = i; + while (i < size && data[i] != '\n') i++; + hoedown_buffer_put(ob, data + mark, i - mark); - if (i >= size) break; + if (i >= size) break; - if (!(i > 0 && data[i-1] == ' ')) - hoedown_buffer_putc(ob, ' '); - i++; - } + if (!(i > 0 && data[i-1] == ' ')) + hoedown_buffer_putc(ob, ' '); + i++; + } } /**************************** @@ -366,237 +366,237 @@ replace_spacing(hoedown_buffer *ob, const uint8_t *data, size_t size) static size_t is_mail_autolink(uint8_t *data, size_t size) { - size_t i = 0, nb = 0; + size_t i = 0, nb = 0; - /* address is assumed to be: [-@._a-zA-Z0-9]+ with exactly one '@' */ - for (i = 0; i < size; ++i) { - if (isalnum(data[i])) - continue; + /* address is assumed to be: [-@._a-zA-Z0-9]+ with exactly one '@' */ + for (i = 0; i < size; ++i) { + if (isalnum(data[i])) + continue; - switch (data[i]) { - case '@': - nb++; + switch (data[i]) { + case '@': + nb++; - case '-': - case '.': - case '_': - break; + case '-': + case '.': + case '_': + break; - case '>': - return (nb == 1) ? i + 1 : 0; + case '>': + return (nb == 1) ? i + 1 : 0; - default: - return 0; - } - } + default: + return 0; + } + } - return 0; + return 0; } /* tag_length • returns the length of the given tag, or 0 is it's not valid */ static size_t tag_length(uint8_t *data, size_t size, hoedown_autolink_type *autolink) { - size_t i, j; + size_t i, j; - /* a valid tag can't be shorter than 3 chars */ - if (size < 3) return 0; + /* a valid tag can't be shorter than 3 chars */ + if (size < 3) return 0; - /* begins with a '<' optionally followed by '/', followed by letter or number */ - if (data[0] != '<') return 0; - i = (data[1] == '/') ? 2 : 1; + /* begins with a '<' optionally followed by '/', followed by letter or number */ + if (data[0] != '<') return 0; + i = (data[1] == '/') ? 2 : 1; - if (!isalnum(data[i])) - return 0; + if (!isalnum(data[i])) + return 0; - /* scheme test */ - *autolink = HOEDOWN_AUTOLINK_NONE; + /* scheme test */ + *autolink = HOEDOWN_AUTOLINK_NONE; - /* try to find the beginning of an URI */ - while (i < size && (isalnum(data[i]) || data[i] == '.' || data[i] == '+' || data[i] == '-')) - i++; + /* try to find the beginning of an URI */ + while (i < size && (isalnum(data[i]) || data[i] == '.' || data[i] == '+' || data[i] == '-')) + i++; - if (i > 1 && data[i] == '@') { - if ((j = is_mail_autolink(data + i, size - i)) != 0) { - *autolink = HOEDOWN_AUTOLINK_EMAIL; - return i + j; - } - } + if (i > 1 && data[i] == '@') { + if ((j = is_mail_autolink(data + i, size - i)) != 0) { + *autolink = HOEDOWN_AUTOLINK_EMAIL; + return i + j; + } + } - if (i > 2 && data[i] == ':') { - *autolink = HOEDOWN_AUTOLINK_NORMAL; - i++; - } + if (i > 2 && data[i] == ':') { + *autolink = HOEDOWN_AUTOLINK_NORMAL; + i++; + } - /* completing autolink test: no spacing or ' or " */ - if (i >= size) - *autolink = HOEDOWN_AUTOLINK_NONE; + /* completing autolink test: no spacing or ' or " */ + if (i >= size) + *autolink = HOEDOWN_AUTOLINK_NONE; - else if (*autolink) { - j = i; + else if (*autolink) { + j = i; - while (i < size) { - if (data[i] == '\\') i += 2; - else if (data[i] == '>' || data[i] == '\'' || - data[i] == '"' || data[i] == ' ' || data[i] == '\n') - break; - else i++; - } + while (i < size) { + if (data[i] == '\\') i += 2; + else if (data[i] == '>' || data[i] == '\'' || + data[i] == '"' || data[i] == ' ' || data[i] == '\n') + break; + else i++; + } - if (i >= size) return 0; - if (i > j && data[i] == '>') return i + 1; - /* one of the forbidden chars has been found */ - *autolink = HOEDOWN_AUTOLINK_NONE; - } + if (i >= size) return 0; + if (i > j && data[i] == '>') return i + 1; + /* one of the forbidden chars has been found */ + *autolink = HOEDOWN_AUTOLINK_NONE; + } - /* looking for something looking like a tag end */ - while (i < size && data[i] != '>') i++; - if (i >= size) return 0; - return i + 1; + /* looking for something looking like a tag end */ + while (i < size && data[i] != '>') i++; + if (i >= size) return 0; + return i + 1; } /* parse_inline • parses inline markdown elements */ static void parse_inline(hoedown_buffer *ob, hoedown_document *doc, uint8_t *data, size_t size) { - size_t i = 0, end = 0, consumed = 0; - hoedown_buffer work = { 0, 0, 0, 0, NULL, NULL, NULL }; - uint8_t *active_char = doc->active_char; - - if (doc->work_bufs[BUFFER_SPAN].size + - doc->work_bufs[BUFFER_BLOCK].size > doc->max_nesting) - return; - - while (i < size) { - /* copying inactive chars into the output */ - while (end < size && active_char[data[end]] == 0) - end++; - - if (doc->md.normal_text) { - work.data = data + i; - work.size = end - i; - doc->md.normal_text(ob, &work, &doc->data); - } - else - hoedown_buffer_put(ob, data + i, end - i); - - if (end >= size) break; - i = end; - - end = markdown_char_ptrs[ (int)active_char[data[end]] ](ob, doc, data + i, i - consumed, size - i); - if (!end) /* no action from the callback */ - end = i + 1; - else { - i += end; - end = i; - consumed = i; - } - } + size_t i = 0, end = 0, consumed = 0; + hoedown_buffer work = { 0, 0, 0, 0, NULL, NULL, NULL }; + uint8_t *active_char = doc->active_char; + + if (doc->work_bufs[BUFFER_SPAN].size + + doc->work_bufs[BUFFER_BLOCK].size > doc->max_nesting) + return; + + while (i < size) { + /* copying inactive chars into the output */ + while (end < size && active_char[data[end]] == 0) + end++; + + if (doc->md.normal_text) { + work.data = data + i; + work.size = end - i; + doc->md.normal_text(ob, &work, &doc->data); + } + else + hoedown_buffer_put(ob, data + i, end - i); + + if (end >= size) break; + i = end; + + end = markdown_char_ptrs[ (int)active_char[data[end]] ](ob, doc, data + i, i - consumed, size - i); + if (!end) /* no action from the callback */ + end = i + 1; + else { + i += end; + end = i; + consumed = i; + } + } } /* is_escaped • returns whether special char at data[loc] is escaped by '\\' */ static int is_escaped(uint8_t *data, size_t loc) { - size_t i = loc; - while (i >= 1 && data[i - 1] == '\\') - i--; + size_t i = loc; + while (i >= 1 && data[i - 1] == '\\') + i--; - /* odd numbers of backslashes escapes data[loc] */ - return (loc - i) % 2; + /* odd numbers of backslashes escapes data[loc] */ + return (loc - i) % 2; } /* find_emph_char • looks for the next emph uint8_t, skipping other constructs */ static size_t find_emph_char(uint8_t *data, size_t size, uint8_t c) { - size_t i = 0; + size_t i = 0; - while (i < size) { - while (i < size && data[i] != c && data[i] != '[' && data[i] != '`') - i++; + while (i < size) { + while (i < size && data[i] != c && data[i] != '[' && data[i] != '`') + i++; - if (i == size) - return 0; + if (i == size) + return 0; - /* not counting escaped chars */ - if (is_escaped(data, i)) { - i++; continue; - } + /* not counting escaped chars */ + if (is_escaped(data, i)) { + i++; continue; + } - if (data[i] == c) - return i; + if (data[i] == c) + return i; - /* skipping a codespan */ - if (data[i] == '`') { - size_t span_nb = 0, bt; - size_t tmp_i = 0; + /* skipping a codespan */ + if (data[i] == '`') { + size_t span_nb = 0, bt; + size_t tmp_i = 0; - /* counting the number of opening backticks */ - while (i < size && data[i] == '`') { - i++; span_nb++; - } + /* counting the number of opening backticks */ + while (i < size && data[i] == '`') { + i++; span_nb++; + } - if (i >= size) return 0; + if (i >= size) return 0; - /* finding the matching closing sequence */ - bt = 0; - while (i < size && bt < span_nb) { - if (!tmp_i && data[i] == c) tmp_i = i; - if (data[i] == '`') bt++; - else bt = 0; - i++; - } + /* finding the matching closing sequence */ + bt = 0; + while (i < size && bt < span_nb) { + if (!tmp_i && data[i] == c) tmp_i = i; + if (data[i] == '`') bt++; + else bt = 0; + i++; + } - /* not a well-formed codespan; use found matching emph char */ - if (i >= size) return tmp_i; - } - /* skipping a link */ - else if (data[i] == '[') { - size_t tmp_i = 0; - uint8_t cc; + /* not a well-formed codespan; use found matching emph char */ + if (i >= size) return tmp_i; + } + /* skipping a link */ + else if (data[i] == '[') { + size_t tmp_i = 0; + uint8_t cc; - i++; - while (i < size && data[i] != ']') { - if (!tmp_i && data[i] == c) tmp_i = i; - i++; - } + i++; + while (i < size && data[i] != ']') { + if (!tmp_i && data[i] == c) tmp_i = i; + i++; + } - i++; - while (i < size && _isspace(data[i])) - i++; + i++; + while (i < size && _isspace(data[i])) + i++; - if (i >= size) - return tmp_i; + if (i >= size) + return tmp_i; - switch (data[i]) { - case '[': - cc = ']'; break; + switch (data[i]) { + case '[': + cc = ']'; break; - case '(': - cc = ')'; break; + case '(': + cc = ')'; break; - default: - if (tmp_i) - return tmp_i; - else - continue; - } + default: + if (tmp_i) + return tmp_i; + else + continue; + } - i++; - while (i < size && data[i] != cc) { - if (!tmp_i && data[i] == c) tmp_i = i; - i++; - } + i++; + while (i < size && data[i] != cc) { + if (!tmp_i && data[i] == c) tmp_i = i; + i++; + } - if (i >= size) - return tmp_i; + if (i >= size) + return tmp_i; - i++; - } - } + i++; + } + } - return 0; + return 0; } /* parse_emph1 • parsing single emphase */ @@ -604,72 +604,72 @@ find_emph_char(uint8_t *data, size_t size, uint8_t c) static size_t parse_emph1(hoedown_buffer *ob, hoedown_document *doc, uint8_t *data, size_t size, uint8_t c) { - size_t i = 0, len; - hoedown_buffer *work = 0; - int r; + size_t i = 0, len; + hoedown_buffer *work = 0; + int r; - /* skipping one symbol if coming from emph3 */ - if (size > 1 && data[0] == c && data[1] == c) i = 1; + /* skipping one symbol if coming from emph3 */ + if (size > 1 && data[0] == c && data[1] == c) i = 1; - while (i < size) { - len = find_emph_char(data + i, size - i, c); - if (!len) return 0; - i += len; - if (i >= size) return 0; + while (i < size) { + len = find_emph_char(data + i, size - i, c); + if (!len) return 0; + i += len; + if (i >= size) return 0; - if (data[i] == c && !_isspace(data[i - 1])) { + if (data[i] == c && !_isspace(data[i - 1])) { - if (doc->ext_flags & HOEDOWN_EXT_NO_INTRA_EMPHASIS) { - if (i + 1 < size && isalnum(data[i + 1])) - continue; - } + if (doc->ext_flags & HOEDOWN_EXT_NO_INTRA_EMPHASIS) { + if (i + 1 < size && isalnum(data[i + 1])) + continue; + } - work = newbuf(doc, BUFFER_SPAN); - parse_inline(work, doc, data, i); + work = newbuf(doc, BUFFER_SPAN); + parse_inline(work, doc, data, i); - if (doc->ext_flags & HOEDOWN_EXT_UNDERLINE && c == '_') - r = doc->md.underline(ob, work, &doc->data); - else - r = doc->md.emphasis(ob, work, &doc->data); + if (doc->ext_flags & HOEDOWN_EXT_UNDERLINE && c == '_') + r = doc->md.underline(ob, work, &doc->data); + else + r = doc->md.emphasis(ob, work, &doc->data); - popbuf(doc, BUFFER_SPAN); - return r ? i + 1 : 0; - } - } + popbuf(doc, BUFFER_SPAN); + return r ? i + 1 : 0; + } + } - return 0; + return 0; } /* parse_emph2 • parsing single emphase */ static size_t parse_emph2(hoedown_buffer *ob, hoedown_document *doc, uint8_t *data, size_t size, uint8_t c) { - size_t i = 0, len; - hoedown_buffer *work = 0; - int r; + size_t i = 0, len; + hoedown_buffer *work = 0; + int r; - while (i < size) { - len = find_emph_char(data + i, size - i, c); - if (!len) return 0; - i += len; + while (i < size) { + len = find_emph_char(data + i, size - i, c); + if (!len) return 0; + i += len; - if (i + 1 < size && data[i] == c && data[i + 1] == c && i && !_isspace(data[i - 1])) { - work = newbuf(doc, BUFFER_SPAN); - parse_inline(work, doc, data, i); + if (i + 1 < size && data[i] == c && data[i + 1] == c && i && !_isspace(data[i - 1])) { + work = newbuf(doc, BUFFER_SPAN); + parse_inline(work, doc, data, i); - if (c == '~') - r = doc->md.strikethrough(ob, work, &doc->data); - else if (c == '=') - r = doc->md.highlight(ob, work, &doc->data); - else - r = doc->md.double_emphasis(ob, work, &doc->data); + if (c == '~') + r = doc->md.strikethrough(ob, work, &doc->data); + else if (c == '=') + r = doc->md.highlight(ob, work, &doc->data); + else + r = doc->md.double_emphasis(ob, work, &doc->data); - popbuf(doc, BUFFER_SPAN); - return r ? i + 2 : 0; - } - i++; - } - return 0; + popbuf(doc, BUFFER_SPAN); + return r ? i + 2 : 0; + } + i++; + } + return 0; } /* parse_emph3 • parsing single emphase */ @@ -677,121 +677,121 @@ parse_emph2(hoedown_buffer *ob, hoedown_document *doc, uint8_t *data, size_t siz static size_t parse_emph3(hoedown_buffer *ob, hoedown_document *doc, uint8_t *data, size_t size, uint8_t c) { - size_t i = 0, len; - int r; + size_t i = 0, len; + int r; - while (i < size) { - len = find_emph_char(data + i, size - i, c); - if (!len) return 0; - i += len; + while (i < size) { + len = find_emph_char(data + i, size - i, c); + if (!len) return 0; + i += len; - /* skip spacing preceded symbols */ - if (data[i] != c || _isspace(data[i - 1])) - continue; + /* skip spacing preceded symbols */ + if (data[i] != c || _isspace(data[i - 1])) + continue; - if (i + 2 < size && data[i + 1] == c && data[i + 2] == c && doc->md.triple_emphasis) { - /* triple symbol found */ - hoedown_buffer *work = newbuf(doc, BUFFER_SPAN); + if (i + 2 < size && data[i + 1] == c && data[i + 2] == c && doc->md.triple_emphasis) { + /* triple symbol found */ + hoedown_buffer *work = newbuf(doc, BUFFER_SPAN); - parse_inline(work, doc, data, i); - r = doc->md.triple_emphasis(ob, work, &doc->data); - popbuf(doc, BUFFER_SPAN); - return r ? i + 3 : 0; + parse_inline(work, doc, data, i); + r = doc->md.triple_emphasis(ob, work, &doc->data); + popbuf(doc, BUFFER_SPAN); + return r ? i + 3 : 0; - } else if (i + 1 < size && data[i + 1] == c) { - /* double symbol found, handing over to emph1 */ - len = parse_emph1(ob, doc, data - 2, size + 2, c); - if (!len) return 0; - else return len - 2; + } else if (i + 1 < size && data[i + 1] == c) { + /* double symbol found, handing over to emph1 */ + len = parse_emph1(ob, doc, data - 2, size + 2, c); + if (!len) return 0; + else return len - 2; - } else { - /* single symbol found, handing over to emph2 */ - len = parse_emph2(ob, doc, data - 1, size + 1, c); - if (!len) return 0; - else return len - 1; - } - } - return 0; + } else { + /* single symbol found, handing over to emph2 */ + len = parse_emph2(ob, doc, data - 1, size + 1, c); + if (!len) return 0; + else return len - 1; + } + } + return 0; } /* parse_math • parses a math span until the given ending delimiter */ static size_t parse_math(hoedown_buffer *ob, hoedown_document *doc, uint8_t *data, size_t offset, size_t size, const char *end, size_t delimsz, int displaymode) { - hoedown_buffer text = { NULL, 0, 0, 0, NULL, NULL, NULL }; - size_t i = delimsz; + hoedown_buffer text = { NULL, 0, 0, 0, NULL, NULL, NULL }; + size_t i = delimsz; - if (!doc->md.math) - return 0; + if (!doc->md.math) + return 0; - /* find ending delimiter */ - while (1) { - while (i < size && data[i] != (uint8_t)end[0]) - i++; + /* find ending delimiter */ + while (1) { + while (i < size && data[i] != (uint8_t)end[0]) + i++; - if (i >= size) - return 0; + if (i >= size) + return 0; - if (!is_escaped(data, i) && !(i + delimsz > size) - && memcmp(data + i, end, delimsz) == 0) - break; + if (!is_escaped(data, i) && !(i + delimsz > size) + && memcmp(data + i, end, delimsz) == 0) + break; - i++; - } + i++; + } - /* prepare buffers */ - text.data = data + delimsz; - text.size = i - delimsz; + /* prepare buffers */ + text.data = data + delimsz; + text.size = i - delimsz; - /* if this is a $$ and MATH_EXPLICIT is not active, - * guess whether displaymode should be enabled from the context */ - i += delimsz; - if (delimsz == 2 && !(doc->ext_flags & HOEDOWN_EXT_MATH_EXPLICIT)) - displaymode = is_empty_all(data - offset, offset) && is_empty_all(data + i, size - i); + /* if this is a $$ and MATH_EXPLICIT is not active, + * guess whether displaymode should be enabled from the context */ + i += delimsz; + if (delimsz == 2 && !(doc->ext_flags & HOEDOWN_EXT_MATH_EXPLICIT)) + displaymode = is_empty_all(data - offset, offset) && is_empty_all(data + i, size - i); - /* call callback */ - if (doc->md.math(ob, &text, displaymode, &doc->data)) - return i; + /* call callback */ + if (doc->md.math(ob, &text, displaymode, &doc->data)) + return i; - return 0; + return 0; } /* char_emphasis • single and double emphasis parsing */ static size_t char_emphasis(hoedown_buffer *ob, hoedown_document *doc, uint8_t *data, size_t offset, size_t size) { - uint8_t c = data[0]; - size_t ret; + uint8_t c = data[0]; + size_t ret; - if (doc->ext_flags & HOEDOWN_EXT_NO_INTRA_EMPHASIS) { - if (offset > 0 && !_isspace(data[-1]) && data[-1] != '>' && data[-1] != '(') - return 0; - } + if (doc->ext_flags & HOEDOWN_EXT_NO_INTRA_EMPHASIS) { + if (offset > 0 && !_isspace(data[-1]) && data[-1] != '>' && data[-1] != '(') + return 0; + } - if (size > 2 && data[1] != c) { - /* spacing cannot follow an opening emphasis; - * strikethrough and highlight only takes two characters '~~' */ - if (c == '~' || c == '=' || _isspace(data[1]) || (ret = parse_emph1(ob, doc, data + 1, size - 1, c)) == 0) - return 0; + if (size > 2 && data[1] != c) { + /* spacing cannot follow an opening emphasis; + * strikethrough and highlight only takes two characters '~~' */ + if (c == '~' || c == '=' || _isspace(data[1]) || (ret = parse_emph1(ob, doc, data + 1, size - 1, c)) == 0) + return 0; - return ret + 1; - } + return ret + 1; + } - if (size > 3 && data[1] == c && data[2] != c) { - if (_isspace(data[2]) || (ret = parse_emph2(ob, doc, data + 2, size - 2, c)) == 0) - return 0; + if (size > 3 && data[1] == c && data[2] != c) { + if (_isspace(data[2]) || (ret = parse_emph2(ob, doc, data + 2, size - 2, c)) == 0) + return 0; - return ret + 2; - } + return ret + 2; + } - if (size > 4 && data[1] == c && data[2] == c && data[3] != c) { - if (c == '~' || c == '=' || _isspace(data[3]) || (ret = parse_emph3(ob, doc, data + 3, size - 3, c)) == 0) - return 0; + if (size > 4 && data[1] == c && data[2] == c && data[3] != c) { + if (c == '~' || c == '=' || _isspace(data[3]) || (ret = parse_emph3(ob, doc, data + 3, size - 3, c)) == 0) + return 0; - return ret + 3; - } + return ret + 3; + } - return 0; + return 0; } @@ -799,14 +799,14 @@ char_emphasis(hoedown_buffer *ob, hoedown_document *doc, uint8_t *data, size_t o static size_t char_linebreak(hoedown_buffer *ob, hoedown_document *doc, uint8_t *data, size_t offset, size_t size) { - if (offset < 2 || data[-1] != ' ' || data[-2] != ' ') - return 0; + if (offset < 2 || data[-1] != ' ' || data[-2] != ' ') + return 0; - /* removing the last space from ob and rendering */ - while (ob->size && ob->data[ob->size - 1] == ' ') - ob->size--; + /* removing the last space from ob and rendering */ + while (ob->size && ob->data[ob->size - 1] == ' ') + ob->size--; - return doc->md.linebreak(ob, &doc->data) ? 1 : 0; + return doc->md.linebreak(ob, &doc->data) ? 1 : 0; } @@ -814,91 +814,91 @@ char_linebreak(hoedown_buffer *ob, hoedown_document *doc, uint8_t *data, size_t static size_t char_codespan(hoedown_buffer *ob, hoedown_document *doc, uint8_t *data, size_t offset, size_t size) { - hoedown_buffer work = { NULL, 0, 0, 0, NULL, NULL, NULL }; - size_t end, nb = 0, i, f_begin, f_end; + hoedown_buffer work = { NULL, 0, 0, 0, NULL, NULL, NULL }; + size_t end, nb = 0, i, f_begin, f_end; - /* counting the number of backticks in the delimiter */ - while (nb < size && data[nb] == '`') - nb++; + /* counting the number of backticks in the delimiter */ + while (nb < size && data[nb] == '`') + nb++; - /* finding the next delimiter */ - i = 0; - for (end = nb; end < size && i < nb; end++) { - if (data[end] == '`') i++; - else i = 0; - } + /* finding the next delimiter */ + i = 0; + for (end = nb; end < size && i < nb; end++) { + if (data[end] == '`') i++; + else i = 0; + } - if (i < nb && end >= size) - return 0; /* no matching delimiter */ + if (i < nb && end >= size) + return 0; /* no matching delimiter */ - /* trimming outside spaces */ - f_begin = nb; - while (f_begin < end && data[f_begin] == ' ') - f_begin++; + /* trimming outside spaces */ + f_begin = nb; + while (f_begin < end && data[f_begin] == ' ') + f_begin++; - f_end = end - nb; - while (f_end > nb && data[f_end-1] == ' ') - f_end--; + f_end = end - nb; + while (f_end > nb && data[f_end-1] == ' ') + f_end--; - /* real code span */ - if (f_begin < f_end) { - work.data = data + f_begin; - work.size = f_end - f_begin; + /* real code span */ + if (f_begin < f_end) { + work.data = data + f_begin; + work.size = f_end - f_begin; - if (!doc->md.codespan(ob, &work, &doc->data)) - end = 0; - } else { - if (!doc->md.codespan(ob, 0, &doc->data)) - end = 0; - } + if (!doc->md.codespan(ob, &work, &doc->data)) + end = 0; + } else { + if (!doc->md.codespan(ob, 0, &doc->data)) + end = 0; + } - return end; + return end; } /* char_quote • '"' parsing a quote */ static size_t char_quote(hoedown_buffer *ob, hoedown_document *doc, uint8_t *data, size_t offset, size_t size) { - size_t end, nq = 0, i, f_begin, f_end; + size_t end, nq = 0, i, f_begin, f_end; - /* counting the number of quotes in the delimiter */ - while (nq < size && data[nq] == '"') - nq++; + /* counting the number of quotes in the delimiter */ + while (nq < size && data[nq] == '"') + nq++; - /* finding the next delimiter */ - end = nq; - while (1) { - i = end; - end += find_emph_char(data + end, size - end, '"'); - if (end == i) return 0; /* no matching delimiter */ - i = end; - while (end < size && data[end] == '"' && end - i < nq) end++; - if (end - i >= nq) break; - } + /* finding the next delimiter */ + end = nq; + while (1) { + i = end; + end += find_emph_char(data + end, size - end, '"'); + if (end == i) return 0; /* no matching delimiter */ + i = end; + while (end < size && data[end] == '"' && end - i < nq) end++; + if (end - i >= nq) break; + } - /* trimming outside spaces */ - f_begin = nq; - while (f_begin < end && data[f_begin] == ' ') - f_begin++; + /* trimming outside spaces */ + f_begin = nq; + while (f_begin < end && data[f_begin] == ' ') + f_begin++; - f_end = end - nq; - while (f_end > nq && data[f_end-1] == ' ') - f_end--; + f_end = end - nq; + while (f_end > nq && data[f_end-1] == ' ') + f_end--; - /* real quote */ - if (f_begin < f_end) { - hoedown_buffer *work = newbuf(doc, BUFFER_SPAN); - parse_inline(work, doc, data + f_begin, f_end - f_begin); + /* real quote */ + if (f_begin < f_end) { + hoedown_buffer *work = newbuf(doc, BUFFER_SPAN); + parse_inline(work, doc, data + f_begin, f_end - f_begin); - if (!doc->md.quote(ob, work, &doc->data)) - end = 0; - popbuf(doc, BUFFER_SPAN); - } else { - if (!doc->md.quote(ob, 0, &doc->data)) - end = 0; - } + if (!doc->md.quote(ob, work, &doc->data)) + end = 0; + popbuf(doc, BUFFER_SPAN); + } else { + if (!doc->md.quote(ob, 0, &doc->data)) + end = 0; + } - return end; + return end; } @@ -906,32 +906,32 @@ char_quote(hoedown_buffer *ob, hoedown_document *doc, uint8_t *data, size_t offs static size_t char_escape(hoedown_buffer *ob, hoedown_document *doc, uint8_t *data, size_t offset, size_t size) { - static const char *escape_chars = "\\`*_{}[]()#+-.!:|&<>^~=\"$"; - hoedown_buffer work = { 0, 0, 0, 0, NULL, NULL, NULL }; - size_t w; + static const char *escape_chars = "\\`*_{}[]()#+-.!:|&<>^~=\"$"; + hoedown_buffer work = { 0, 0, 0, 0, NULL, NULL, NULL }; + size_t w; - if (size > 1) { - if (data[1] == '\\' && (doc->ext_flags & HOEDOWN_EXT_MATH) && - size > 2 && (data[2] == '(' || data[2] == '[')) { - const char *end = (data[2] == '[') ? "\\\\]" : "\\\\)"; - w = parse_math(ob, doc, data, offset, size, end, 3, data[2] == '['); - if (w) return w; - } + if (size > 1) { + if (data[1] == '\\' && (doc->ext_flags & HOEDOWN_EXT_MATH) && + size > 2 && (data[2] == '(' || data[2] == '[')) { + const char *end = (data[2] == '[') ? "\\\\]" : "\\\\)"; + w = parse_math(ob, doc, data, offset, size, end, 3, data[2] == '['); + if (w) return w; + } - if (strchr(escape_chars, data[1]) == NULL) - return 0; + if (strchr(escape_chars, data[1]) == NULL) + return 0; - if (doc->md.normal_text) { - work.data = data + 1; - work.size = 1; - doc->md.normal_text(ob, &work, &doc->data); - } - else hoedown_buffer_putc(ob, data[1]); - } else if (size == 1) { - hoedown_buffer_putc(ob, data[0]); - } + if (doc->md.normal_text) { + work.data = data + 1; + work.size = 1; + doc->md.normal_text(ob, &work, &doc->data); + } + else hoedown_buffer_putc(ob, data[1]); + } else if (size == 1) { + hoedown_buffer_putc(ob, data[0]); + } - return 2; + return 2; } /* char_entity • '&' escaped when it doesn't belong to an entity */ @@ -939,403 +939,403 @@ char_escape(hoedown_buffer *ob, hoedown_document *doc, uint8_t *data, size_t off static size_t char_entity(hoedown_buffer *ob, hoedown_document *doc, uint8_t *data, size_t offset, size_t size) { - size_t end = 1; - hoedown_buffer work = { 0, 0, 0, 0, NULL, NULL, NULL }; + size_t end = 1; + hoedown_buffer work = { 0, 0, 0, 0, NULL, NULL, NULL }; - if (end < size && data[end] == '#') - end++; + if (end < size && data[end] == '#') + end++; - while (end < size && isalnum(data[end])) - end++; + while (end < size && isalnum(data[end])) + end++; - if (end < size && data[end] == ';') - end++; /* real entity */ - else - return 0; /* lone '&' */ + if (end < size && data[end] == ';') + end++; /* real entity */ + else + return 0; /* lone '&' */ - if (doc->md.entity) { - work.data = data; - work.size = end; - doc->md.entity(ob, &work, &doc->data); - } - else hoedown_buffer_put(ob, data, end); + if (doc->md.entity) { + work.data = data; + work.size = end; + doc->md.entity(ob, &work, &doc->data); + } + else hoedown_buffer_put(ob, data, end); - return end; + return end; } /* char_langle_tag • '<' when tags or autolinks are allowed */ static size_t char_langle_tag(hoedown_buffer *ob, hoedown_document *doc, uint8_t *data, size_t offset, size_t size) { - hoedown_buffer work = { NULL, 0, 0, 0, NULL, NULL, NULL }; - hoedown_autolink_type altype = HOEDOWN_AUTOLINK_NONE; - size_t end = tag_length(data, size, &altype); - int ret = 0; + hoedown_buffer work = { NULL, 0, 0, 0, NULL, NULL, NULL }; + hoedown_autolink_type altype = HOEDOWN_AUTOLINK_NONE; + size_t end = tag_length(data, size, &altype); + int ret = 0; - work.data = data; - work.size = end; + work.data = data; + work.size = end; - if (end > 2) { - if (doc->md.autolink && altype != HOEDOWN_AUTOLINK_NONE) { - hoedown_buffer *u_link = newbuf(doc, BUFFER_SPAN); - work.data = data + 1; - work.size = end - 2; - unscape_text(u_link, &work); - ret = doc->md.autolink(ob, u_link, altype, &doc->data); - popbuf(doc, BUFFER_SPAN); - } - else if (doc->md.raw_html) - ret = doc->md.raw_html(ob, &work, &doc->data); - } + if (end > 2) { + if (doc->md.autolink && altype != HOEDOWN_AUTOLINK_NONE) { + hoedown_buffer *u_link = newbuf(doc, BUFFER_SPAN); + work.data = data + 1; + work.size = end - 2; + unscape_text(u_link, &work); + ret = doc->md.autolink(ob, u_link, altype, &doc->data); + popbuf(doc, BUFFER_SPAN); + } + else if (doc->md.raw_html) + ret = doc->md.raw_html(ob, &work, &doc->data); + } - if (!ret) return 0; - else return end; + if (!ret) return 0; + else return end; } static size_t char_autolink_www(hoedown_buffer *ob, hoedown_document *doc, uint8_t *data, size_t offset, size_t size) { - hoedown_buffer *link, *link_url, *link_text; - size_t link_len, rewind; + hoedown_buffer *link, *link_url, *link_text; + size_t link_len, rewind; - if (!doc->md.link || doc->in_link_body) - return 0; + if (!doc->md.link || doc->in_link_body) + return 0; - link = newbuf(doc, BUFFER_SPAN); + link = newbuf(doc, BUFFER_SPAN); - if ((link_len = hoedown_autolink__www(&rewind, link, data, offset, size, HOEDOWN_AUTOLINK_SHORT_DOMAINS)) > 0) { - link_url = newbuf(doc, BUFFER_SPAN); - HOEDOWN_BUFPUTSL(link_url, "http://"); - hoedown_buffer_put(link_url, link->data, link->size); + if ((link_len = hoedown_autolink__www(&rewind, link, data, offset, size, HOEDOWN_AUTOLINK_SHORT_DOMAINS)) > 0) { + link_url = newbuf(doc, BUFFER_SPAN); + HOEDOWN_BUFPUTSL(link_url, "http://"); + hoedown_buffer_put(link_url, link->data, link->size); - ob->size -= rewind; - if (doc->md.normal_text) { - link_text = newbuf(doc, BUFFER_SPAN); - doc->md.normal_text(link_text, link, &doc->data); - doc->md.link(ob, link_text, link_url, NULL, &doc->data); - popbuf(doc, BUFFER_SPAN); - } else { - doc->md.link(ob, link, link_url, NULL, &doc->data); - } - popbuf(doc, BUFFER_SPAN); - } + ob->size -= rewind; + if (doc->md.normal_text) { + link_text = newbuf(doc, BUFFER_SPAN); + doc->md.normal_text(link_text, link, &doc->data); + doc->md.link(ob, link_text, link_url, NULL, &doc->data); + popbuf(doc, BUFFER_SPAN); + } else { + doc->md.link(ob, link, link_url, NULL, &doc->data); + } + popbuf(doc, BUFFER_SPAN); + } - popbuf(doc, BUFFER_SPAN); - return link_len; + popbuf(doc, BUFFER_SPAN); + return link_len; } static size_t char_autolink_email(hoedown_buffer *ob, hoedown_document *doc, uint8_t *data, size_t offset, size_t size) { - hoedown_buffer *link; - size_t link_len, rewind; + hoedown_buffer *link; + size_t link_len, rewind; - if (!doc->md.autolink || doc->in_link_body) - return 0; + if (!doc->md.autolink || doc->in_link_body) + return 0; - link = newbuf(doc, BUFFER_SPAN); + link = newbuf(doc, BUFFER_SPAN); - if ((link_len = hoedown_autolink__email(&rewind, link, data, offset, size, 0)) > 0) { - ob->size -= rewind; - doc->md.autolink(ob, link, HOEDOWN_AUTOLINK_EMAIL, &doc->data); - } + if ((link_len = hoedown_autolink__email(&rewind, link, data, offset, size, 0)) > 0) { + ob->size -= rewind; + doc->md.autolink(ob, link, HOEDOWN_AUTOLINK_EMAIL, &doc->data); + } - popbuf(doc, BUFFER_SPAN); - return link_len; + popbuf(doc, BUFFER_SPAN); + return link_len; } static size_t char_autolink_url(hoedown_buffer *ob, hoedown_document *doc, uint8_t *data, size_t offset, size_t size) { - hoedown_buffer *link; - size_t link_len, rewind; + hoedown_buffer *link; + size_t link_len, rewind; - if (!doc->md.autolink || doc->in_link_body) - return 0; + if (!doc->md.autolink || doc->in_link_body) + return 0; - link = newbuf(doc, BUFFER_SPAN); + link = newbuf(doc, BUFFER_SPAN); - if ((link_len = hoedown_autolink__url(&rewind, link, data, offset, size, 0)) > 0) { - ob->size -= rewind; - doc->md.autolink(ob, link, HOEDOWN_AUTOLINK_NORMAL, &doc->data); - } + if ((link_len = hoedown_autolink__url(&rewind, link, data, offset, size, 0)) > 0) { + ob->size -= rewind; + doc->md.autolink(ob, link, HOEDOWN_AUTOLINK_NORMAL, &doc->data); + } - popbuf(doc, BUFFER_SPAN); - return link_len; + popbuf(doc, BUFFER_SPAN); + return link_len; } /* char_link • '[': parsing a link, a footnote or an image */ static size_t char_link(hoedown_buffer *ob, hoedown_document *doc, uint8_t *data, size_t offset, size_t size) { - int is_img = (offset && data[-1] == '!' && !is_escaped(data - offset, offset - 1)); - int is_footnote = (doc->ext_flags & HOEDOWN_EXT_FOOTNOTES && data[1] == '^'); - size_t i = 1, txt_e, link_b = 0, link_e = 0, title_b = 0, title_e = 0; - hoedown_buffer *content = NULL; - hoedown_buffer *link = NULL; - hoedown_buffer *title = NULL; - hoedown_buffer *u_link = NULL; - size_t org_work_size = doc->work_bufs[BUFFER_SPAN].size; - int ret = 0, in_title = 0, qtype = 0; - - /* checking whether the correct renderer exists */ - if ((is_footnote && !doc->md.footnote_ref) || (is_img && !doc->md.image) - || (!is_img && !is_footnote && !doc->md.link)) - goto cleanup; - - /* looking for the matching closing bracket */ - i += find_emph_char(data + i, size - i, ']'); - txt_e = i; - - if (i < size && data[i] == ']') i++; - else goto cleanup; - - /* footnote link */ - if (is_footnote) { - hoedown_buffer id = { NULL, 0, 0, 0, NULL, NULL, NULL }; - struct footnote_ref *fr; - - if (txt_e < 3) - goto cleanup; - - id.data = data + 2; - id.size = txt_e - 2; - - fr = find_footnote_ref(&doc->footnotes_found, id.data, id.size); - - /* mark footnote used */ - if (fr && !fr->is_used) { - if(!add_footnote_ref(&doc->footnotes_used, fr)) - goto cleanup; - fr->is_used = 1; - fr->num = doc->footnotes_used.count; - - /* render */ - if (doc->md.footnote_ref) - ret = doc->md.footnote_ref(ob, fr->num, &doc->data); - } - - goto cleanup; - } - - /* skip any amount of spacing */ - /* (this is much more laxist than original markdown syntax) */ - while (i < size && _isspace(data[i])) - i++; - - /* inline style link */ - if (i < size && data[i] == '(') { - size_t nb_p; - - /* skipping initial spacing */ - i++; - - while (i < size && _isspace(data[i])) - i++; - - link_b = i; - - /* looking for link end: ' " ) */ - /* Count the number of open parenthesis */ - nb_p = 0; - - while (i < size) { - if (data[i] == '\\') i += 2; - else if (data[i] == '(' && i != 0) { - nb_p++; i++; - } - else if (data[i] == ')') { - if (nb_p == 0) break; - else nb_p--; i++; - } else if (i >= 1 && _isspace(data[i-1]) && (data[i] == '\'' || data[i] == '"')) break; - else i++; - } - - if (i >= size) goto cleanup; - link_e = i; - - /* looking for title end if present */ - if (data[i] == '\'' || data[i] == '"') { - qtype = data[i]; - in_title = 1; - i++; - title_b = i; - - while (i < size) { - if (data[i] == '\\') i += 2; - else if (data[i] == qtype) {in_title = 0; i++;} - else if ((data[i] == ')') && !in_title) break; - else i++; - } - - if (i >= size) goto cleanup; - - /* skipping spacing after title */ - title_e = i - 1; - while (title_e > title_b && _isspace(data[title_e])) - title_e--; - - /* checking for closing quote presence */ - if (data[title_e] != '\'' && data[title_e] != '"') { - title_b = title_e = 0; - link_e = i; - } - } - - /* remove spacing at the end of the link */ - while (link_e > link_b && _isspace(data[link_e - 1])) - link_e--; - - /* remove optional angle brackets around the link */ - if (data[link_b] == '<') link_b++; - if (data[link_e - 1] == '>') link_e--; - - /* building escaped link and title */ - if (link_e > link_b) { - link = newbuf(doc, BUFFER_SPAN); - hoedown_buffer_put(link, data + link_b, link_e - link_b); - } - - if (title_e > title_b) { - title = newbuf(doc, BUFFER_SPAN); - hoedown_buffer_put(title, data + title_b, title_e - title_b); - } - - i++; - } - - /* reference style link */ - else if (i < size && data[i] == '[') { - hoedown_buffer *id = newbuf(doc, BUFFER_SPAN); - struct link_ref *lr; - - /* looking for the id */ - i++; - link_b = i; - while (i < size && data[i] != ']') i++; - if (i >= size) goto cleanup; - link_e = i; - - /* finding the link_ref */ - if (link_b == link_e) - replace_spacing(id, data + 1, txt_e - 1); - else - hoedown_buffer_put(id, data + link_b, link_e - link_b); - - lr = find_link_ref(doc->refs, id->data, id->size); - if (!lr) - goto cleanup; - - /* keeping link and title from link_ref */ - link = lr->link; - title = lr->title; - i++; - } - - /* shortcut reference style link */ - else { - hoedown_buffer *id = newbuf(doc, BUFFER_SPAN); - struct link_ref *lr; - - /* crafting the id */ - replace_spacing(id, data + 1, txt_e - 1); - - /* finding the link_ref */ - lr = find_link_ref(doc->refs, id->data, id->size); - if (!lr) - goto cleanup; - - /* keeping link and title from link_ref */ - link = lr->link; - title = lr->title; - - /* rewinding the spacing */ - i = txt_e + 1; - } - - /* building content: img alt is kept, only link content is parsed */ - if (txt_e > 1) { - content = newbuf(doc, BUFFER_SPAN); - if (is_img) { - hoedown_buffer_put(content, data + 1, txt_e - 1); - } else { - /* disable autolinking when parsing inline the - * content of a link */ - doc->in_link_body = 1; - parse_inline(content, doc, data + 1, txt_e - 1); - doc->in_link_body = 0; - } - } - - if (link) { - u_link = newbuf(doc, BUFFER_SPAN); - unscape_text(u_link, link); - } - - /* calling the relevant rendering function */ - if (is_img) { - if (ob->size && ob->data[ob->size - 1] == '!') - ob->size -= 1; - - ret = doc->md.image(ob, u_link, title, content, &doc->data); - } else { - ret = doc->md.link(ob, content, u_link, title, &doc->data); - } - - /* cleanup */ + int is_img = (offset && data[-1] == '!' && !is_escaped(data - offset, offset - 1)); + int is_footnote = (doc->ext_flags & HOEDOWN_EXT_FOOTNOTES && data[1] == '^'); + size_t i = 1, txt_e, link_b = 0, link_e = 0, title_b = 0, title_e = 0; + hoedown_buffer *content = NULL; + hoedown_buffer *link = NULL; + hoedown_buffer *title = NULL; + hoedown_buffer *u_link = NULL; + size_t org_work_size = doc->work_bufs[BUFFER_SPAN].size; + int ret = 0, in_title = 0, qtype = 0; + + /* checking whether the correct renderer exists */ + if ((is_footnote && !doc->md.footnote_ref) || (is_img && !doc->md.image) + || (!is_img && !is_footnote && !doc->md.link)) + goto cleanup; + + /* looking for the matching closing bracket */ + i += find_emph_char(data + i, size - i, ']'); + txt_e = i; + + if (i < size && data[i] == ']') i++; + else goto cleanup; + + /* footnote link */ + if (is_footnote) { + hoedown_buffer id = { NULL, 0, 0, 0, NULL, NULL, NULL }; + struct footnote_ref *fr; + + if (txt_e < 3) + goto cleanup; + + id.data = data + 2; + id.size = txt_e - 2; + + fr = find_footnote_ref(&doc->footnotes_found, id.data, id.size); + + /* mark footnote used */ + if (fr && !fr->is_used) { + if(!add_footnote_ref(&doc->footnotes_used, fr)) + goto cleanup; + fr->is_used = 1; + fr->num = doc->footnotes_used.count; + + /* render */ + if (doc->md.footnote_ref) + ret = doc->md.footnote_ref(ob, fr->num, &doc->data); + } + + goto cleanup; + } + + /* skip any amount of spacing */ + /* (this is much more laxist than original markdown syntax) */ + while (i < size && _isspace(data[i])) + i++; + + /* inline style link */ + if (i < size && data[i] == '(') { + size_t nb_p; + + /* skipping initial spacing */ + i++; + + while (i < size && _isspace(data[i])) + i++; + + link_b = i; + + /* looking for link end: ' " ) */ + /* Count the number of open parenthesis */ + nb_p = 0; + + while (i < size) { + if (data[i] == '\\') i += 2; + else if (data[i] == '(' && i != 0) { + nb_p++; i++; + } + else if (data[i] == ')') { + if (nb_p == 0) break; + else nb_p--; i++; + } else if (i >= 1 && _isspace(data[i-1]) && (data[i] == '\'' || data[i] == '"')) break; + else i++; + } + + if (i >= size) goto cleanup; + link_e = i; + + /* looking for title end if present */ + if (data[i] == '\'' || data[i] == '"') { + qtype = data[i]; + in_title = 1; + i++; + title_b = i; + + while (i < size) { + if (data[i] == '\\') i += 2; + else if (data[i] == qtype) {in_title = 0; i++;} + else if ((data[i] == ')') && !in_title) break; + else i++; + } + + if (i >= size) goto cleanup; + + /* skipping spacing after title */ + title_e = i - 1; + while (title_e > title_b && _isspace(data[title_e])) + title_e--; + + /* checking for closing quote presence */ + if (data[title_e] != '\'' && data[title_e] != '"') { + title_b = title_e = 0; + link_e = i; + } + } + + /* remove spacing at the end of the link */ + while (link_e > link_b && _isspace(data[link_e - 1])) + link_e--; + + /* remove optional angle brackets around the link */ + if (data[link_b] == '<') link_b++; + if (data[link_e - 1] == '>') link_e--; + + /* building escaped link and title */ + if (link_e > link_b) { + link = newbuf(doc, BUFFER_SPAN); + hoedown_buffer_put(link, data + link_b, link_e - link_b); + } + + if (title_e > title_b) { + title = newbuf(doc, BUFFER_SPAN); + hoedown_buffer_put(title, data + title_b, title_e - title_b); + } + + i++; + } + + /* reference style link */ + else if (i < size && data[i] == '[') { + hoedown_buffer *id = newbuf(doc, BUFFER_SPAN); + struct link_ref *lr; + + /* looking for the id */ + i++; + link_b = i; + while (i < size && data[i] != ']') i++; + if (i >= size) goto cleanup; + link_e = i; + + /* finding the link_ref */ + if (link_b == link_e) + replace_spacing(id, data + 1, txt_e - 1); + else + hoedown_buffer_put(id, data + link_b, link_e - link_b); + + lr = find_link_ref(doc->refs, id->data, id->size); + if (!lr) + goto cleanup; + + /* keeping link and title from link_ref */ + link = lr->link; + title = lr->title; + i++; + } + + /* shortcut reference style link */ + else { + hoedown_buffer *id = newbuf(doc, BUFFER_SPAN); + struct link_ref *lr; + + /* crafting the id */ + replace_spacing(id, data + 1, txt_e - 1); + + /* finding the link_ref */ + lr = find_link_ref(doc->refs, id->data, id->size); + if (!lr) + goto cleanup; + + /* keeping link and title from link_ref */ + link = lr->link; + title = lr->title; + + /* rewinding the spacing */ + i = txt_e + 1; + } + + /* building content: img alt is kept, only link content is parsed */ + if (txt_e > 1) { + content = newbuf(doc, BUFFER_SPAN); + if (is_img) { + hoedown_buffer_put(content, data + 1, txt_e - 1); + } else { + /* disable autolinking when parsing inline the + * content of a link */ + doc->in_link_body = 1; + parse_inline(content, doc, data + 1, txt_e - 1); + doc->in_link_body = 0; + } + } + + if (link) { + u_link = newbuf(doc, BUFFER_SPAN); + unscape_text(u_link, link); + } + + /* calling the relevant rendering function */ + if (is_img) { + if (ob->size && ob->data[ob->size - 1] == '!') + ob->size -= 1; + + ret = doc->md.image(ob, u_link, title, content, &doc->data); + } else { + ret = doc->md.link(ob, content, u_link, title, &doc->data); + } + + /* cleanup */ cleanup: - doc->work_bufs[BUFFER_SPAN].size = (int)org_work_size; - return ret ? i : 0; + doc->work_bufs[BUFFER_SPAN].size = (int)org_work_size; + return ret ? i : 0; } static size_t char_superscript(hoedown_buffer *ob, hoedown_document *doc, uint8_t *data, size_t offset, size_t size) { - size_t sup_start, sup_len; - hoedown_buffer *sup; + size_t sup_start, sup_len; + hoedown_buffer *sup; - if (!doc->md.superscript) - return 0; + if (!doc->md.superscript) + return 0; - if (size < 2) - return 0; + if (size < 2) + return 0; - if (data[1] == '(') { - sup_start = 2; - sup_len = find_emph_char(data + 2, size - 2, ')') + 2; + if (data[1] == '(') { + sup_start = 2; + sup_len = find_emph_char(data + 2, size - 2, ')') + 2; - if (sup_len == size) - return 0; - } else { - sup_start = sup_len = 1; + if (sup_len == size) + return 0; + } else { + sup_start = sup_len = 1; - while (sup_len < size && !_isspace(data[sup_len])) - sup_len++; - } + while (sup_len < size && !_isspace(data[sup_len])) + sup_len++; + } - if (sup_len - sup_start == 0) - return (sup_start == 2) ? 3 : 0; + if (sup_len - sup_start == 0) + return (sup_start == 2) ? 3 : 0; - sup = newbuf(doc, BUFFER_SPAN); - parse_inline(sup, doc, data + sup_start, sup_len - sup_start); - doc->md.superscript(ob, sup, &doc->data); - popbuf(doc, BUFFER_SPAN); + sup = newbuf(doc, BUFFER_SPAN); + parse_inline(sup, doc, data + sup_start, sup_len - sup_start); + doc->md.superscript(ob, sup, &doc->data); + popbuf(doc, BUFFER_SPAN); - return (sup_start == 2) ? sup_len + 1 : sup_len; + return (sup_start == 2) ? sup_len + 1 : sup_len; } static size_t char_math(hoedown_buffer *ob, hoedown_document *doc, uint8_t *data, size_t offset, size_t size) { - /* double dollar */ - if (size > 1 && data[1] == '$') - return parse_math(ob, doc, data, offset, size, "$$", 2, 1); + /* double dollar */ + if (size > 1 && data[1] == '$') + return parse_math(ob, doc, data, offset, size, "$$", 2, 1); - /* single dollar allowed only with MATH_EXPLICIT flag */ - if (doc->ext_flags & HOEDOWN_EXT_MATH_EXPLICIT) - return parse_math(ob, doc, data, offset, size, "$", 1, 0); + /* single dollar allowed only with MATH_EXPLICIT flag */ + if (doc->ext_flags & HOEDOWN_EXT_MATH_EXPLICIT) + return parse_math(ob, doc, data, offset, size, "$", 1, 0); - return 0; + return 0; } /********************************* @@ -1346,44 +1346,44 @@ char_math(hoedown_buffer *ob, hoedown_document *doc, uint8_t *data, size_t offse static size_t is_empty(const uint8_t *data, size_t size) { - size_t i; + size_t i; - for (i = 0; i < size && data[i] != '\n'; i++) - if (data[i] != ' ') - return 0; + for (i = 0; i < size && data[i] != '\n'; i++) + if (data[i] != ' ') + return 0; - return i + 1; + return i + 1; } /* is_hrule • returns whether a line is a horizontal rule */ static int is_hrule(uint8_t *data, size_t size) { - size_t i = 0, n = 0; - uint8_t c; + size_t i = 0, n = 0; + uint8_t c; - /* skipping initial spaces */ - if (size < 3) return 0; - if (data[0] == ' ') { i++; - if (data[1] == ' ') { i++; - if (data[2] == ' ') { i++; } } } + /* skipping initial spaces */ + if (size < 3) return 0; + if (data[0] == ' ') { i++; + if (data[1] == ' ') { i++; + if (data[2] == ' ') { i++; } } } - /* looking at the hrule uint8_t */ - if (i + 2 >= size - || (data[i] != '*' && data[i] != '-' && data[i] != '_')) - return 0; - c = data[i]; + /* looking at the hrule uint8_t */ + if (i + 2 >= size + || (data[i] != '*' && data[i] != '-' && data[i] != '_')) + return 0; + c = data[i]; - /* the whole line must be the char or space */ - while (i < size && data[i] != '\n') { - if (data[i] == c) n++; - else if (data[i] != ' ') - return 0; + /* the whole line must be the char or space */ + while (i < size && data[i] != '\n') { + if (data[i] == c) n++; + else if (data[i] != ' ') + return 0; - i++; - } + i++; + } - return n >= 3; + return n >= 3; } /* check if a line is a code fence; return the @@ -1392,239 +1392,239 @@ is_hrule(uint8_t *data, size_t size) static size_t is_codefence(uint8_t *data, size_t size, size_t *width, uint8_t *chr) { - size_t i = 0, n = 1; - uint8_t c; + size_t i = 0, n = 1; + uint8_t c; - /* skipping initial spaces */ - if (size < 3) - return 0; + /* skipping initial spaces */ + if (size < 3) + return 0; - if (data[0] == ' ') { i++; - if (data[1] == ' ') { i++; - if (data[2] == ' ') { i++; } } } + if (data[0] == ' ') { i++; + if (data[1] == ' ') { i++; + if (data[2] == ' ') { i++; } } } - /* looking at the hrule uint8_t */ - c = data[i]; - if (i + 2 >= size || !(c=='~' || c=='`')) - return 0; + /* looking at the hrule uint8_t */ + c = data[i]; + if (i + 2 >= size || !(c=='~' || c=='`')) + return 0; - /* the fence must be that same character */ - while (++i < size && data[i] == c) - ++n; + /* the fence must be that same character */ + while (++i < size && data[i] == c) + ++n; - if (n < 3) - return 0; + if (n < 3) + return 0; - if (width) *width = n; - if (chr) *chr = c; - return i; + if (width) *width = n; + if (chr) *chr = c; + return i; } /* expects single line, checks if it's a codefence and extracts language */ static size_t parse_codefence(uint8_t *data, size_t size, hoedown_buffer *lang, size_t *width, uint8_t *chr) { - size_t i, w, lang_start; + size_t i, w, lang_start; - i = w = is_codefence(data, size, width, chr); - if (i == 0) - return 0; + i = w = is_codefence(data, size, width, chr); + if (i == 0) + return 0; - while (i < size && _isspace(data[i])) - i++; + while (i < size && _isspace(data[i])) + i++; - lang_start = i; + lang_start = i; - while (i < size && !_isspace(data[i])) - i++; + while (i < size && !_isspace(data[i])) + i++; - lang->data = data + lang_start; - lang->size = i - lang_start; + lang->data = data + lang_start; + lang->size = i - lang_start; - /* Avoid parsing a codespan as a fence */ - i = lang_start + 2; - while (i < size && !(data[i] == *chr && data[i-1] == *chr && data[i-2] == *chr)) i++; - if (i < size) return 0; + /* Avoid parsing a codespan as a fence */ + i = lang_start + 2; + while (i < size && !(data[i] == *chr && data[i-1] == *chr && data[i-2] == *chr)) i++; + if (i < size) return 0; - return w; + return w; } /* is_atxheader • returns whether the line is a hash-prefixed header */ static int is_atxheader(hoedown_document *doc, uint8_t *data, size_t size) { - if (data[0] != '#') - return 0; + if (data[0] != '#') + return 0; - if (doc->ext_flags & HOEDOWN_EXT_SPACE_HEADERS) { - size_t level = 0; + if (doc->ext_flags & HOEDOWN_EXT_SPACE_HEADERS) { + size_t level = 0; - while (level < size && level < 6 && data[level] == '#') - level++; + while (level < size && level < 6 && data[level] == '#') + level++; - if (level < size && data[level] != ' ') - return 0; - } + if (level < size && data[level] != ' ') + return 0; + } - return 1; + return 1; } /* is_headerline • returns whether the line is a setext-style hdr underline */ static int is_headerline(uint8_t *data, size_t size) { - size_t i = 0; + size_t i = 0; - /* test of level 1 header */ - if (data[i] == '=') { - for (i = 1; i < size && data[i] == '='; i++); - while (i < size && data[i] == ' ') i++; - return (i >= size || data[i] == '\n') ? 1 : 0; } + /* test of level 1 header */ + if (data[i] == '=') { + for (i = 1; i < size && data[i] == '='; i++); + while (i < size && data[i] == ' ') i++; + return (i >= size || data[i] == '\n') ? 1 : 0; } - /* test of level 2 header */ - if (data[i] == '-') { - for (i = 1; i < size && data[i] == '-'; i++); - while (i < size && data[i] == ' ') i++; - return (i >= size || data[i] == '\n') ? 2 : 0; } + /* test of level 2 header */ + if (data[i] == '-') { + for (i = 1; i < size && data[i] == '-'; i++); + while (i < size && data[i] == ' ') i++; + return (i >= size || data[i] == '\n') ? 2 : 0; } - return 0; + return 0; } static int is_next_headerline(uint8_t *data, size_t size) { - size_t i = 0; + size_t i = 0; - while (i < size && data[i] != '\n') - i++; + while (i < size && data[i] != '\n') + i++; - if (++i >= size) - return 0; + if (++i >= size) + return 0; - return is_headerline(data + i, size - i); + return is_headerline(data + i, size - i); } /* prefix_quote • returns blockquote prefix length */ static size_t prefix_quote(uint8_t *data, size_t size) { - size_t i = 0; - if (i < size && data[i] == ' ') i++; - if (i < size && data[i] == ' ') i++; - if (i < size && data[i] == ' ') i++; + size_t i = 0; + if (i < size && data[i] == ' ') i++; + if (i < size && data[i] == ' ') i++; + if (i < size && data[i] == ' ') i++; - if (i < size && data[i] == '>') { - if (i + 1 < size && data[i + 1] == ' ') - return i + 2; + if (i < size && data[i] == '>') { + if (i + 1 < size && data[i + 1] == ' ') + return i + 2; - return i + 1; - } + return i + 1; + } - return 0; + return 0; } /* prefix_code • returns prefix length for block code*/ static size_t prefix_code(uint8_t *data, size_t size) { - if (size > 3 && data[0] == ' ' && data[1] == ' ' - && data[2] == ' ' && data[3] == ' ') return 4; + if (size > 3 && data[0] == ' ' && data[1] == ' ' + && data[2] == ' ' && data[3] == ' ') return 4; - return 0; + return 0; } /* prefix_oli • returns ordered list item prefix */ static size_t prefix_oli(uint8_t *data, size_t size) { - size_t i = 0; + size_t i = 0; - if (i < size && data[i] == ' ') i++; - if (i < size && data[i] == ' ') i++; - if (i < size && data[i] == ' ') i++; + if (i < size && data[i] == ' ') i++; + if (i < size && data[i] == ' ') i++; + if (i < size && data[i] == ' ') i++; - if (i >= size || data[i] < '0' || data[i] > '9') - return 0; + if (i >= size || data[i] < '0' || data[i] > '9') + return 0; - while (i < size && data[i] >= '0' && data[i] <= '9') - i++; + while (i < size && data[i] >= '0' && data[i] <= '9') + i++; - if (i + 1 >= size || data[i] != '.' || data[i + 1] != ' ') - return 0; + if (i + 1 >= size || data[i] != '.' || data[i + 1] != ' ') + return 0; - if (is_next_headerline(data + i, size - i)) - return 0; + if (is_next_headerline(data + i, size - i)) + return 0; - return i + 2; + return i + 2; } /* prefix_uli • returns ordered list item prefix */ static size_t prefix_uli(uint8_t *data, size_t size) { - size_t i = 0; + size_t i = 0; - if (i < size && data[i] == ' ') i++; - if (i < size && data[i] == ' ') i++; - if (i < size && data[i] == ' ') i++; + if (i < size && data[i] == ' ') i++; + if (i < size && data[i] == ' ') i++; + if (i < size && data[i] == ' ') i++; - if (i + 1 >= size || - (data[i] != '*' && data[i] != '+' && data[i] != '-') || - data[i + 1] != ' ') - return 0; + if (i + 1 >= size || + (data[i] != '*' && data[i] != '+' && data[i] != '-') || + data[i + 1] != ' ') + return 0; - if (is_next_headerline(data + i, size - i)) - return 0; + if (is_next_headerline(data + i, size - i)) + return 0; - return i + 2; + return i + 2; } /* parse_block • parsing of one block, returning next uint8_t to parse */ static void parse_block(hoedown_buffer *ob, hoedown_document *doc, - uint8_t *data, size_t size); + uint8_t *data, size_t size); /* parse_blockquote • handles parsing of a blockquote fragment */ static size_t parse_blockquote(hoedown_buffer *ob, hoedown_document *doc, uint8_t *data, size_t size) { - size_t beg, end = 0, pre, work_size = 0; - uint8_t *work_data = 0; - hoedown_buffer *out = 0; + size_t beg, end = 0, pre, work_size = 0; + uint8_t *work_data = 0; + hoedown_buffer *out = 0; - out = newbuf(doc, BUFFER_BLOCK); - beg = 0; - while (beg < size) { - for (end = beg + 1; end < size && data[end - 1] != '\n'; end++); + out = newbuf(doc, BUFFER_BLOCK); + beg = 0; + while (beg < size) { + for (end = beg + 1; end < size && data[end - 1] != '\n'; end++); - pre = prefix_quote(data + beg, end - beg); + pre = prefix_quote(data + beg, end - beg); - if (pre) - beg += pre; /* skipping prefix */ + if (pre) + beg += pre; /* skipping prefix */ - /* empty line followed by non-quote line */ - else if (is_empty(data + beg, end - beg) && - (end >= size || (prefix_quote(data + end, size - end) == 0 && - !is_empty(data + end, size - end)))) - break; + /* empty line followed by non-quote line */ + else if (is_empty(data + beg, end - beg) && + (end >= size || (prefix_quote(data + end, size - end) == 0 && + !is_empty(data + end, size - end)))) + break; - if (beg < end) { /* copy into the in-place working buffer */ - /* hoedown_buffer_put(work, data + beg, end - beg); */ - if (!work_data) - work_data = data + beg; - else if (data + beg != work_data + work_size) - memmove(work_data + work_size, data + beg, end - beg); - work_size += end - beg; - } - beg = end; - } + if (beg < end) { /* copy into the in-place working buffer */ + /* hoedown_buffer_put(work, data + beg, end - beg); */ + if (!work_data) + work_data = data + beg; + else if (data + beg != work_data + work_size) + memmove(work_data + work_size, data + beg, end - beg); + work_size += end - beg; + } + beg = end; + } - parse_block(out, doc, work_data, work_size); - if (doc->md.blockquote) - doc->md.blockquote(ob, out, &doc->data); - popbuf(doc, BUFFER_BLOCK); - return end; + parse_block(out, doc, work_data, work_size); + if (doc->md.blockquote) + doc->md.blockquote(ob, out, &doc->data); + popbuf(doc, BUFFER_BLOCK); + return end; } static size_t @@ -1634,301 +1634,301 @@ parse_htmlblock(hoedown_buffer *ob, hoedown_document *doc, uint8_t *data, size_t static size_t parse_paragraph(hoedown_buffer *ob, hoedown_document *doc, uint8_t *data, size_t size) { - hoedown_buffer work = { NULL, 0, 0, 0, NULL, NULL, NULL }; - size_t i = 0, end = 0; - int level = 0; + hoedown_buffer work = { NULL, 0, 0, 0, NULL, NULL, NULL }; + size_t i = 0, end = 0; + int level = 0; - work.data = data; + work.data = data; - while (i < size) { - for (end = i + 1; end < size && data[end - 1] != '\n'; end++) /* empty */; + while (i < size) { + for (end = i + 1; end < size && data[end - 1] != '\n'; end++) /* empty */; - if (is_empty(data + i, size - i)) - break; + if (is_empty(data + i, size - i)) + break; - if ((level = is_headerline(data + i, size - i)) != 0) - break; + if ((level = is_headerline(data + i, size - i)) != 0) + break; - if (is_atxheader(doc, data + i, size - i) || - is_hrule(data + i, size - i) || - prefix_quote(data + i, size - i)) { - end = i; - break; - } + if (is_atxheader(doc, data + i, size - i) || + is_hrule(data + i, size - i) || + prefix_quote(data + i, size - i)) { + end = i; + break; + } - i = end; - } + i = end; + } - work.size = i; - while (work.size && data[work.size - 1] == '\n') - work.size--; + work.size = i; + while (work.size && data[work.size - 1] == '\n') + work.size--; - if (!level) { - hoedown_buffer *tmp = newbuf(doc, BUFFER_BLOCK); - parse_inline(tmp, doc, work.data, work.size); - if (doc->md.paragraph) - doc->md.paragraph(ob, tmp, &doc->data); - popbuf(doc, BUFFER_BLOCK); - } else { - hoedown_buffer *header_work; + if (!level) { + hoedown_buffer *tmp = newbuf(doc, BUFFER_BLOCK); + parse_inline(tmp, doc, work.data, work.size); + if (doc->md.paragraph) + doc->md.paragraph(ob, tmp, &doc->data); + popbuf(doc, BUFFER_BLOCK); + } else { + hoedown_buffer *header_work; - if (work.size) { - size_t beg; - i = work.size; - work.size -= 1; + if (work.size) { + size_t beg; + i = work.size; + work.size -= 1; - while (work.size && data[work.size] != '\n') - work.size -= 1; + while (work.size && data[work.size] != '\n') + work.size -= 1; - beg = work.size + 1; - while (work.size && data[work.size - 1] == '\n') - work.size -= 1; + beg = work.size + 1; + while (work.size && data[work.size - 1] == '\n') + work.size -= 1; - if (work.size > 0) { - hoedown_buffer *tmp = newbuf(doc, BUFFER_BLOCK); - parse_inline(tmp, doc, work.data, work.size); + if (work.size > 0) { + hoedown_buffer *tmp = newbuf(doc, BUFFER_BLOCK); + parse_inline(tmp, doc, work.data, work.size); - if (doc->md.paragraph) - doc->md.paragraph(ob, tmp, &doc->data); + if (doc->md.paragraph) + doc->md.paragraph(ob, tmp, &doc->data); - popbuf(doc, BUFFER_BLOCK); - work.data += beg; - work.size = i - beg; - } - else work.size = i; - } + popbuf(doc, BUFFER_BLOCK); + work.data += beg; + work.size = i - beg; + } + else work.size = i; + } - header_work = newbuf(doc, BUFFER_SPAN); - parse_inline(header_work, doc, work.data, work.size); + header_work = newbuf(doc, BUFFER_SPAN); + parse_inline(header_work, doc, work.data, work.size); - if (doc->md.header) - doc->md.header(ob, header_work, (int)level, &doc->data); + if (doc->md.header) + doc->md.header(ob, header_work, (int)level, &doc->data); - popbuf(doc, BUFFER_SPAN); - } + popbuf(doc, BUFFER_SPAN); + } - return end; + return end; } /* parse_fencedcode • handles parsing of a block-level code fragment */ static size_t parse_fencedcode(hoedown_buffer *ob, hoedown_document *doc, uint8_t *data, size_t size) { - hoedown_buffer text = { 0, 0, 0, 0, NULL, NULL, NULL }; - hoedown_buffer lang = { 0, 0, 0, 0, NULL, NULL, NULL }; - size_t i = 0, text_start, line_start; - size_t w, w2; - size_t width, width2; - uint8_t chr, chr2; + hoedown_buffer text = { 0, 0, 0, 0, NULL, NULL, NULL }; + hoedown_buffer lang = { 0, 0, 0, 0, NULL, NULL, NULL }; + size_t i = 0, text_start, line_start; + size_t w, w2; + size_t width, width2; + uint8_t chr, chr2; - /* parse codefence line */ - while (i < size && data[i] != '\n') - i++; + /* parse codefence line */ + while (i < size && data[i] != '\n') + i++; - w = parse_codefence(data, i, &lang, &width, &chr); - if (!w) - return 0; + w = parse_codefence(data, i, &lang, &width, &chr); + if (!w) + return 0; - /* search for end */ - i++; - text_start = i; - while ((line_start = i) < size) { - while (i < size && data[i] != '\n') - i++; + /* search for end */ + i++; + text_start = i; + while ((line_start = i) < size) { + while (i < size && data[i] != '\n') + i++; - w2 = is_codefence(data + line_start, i - line_start, &width2, &chr2); - if (w == w2 && width == width2 && chr == chr2 && - is_empty(data + (line_start+w), i - (line_start+w))) - break; + w2 = is_codefence(data + line_start, i - line_start, &width2, &chr2); + if (w == w2 && width == width2 && chr == chr2 && + is_empty(data + (line_start+w), i - (line_start+w))) + break; - i++; - } + i++; + } - text.data = data + text_start; - text.size = line_start - text_start; + text.data = data + text_start; + text.size = line_start - text_start; - if (doc->md.blockcode) - doc->md.blockcode(ob, text.size ? &text : NULL, lang.size ? &lang : NULL, &doc->data); + if (doc->md.blockcode) + doc->md.blockcode(ob, text.size ? &text : NULL, lang.size ? &lang : NULL, &doc->data); - return i; + return i; } static size_t parse_blockcode(hoedown_buffer *ob, hoedown_document *doc, uint8_t *data, size_t size) { - size_t beg, end, pre; - hoedown_buffer *work = 0; + size_t beg, end, pre; + hoedown_buffer *work = 0; - work = newbuf(doc, BUFFER_BLOCK); + work = newbuf(doc, BUFFER_BLOCK); - beg = 0; - while (beg < size) { - for (end = beg + 1; end < size && data[end - 1] != '\n'; end++) {}; - pre = prefix_code(data + beg, end - beg); + beg = 0; + while (beg < size) { + for (end = beg + 1; end < size && data[end - 1] != '\n'; end++) {}; + pre = prefix_code(data + beg, end - beg); - if (pre) - beg += pre; /* skipping prefix */ - else if (!is_empty(data + beg, end - beg)) - /* non-empty non-prefixed line breaks the pre */ - break; + if (pre) + beg += pre; /* skipping prefix */ + else if (!is_empty(data + beg, end - beg)) + /* non-empty non-prefixed line breaks the pre */ + break; - if (beg < end) { - /* verbatim copy to the working buffer, - escaping entities */ - if (is_empty(data + beg, end - beg)) - hoedown_buffer_putc(work, '\n'); - else hoedown_buffer_put(work, data + beg, end - beg); - } - beg = end; - } + if (beg < end) { + /* verbatim copy to the working buffer, + escaping entities */ + if (is_empty(data + beg, end - beg)) + hoedown_buffer_putc(work, '\n'); + else hoedown_buffer_put(work, data + beg, end - beg); + } + beg = end; + } - while (work->size && work->data[work->size - 1] == '\n') - work->size -= 1; + while (work->size && work->data[work->size - 1] == '\n') + work->size -= 1; - hoedown_buffer_putc(work, '\n'); + hoedown_buffer_putc(work, '\n'); - if (doc->md.blockcode) - doc->md.blockcode(ob, work, NULL, &doc->data); + if (doc->md.blockcode) + doc->md.blockcode(ob, work, NULL, &doc->data); - popbuf(doc, BUFFER_BLOCK); - return beg; + popbuf(doc, BUFFER_BLOCK); + return beg; } /* parse_listitem • parsing of a single list item */ -/* assuming initial prefix is already removed */ +/* assuming initial prefix is already removed */ static size_t parse_listitem(hoedown_buffer *ob, hoedown_document *doc, uint8_t *data, size_t size, hoedown_list_flags *flags) { - hoedown_buffer *work = 0, *inter = 0; - size_t beg = 0, end, pre, sublist = 0, orgpre = 0, i; - int in_empty = 0, has_inside_empty = 0, in_fence = 0; - - /* keeping track of the first indentation prefix */ - while (orgpre < 3 && orgpre < size && data[orgpre] == ' ') - orgpre++; - - beg = prefix_uli(data, size); - if (!beg) - beg = prefix_oli(data, size); - - if (!beg) - return 0; - - /* skipping to the beginning of the following line */ - end = beg; - while (end < size && data[end - 1] != '\n') - end++; - - /* getting working buffers */ - work = newbuf(doc, BUFFER_SPAN); - inter = newbuf(doc, BUFFER_SPAN); - - /* putting the first line into the working buffer */ - hoedown_buffer_put(work, data + beg, end - beg); - beg = end; - - /* process the following lines */ - while (beg < size) { - size_t has_next_uli = 0, has_next_oli = 0; - - end++; - - while (end < size && data[end - 1] != '\n') - end++; - - /* process an empty line */ - if (is_empty(data + beg, end - beg)) { - in_empty = 1; - beg = end; - continue; - } - - /* calculating the indentation */ - i = 0; - while (i < 4 && beg + i < end && data[beg + i] == ' ') - i++; - - pre = i; - - if (doc->ext_flags & HOEDOWN_EXT_FENCED_CODE) { - if (is_codefence(data + beg + i, end - beg - i, NULL, NULL)) - in_fence = !in_fence; - } - - /* Only check for new list items if we are **not** inside - * a fenced code block */ - if (!in_fence) { - has_next_uli = prefix_uli(data + beg + i, end - beg - i); - has_next_oli = prefix_oli(data + beg + i, end - beg - i); - } - - /* checking for a new item */ - if ((has_next_uli && !is_hrule(data + beg + i, end - beg - i)) || has_next_oli) { - if (in_empty) - has_inside_empty = 1; - - /* the following item must have the same (or less) indentation */ - if (pre <= orgpre) { - /* if the following item has different list type, we end this list */ - if (in_empty && ( - ((*flags & HOEDOWN_LIST_ORDERED) && has_next_uli) || - (!(*flags & HOEDOWN_LIST_ORDERED) && has_next_oli))) - *flags |= HOEDOWN_LI_END; - - break; - } - - if (!sublist) - sublist = work->size; - } - /* joining only indented stuff after empty lines; - * note that now we only require 1 space of indentation - * to continue a list */ - else if (in_empty && pre == 0) { - *flags |= HOEDOWN_LI_END; - break; - } - - if (in_empty) { - hoedown_buffer_putc(work, '\n'); - has_inside_empty = 1; - in_empty = 0; - } - - /* adding the line without prefix into the working buffer */ - hoedown_buffer_put(work, data + beg + i, end - beg - i); - beg = end; - } - - /* render of li contents */ - if (has_inside_empty) - *flags |= HOEDOWN_LI_BLOCK; - - if (*flags & HOEDOWN_LI_BLOCK) { - /* intermediate render of block li */ - if (sublist && sublist < work->size) { - parse_block(inter, doc, work->data, sublist); - parse_block(inter, doc, work->data + sublist, work->size - sublist); - } - else - parse_block(inter, doc, work->data, work->size); - } else { - /* intermediate render of inline li */ - if (sublist && sublist < work->size) { - parse_inline(inter, doc, work->data, sublist); - parse_block(inter, doc, work->data + sublist, work->size - sublist); - } - else - parse_inline(inter, doc, work->data, work->size); - } - - /* render of li itself */ - if (doc->md.listitem) - doc->md.listitem(ob, inter, *flags, &doc->data); - - popbuf(doc, BUFFER_SPAN); - popbuf(doc, BUFFER_SPAN); - return beg; + hoedown_buffer *work = 0, *inter = 0; + size_t beg = 0, end, pre, sublist = 0, orgpre = 0, i; + int in_empty = 0, has_inside_empty = 0, in_fence = 0; + + /* keeping track of the first indentation prefix */ + while (orgpre < 3 && orgpre < size && data[orgpre] == ' ') + orgpre++; + + beg = prefix_uli(data, size); + if (!beg) + beg = prefix_oli(data, size); + + if (!beg) + return 0; + + /* skipping to the beginning of the following line */ + end = beg; + while (end < size && data[end - 1] != '\n') + end++; + + /* getting working buffers */ + work = newbuf(doc, BUFFER_SPAN); + inter = newbuf(doc, BUFFER_SPAN); + + /* putting the first line into the working buffer */ + hoedown_buffer_put(work, data + beg, end - beg); + beg = end; + + /* process the following lines */ + while (beg < size) { + size_t has_next_uli = 0, has_next_oli = 0; + + end++; + + while (end < size && data[end - 1] != '\n') + end++; + + /* process an empty line */ + if (is_empty(data + beg, end - beg)) { + in_empty = 1; + beg = end; + continue; + } + + /* calculating the indentation */ + i = 0; + while (i < 4 && beg + i < end && data[beg + i] == ' ') + i++; + + pre = i; + + if (doc->ext_flags & HOEDOWN_EXT_FENCED_CODE) { + if (is_codefence(data + beg + i, end - beg - i, NULL, NULL)) + in_fence = !in_fence; + } + + /* Only check for new list items if we are **not** inside + * a fenced code block */ + if (!in_fence) { + has_next_uli = prefix_uli(data + beg + i, end - beg - i); + has_next_oli = prefix_oli(data + beg + i, end - beg - i); + } + + /* checking for a new item */ + if ((has_next_uli && !is_hrule(data + beg + i, end - beg - i)) || has_next_oli) { + if (in_empty) + has_inside_empty = 1; + + /* the following item must have the same (or less) indentation */ + if (pre <= orgpre) { + /* if the following item has different list type, we end this list */ + if (in_empty && ( + ((*flags & HOEDOWN_LIST_ORDERED) && has_next_uli) || + (!(*flags & HOEDOWN_LIST_ORDERED) && has_next_oli))) + *flags |= HOEDOWN_LI_END; + + break; + } + + if (!sublist) + sublist = work->size; + } + /* joining only indented stuff after empty lines; + * note that now we only require 1 space of indentation + * to continue a list */ + else if (in_empty && pre == 0) { + *flags |= HOEDOWN_LI_END; + break; + } + + if (in_empty) { + hoedown_buffer_putc(work, '\n'); + has_inside_empty = 1; + in_empty = 0; + } + + /* adding the line without prefix into the working buffer */ + hoedown_buffer_put(work, data + beg + i, end - beg - i); + beg = end; + } + + /* render of li contents */ + if (has_inside_empty) + *flags |= HOEDOWN_LI_BLOCK; + + if (*flags & HOEDOWN_LI_BLOCK) { + /* intermediate render of block li */ + if (sublist && sublist < work->size) { + parse_block(inter, doc, work->data, sublist); + parse_block(inter, doc, work->data + sublist, work->size - sublist); + } + else + parse_block(inter, doc, work->data, work->size); + } else { + /* intermediate render of inline li */ + if (sublist && sublist < work->size) { + parse_inline(inter, doc, work->data, sublist); + parse_block(inter, doc, work->data + sublist, work->size - sublist); + } + else + parse_inline(inter, doc, work->data, work->size); + } + + /* render of li itself */ + if (doc->md.listitem) + doc->md.listitem(ob, inter, *flags, &doc->data); + + popbuf(doc, BUFFER_SPAN); + popbuf(doc, BUFFER_SPAN); + return beg; } @@ -1936,460 +1936,460 @@ parse_listitem(hoedown_buffer *ob, hoedown_document *doc, uint8_t *data, size_t static size_t parse_list(hoedown_buffer *ob, hoedown_document *doc, uint8_t *data, size_t size, hoedown_list_flags flags) { - hoedown_buffer *work = 0; - size_t i = 0, j; + hoedown_buffer *work = 0; + size_t i = 0, j; - work = newbuf(doc, BUFFER_BLOCK); + work = newbuf(doc, BUFFER_BLOCK); - while (i < size) { - j = parse_listitem(work, doc, data + i, size - i, &flags); - i += j; + while (i < size) { + j = parse_listitem(work, doc, data + i, size - i, &flags); + i += j; - if (!j || (flags & HOEDOWN_LI_END)) - break; - } + if (!j || (flags & HOEDOWN_LI_END)) + break; + } - if (doc->md.list) - doc->md.list(ob, work, flags, &doc->data); - popbuf(doc, BUFFER_BLOCK); - return i; + if (doc->md.list) + doc->md.list(ob, work, flags, &doc->data); + popbuf(doc, BUFFER_BLOCK); + return i; } /* parse_atxheader • parsing of atx-style headers */ static size_t parse_atxheader(hoedown_buffer *ob, hoedown_document *doc, uint8_t *data, size_t size) { - size_t level = 0; - size_t i, end, skip; + size_t level = 0; + size_t i, end, skip; - while (level < size && level < 6 && data[level] == '#') - level++; + while (level < size && level < 6 && data[level] == '#') + level++; - for (i = level; i < size && data[i] == ' '; i++); + for (i = level; i < size && data[i] == ' '; i++); - for (end = i; end < size && data[end] != '\n'; end++); - skip = end; + for (end = i; end < size && data[end] != '\n'; end++); + skip = end; - while (end && data[end - 1] == '#') - end--; + while (end && data[end - 1] == '#') + end--; - while (end && data[end - 1] == ' ') - end--; + while (end && data[end - 1] == ' ') + end--; - if (end > i) { - hoedown_buffer *work = newbuf(doc, BUFFER_SPAN); + if (end > i) { + hoedown_buffer *work = newbuf(doc, BUFFER_SPAN); - parse_inline(work, doc, data + i, end - i); + parse_inline(work, doc, data + i, end - i); - if (doc->md.header) - doc->md.header(ob, work, (int)level, &doc->data); + if (doc->md.header) + doc->md.header(ob, work, (int)level, &doc->data); - popbuf(doc, BUFFER_SPAN); - } + popbuf(doc, BUFFER_SPAN); + } - return skip; + return skip; } /* parse_footnote_def • parse a single footnote definition */ static void parse_footnote_def(hoedown_buffer *ob, hoedown_document *doc, unsigned int num, uint8_t *data, size_t size) { - hoedown_buffer *work = 0; - work = newbuf(doc, BUFFER_SPAN); + hoedown_buffer *work = 0; + work = newbuf(doc, BUFFER_SPAN); - parse_block(work, doc, data, size); + parse_block(work, doc, data, size); - if (doc->md.footnote_def) - doc->md.footnote_def(ob, work, num, &doc->data); - popbuf(doc, BUFFER_SPAN); + if (doc->md.footnote_def) + doc->md.footnote_def(ob, work, num, &doc->data); + popbuf(doc, BUFFER_SPAN); } /* parse_footnote_list • render the contents of the footnotes */ static void parse_footnote_list(hoedown_buffer *ob, hoedown_document *doc, struct footnote_list *footnotes) { - hoedown_buffer *work = 0; - struct footnote_item *item; - struct footnote_ref *ref; + hoedown_buffer *work = 0; + struct footnote_item *item; + struct footnote_ref *ref; - if (footnotes->count == 0) - return; + if (footnotes->count == 0) + return; - work = newbuf(doc, BUFFER_BLOCK); + work = newbuf(doc, BUFFER_BLOCK); - item = footnotes->head; - while (item) { - ref = item->ref; - parse_footnote_def(work, doc, ref->num, ref->contents->data, ref->contents->size); - item = item->next; - } + item = footnotes->head; + while (item) { + ref = item->ref; + parse_footnote_def(work, doc, ref->num, ref->contents->data, ref->contents->size); + item = item->next; + } - if (doc->md.footnotes) - doc->md.footnotes(ob, work, &doc->data); - popbuf(doc, BUFFER_BLOCK); + if (doc->md.footnotes) + doc->md.footnotes(ob, work, &doc->data); + popbuf(doc, BUFFER_BLOCK); } /* htmlblock_is_end • check for end of HTML block : </tag>( *)\n */ -/* returns tag length on match, 0 otherwise */ -/* assumes data starts with "<" */ +/* returns tag length on match, 0 otherwise */ +/* assumes data starts with "<" */ static size_t htmlblock_is_end( - const char *tag, - size_t tag_len, - hoedown_document *doc, - uint8_t *data, - size_t size) + const char *tag, + size_t tag_len, + hoedown_document *doc, + uint8_t *data, + size_t size) { - size_t i = tag_len + 3, w; + size_t i = tag_len + 3, w; - /* try to match the end tag */ - /* note: we're not considering tags like "</tag >" which are still valid */ - if (i > size || - data[1] != '/' || - strncasecmp((char *)data + 2, tag, tag_len) != 0 || - data[tag_len + 2] != '>') - return 0; + /* try to match the end tag */ + /* note: we're not considering tags like "</tag >" which are still valid */ + if (i > size || + data[1] != '/' || + strncasecmp((char *)data + 2, tag, tag_len) != 0 || + data[tag_len + 2] != '>') + return 0; - /* rest of the line must be empty */ - if ((w = is_empty(data + i, size - i)) == 0 && i < size) - return 0; + /* rest of the line must be empty */ + if ((w = is_empty(data + i, size - i)) == 0 && i < size) + return 0; - return i + w; + return i + w; } /* htmlblock_find_end • try to find HTML block ending tag */ -/* returns the length on match, 0 otherwise */ +/* returns the length on match, 0 otherwise */ static size_t htmlblock_find_end( - const char *tag, - size_t tag_len, - hoedown_document *doc, - uint8_t *data, - size_t size) + const char *tag, + size_t tag_len, + hoedown_document *doc, + uint8_t *data, + size_t size) { - size_t i = 0, w; + size_t i = 0, w; - while (1) { - while (i < size && data[i] != '<') i++; - if (i >= size) return 0; + while (1) { + while (i < size && data[i] != '<') i++; + if (i >= size) return 0; - w = htmlblock_is_end(tag, tag_len, doc, data + i, size - i); - if (w) return i + w; - i++; - } + w = htmlblock_is_end(tag, tag_len, doc, data + i, size - i); + if (w) return i + w; + i++; + } } /* htmlblock_find_end_strict • try to find end of HTML block in strict mode */ -/* (it must be an unindented line, and have a blank line afterwads) */ -/* returns the length on match, 0 otherwise */ +/* (it must be an unindented line, and have a blank line afterwads) */ +/* returns the length on match, 0 otherwise */ static size_t htmlblock_find_end_strict( - const char *tag, - size_t tag_len, - hoedown_document *doc, - uint8_t *data, - size_t size) + const char *tag, + size_t tag_len, + hoedown_document *doc, + uint8_t *data, + size_t size) { - size_t i = 0, mark; + size_t i = 0, mark; - while (1) { - mark = i; - while (i < size && data[i] != '\n') i++; - if (i < size) i++; - if (i == mark) return 0; + while (1) { + mark = i; + while (i < size && data[i] != '\n') i++; + if (i < size) i++; + if (i == mark) return 0; - if (data[mark] == ' ' && mark > 0) continue; - mark += htmlblock_find_end(tag, tag_len, doc, data + mark, i - mark); - if (mark == i && (is_empty(data + i, size - i) || i >= size)) break; - } + if (data[mark] == ' ' && mark > 0) continue; + mark += htmlblock_find_end(tag, tag_len, doc, data + mark, i - mark); + if (mark == i && (is_empty(data + i, size - i) || i >= size)) break; + } - return i; + return i; } /* parse_htmlblock • parsing of inline HTML block */ static size_t parse_htmlblock(hoedown_buffer *ob, hoedown_document *doc, uint8_t *data, size_t size, int do_render) { - hoedown_buffer work = { NULL, 0, 0, 0, NULL, NULL, NULL }; - size_t i, j = 0, tag_len, tag_end; - const char *curtag = NULL; + hoedown_buffer work = { NULL, 0, 0, 0, NULL, NULL, NULL }; + size_t i, j = 0, tag_len, tag_end; + const char *curtag = NULL; - work.data = data; + work.data = data; - /* identification of the opening tag */ - if (size < 2 || data[0] != '<') - return 0; + /* identification of the opening tag */ + if (size < 2 || data[0] != '<') + return 0; - i = 1; - while (i < size && data[i] != '>' && data[i] != ' ') - i++; + i = 1; + while (i < size && data[i] != '>' && data[i] != ' ') + i++; - if (i < size) - curtag = hoedown_find_block_tag((char *)data + 1, (int)i - 1); + if (i < size) + curtag = hoedown_find_block_tag((char *)data + 1, (int)i - 1); - /* handling of special cases */ - if (!curtag) { + /* handling of special cases */ + if (!curtag) { - /* HTML comment, laxist form */ - if (size > 5 && data[1] == '!' && data[2] == '-' && data[3] == '-') { - i = 5; + /* HTML comment, laxist form */ + if (size > 5 && data[1] == '!' && data[2] == '-' && data[3] == '-') { + i = 5; - while (i < size && !(data[i - 2] == '-' && data[i - 1] == '-' && data[i] == '>')) - i++; + while (i < size && !(data[i - 2] == '-' && data[i - 1] == '-' && data[i] == '>')) + i++; - i++; + i++; - if (i < size) - j = is_empty(data + i, size - i); + if (i < size) + j = is_empty(data + i, size - i); - if (j) { - work.size = i + j; - if (do_render && doc->md.blockhtml) - doc->md.blockhtml(ob, &work, &doc->data); - return work.size; - } - } + if (j) { + work.size = i + j; + if (do_render && doc->md.blockhtml) + doc->md.blockhtml(ob, &work, &doc->data); + return work.size; + } + } - /* HR, which is the only self-closing block tag considered */ - if (size > 4 && (data[1] == 'h' || data[1] == 'H') && (data[2] == 'r' || data[2] == 'R')) { - i = 3; - while (i < size && data[i] != '>') - i++; + /* HR, which is the only self-closing block tag considered */ + if (size > 4 && (data[1] == 'h' || data[1] == 'H') && (data[2] == 'r' || data[2] == 'R')) { + i = 3; + while (i < size && data[i] != '>') + i++; - if (i + 1 < size) { - i++; - j = is_empty(data + i, size - i); - if (j) { - work.size = i + j; - if (do_render && doc->md.blockhtml) - doc->md.blockhtml(ob, &work, &doc->data); - return work.size; - } - } - } + if (i + 1 < size) { + i++; + j = is_empty(data + i, size - i); + if (j) { + work.size = i + j; + if (do_render && doc->md.blockhtml) + doc->md.blockhtml(ob, &work, &doc->data); + return work.size; + } + } + } - /* no special case recognised */ - return 0; - } + /* no special case recognised */ + return 0; + } - /* looking for a matching closing tag in strict mode */ - tag_len = strlen(curtag); - tag_end = htmlblock_find_end_strict(curtag, tag_len, doc, data, size); + /* looking for a matching closing tag in strict mode */ + tag_len = strlen(curtag); + tag_end = htmlblock_find_end_strict(curtag, tag_len, doc, data, size); - /* if not found, trying a second pass looking for indented match */ - /* but not if tag is "ins" or "del" (following original Markdown.pl) */ - if (!tag_end && strcmp(curtag, "ins") != 0 && strcmp(curtag, "del") != 0) - tag_end = htmlblock_find_end(curtag, tag_len, doc, data, size); + /* if not found, trying a second pass looking for indented match */ + /* but not if tag is "ins" or "del" (following original Markdown.pl) */ + if (!tag_end && strcmp(curtag, "ins") != 0 && strcmp(curtag, "del") != 0) + tag_end = htmlblock_find_end(curtag, tag_len, doc, data, size); - if (!tag_end) - return 0; + if (!tag_end) + return 0; - /* the end of the block has been found */ - work.size = tag_end; - if (do_render && doc->md.blockhtml) - doc->md.blockhtml(ob, &work, &doc->data); + /* the end of the block has been found */ + work.size = tag_end; + if (do_render && doc->md.blockhtml) + doc->md.blockhtml(ob, &work, &doc->data); - return tag_end; + return tag_end; } static void parse_table_row( - hoedown_buffer *ob, - hoedown_document *doc, - uint8_t *data, - size_t size, - size_t columns, - hoedown_table_flags *col_data, - hoedown_table_flags header_flag) + hoedown_buffer *ob, + hoedown_document *doc, + uint8_t *data, + size_t size, + size_t columns, + hoedown_table_flags *col_data, + hoedown_table_flags header_flag) { - size_t i = 0, col, len; - hoedown_buffer *row_work = 0; + size_t i = 0, col, len; + hoedown_buffer *row_work = 0; - if (!doc->md.table_cell || !doc->md.table_row) - return; + if (!doc->md.table_cell || !doc->md.table_row) + return; - row_work = newbuf(doc, BUFFER_SPAN); + row_work = newbuf(doc, BUFFER_SPAN); - if (i < size && data[i] == '|') - i++; + if (i < size && data[i] == '|') + i++; - for (col = 0; col < columns && i < size; ++col) { - size_t cell_start, cell_end; - hoedown_buffer *cell_work; + for (col = 0; col < columns && i < size; ++col) { + size_t cell_start, cell_end; + hoedown_buffer *cell_work; - cell_work = newbuf(doc, BUFFER_SPAN); + cell_work = newbuf(doc, BUFFER_SPAN); - while (i < size && _isspace(data[i])) - i++; + while (i < size && _isspace(data[i])) + i++; - cell_start = i; + cell_start = i; - len = find_emph_char(data + i, size - i, '|'); - i += len ? len : size - i; + len = find_emph_char(data + i, size - i, '|'); + i += len ? len : size - i; - cell_end = i - 1; + cell_end = i - 1; - while (cell_end > cell_start && _isspace(data[cell_end])) - cell_end--; + while (cell_end > cell_start && _isspace(data[cell_end])) + cell_end--; - parse_inline(cell_work, doc, data + cell_start, 1 + cell_end - cell_start); - doc->md.table_cell(row_work, cell_work, col_data[col] | header_flag, &doc->data); + parse_inline(cell_work, doc, data + cell_start, 1 + cell_end - cell_start); + doc->md.table_cell(row_work, cell_work, col_data[col] | header_flag, &doc->data); - popbuf(doc, BUFFER_SPAN); - i++; - } + popbuf(doc, BUFFER_SPAN); + i++; + } - for (; col < columns; ++col) { - hoedown_buffer empty_cell = { 0, 0, 0, 0, NULL, NULL, NULL }; - doc->md.table_cell(row_work, &empty_cell, col_data[col] | header_flag, &doc->data); - } + for (; col < columns; ++col) { + hoedown_buffer empty_cell = { 0, 0, 0, 0, NULL, NULL, NULL }; + doc->md.table_cell(row_work, &empty_cell, col_data[col] | header_flag, &doc->data); + } - doc->md.table_row(ob, row_work, &doc->data); + doc->md.table_row(ob, row_work, &doc->data); - popbuf(doc, BUFFER_SPAN); + popbuf(doc, BUFFER_SPAN); } static size_t parse_table_header( - hoedown_buffer *ob, - hoedown_document *doc, - uint8_t *data, - size_t size, - size_t *columns, - hoedown_table_flags **column_data) + hoedown_buffer *ob, + hoedown_document *doc, + uint8_t *data, + size_t size, + size_t *columns, + hoedown_table_flags **column_data) { - int pipes; - size_t i = 0, col, header_end, under_end; + int pipes; + size_t i = 0, col, header_end, under_end; - pipes = 0; - while (i < size && data[i] != '\n') - if (data[i++] == '|') - pipes++; + pipes = 0; + while (i < size && data[i] != '\n') + if (data[i++] == '|') + pipes++; - if (i == size || pipes == 0) - return 0; + if (i == size || pipes == 0) + return 0; - header_end = i; + header_end = i; - while (header_end > 0 && _isspace(data[header_end - 1])) - header_end--; + while (header_end > 0 && _isspace(data[header_end - 1])) + header_end--; - if (data[0] == '|') - pipes--; + if (data[0] == '|') + pipes--; - if (header_end && data[header_end - 1] == '|') - pipes--; + if (header_end && data[header_end - 1] == '|') + pipes--; - if (pipes < 0) - return 0; + if (pipes < 0) + return 0; - *columns = pipes + 1; - *column_data = hoedown_calloc(*columns, sizeof(hoedown_table_flags)); + *columns = pipes + 1; + *column_data = hoedown_calloc(*columns, sizeof(hoedown_table_flags)); - /* Parse the header underline */ - i++; - if (i < size && data[i] == '|') - i++; + /* Parse the header underline */ + i++; + if (i < size && data[i] == '|') + i++; - under_end = i; - while (under_end < size && data[under_end] != '\n') - under_end++; + under_end = i; + while (under_end < size && data[under_end] != '\n') + under_end++; - for (col = 0; col < *columns && i < under_end; ++col) { - size_t dashes = 0; + for (col = 0; col < *columns && i < under_end; ++col) { + size_t dashes = 0; - while (i < under_end && data[i] == ' ') - i++; + while (i < under_end && data[i] == ' ') + i++; - if (data[i] == ':') { - i++; (*column_data)[col] |= HOEDOWN_TABLE_ALIGN_LEFT; - dashes++; - } + if (data[i] == ':') { + i++; (*column_data)[col] |= HOEDOWN_TABLE_ALIGN_LEFT; + dashes++; + } - while (i < under_end && data[i] == '-') { - i++; dashes++; - } + while (i < under_end && data[i] == '-') { + i++; dashes++; + } - if (i < under_end && data[i] == ':') { - i++; (*column_data)[col] |= HOEDOWN_TABLE_ALIGN_RIGHT; - dashes++; - } + if (i < under_end && data[i] == ':') { + i++; (*column_data)[col] |= HOEDOWN_TABLE_ALIGN_RIGHT; + dashes++; + } - while (i < under_end && data[i] == ' ') - i++; + while (i < under_end && data[i] == ' ') + i++; - if (i < under_end && data[i] != '|' && data[i] != '+') - break; + if (i < under_end && data[i] != '|' && data[i] != '+') + break; - if (dashes < 3) - break; + if (dashes < 3) + break; - i++; - } + i++; + } - if (col < *columns) - return 0; + if (col < *columns) + return 0; - parse_table_row( - ob, doc, data, - header_end, - *columns, - *column_data, - HOEDOWN_TABLE_HEADER - ); + parse_table_row( + ob, doc, data, + header_end, + *columns, + *column_data, + HOEDOWN_TABLE_HEADER + ); - return under_end + 1; + return under_end + 1; } static size_t parse_table( - hoedown_buffer *ob, - hoedown_document *doc, - uint8_t *data, - size_t size) + hoedown_buffer *ob, + hoedown_document *doc, + uint8_t *data, + size_t size) { - size_t i; + size_t i; - hoedown_buffer *work = 0; - hoedown_buffer *header_work = 0; - hoedown_buffer *body_work = 0; + hoedown_buffer *work = 0; + hoedown_buffer *header_work = 0; + hoedown_buffer *body_work = 0; - size_t columns; - hoedown_table_flags *col_data = NULL; + size_t columns; + hoedown_table_flags *col_data = NULL; - work = newbuf(doc, BUFFER_BLOCK); - header_work = newbuf(doc, BUFFER_SPAN); - body_work = newbuf(doc, BUFFER_BLOCK); + work = newbuf(doc, BUFFER_BLOCK); + header_work = newbuf(doc, BUFFER_SPAN); + body_work = newbuf(doc, BUFFER_BLOCK); - i = parse_table_header(header_work, doc, data, size, &columns, &col_data); - if (i > 0) { + i = parse_table_header(header_work, doc, data, size, &columns, &col_data); + if (i > 0) { - while (i < size) { - size_t row_start; - int pipes = 0; + while (i < size) { + size_t row_start; + int pipes = 0; - row_start = i; + row_start = i; - while (i < size && data[i] != '\n') - if (data[i++] == '|') - pipes++; + while (i < size && data[i] != '\n') + if (data[i++] == '|') + pipes++; - if (pipes == 0 || i == size) { - i = row_start; - break; - } + if (pipes == 0 || i == size) { + i = row_start; + break; + } - parse_table_row( - body_work, - doc, - data + row_start, - i - row_start, - columns, - col_data, 0 - ); + parse_table_row( + body_work, + doc, + data + row_start, + i - row_start, + columns, + col_data, 0 + ); - i++; - } + i++; + } if (doc->md.table_header) doc->md.table_header(work, header_work, &doc->data); @@ -2397,76 +2397,76 @@ parse_table( if (doc->md.table_body) doc->md.table_body(work, body_work, &doc->data); - if (doc->md.table) - doc->md.table(ob, work, &doc->data); - } + if (doc->md.table) + doc->md.table(ob, work, &doc->data); + } - free(col_data); - popbuf(doc, BUFFER_SPAN); - popbuf(doc, BUFFER_BLOCK); - popbuf(doc, BUFFER_BLOCK); - return i; + free(col_data); + popbuf(doc, BUFFER_SPAN); + popbuf(doc, BUFFER_BLOCK); + popbuf(doc, BUFFER_BLOCK); + return i; } /* parse_block • parsing of one block, returning next uint8_t to parse */ static void parse_block(hoedown_buffer *ob, hoedown_document *doc, uint8_t *data, size_t size) { - size_t beg, end, i; - uint8_t *txt_data; - beg = 0; + size_t beg, end, i; + uint8_t *txt_data; + beg = 0; - if (doc->work_bufs[BUFFER_SPAN].size + - doc->work_bufs[BUFFER_BLOCK].size > doc->max_nesting) - return; + if (doc->work_bufs[BUFFER_SPAN].size + + doc->work_bufs[BUFFER_BLOCK].size > doc->max_nesting) + return; - while (beg < size) { - txt_data = data + beg; - end = size - beg; + while (beg < size) { + txt_data = data + beg; + end = size - beg; - if (is_atxheader(doc, txt_data, end)) - beg += parse_atxheader(ob, doc, txt_data, end); + if (is_atxheader(doc, txt_data, end)) + beg += parse_atxheader(ob, doc, txt_data, end); - else if (data[beg] == '<' && doc->md.blockhtml && - (i = parse_htmlblock(ob, doc, txt_data, end, 1)) != 0) - beg += i; + else if (data[beg] == '<' && doc->md.blockhtml && + (i = parse_htmlblock(ob, doc, txt_data, end, 1)) != 0) + beg += i; - else if ((i = is_empty(txt_data, end)) != 0) - beg += i; + else if ((i = is_empty(txt_data, end)) != 0) + beg += i; - else if (is_hrule(txt_data, end)) { - if (doc->md.hrule) - doc->md.hrule(ob, &doc->data); + else if (is_hrule(txt_data, end)) { + if (doc->md.hrule) + doc->md.hrule(ob, &doc->data); - while (beg < size && data[beg] != '\n') - beg++; + while (beg < size && data[beg] != '\n') + beg++; - beg++; - } + beg++; + } - else if ((doc->ext_flags & HOEDOWN_EXT_FENCED_CODE) != 0 && - (i = parse_fencedcode(ob, doc, txt_data, end)) != 0) - beg += i; + else if ((doc->ext_flags & HOEDOWN_EXT_FENCED_CODE) != 0 && + (i = parse_fencedcode(ob, doc, txt_data, end)) != 0) + beg += i; - else if ((doc->ext_flags & HOEDOWN_EXT_TABLES) != 0 && - (i = parse_table(ob, doc, txt_data, end)) != 0) - beg += i; + else if ((doc->ext_flags & HOEDOWN_EXT_TABLES) != 0 && + (i = parse_table(ob, doc, txt_data, end)) != 0) + beg += i; - else if (prefix_quote(txt_data, end)) - beg += parse_blockquote(ob, doc, txt_data, end); + else if (prefix_quote(txt_data, end)) + beg += parse_blockquote(ob, doc, txt_data, end); - else if (!(doc->ext_flags & HOEDOWN_EXT_DISABLE_INDENTED_CODE) && prefix_code(txt_data, end)) - beg += parse_blockcode(ob, doc, txt_data, end); + else if (!(doc->ext_flags & HOEDOWN_EXT_DISABLE_INDENTED_CODE) && prefix_code(txt_data, end)) + beg += parse_blockcode(ob, doc, txt_data, end); - else if (prefix_uli(txt_data, end)) - beg += parse_list(ob, doc, txt_data, end, 0); + else if (prefix_uli(txt_data, end)) + beg += parse_list(ob, doc, txt_data, end, 0); - else if (prefix_oli(txt_data, end)) - beg += parse_list(ob, doc, txt_data, end, HOEDOWN_LIST_ORDERED); + else if (prefix_oli(txt_data, end)) + beg += parse_list(ob, doc, txt_data, end, HOEDOWN_LIST_ORDERED); - else - beg += parse_paragraph(ob, doc, txt_data, end); - } + else + beg += parse_paragraph(ob, doc, txt_data, end); + } } @@ -2479,253 +2479,253 @@ parse_block(hoedown_buffer *ob, hoedown_document *doc, uint8_t *data, size_t siz static int is_footnote(const uint8_t *data, size_t beg, size_t end, size_t *last, struct footnote_list *list) { - size_t i = 0; - hoedown_buffer *contents = 0; - size_t ind = 0; - int in_empty = 0; - size_t start = 0; - - size_t id_offset, id_end; - - /* up to 3 optional leading spaces */ - if (beg + 3 >= end) return 0; - if (data[beg] == ' ') { i = 1; - if (data[beg + 1] == ' ') { i = 2; - if (data[beg + 2] == ' ') { i = 3; - if (data[beg + 3] == ' ') return 0; } } } - i += beg; - - /* id part: caret followed by anything between brackets */ - if (data[i] != '[') return 0; - i++; - if (i >= end || data[i] != '^') return 0; - i++; - id_offset = i; - while (i < end && data[i] != '\n' && data[i] != '\r' && data[i] != ']') - i++; - if (i >= end || data[i] != ']') return 0; - id_end = i; - - /* spacer: colon (space | tab)* newline? (space | tab)* */ - i++; - if (i >= end || data[i] != ':') return 0; - i++; - - /* getting content buffer */ - contents = hoedown_buffer_new(64); - - start = i; - - /* process lines similar to a list item */ - while (i < end) { - while (i < end && data[i] != '\n' && data[i] != '\r') i++; - - /* process an empty line */ - if (is_empty(data + start, i - start)) { - in_empty = 1; - if (i < end && (data[i] == '\n' || data[i] == '\r')) { - i++; - if (i < end && data[i] == '\n' && data[i - 1] == '\r') i++; - } - start = i; - continue; - } - - /* calculating the indentation */ - ind = 0; - while (ind < 4 && start + ind < end && data[start + ind] == ' ') - ind++; - - /* joining only indented stuff after empty lines; - * note that now we only require 1 space of indentation - * to continue, just like lists */ - if (ind == 0) { - if (start == id_end + 2 && data[start] == '\t') {} - else break; - } - else if (in_empty) { - hoedown_buffer_putc(contents, '\n'); - } - - in_empty = 0; - - /* adding the line into the content buffer */ - hoedown_buffer_put(contents, data + start + ind, i - start - ind); - /* add carriage return */ - if (i < end) { - hoedown_buffer_putc(contents, '\n'); - if (i < end && (data[i] == '\n' || data[i] == '\r')) { - i++; - if (i < end && data[i] == '\n' && data[i - 1] == '\r') i++; - } - } - start = i; - } - - if (last) - *last = start; - - if (list) { - struct footnote_ref *ref; - ref = create_footnote_ref(list, data + id_offset, id_end - id_offset); - if (!ref) - return 0; - if (!add_footnote_ref(list, ref)) { - free_footnote_ref(ref); - return 0; - } - ref->contents = contents; - } - - return 1; + size_t i = 0; + hoedown_buffer *contents = 0; + size_t ind = 0; + int in_empty = 0; + size_t start = 0; + + size_t id_offset, id_end; + + /* up to 3 optional leading spaces */ + if (beg + 3 >= end) return 0; + if (data[beg] == ' ') { i = 1; + if (data[beg + 1] == ' ') { i = 2; + if (data[beg + 2] == ' ') { i = 3; + if (data[beg + 3] == ' ') return 0; } } } + i += beg; + + /* id part: caret followed by anything between brackets */ + if (data[i] != '[') return 0; + i++; + if (i >= end || data[i] != '^') return 0; + i++; + id_offset = i; + while (i < end && data[i] != '\n' && data[i] != '\r' && data[i] != ']') + i++; + if (i >= end || data[i] != ']') return 0; + id_end = i; + + /* spacer: colon (space | tab)* newline? (space | tab)* */ + i++; + if (i >= end || data[i] != ':') return 0; + i++; + + /* getting content buffer */ + contents = hoedown_buffer_new(64); + + start = i; + + /* process lines similar to a list item */ + while (i < end) { + while (i < end && data[i] != '\n' && data[i] != '\r') i++; + + /* process an empty line */ + if (is_empty(data + start, i - start)) { + in_empty = 1; + if (i < end && (data[i] == '\n' || data[i] == '\r')) { + i++; + if (i < end && data[i] == '\n' && data[i - 1] == '\r') i++; + } + start = i; + continue; + } + + /* calculating the indentation */ + ind = 0; + while (ind < 4 && start + ind < end && data[start + ind] == ' ') + ind++; + + /* joining only indented stuff after empty lines; + * note that now we only require 1 space of indentation + * to continue, just like lists */ + if (ind == 0) { + if (start == id_end + 2 && data[start] == '\t') {} + else break; + } + else if (in_empty) { + hoedown_buffer_putc(contents, '\n'); + } + + in_empty = 0; + + /* adding the line into the content buffer */ + hoedown_buffer_put(contents, data + start + ind, i - start - ind); + /* add carriage return */ + if (i < end) { + hoedown_buffer_putc(contents, '\n'); + if (i < end && (data[i] == '\n' || data[i] == '\r')) { + i++; + if (i < end && data[i] == '\n' && data[i - 1] == '\r') i++; + } + } + start = i; + } + + if (last) + *last = start; + + if (list) { + struct footnote_ref *ref; + ref = create_footnote_ref(list, data + id_offset, id_end - id_offset); + if (!ref) + return 0; + if (!add_footnote_ref(list, ref)) { + free_footnote_ref(ref); + return 0; + } + ref->contents = contents; + } + + return 1; } /* is_ref • returns whether a line is a reference or not */ static int is_ref(const uint8_t *data, size_t beg, size_t end, size_t *last, struct link_ref **refs) { -/* int n; */ - size_t i = 0; - size_t id_offset, id_end; - size_t link_offset, link_end; - size_t title_offset, title_end; - size_t line_end; - - /* up to 3 optional leading spaces */ - if (beg + 3 >= end) return 0; - if (data[beg] == ' ') { i = 1; - if (data[beg + 1] == ' ') { i = 2; - if (data[beg + 2] == ' ') { i = 3; - if (data[beg + 3] == ' ') return 0; } } } - i += beg; - - /* id part: anything but a newline between brackets */ - if (data[i] != '[') return 0; - i++; - id_offset = i; - while (i < end && data[i] != '\n' && data[i] != '\r' && data[i] != ']') - i++; - if (i >= end || data[i] != ']') return 0; - id_end = i; - - /* spacer: colon (space | tab)* newline? (space | tab)* */ - i++; - if (i >= end || data[i] != ':') return 0; - i++; - while (i < end && data[i] == ' ') i++; - if (i < end && (data[i] == '\n' || data[i] == '\r')) { - i++; - if (i < end && data[i] == '\r' && data[i - 1] == '\n') i++; } - while (i < end && data[i] == ' ') i++; - if (i >= end) return 0; - - /* link: spacing-free sequence, optionally between angle brackets */ - if (data[i] == '<') - i++; - - link_offset = i; - - while (i < end && data[i] != ' ' && data[i] != '\n' && data[i] != '\r') - i++; - - if (data[i - 1] == '>') link_end = i - 1; - else link_end = i; - - /* optional spacer: (space | tab)* (newline | '\'' | '"' | '(' ) */ - while (i < end && data[i] == ' ') i++; - if (i < end && data[i] != '\n' && data[i] != '\r' - && data[i] != '\'' && data[i] != '"' && data[i] != '(') - return 0; - line_end = 0; - /* computing end-of-line */ - if (i >= end || data[i] == '\r' || data[i] == '\n') line_end = i; - if (i + 1 < end && data[i] == '\n' && data[i + 1] == '\r') - line_end = i + 1; - - /* optional (space|tab)* spacer after a newline */ - if (line_end) { - i = line_end + 1; - while (i < end && data[i] == ' ') i++; } - - /* optional title: any non-newline sequence enclosed in '"() - alone on its line */ - title_offset = title_end = 0; - if (i + 1 < end - && (data[i] == '\'' || data[i] == '"' || data[i] == '(')) { - i++; - title_offset = i; - /* looking for EOL */ - while (i < end && data[i] != '\n' && data[i] != '\r') i++; - if (i + 1 < end && data[i] == '\n' && data[i + 1] == '\r') - title_end = i + 1; - else title_end = i; - /* stepping back */ - i -= 1; - while (i > title_offset && data[i] == ' ') - i -= 1; - if (i > title_offset - && (data[i] == '\'' || data[i] == '"' || data[i] == ')')) { - line_end = title_end; - title_end = i; } } - - if (!line_end || link_end == link_offset) - return 0; /* garbage after the link empty link */ - - /* a valid ref has been found, filling-in return structures */ - if (last) - *last = line_end; - - if (refs) { - struct link_ref *ref; - - ref = add_link_ref(refs, data + id_offset, id_end - id_offset); - if (!ref) - return 0; - - ref->link = hoedown_buffer_new(link_end - link_offset); - hoedown_buffer_put(ref->link, data + link_offset, link_end - link_offset); - - if (title_end > title_offset) { - ref->title = hoedown_buffer_new(title_end - title_offset); - hoedown_buffer_put(ref->title, data + title_offset, title_end - title_offset); - } - } - - return 1; +/* int n; */ + size_t i = 0; + size_t id_offset, id_end; + size_t link_offset, link_end; + size_t title_offset, title_end; + size_t line_end; + + /* up to 3 optional leading spaces */ + if (beg + 3 >= end) return 0; + if (data[beg] == ' ') { i = 1; + if (data[beg + 1] == ' ') { i = 2; + if (data[beg + 2] == ' ') { i = 3; + if (data[beg + 3] == ' ') return 0; } } } + i += beg; + + /* id part: anything but a newline between brackets */ + if (data[i] != '[') return 0; + i++; + id_offset = i; + while (i < end && data[i] != '\n' && data[i] != '\r' && data[i] != ']') + i++; + if (i >= end || data[i] != ']') return 0; + id_end = i; + + /* spacer: colon (space | tab)* newline? (space | tab)* */ + i++; + if (i >= end || data[i] != ':') return 0; + i++; + while (i < end && data[i] == ' ') i++; + if (i < end && (data[i] == '\n' || data[i] == '\r')) { + i++; + if (i < end && data[i] == '\r' && data[i - 1] == '\n') i++; } + while (i < end && data[i] == ' ') i++; + if (i >= end) return 0; + + /* link: spacing-free sequence, optionally between angle brackets */ + if (data[i] == '<') + i++; + + link_offset = i; + + while (i < end && data[i] != ' ' && data[i] != '\n' && data[i] != '\r') + i++; + + if (data[i - 1] == '>') link_end = i - 1; + else link_end = i; + + /* optional spacer: (space | tab)* (newline | '\'' | '"' | '(' ) */ + while (i < end && data[i] == ' ') i++; + if (i < end && data[i] != '\n' && data[i] != '\r' + && data[i] != '\'' && data[i] != '"' && data[i] != '(') + return 0; + line_end = 0; + /* computing end-of-line */ + if (i >= end || data[i] == '\r' || data[i] == '\n') line_end = i; + if (i + 1 < end && data[i] == '\n' && data[i + 1] == '\r') + line_end = i + 1; + + /* optional (space|tab)* spacer after a newline */ + if (line_end) { + i = line_end + 1; + while (i < end && data[i] == ' ') i++; } + + /* optional title: any non-newline sequence enclosed in '"() + alone on its line */ + title_offset = title_end = 0; + if (i + 1 < end + && (data[i] == '\'' || data[i] == '"' || data[i] == '(')) { + i++; + title_offset = i; + /* looking for EOL */ + while (i < end && data[i] != '\n' && data[i] != '\r') i++; + if (i + 1 < end && data[i] == '\n' && data[i + 1] == '\r') + title_end = i + 1; + else title_end = i; + /* stepping back */ + i -= 1; + while (i > title_offset && data[i] == ' ') + i -= 1; + if (i > title_offset + && (data[i] == '\'' || data[i] == '"' || data[i] == ')')) { + line_end = title_end; + title_end = i; } } + + if (!line_end || link_end == link_offset) + return 0; /* garbage after the link empty link */ + + /* a valid ref has been found, filling-in return structures */ + if (last) + *last = line_end; + + if (refs) { + struct link_ref *ref; + + ref = add_link_ref(refs, data + id_offset, id_end - id_offset); + if (!ref) + return 0; + + ref->link = hoedown_buffer_new(link_end - link_offset); + hoedown_buffer_put(ref->link, data + link_offset, link_end - link_offset); + + if (title_end > title_offset) { + ref->title = hoedown_buffer_new(title_end - title_offset); + hoedown_buffer_put(ref->title, data + title_offset, title_end - title_offset); + } + } + + return 1; } static void expand_tabs(hoedown_buffer *ob, const uint8_t *line, size_t size) { - /* This code makes two assumptions: - * - Input is valid UTF-8. (Any byte with top two bits 10 is skipped, - * whether or not it is a valid UTF-8 continuation byte.) - * - Input contains no combining characters. (Combining characters - * should be skipped but are not.) - */ - size_t i = 0, tab = 0; + /* This code makes two assumptions: + * - Input is valid UTF-8. (Any byte with top two bits 10 is skipped, + * whether or not it is a valid UTF-8 continuation byte.) + * - Input contains no combining characters. (Combining characters + * should be skipped but are not.) + */ + size_t i = 0, tab = 0; - while (i < size) { - size_t org = i; + while (i < size) { + size_t org = i; - while (i < size && line[i] != '\t') { - /* ignore UTF-8 continuation bytes */ - if ((line[i] & 0xc0) != 0x80) - tab++; - i++; - } + while (i < size && line[i] != '\t') { + /* ignore UTF-8 continuation bytes */ + if ((line[i] & 0xc0) != 0x80) + tab++; + i++; + } - if (i > org) - hoedown_buffer_put(ob, line + org, i - org); + if (i > org) + hoedown_buffer_put(ob, line + org, i - org); - if (i >= size) - break; + if (i >= size) + break; - do { - hoedown_buffer_putc(ob, ' '); tab++; - } while (tab % 4); + do { + hoedown_buffer_putc(ob, ' '); tab++; + } while (tab % 4); - i++; - } + i++; + } } /********************** @@ -2734,225 +2734,225 @@ static void expand_tabs(hoedown_buffer *ob, const uint8_t *line, size_t size) hoedown_document * hoedown_document_new( - const hoedown_renderer *renderer, - hoedown_extensions extensions, - size_t max_nesting) + const hoedown_renderer *renderer, + hoedown_extensions extensions, + size_t max_nesting) { - hoedown_document *doc = NULL; + hoedown_document *doc = NULL; - assert(max_nesting > 0 && renderer); + assert(max_nesting > 0 && renderer); - doc = hoedown_malloc(sizeof(hoedown_document)); - memcpy(&doc->md, renderer, sizeof(hoedown_renderer)); + doc = hoedown_malloc(sizeof(hoedown_document)); + memcpy(&doc->md, renderer, sizeof(hoedown_renderer)); - doc->data.opaque = renderer->opaque; + doc->data.opaque = renderer->opaque; - hoedown_stack_init(&doc->work_bufs[BUFFER_BLOCK], 4); - hoedown_stack_init(&doc->work_bufs[BUFFER_SPAN], 8); + hoedown_stack_init(&doc->work_bufs[BUFFER_BLOCK], 4); + hoedown_stack_init(&doc->work_bufs[BUFFER_SPAN], 8); - memset(doc->active_char, 0x0, 256); + memset(doc->active_char, 0x0, 256); - if (extensions & HOEDOWN_EXT_UNDERLINE && doc->md.underline) { - doc->active_char['_'] = MD_CHAR_EMPHASIS; - } + if (extensions & HOEDOWN_EXT_UNDERLINE && doc->md.underline) { + doc->active_char['_'] = MD_CHAR_EMPHASIS; + } - if (doc->md.emphasis || doc->md.double_emphasis || doc->md.triple_emphasis) { - doc->active_char['*'] = MD_CHAR_EMPHASIS; - doc->active_char['_'] = MD_CHAR_EMPHASIS; - if (extensions & HOEDOWN_EXT_STRIKETHROUGH) - doc->active_char['~'] = MD_CHAR_EMPHASIS; - if (extensions & HOEDOWN_EXT_HIGHLIGHT) - doc->active_char['='] = MD_CHAR_EMPHASIS; - } + if (doc->md.emphasis || doc->md.double_emphasis || doc->md.triple_emphasis) { + doc->active_char['*'] = MD_CHAR_EMPHASIS; + doc->active_char['_'] = MD_CHAR_EMPHASIS; + if (extensions & HOEDOWN_EXT_STRIKETHROUGH) + doc->active_char['~'] = MD_CHAR_EMPHASIS; + if (extensions & HOEDOWN_EXT_HIGHLIGHT) + doc->active_char['='] = MD_CHAR_EMPHASIS; + } - if (doc->md.codespan) - doc->active_char['`'] = MD_CHAR_CODESPAN; + if (doc->md.codespan) + doc->active_char['`'] = MD_CHAR_CODESPAN; - if (doc->md.linebreak) - doc->active_char['\n'] = MD_CHAR_LINEBREAK; + if (doc->md.linebreak) + doc->active_char['\n'] = MD_CHAR_LINEBREAK; - if (doc->md.image || doc->md.link || doc->md.footnotes || doc->md.footnote_ref) - doc->active_char['['] = MD_CHAR_LINK; + if (doc->md.image || doc->md.link || doc->md.footnotes || doc->md.footnote_ref) + doc->active_char['['] = MD_CHAR_LINK; - doc->active_char['<'] = MD_CHAR_LANGLE; - doc->active_char['\\'] = MD_CHAR_ESCAPE; - doc->active_char['&'] = MD_CHAR_ENTITY; + doc->active_char['<'] = MD_CHAR_LANGLE; + doc->active_char['\\'] = MD_CHAR_ESCAPE; + doc->active_char['&'] = MD_CHAR_ENTITY; - if (extensions & HOEDOWN_EXT_AUTOLINK) { - doc->active_char[':'] = MD_CHAR_AUTOLINK_URL; - doc->active_char['@'] = MD_CHAR_AUTOLINK_EMAIL; - doc->active_char['w'] = MD_CHAR_AUTOLINK_WWW; - } + if (extensions & HOEDOWN_EXT_AUTOLINK) { + doc->active_char[':'] = MD_CHAR_AUTOLINK_URL; + doc->active_char['@'] = MD_CHAR_AUTOLINK_EMAIL; + doc->active_char['w'] = MD_CHAR_AUTOLINK_WWW; + } - if (extensions & HOEDOWN_EXT_SUPERSCRIPT) - doc->active_char['^'] = MD_CHAR_SUPERSCRIPT; + if (extensions & HOEDOWN_EXT_SUPERSCRIPT) + doc->active_char['^'] = MD_CHAR_SUPERSCRIPT; - if (extensions & HOEDOWN_EXT_QUOTE) - doc->active_char['"'] = MD_CHAR_QUOTE; + if (extensions & HOEDOWN_EXT_QUOTE) + doc->active_char['"'] = MD_CHAR_QUOTE; - if (extensions & HOEDOWN_EXT_MATH) - doc->active_char['$'] = MD_CHAR_MATH; + if (extensions & HOEDOWN_EXT_MATH) + doc->active_char['$'] = MD_CHAR_MATH; - /* Extension data */ - doc->ext_flags = extensions; - doc->max_nesting = max_nesting; - doc->in_link_body = 0; + /* Extension data */ + doc->ext_flags = extensions; + doc->max_nesting = max_nesting; + doc->in_link_body = 0; - return doc; + return doc; } void hoedown_document_render(hoedown_document *doc, hoedown_buffer *ob, const uint8_t *data, size_t size) { - static const uint8_t UTF8_BOM[] = {0xEF, 0xBB, 0xBF}; + static const uint8_t UTF8_BOM[] = {0xEF, 0xBB, 0xBF}; - hoedown_buffer *text; - size_t beg, end; + hoedown_buffer *text; + size_t beg, end; - int footnotes_enabled; + int footnotes_enabled; - text = hoedown_buffer_new(64); + text = hoedown_buffer_new(64); - /* Preallocate enough space for our buffer to avoid expanding while copying */ - hoedown_buffer_grow(text, size); + /* Preallocate enough space for our buffer to avoid expanding while copying */ + hoedown_buffer_grow(text, size); - /* reset the references table */ - memset(&doc->refs, 0x0, REF_TABLE_SIZE * sizeof(void *)); + /* reset the references table */ + memset(&doc->refs, 0x0, REF_TABLE_SIZE * sizeof(void *)); - footnotes_enabled = doc->ext_flags & HOEDOWN_EXT_FOOTNOTES; + footnotes_enabled = doc->ext_flags & HOEDOWN_EXT_FOOTNOTES; - /* reset the footnotes lists */ - if (footnotes_enabled) { - memset(&doc->footnotes_found, 0x0, sizeof(doc->footnotes_found)); - memset(&doc->footnotes_used, 0x0, sizeof(doc->footnotes_used)); - } + /* reset the footnotes lists */ + if (footnotes_enabled) { + memset(&doc->footnotes_found, 0x0, sizeof(doc->footnotes_found)); + memset(&doc->footnotes_used, 0x0, sizeof(doc->footnotes_used)); + } - /* first pass: looking for references, copying everything else */ - beg = 0; + /* first pass: looking for references, copying everything else */ + beg = 0; - /* Skip a possible UTF-8 BOM, even though the Unicode standard - * discourages having these in UTF-8 documents */ - if (size >= 3 && memcmp(data, UTF8_BOM, 3) == 0) - beg += 3; + /* Skip a possible UTF-8 BOM, even though the Unicode standard + * discourages having these in UTF-8 documents */ + if (size >= 3 && memcmp(data, UTF8_BOM, 3) == 0) + beg += 3; - while (beg < size) /* iterating over lines */ - if (footnotes_enabled && is_footnote(data, beg, size, &end, &doc->footnotes_found)) - beg = end; - else if (is_ref(data, beg, size, &end, doc->refs)) - beg = end; - else { /* skipping to the next line */ - end = beg; - while (end < size && data[end] != '\n' && data[end] != '\r') - end++; + while (beg < size) /* iterating over lines */ + if (footnotes_enabled && is_footnote(data, beg, size, &end, &doc->footnotes_found)) + beg = end; + else if (is_ref(data, beg, size, &end, doc->refs)) + beg = end; + else { /* skipping to the next line */ + end = beg; + while (end < size && data[end] != '\n' && data[end] != '\r') + end++; - /* adding the line body if present */ - if (end > beg) - expand_tabs(text, data + beg, end - beg); + /* adding the line body if present */ + if (end > beg) + expand_tabs(text, data + beg, end - beg); - while (end < size && (data[end] == '\n' || data[end] == '\r')) { - /* add one \n per newline */ - if (data[end] == '\n' || (end + 1 < size && data[end + 1] != '\n')) - hoedown_buffer_putc(text, '\n'); - end++; - } + while (end < size && (data[end] == '\n' || data[end] == '\r')) { + /* add one \n per newline */ + if (data[end] == '\n' || (end + 1 < size && data[end + 1] != '\n')) + hoedown_buffer_putc(text, '\n'); + end++; + } - beg = end; - } + beg = end; + } - /* pre-grow the output buffer to minimize allocations */ - hoedown_buffer_grow(ob, text->size + (text->size >> 1)); + /* pre-grow the output buffer to minimize allocations */ + hoedown_buffer_grow(ob, text->size + (text->size >> 1)); - /* second pass: actual rendering */ - if (doc->md.doc_header) - doc->md.doc_header(ob, 0, &doc->data); + /* second pass: actual rendering */ + if (doc->md.doc_header) + doc->md.doc_header(ob, 0, &doc->data); - if (text->size) { - /* adding a final newline if not already present */ - if (text->data[text->size - 1] != '\n' && text->data[text->size - 1] != '\r') - hoedown_buffer_putc(text, '\n'); + if (text->size) { + /* adding a final newline if not already present */ + if (text->data[text->size - 1] != '\n' && text->data[text->size - 1] != '\r') + hoedown_buffer_putc(text, '\n'); - parse_block(ob, doc, text->data, text->size); - } + parse_block(ob, doc, text->data, text->size); + } - /* footnotes */ - if (footnotes_enabled) - parse_footnote_list(ob, doc, &doc->footnotes_used); + /* footnotes */ + if (footnotes_enabled) + parse_footnote_list(ob, doc, &doc->footnotes_used); - if (doc->md.doc_footer) - doc->md.doc_footer(ob, 0, &doc->data); + if (doc->md.doc_footer) + doc->md.doc_footer(ob, 0, &doc->data); - /* clean-up */ - hoedown_buffer_free(text); - free_link_refs(doc->refs); - if (footnotes_enabled) { - free_footnote_list(&doc->footnotes_found, 1); - free_footnote_list(&doc->footnotes_used, 0); - } + /* clean-up */ + hoedown_buffer_free(text); + free_link_refs(doc->refs); + if (footnotes_enabled) { + free_footnote_list(&doc->footnotes_found, 1); + free_footnote_list(&doc->footnotes_used, 0); + } - assert(doc->work_bufs[BUFFER_SPAN].size == 0); - assert(doc->work_bufs[BUFFER_BLOCK].size == 0); + assert(doc->work_bufs[BUFFER_SPAN].size == 0); + assert(doc->work_bufs[BUFFER_BLOCK].size == 0); } void hoedown_document_render_inline(hoedown_document *doc, hoedown_buffer *ob, const uint8_t *data, size_t size) { - size_t i = 0, mark; - hoedown_buffer *text = hoedown_buffer_new(64); + size_t i = 0, mark; + hoedown_buffer *text = hoedown_buffer_new(64); - /* reset the references table */ - memset(&doc->refs, 0x0, REF_TABLE_SIZE * sizeof(void *)); + /* reset the references table */ + memset(&doc->refs, 0x0, REF_TABLE_SIZE * sizeof(void *)); - /* first pass: expand tabs and process newlines */ - hoedown_buffer_grow(text, size); - while (1) { - mark = i; - while (i < size && data[i] != '\n' && data[i] != '\r') - i++; + /* first pass: expand tabs and process newlines */ + hoedown_buffer_grow(text, size); + while (1) { + mark = i; + while (i < size && data[i] != '\n' && data[i] != '\r') + i++; - expand_tabs(text, data + mark, i - mark); + expand_tabs(text, data + mark, i - mark); - if (i >= size) - break; + if (i >= size) + break; - while (i < size && (data[i] == '\n' || data[i] == '\r')) { - /* add one \n per newline */ - if (data[i] == '\n' || (i + 1 < size && data[i + 1] != '\n')) - hoedown_buffer_putc(text, '\n'); - i++; - } - } + while (i < size && (data[i] == '\n' || data[i] == '\r')) { + /* add one \n per newline */ + if (data[i] == '\n' || (i + 1 < size && data[i + 1] != '\n')) + hoedown_buffer_putc(text, '\n'); + i++; + } + } - /* second pass: actual rendering */ - hoedown_buffer_grow(ob, text->size + (text->size >> 1)); + /* second pass: actual rendering */ + hoedown_buffer_grow(ob, text->size + (text->size >> 1)); - if (doc->md.doc_header) - doc->md.doc_header(ob, 1, &doc->data); + if (doc->md.doc_header) + doc->md.doc_header(ob, 1, &doc->data); - parse_inline(ob, doc, text->data, text->size); + parse_inline(ob, doc, text->data, text->size); - if (doc->md.doc_footer) - doc->md.doc_footer(ob, 1, &doc->data); + if (doc->md.doc_footer) + doc->md.doc_footer(ob, 1, &doc->data); - /* clean-up */ - hoedown_buffer_free(text); + /* clean-up */ + hoedown_buffer_free(text); - assert(doc->work_bufs[BUFFER_SPAN].size == 0); - assert(doc->work_bufs[BUFFER_BLOCK].size == 0); + assert(doc->work_bufs[BUFFER_SPAN].size == 0); + assert(doc->work_bufs[BUFFER_BLOCK].size == 0); } void hoedown_document_free(hoedown_document *doc) { - size_t i; + size_t i; - for (i = 0; i < (size_t)doc->work_bufs[BUFFER_SPAN].asize; ++i) - hoedown_buffer_free(doc->work_bufs[BUFFER_SPAN].item[i]); + for (i = 0; i < (size_t)doc->work_bufs[BUFFER_SPAN].asize; ++i) + hoedown_buffer_free(doc->work_bufs[BUFFER_SPAN].item[i]); - for (i = 0; i < (size_t)doc->work_bufs[BUFFER_BLOCK].asize; ++i) - hoedown_buffer_free(doc->work_bufs[BUFFER_BLOCK].item[i]); + for (i = 0; i < (size_t)doc->work_bufs[BUFFER_BLOCK].asize; ++i) + hoedown_buffer_free(doc->work_bufs[BUFFER_BLOCK].item[i]); - hoedown_stack_uninit(&doc->work_bufs[BUFFER_SPAN]); - hoedown_stack_uninit(&doc->work_bufs[BUFFER_BLOCK]); + hoedown_stack_uninit(&doc->work_bufs[BUFFER_SPAN]); + hoedown_stack_uninit(&doc->work_bufs[BUFFER_BLOCK]); - free(doc); + free(doc); } diff --git a/libraries/hoedown/src/escape.c b/libraries/hoedown/src/escape.c index b4a275ba..ce25dd54 100644 --- a/libraries/hoedown/src/escape.c +++ b/libraries/hoedown/src/escape.c @@ -12,13 +12,13 @@ /* * The following characters will not be escaped: * - * -_.+!*'(),%#@?=;:/,+&$ alphanum + * -_.+!*'(),%#@?=;:/,+&$ alphanum * * Note that this character set is the addition of: * - * - The characters which are safe to be in an URL - * - The characters which are *not* safe to be in - * an URL because they are RESERVED characters. + * - The characters which are safe to be in an URL + * - The characters which are *not* safe to be in + * an URL because they are RESERVED characters. * * We assume (lazily) that any RESERVED char that * appears inside an URL is actually meant to @@ -35,84 +35,84 @@ * */ static const uint8_t HREF_SAFE[UINT8_MAX+1] = { - 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, - 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, - 0, 1, 0, 1, 1, 1, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, - 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 1, 0, 1, - 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, - 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 1, - 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, - 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, - 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, - 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, - 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, - 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, - 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, - 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, - 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, - 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 1, 0, 1, 1, 1, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 1, 0, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 1, + 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, }; void hoedown_escape_href(hoedown_buffer *ob, const uint8_t *data, size_t size) { - static const char hex_chars[] = "0123456789ABCDEF"; - size_t i = 0, mark; - char hex_str[3]; - - hex_str[0] = '%'; - - while (i < size) { - mark = i; - while (i < size && HREF_SAFE[data[i]]) i++; - - /* Optimization for cases where there's nothing to escape */ - if (mark == 0 && i >= size) { - hoedown_buffer_put(ob, data, size); - return; - } - - if (likely(i > mark)) { - hoedown_buffer_put(ob, data + mark, i - mark); - } - - /* escaping */ - if (i >= size) - break; - - switch (data[i]) { - /* amp appears all the time in URLs, but needs - * HTML-entity escaping to be inside an href */ - case '&': - HOEDOWN_BUFPUTSL(ob, "&"); - break; - - /* the single quote is a valid URL character - * according to the standard; it needs HTML - * entity escaping too */ - case '\'': - HOEDOWN_BUFPUTSL(ob, "'"); - break; - - /* the space can be escaped to %20 or a plus - * sign. we're going with the generic escape - * for now. the plus thing is more commonly seen - * when building GET strings */ + static const char hex_chars[] = "0123456789ABCDEF"; + size_t i = 0, mark; + char hex_str[3]; + + hex_str[0] = '%'; + + while (i < size) { + mark = i; + while (i < size && HREF_SAFE[data[i]]) i++; + + /* Optimization for cases where there's nothing to escape */ + if (mark == 0 && i >= size) { + hoedown_buffer_put(ob, data, size); + return; + } + + if (likely(i > mark)) { + hoedown_buffer_put(ob, data + mark, i - mark); + } + + /* escaping */ + if (i >= size) + break; + + switch (data[i]) { + /* amp appears all the time in URLs, but needs + * HTML-entity escaping to be inside an href */ + case '&': + HOEDOWN_BUFPUTSL(ob, "&"); + break; + + /* the single quote is a valid URL character + * according to the standard; it needs HTML + * entity escaping too */ + case '\'': + HOEDOWN_BUFPUTSL(ob, "'"); + break; + + /* the space can be escaped to %20 or a plus + * sign. we're going with the generic escape + * for now. the plus thing is more commonly seen + * when building GET strings */ #if 0 - case ' ': - hoedown_buffer_putc(ob, '+'); - break; + case ' ': + hoedown_buffer_putc(ob, '+'); + break; #endif - /* every other character goes with a %XX escaping */ - default: - hex_str[1] = hex_chars[(data[i] >> 4) & 0xF]; - hex_str[2] = hex_chars[data[i] & 0xF]; - hoedown_buffer_put(ob, (uint8_t *)hex_str, 3); - } + /* every other character goes with a %XX escaping */ + default: + hex_str[1] = hex_chars[(data[i] >> 4) & 0xF]; + hex_str[2] = hex_chars[data[i] & 0xF]; + hoedown_buffer_put(ob, (uint8_t *)hex_str, 3); + } - i++; - } + i++; + } } @@ -128,22 +128,22 @@ hoedown_escape_href(hoedown_buffer *ob, const uint8_t *data, size_t size) * */ static const uint8_t HTML_ESCAPE_TABLE[UINT8_MAX+1] = { - 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, - 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, - 0, 0, 1, 0, 0, 0, 2, 3, 0, 0, 0, 0, 0, 0, 0, 4, - 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 5, 0, 6, 0, - 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, - 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, - 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, - 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, - 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, - 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, - 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, - 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, - 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, - 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, - 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, - 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 1, 0, 0, 0, 2, 3, 0, 0, 0, 0, 0, 0, 0, 4, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 5, 0, 6, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, }; static const char *HTML_ESCAPES[] = { @@ -159,30 +159,30 @@ static const char *HTML_ESCAPES[] = { void hoedown_escape_html(hoedown_buffer *ob, const uint8_t *data, size_t size, int secure) { - size_t i = 0, mark; + size_t i = 0, mark; - while (1) { - mark = i; - while (i < size && HTML_ESCAPE_TABLE[data[i]] == 0) i++; + while (1) { + mark = i; + while (i < size && HTML_ESCAPE_TABLE[data[i]] == 0) i++; - /* Optimization for cases where there's nothing to escape */ - if (mark == 0 && i >= size) { - hoedown_buffer_put(ob, data, size); - return; - } + /* Optimization for cases where there's nothing to escape */ + if (mark == 0 && i >= size) { + hoedown_buffer_put(ob, data, size); + return; + } - if (likely(i > mark)) - hoedown_buffer_put(ob, data + mark, i - mark); + if (likely(i > mark)) + hoedown_buffer_put(ob, data + mark, i - mark); - if (i >= size) break; + if (i >= size) break; - /* The forward slash is only escaped in secure mode */ - if (!secure && data[i] == '/') { - hoedown_buffer_putc(ob, '/'); - } else { - hoedown_buffer_puts(ob, HTML_ESCAPES[HTML_ESCAPE_TABLE[data[i]]]); - } + /* The forward slash is only escaped in secure mode */ + if (!secure && data[i] == '/') { + hoedown_buffer_putc(ob, '/'); + } else { + hoedown_buffer_puts(ob, HTML_ESCAPES[HTML_ESCAPE_TABLE[data[i]]]); + } - i++; - } + i++; + } } diff --git a/libraries/hoedown/src/html.c b/libraries/hoedown/src/html.c index 4b18d804..8bf3358e 100644 --- a/libraries/hoedown/src/html.c +++ b/libraries/hoedown/src/html.c @@ -12,44 +12,44 @@ hoedown_html_tag hoedown_html_is_tag(const uint8_t *data, size_t size, const char *tagname) { - size_t i; - int closed = 0; + size_t i; + int closed = 0; - if (size < 3 || data[0] != '<') - return HOEDOWN_HTML_TAG_NONE; + if (size < 3 || data[0] != '<') + return HOEDOWN_HTML_TAG_NONE; - i = 1; + i = 1; - if (data[i] == '/') { - closed = 1; - i++; - } + if (data[i] == '/') { + closed = 1; + i++; + } - for (; i < size; ++i, ++tagname) { - if (*tagname == 0) - break; + for (; i < size; ++i, ++tagname) { + if (*tagname == 0) + break; - if (data[i] != *tagname) - return HOEDOWN_HTML_TAG_NONE; - } + if (data[i] != *tagname) + return HOEDOWN_HTML_TAG_NONE; + } - if (i == size) - return HOEDOWN_HTML_TAG_NONE; + if (i == size) + return HOEDOWN_HTML_TAG_NONE; - if (isspace(data[i]) || data[i] == '>') - return closed ? HOEDOWN_HTML_TAG_CLOSE : HOEDOWN_HTML_TAG_OPEN; + if (isspace(data[i]) || data[i] == '>') + return closed ? HOEDOWN_HTML_TAG_CLOSE : HOEDOWN_HTML_TAG_OPEN; - return HOEDOWN_HTML_TAG_NONE; + return HOEDOWN_HTML_TAG_NONE; } static void escape_html(hoedown_buffer *ob, const uint8_t *source, size_t length) { - hoedown_escape_html(ob, source, length, 0); + hoedown_escape_html(ob, source, length, 0); } static void escape_href(hoedown_buffer *ob, const uint8_t *source, size_t length) { - hoedown_escape_href(ob, source, length); + hoedown_escape_href(ob, source, length); } /******************** @@ -58,353 +58,353 @@ static void escape_href(hoedown_buffer *ob, const uint8_t *source, size_t length static int rndr_autolink(hoedown_buffer *ob, const hoedown_buffer *link, hoedown_autolink_type type, const hoedown_renderer_data *data) { - hoedown_html_renderer_state *state = data->opaque; + hoedown_html_renderer_state *state = data->opaque; - if (!link || !link->size) - return 0; + if (!link || !link->size) + return 0; - HOEDOWN_BUFPUTSL(ob, "<a href=\""); - if (type == HOEDOWN_AUTOLINK_EMAIL) - HOEDOWN_BUFPUTSL(ob, "mailto:"); - escape_href(ob, link->data, link->size); + HOEDOWN_BUFPUTSL(ob, "<a href=\""); + if (type == HOEDOWN_AUTOLINK_EMAIL) + HOEDOWN_BUFPUTSL(ob, "mailto:"); + escape_href(ob, link->data, link->size); - if (state->link_attributes) { - hoedown_buffer_putc(ob, '\"'); - state->link_attributes(ob, link, data); - hoedown_buffer_putc(ob, '>'); - } else { - HOEDOWN_BUFPUTSL(ob, "\">"); - } + if (state->link_attributes) { + hoedown_buffer_putc(ob, '\"'); + state->link_attributes(ob, link, data); + hoedown_buffer_putc(ob, '>'); + } else { + HOEDOWN_BUFPUTSL(ob, "\">"); + } - /* - * Pretty printing: if we get an email address as - * an actual URI, e.g. `mailto:foo@bar.com`, we don't - * want to print the `mailto:` prefix - */ - if (hoedown_buffer_prefix(link, "mailto:") == 0) { - escape_html(ob, link->data + 7, link->size - 7); - } else { - escape_html(ob, link->data, link->size); - } + /* + * Pretty printing: if we get an email address as + * an actual URI, e.g. `mailto:foo@bar.com`, we don't + * want to print the `mailto:` prefix + */ + if (hoedown_buffer_prefix(link, "mailto:") == 0) { + escape_html(ob, link->data + 7, link->size - 7); + } else { + escape_html(ob, link->data, link->size); + } - HOEDOWN_BUFPUTSL(ob, "</a>"); + HOEDOWN_BUFPUTSL(ob, "</a>"); - return 1; + return 1; } static void rndr_blockcode(hoedown_buffer *ob, const hoedown_buffer *text, const hoedown_buffer *lang, const hoedown_renderer_data *data) { - if (ob->size) hoedown_buffer_putc(ob, '\n'); + if (ob->size) hoedown_buffer_putc(ob, '\n'); - if (lang) { - HOEDOWN_BUFPUTSL(ob, "<pre><code class=\"language-"); - escape_html(ob, lang->data, lang->size); - HOEDOWN_BUFPUTSL(ob, "\">"); - } else { - HOEDOWN_BUFPUTSL(ob, "<pre><code>"); - } + if (lang) { + HOEDOWN_BUFPUTSL(ob, "<pre><code class=\"language-"); + escape_html(ob, lang->data, lang->size); + HOEDOWN_BUFPUTSL(ob, "\">"); + } else { + HOEDOWN_BUFPUTSL(ob, "<pre><code>"); + } - if (text) - escape_html(ob, text->data, text->size); + if (text) + escape_html(ob, text->data, text->size); - HOEDOWN_BUFPUTSL(ob, "</code></pre>\n"); + HOEDOWN_BUFPUTSL(ob, "</code></pre>\n"); } static void rndr_blockquote(hoedown_buffer *ob, const hoedown_buffer *content, const hoedown_renderer_data *data) { - if (ob->size) hoedown_buffer_putc(ob, '\n'); - HOEDOWN_BUFPUTSL(ob, "<blockquote>\n"); - if (content) hoedown_buffer_put(ob, content->data, content->size); - HOEDOWN_BUFPUTSL(ob, "</blockquote>\n"); + if (ob->size) hoedown_buffer_putc(ob, '\n'); + HOEDOWN_BUFPUTSL(ob, "<blockquote>\n"); + if (content) hoedown_buffer_put(ob, content->data, content->size); + HOEDOWN_BUFPUTSL(ob, "</blockquote>\n"); } static int rndr_codespan(hoedown_buffer *ob, const hoedown_buffer *text, const hoedown_renderer_data *data) { - HOEDOWN_BUFPUTSL(ob, "<code>"); - if (text) escape_html(ob, text->data, text->size); - HOEDOWN_BUFPUTSL(ob, "</code>"); - return 1; + HOEDOWN_BUFPUTSL(ob, "<code>"); + if (text) escape_html(ob, text->data, text->size); + HOEDOWN_BUFPUTSL(ob, "</code>"); + return 1; } static int rndr_strikethrough(hoedown_buffer *ob, const hoedown_buffer *content, const hoedown_renderer_data *data) { - if (!content || !content->size) - return 0; + if (!content || !content->size) + return 0; - HOEDOWN_BUFPUTSL(ob, "<del>"); - hoedown_buffer_put(ob, content->data, content->size); - HOEDOWN_BUFPUTSL(ob, "</del>"); - return 1; + HOEDOWN_BUFPUTSL(ob, "<del>"); + hoedown_buffer_put(ob, content->data, content->size); + HOEDOWN_BUFPUTSL(ob, "</del>"); + return 1; } static int rndr_double_emphasis(hoedown_buffer *ob, const hoedown_buffer *content, const hoedown_renderer_data *data) { - if (!content || !content->size) - return 0; + if (!content || !content->size) + return 0; - HOEDOWN_BUFPUTSL(ob, "<strong>"); - hoedown_buffer_put(ob, content->data, content->size); - HOEDOWN_BUFPUTSL(ob, "</strong>"); + HOEDOWN_BUFPUTSL(ob, "<strong>"); + hoedown_buffer_put(ob, content->data, content->size); + HOEDOWN_BUFPUTSL(ob, "</strong>"); - return 1; + return 1; } static int rndr_emphasis(hoedown_buffer *ob, const hoedown_buffer *content, const hoedown_renderer_data *data) { - if (!content || !content->size) return 0; - HOEDOWN_BUFPUTSL(ob, "<em>"); - if (content) hoedown_buffer_put(ob, content->data, content->size); - HOEDOWN_BUFPUTSL(ob, "</em>"); - return 1; + if (!content || !content->size) return 0; + HOEDOWN_BUFPUTSL(ob, "<em>"); + if (content) hoedown_buffer_put(ob, content->data, content->size); + HOEDOWN_BUFPUTSL(ob, "</em>"); + return 1; } static int rndr_underline(hoedown_buffer *ob, const hoedown_buffer *content, const hoedown_renderer_data *data) { - if (!content || !content->size) - return 0; + if (!content || !content->size) + return 0; - HOEDOWN_BUFPUTSL(ob, "<u>"); - hoedown_buffer_put(ob, content->data, content->size); - HOEDOWN_BUFPUTSL(ob, "</u>"); + HOEDOWN_BUFPUTSL(ob, "<u>"); + hoedown_buffer_put(ob, content->data, content->size); + HOEDOWN_BUFPUTSL(ob, "</u>"); - return 1; + return 1; } static int rndr_highlight(hoedown_buffer *ob, const hoedown_buffer *content, const hoedown_renderer_data *data) { - if (!content || !content->size) - return 0; + if (!content || !content->size) + return 0; - HOEDOWN_BUFPUTSL(ob, "<mark>"); - hoedown_buffer_put(ob, content->data, content->size); - HOEDOWN_BUFPUTSL(ob, "</mark>"); + HOEDOWN_BUFPUTSL(ob, "<mark>"); + hoedown_buffer_put(ob, content->data, content->size); + HOEDOWN_BUFPUTSL(ob, "</mark>"); - return 1; + return 1; } static int rndr_quote(hoedown_buffer *ob, const hoedown_buffer *content, const hoedown_renderer_data *data) { - if (!content || !content->size) - return 0; + if (!content || !content->size) + return 0; - HOEDOWN_BUFPUTSL(ob, "<q>"); - hoedown_buffer_put(ob, content->data, content->size); - HOEDOWN_BUFPUTSL(ob, "</q>"); + HOEDOWN_BUFPUTSL(ob, "<q>"); + hoedown_buffer_put(ob, content->data, content->size); + HOEDOWN_BUFPUTSL(ob, "</q>"); - return 1; + return 1; } static int rndr_linebreak(hoedown_buffer *ob, const hoedown_renderer_data *data) { - hoedown_html_renderer_state *state = data->opaque; - hoedown_buffer_puts(ob, USE_XHTML(state) ? "<br/>\n" : "<br>\n"); - return 1; + hoedown_html_renderer_state *state = data->opaque; + hoedown_buffer_puts(ob, USE_XHTML(state) ? "<br/>\n" : "<br>\n"); + return 1; } static void rndr_header(hoedown_buffer *ob, const hoedown_buffer *content, int level, const hoedown_renderer_data *data) { - hoedown_html_renderer_state *state = data->opaque; + hoedown_html_renderer_state *state = data->opaque; - if (ob->size) - hoedown_buffer_putc(ob, '\n'); + if (ob->size) + hoedown_buffer_putc(ob, '\n'); - if (level <= state->toc_data.nesting_level) - hoedown_buffer_printf(ob, "<h%d id=\"toc_%d\">", level, state->toc_data.header_count++); - else - hoedown_buffer_printf(ob, "<h%d>", level); + if (level <= state->toc_data.nesting_level) + hoedown_buffer_printf(ob, "<h%d id=\"toc_%d\">", level, state->toc_data.header_count++); + else + hoedown_buffer_printf(ob, "<h%d>", level); - if (content) hoedown_buffer_put(ob, content->data, content->size); - hoedown_buffer_printf(ob, "</h%d>\n", level); + if (content) hoedown_buffer_put(ob, content->data, content->size); + hoedown_buffer_printf(ob, "</h%d>\n", level); } static int rndr_link(hoedown_buffer *ob, const hoedown_buffer *content, const hoedown_buffer *link, const hoedown_buffer *title, const hoedown_renderer_data *data) { - hoedown_html_renderer_state *state = data->opaque; + hoedown_html_renderer_state *state = data->opaque; - HOEDOWN_BUFPUTSL(ob, "<a href=\""); + HOEDOWN_BUFPUTSL(ob, "<a href=\""); - if (link && link->size) - escape_href(ob, link->data, link->size); + if (link && link->size) + escape_href(ob, link->data, link->size); - if (title && title->size) { - HOEDOWN_BUFPUTSL(ob, "\" title=\""); - escape_html(ob, title->data, title->size); - } + if (title && title->size) { + HOEDOWN_BUFPUTSL(ob, "\" title=\""); + escape_html(ob, title->data, title->size); + } - if (state->link_attributes) { - hoedown_buffer_putc(ob, '\"'); - state->link_attributes(ob, link, data); - hoedown_buffer_putc(ob, '>'); - } else { - HOEDOWN_BUFPUTSL(ob, "\">"); - } + if (state->link_attributes) { + hoedown_buffer_putc(ob, '\"'); + state->link_attributes(ob, link, data); + hoedown_buffer_putc(ob, '>'); + } else { + HOEDOWN_BUFPUTSL(ob, "\">"); + } - if (content && content->size) hoedown_buffer_put(ob, content->data, content->size); - HOEDOWN_BUFPUTSL(ob, "</a>"); - return 1; + if (content && content->size) hoedown_buffer_put(ob, content->data, content->size); + HOEDOWN_BUFPUTSL(ob, "</a>"); + return 1; } static void rndr_list(hoedown_buffer *ob, const hoedown_buffer *content, hoedown_list_flags flags, const hoedown_renderer_data *data) { - if (ob->size) hoedown_buffer_putc(ob, '\n'); - hoedown_buffer_put(ob, (const uint8_t *)(flags & HOEDOWN_LIST_ORDERED ? "<ol>\n" : "<ul>\n"), 5); - if (content) hoedown_buffer_put(ob, content->data, content->size); - hoedown_buffer_put(ob, (const uint8_t *)(flags & HOEDOWN_LIST_ORDERED ? "</ol>\n" : "</ul>\n"), 6); + if (ob->size) hoedown_buffer_putc(ob, '\n'); + hoedown_buffer_put(ob, (const uint8_t *)(flags & HOEDOWN_LIST_ORDERED ? "<ol>\n" : "<ul>\n"), 5); + if (content) hoedown_buffer_put(ob, content->data, content->size); + hoedown_buffer_put(ob, (const uint8_t *)(flags & HOEDOWN_LIST_ORDERED ? "</ol>\n" : "</ul>\n"), 6); } static void rndr_listitem(hoedown_buffer *ob, const hoedown_buffer *content, hoedown_list_flags flags, const hoedown_renderer_data *data) { - HOEDOWN_BUFPUTSL(ob, "<li>"); - if (content) { - size_t size = content->size; - while (size && content->data[size - 1] == '\n') - size--; + HOEDOWN_BUFPUTSL(ob, "<li>"); + if (content) { + size_t size = content->size; + while (size && content->data[size - 1] == '\n') + size--; - hoedown_buffer_put(ob, content->data, size); - } - HOEDOWN_BUFPUTSL(ob, "</li>\n"); + hoedown_buffer_put(ob, content->data, size); + } + HOEDOWN_BUFPUTSL(ob, "</li>\n"); } static void rndr_paragraph(hoedown_buffer *ob, const hoedown_buffer *content, const hoedown_renderer_data *data) { - hoedown_html_renderer_state *state = data->opaque; - size_t i = 0; + hoedown_html_renderer_state *state = data->opaque; + size_t i = 0; - if (ob->size) hoedown_buffer_putc(ob, '\n'); + if (ob->size) hoedown_buffer_putc(ob, '\n'); - if (!content || !content->size) - return; + if (!content || !content->size) + return; - while (i < content->size && isspace(content->data[i])) i++; + while (i < content->size && isspace(content->data[i])) i++; - if (i == content->size) - return; + if (i == content->size) + return; - HOEDOWN_BUFPUTSL(ob, "<p>"); - if (state->flags & HOEDOWN_HTML_HARD_WRAP) { - size_t org; - while (i < content->size) { - org = i; - while (i < content->size && content->data[i] != '\n') - i++; + HOEDOWN_BUFPUTSL(ob, "<p>"); + if (state->flags & HOEDOWN_HTML_HARD_WRAP) { + size_t org; + while (i < content->size) { + org = i; + while (i < content->size && content->data[i] != '\n') + i++; - if (i > org) - hoedown_buffer_put(ob, content->data + org, i - org); + if (i > org) + hoedown_buffer_put(ob, content->data + org, i - org); - /* - * do not insert a line break if this newline - * is the last character on the paragraph - */ - if (i >= content->size - 1) - break; + /* + * do not insert a line break if this newline + * is the last character on the paragraph + */ + if (i >= content->size - 1) + break; - rndr_linebreak(ob, data); - i++; - } - } else { - hoedown_buffer_put(ob, content->data + i, content->size - i); - } - HOEDOWN_BUFPUTSL(ob, "</p>\n"); + rndr_linebreak(ob, data); + i++; + } + } else { + hoedown_buffer_put(ob, content->data + i, content->size - i); + } + HOEDOWN_BUFPUTSL(ob, "</p>\n"); } static void rndr_raw_block(hoedown_buffer *ob, const hoedown_buffer *text, const hoedown_renderer_data *data) { - size_t org, sz; + size_t org, sz; - if (!text) - return; + if (!text) + return; - /* FIXME: Do we *really* need to trim the HTML? How does that make a difference? */ - sz = text->size; - while (sz > 0 && text->data[sz - 1] == '\n') - sz--; + /* FIXME: Do we *really* need to trim the HTML? How does that make a difference? */ + sz = text->size; + while (sz > 0 && text->data[sz - 1] == '\n') + sz--; - org = 0; - while (org < sz && text->data[org] == '\n') - org++; + org = 0; + while (org < sz && text->data[org] == '\n') + org++; - if (org >= sz) - return; + if (org >= sz) + return; - if (ob->size) - hoedown_buffer_putc(ob, '\n'); + if (ob->size) + hoedown_buffer_putc(ob, '\n'); - hoedown_buffer_put(ob, text->data + org, sz - org); - hoedown_buffer_putc(ob, '\n'); + hoedown_buffer_put(ob, text->data + org, sz - org); + hoedown_buffer_putc(ob, '\n'); } static int rndr_triple_emphasis(hoedown_buffer *ob, const hoedown_buffer *content, const hoedown_renderer_data *data) { - if (!content || !content->size) return 0; - HOEDOWN_BUFPUTSL(ob, "<strong><em>"); - hoedown_buffer_put(ob, content->data, content->size); - HOEDOWN_BUFPUTSL(ob, "</em></strong>"); - return 1; + if (!content || !content->size) return 0; + HOEDOWN_BUFPUTSL(ob, "<strong><em>"); + hoedown_buffer_put(ob, content->data, content->size); + HOEDOWN_BUFPUTSL(ob, "</em></strong>"); + return 1; } static void rndr_hrule(hoedown_buffer *ob, const hoedown_renderer_data *data) { - hoedown_html_renderer_state *state = data->opaque; - if (ob->size) hoedown_buffer_putc(ob, '\n'); - hoedown_buffer_puts(ob, USE_XHTML(state) ? "<hr/>\n" : "<hr>\n"); + hoedown_html_renderer_state *state = data->opaque; + if (ob->size) hoedown_buffer_putc(ob, '\n'); + hoedown_buffer_puts(ob, USE_XHTML(state) ? "<hr/>\n" : "<hr>\n"); } static int rndr_image(hoedown_buffer *ob, const hoedown_buffer *link, const hoedown_buffer *title, const hoedown_buffer *alt, const hoedown_renderer_data *data) { - hoedown_html_renderer_state *state = data->opaque; - if (!link || !link->size) return 0; + hoedown_html_renderer_state *state = data->opaque; + if (!link || !link->size) return 0; - HOEDOWN_BUFPUTSL(ob, "<img src=\""); - escape_href(ob, link->data, link->size); - HOEDOWN_BUFPUTSL(ob, "\" alt=\""); + HOEDOWN_BUFPUTSL(ob, "<img src=\""); + escape_href(ob, link->data, link->size); + HOEDOWN_BUFPUTSL(ob, "\" alt=\""); - if (alt && alt->size) - escape_html(ob, alt->data, alt->size); + if (alt && alt->size) + escape_html(ob, alt->data, alt->size); - if (title && title->size) { - HOEDOWN_BUFPUTSL(ob, "\" title=\""); - escape_html(ob, title->data, title->size); } + if (title && title->size) { + HOEDOWN_BUFPUTSL(ob, "\" title=\""); + escape_html(ob, title->data, title->size); } - hoedown_buffer_puts(ob, USE_XHTML(state) ? "\"/>" : "\">"); - return 1; + hoedown_buffer_puts(ob, USE_XHTML(state) ? "\"/>" : "\">"); + return 1; } static int rndr_raw_html(hoedown_buffer *ob, const hoedown_buffer *text, const hoedown_renderer_data *data) { - hoedown_html_renderer_state *state = data->opaque; + hoedown_html_renderer_state *state = data->opaque; - /* ESCAPE overrides SKIP_HTML. It doesn't look to see if - * there are any valid tags, just escapes all of them. */ - if((state->flags & HOEDOWN_HTML_ESCAPE) != 0) { - escape_html(ob, text->data, text->size); - return 1; - } + /* ESCAPE overrides SKIP_HTML. It doesn't look to see if + * there are any valid tags, just escapes all of them. */ + if((state->flags & HOEDOWN_HTML_ESCAPE) != 0) { + escape_html(ob, text->data, text->size); + return 1; + } - if ((state->flags & HOEDOWN_HTML_SKIP_HTML) != 0) - return 1; + if ((state->flags & HOEDOWN_HTML_SKIP_HTML) != 0) + return 1; - hoedown_buffer_put(ob, text->data, text->size); - return 1; + hoedown_buffer_put(ob, text->data, text->size); + return 1; } static void @@ -437,318 +437,318 @@ rndr_table_body(hoedown_buffer *ob, const hoedown_buffer *content, const hoedown static void rndr_tablerow(hoedown_buffer *ob, const hoedown_buffer *content, const hoedown_renderer_data *data) { - HOEDOWN_BUFPUTSL(ob, "<tr>\n"); - if (content) hoedown_buffer_put(ob, content->data, content->size); - HOEDOWN_BUFPUTSL(ob, "</tr>\n"); + HOEDOWN_BUFPUTSL(ob, "<tr>\n"); + if (content) hoedown_buffer_put(ob, content->data, content->size); + HOEDOWN_BUFPUTSL(ob, "</tr>\n"); } static void rndr_tablecell(hoedown_buffer *ob, const hoedown_buffer *content, hoedown_table_flags flags, const hoedown_renderer_data *data) { - if (flags & HOEDOWN_TABLE_HEADER) { - HOEDOWN_BUFPUTSL(ob, "<th"); - } else { - HOEDOWN_BUFPUTSL(ob, "<td"); - } + if (flags & HOEDOWN_TABLE_HEADER) { + HOEDOWN_BUFPUTSL(ob, "<th"); + } else { + HOEDOWN_BUFPUTSL(ob, "<td"); + } - switch (flags & HOEDOWN_TABLE_ALIGNMASK) { - case HOEDOWN_TABLE_ALIGN_CENTER: - HOEDOWN_BUFPUTSL(ob, " style=\"text-align: center\">"); - break; + switch (flags & HOEDOWN_TABLE_ALIGNMASK) { + case HOEDOWN_TABLE_ALIGN_CENTER: + HOEDOWN_BUFPUTSL(ob, " style=\"text-align: center\">"); + break; - case HOEDOWN_TABLE_ALIGN_LEFT: - HOEDOWN_BUFPUTSL(ob, " style=\"text-align: left\">"); - break; + case HOEDOWN_TABLE_ALIGN_LEFT: + HOEDOWN_BUFPUTSL(ob, " style=\"text-align: left\">"); + break; - case HOEDOWN_TABLE_ALIGN_RIGHT: - HOEDOWN_BUFPUTSL(ob, " style=\"text-align: right\">"); - break; + case HOEDOWN_TABLE_ALIGN_RIGHT: + HOEDOWN_BUFPUTSL(ob, " style=\"text-align: right\">"); + break; - default: - HOEDOWN_BUFPUTSL(ob, ">"); - } + default: + HOEDOWN_BUFPUTSL(ob, ">"); + } - if (content) - hoedown_buffer_put(ob, content->data, content->size); + if (content) + hoedown_buffer_put(ob, content->data, content->size); - if (flags & HOEDOWN_TABLE_HEADER) { - HOEDOWN_BUFPUTSL(ob, "</th>\n"); - } else { - HOEDOWN_BUFPUTSL(ob, "</td>\n"); - } + if (flags & HOEDOWN_TABLE_HEADER) { + HOEDOWN_BUFPUTSL(ob, "</th>\n"); + } else { + HOEDOWN_BUFPUTSL(ob, "</td>\n"); + } } static int rndr_superscript(hoedown_buffer *ob, const hoedown_buffer *content, const hoedown_renderer_data *data) { - if (!content || !content->size) return 0; - HOEDOWN_BUFPUTSL(ob, "<sup>"); - hoedown_buffer_put(ob, content->data, content->size); - HOEDOWN_BUFPUTSL(ob, "</sup>"); - return 1; + if (!content || !content->size) return 0; + HOEDOWN_BUFPUTSL(ob, "<sup>"); + hoedown_buffer_put(ob, content->data, content->size); + HOEDOWN_BUFPUTSL(ob, "</sup>"); + return 1; } static void rndr_normal_text(hoedown_buffer *ob, const hoedown_buffer *content, const hoedown_renderer_data *data) { - if (content) - escape_html(ob, content->data, content->size); + if (content) + escape_html(ob, content->data, content->size); } static void rndr_footnotes(hoedown_buffer *ob, const hoedown_buffer *content, const hoedown_renderer_data *data) { - hoedown_html_renderer_state *state = data->opaque; + hoedown_html_renderer_state *state = data->opaque; - if (ob->size) hoedown_buffer_putc(ob, '\n'); - HOEDOWN_BUFPUTSL(ob, "<div class=\"footnotes\">\n"); - hoedown_buffer_puts(ob, USE_XHTML(state) ? "<hr/>\n" : "<hr>\n"); - HOEDOWN_BUFPUTSL(ob, "<ol>\n"); + if (ob->size) hoedown_buffer_putc(ob, '\n'); + HOEDOWN_BUFPUTSL(ob, "<div class=\"footnotes\">\n"); + hoedown_buffer_puts(ob, USE_XHTML(state) ? "<hr/>\n" : "<hr>\n"); + HOEDOWN_BUFPUTSL(ob, "<ol>\n"); - if (content) hoedown_buffer_put(ob, content->data, content->size); + if (content) hoedown_buffer_put(ob, content->data, content->size); - HOEDOWN_BUFPUTSL(ob, "\n</ol>\n</div>\n"); + HOEDOWN_BUFPUTSL(ob, "\n</ol>\n</div>\n"); } static void rndr_footnote_def(hoedown_buffer *ob, const hoedown_buffer *content, unsigned int num, const hoedown_renderer_data *data) { - size_t i = 0; - int pfound = 0; - - /* insert anchor at the end of first paragraph block */ - if (content) { - while ((i+3) < content->size) { - if (content->data[i++] != '<') continue; - if (content->data[i++] != '/') continue; - if (content->data[i++] != 'p' && content->data[i] != 'P') continue; - if (content->data[i] != '>') continue; - i -= 3; - pfound = 1; - break; - } - } - - hoedown_buffer_printf(ob, "\n<li id=\"fn%d\">\n", num); - if (pfound) { - hoedown_buffer_put(ob, content->data, i); - hoedown_buffer_printf(ob, " <a href=\"#fnref%d\" rev=\"footnote\">↩</a>", num); - hoedown_buffer_put(ob, content->data + i, content->size - i); - } else if (content) { - hoedown_buffer_put(ob, content->data, content->size); - } - HOEDOWN_BUFPUTSL(ob, "</li>\n"); + size_t i = 0; + int pfound = 0; + + /* insert anchor at the end of first paragraph block */ + if (content) { + while ((i+3) < content->size) { + if (content->data[i++] != '<') continue; + if (content->data[i++] != '/') continue; + if (content->data[i++] != 'p' && content->data[i] != 'P') continue; + if (content->data[i] != '>') continue; + i -= 3; + pfound = 1; + break; + } + } + + hoedown_buffer_printf(ob, "\n<li id=\"fn%d\">\n", num); + if (pfound) { + hoedown_buffer_put(ob, content->data, i); + hoedown_buffer_printf(ob, " <a href=\"#fnref%d\" rev=\"footnote\">↩</a>", num); + hoedown_buffer_put(ob, content->data + i, content->size - i); + } else if (content) { + hoedown_buffer_put(ob, content->data, content->size); + } + HOEDOWN_BUFPUTSL(ob, "</li>\n"); } static int rndr_footnote_ref(hoedown_buffer *ob, unsigned int num, const hoedown_renderer_data *data) { - hoedown_buffer_printf(ob, "<sup id=\"fnref%d\"><a href=\"#fn%d\" rel=\"footnote\">%d</a></sup>", num, num, num); - return 1; + hoedown_buffer_printf(ob, "<sup id=\"fnref%d\"><a href=\"#fn%d\" rel=\"footnote\">%d</a></sup>", num, num, num); + return 1; } static int rndr_math(hoedown_buffer *ob, const hoedown_buffer *text, int displaymode, const hoedown_renderer_data *data) { - hoedown_buffer_put(ob, (const uint8_t *)(displaymode ? "\\[" : "\\("), 2); - escape_html(ob, text->data, text->size); - hoedown_buffer_put(ob, (const uint8_t *)(displaymode ? "\\]" : "\\)"), 2); - return 1; + hoedown_buffer_put(ob, (const uint8_t *)(displaymode ? "\\[" : "\\("), 2); + escape_html(ob, text->data, text->size); + hoedown_buffer_put(ob, (const uint8_t *)(displaymode ? "\\]" : "\\)"), 2); + return 1; } static void toc_header(hoedown_buffer *ob, const hoedown_buffer *content, int level, const hoedown_renderer_data *data) { - hoedown_html_renderer_state *state = data->opaque; - - if (level <= state->toc_data.nesting_level) { - /* set the level offset if this is the first header - * we're parsing for the document */ - if (state->toc_data.current_level == 0) - state->toc_data.level_offset = level - 1; - - level -= state->toc_data.level_offset; - - if (level > state->toc_data.current_level) { - while (level > state->toc_data.current_level) { - HOEDOWN_BUFPUTSL(ob, "<ul>\n<li>\n"); - state->toc_data.current_level++; - } - } else if (level < state->toc_data.current_level) { - HOEDOWN_BUFPUTSL(ob, "</li>\n"); - while (level < state->toc_data.current_level) { - HOEDOWN_BUFPUTSL(ob, "</ul>\n</li>\n"); - state->toc_data.current_level--; - } - HOEDOWN_BUFPUTSL(ob,"<li>\n"); - } else { - HOEDOWN_BUFPUTSL(ob,"</li>\n<li>\n"); - } - - hoedown_buffer_printf(ob, "<a href=\"#toc_%d\">", state->toc_data.header_count++); - if (content) hoedown_buffer_put(ob, content->data, content->size); - HOEDOWN_BUFPUTSL(ob, "</a>\n"); - } + hoedown_html_renderer_state *state = data->opaque; + + if (level <= state->toc_data.nesting_level) { + /* set the level offset if this is the first header + * we're parsing for the document */ + if (state->toc_data.current_level == 0) + state->toc_data.level_offset = level - 1; + + level -= state->toc_data.level_offset; + + if (level > state->toc_data.current_level) { + while (level > state->toc_data.current_level) { + HOEDOWN_BUFPUTSL(ob, "<ul>\n<li>\n"); + state->toc_data.current_level++; + } + } else if (level < state->toc_data.current_level) { + HOEDOWN_BUFPUTSL(ob, "</li>\n"); + while (level < state->toc_data.current_level) { + HOEDOWN_BUFPUTSL(ob, "</ul>\n</li>\n"); + state->toc_data.current_level--; + } + HOEDOWN_BUFPUTSL(ob,"<li>\n"); + } else { + HOEDOWN_BUFPUTSL(ob,"</li>\n<li>\n"); + } + + hoedown_buffer_printf(ob, "<a href=\"#toc_%d\">", state->toc_data.header_count++); + if (content) hoedown_buffer_put(ob, content->data, content->size); + HOEDOWN_BUFPUTSL(ob, "</a>\n"); + } } static int toc_link(hoedown_buffer *ob, const hoedown_buffer *content, const hoedown_buffer *link, const hoedown_buffer *title, const hoedown_renderer_data *data) { - if (content && content->size) hoedown_buffer_put(ob, content->data, content->size); - return 1; + if (content && content->size) hoedown_buffer_put(ob, content->data, content->size); + return 1; } static void toc_finalize(hoedown_buffer *ob, int inline_render, const hoedown_renderer_data *data) { - hoedown_html_renderer_state *state; + hoedown_html_renderer_state *state; - if (inline_render) - return; + if (inline_render) + return; - state = data->opaque; + state = data->opaque; - while (state->toc_data.current_level > 0) { - HOEDOWN_BUFPUTSL(ob, "</li>\n</ul>\n"); - state->toc_data.current_level--; - } + while (state->toc_data.current_level > 0) { + HOEDOWN_BUFPUTSL(ob, "</li>\n</ul>\n"); + state->toc_data.current_level--; + } - state->toc_data.header_count = 0; + state->toc_data.header_count = 0; } hoedown_renderer * hoedown_html_toc_renderer_new(int nesting_level) { - static const hoedown_renderer cb_default = { - NULL, - - NULL, - NULL, - toc_header, - NULL, - NULL, - NULL, - NULL, - NULL, - NULL, - NULL, - NULL, - NULL, - NULL, - NULL, - NULL, - - NULL, - rndr_codespan, - rndr_double_emphasis, - rndr_emphasis, - rndr_underline, - rndr_highlight, - rndr_quote, - NULL, - NULL, - toc_link, - rndr_triple_emphasis, - rndr_strikethrough, - rndr_superscript, - NULL, - NULL, - NULL, - - NULL, - rndr_normal_text, - - NULL, - toc_finalize - }; - - hoedown_html_renderer_state *state; - hoedown_renderer *renderer; - - /* Prepare the state pointer */ - state = hoedown_malloc(sizeof(hoedown_html_renderer_state)); - memset(state, 0x0, sizeof(hoedown_html_renderer_state)); - - state->toc_data.nesting_level = nesting_level; - - /* Prepare the renderer */ - renderer = hoedown_malloc(sizeof(hoedown_renderer)); - memcpy(renderer, &cb_default, sizeof(hoedown_renderer)); - - renderer->opaque = state; - return renderer; + static const hoedown_renderer cb_default = { + NULL, + + NULL, + NULL, + toc_header, + NULL, + NULL, + NULL, + NULL, + NULL, + NULL, + NULL, + NULL, + NULL, + NULL, + NULL, + NULL, + + NULL, + rndr_codespan, + rndr_double_emphasis, + rndr_emphasis, + rndr_underline, + rndr_highlight, + rndr_quote, + NULL, + NULL, + toc_link, + rndr_triple_emphasis, + rndr_strikethrough, + rndr_superscript, + NULL, + NULL, + NULL, + + NULL, + rndr_normal_text, + + NULL, + toc_finalize + }; + + hoedown_html_renderer_state *state; + hoedown_renderer *renderer; + + /* Prepare the state pointer */ + state = hoedown_malloc(sizeof(hoedown_html_renderer_state)); + memset(state, 0x0, sizeof(hoedown_html_renderer_state)); + + state->toc_data.nesting_level = nesting_level; + + /* Prepare the renderer */ + renderer = hoedown_malloc(sizeof(hoedown_renderer)); + memcpy(renderer, &cb_default, sizeof(hoedown_renderer)); + + renderer->opaque = state; + return renderer; } hoedown_renderer * hoedown_html_renderer_new(hoedown_html_flags render_flags, int nesting_level) { - static const hoedown_renderer cb_default = { - NULL, - - rndr_blockcode, - rndr_blockquote, - rndr_header, - rndr_hrule, - rndr_list, - rndr_listitem, - rndr_paragraph, - rndr_table, - rndr_table_header, - rndr_table_body, - rndr_tablerow, - rndr_tablecell, - rndr_footnotes, - rndr_footnote_def, - rndr_raw_block, - - rndr_autolink, - rndr_codespan, - rndr_double_emphasis, - rndr_emphasis, - rndr_underline, - rndr_highlight, - rndr_quote, - rndr_image, - rndr_linebreak, - rndr_link, - rndr_triple_emphasis, - rndr_strikethrough, - rndr_superscript, - rndr_footnote_ref, - rndr_math, - rndr_raw_html, - - NULL, - rndr_normal_text, - - NULL, - NULL - }; - - hoedown_html_renderer_state *state; - hoedown_renderer *renderer; - - /* Prepare the state pointer */ - state = hoedown_malloc(sizeof(hoedown_html_renderer_state)); - memset(state, 0x0, sizeof(hoedown_html_renderer_state)); - - state->flags = render_flags; - state->toc_data.nesting_level = nesting_level; - - /* Prepare the renderer */ - renderer = hoedown_malloc(sizeof(hoedown_renderer)); - memcpy(renderer, &cb_default, sizeof(hoedown_renderer)); - - if (render_flags & HOEDOWN_HTML_SKIP_HTML || render_flags & HOEDOWN_HTML_ESCAPE) - renderer->blockhtml = NULL; - - renderer->opaque = state; - return renderer; + static const hoedown_renderer cb_default = { + NULL, + + rndr_blockcode, + rndr_blockquote, + rndr_header, + rndr_hrule, + rndr_list, + rndr_listitem, + rndr_paragraph, + rndr_table, + rndr_table_header, + rndr_table_body, + rndr_tablerow, + rndr_tablecell, + rndr_footnotes, + rndr_footnote_def, + rndr_raw_block, + + rndr_autolink, + rndr_codespan, + rndr_double_emphasis, + rndr_emphasis, + rndr_underline, + rndr_highlight, + rndr_quote, + rndr_image, + rndr_linebreak, + rndr_link, + rndr_triple_emphasis, + rndr_strikethrough, + rndr_superscript, + rndr_footnote_ref, + rndr_math, + rndr_raw_html, + + NULL, + rndr_normal_text, + + NULL, + NULL + }; + + hoedown_html_renderer_state *state; + hoedown_renderer *renderer; + + /* Prepare the state pointer */ + state = hoedown_malloc(sizeof(hoedown_html_renderer_state)); + memset(state, 0x0, sizeof(hoedown_html_renderer_state)); + + state->flags = render_flags; + state->toc_data.nesting_level = nesting_level; + + /* Prepare the renderer */ + renderer = hoedown_malloc(sizeof(hoedown_renderer)); + memcpy(renderer, &cb_default, sizeof(hoedown_renderer)); + + if (render_flags & HOEDOWN_HTML_SKIP_HTML || render_flags & HOEDOWN_HTML_ESCAPE) + renderer->blockhtml = NULL; + + renderer->opaque = state; + return renderer; } void hoedown_html_renderer_free(hoedown_renderer *renderer) { - free(renderer->opaque); - free(renderer); + free(renderer->opaque); + free(renderer); } diff --git a/libraries/hoedown/src/html_smartypants.c b/libraries/hoedown/src/html_smartypants.c index e24b6bf0..d89624f3 100644 --- a/libraries/hoedown/src/html_smartypants.c +++ b/libraries/hoedown/src/html_smartypants.c @@ -10,8 +10,8 @@ #endif struct smartypants_data { - int in_squote; - int in_dquote; + int in_squote; + int in_dquote; }; static size_t smartypants_cb__ltag(hoedown_buffer *ob, struct smartypants_data *smrt, uint8_t previous_char, const uint8_t *text, size_t size); @@ -26,353 +26,353 @@ static size_t smartypants_cb__backtick(hoedown_buffer *ob, struct smartypants_da static size_t smartypants_cb__escape(hoedown_buffer *ob, struct smartypants_data *smrt, uint8_t previous_char, const uint8_t *text, size_t size); static size_t (*smartypants_cb_ptrs[]) - (hoedown_buffer *, struct smartypants_data *, uint8_t, const uint8_t *, size_t) = + (hoedown_buffer *, struct smartypants_data *, uint8_t, const uint8_t *, size_t) = { - NULL, /* 0 */ - smartypants_cb__dash, /* 1 */ - smartypants_cb__parens, /* 2 */ - smartypants_cb__squote, /* 3 */ - smartypants_cb__dquote, /* 4 */ - smartypants_cb__amp, /* 5 */ - smartypants_cb__period, /* 6 */ - smartypants_cb__number, /* 7 */ - smartypants_cb__ltag, /* 8 */ - smartypants_cb__backtick, /* 9 */ - smartypants_cb__escape, /* 10 */ + NULL, /* 0 */ + smartypants_cb__dash, /* 1 */ + smartypants_cb__parens, /* 2 */ + smartypants_cb__squote, /* 3 */ + smartypants_cb__dquote, /* 4 */ + smartypants_cb__amp, /* 5 */ + smartypants_cb__period, /* 6 */ + smartypants_cb__number, /* 7 */ + smartypants_cb__ltag, /* 8 */ + smartypants_cb__backtick, /* 9 */ + smartypants_cb__escape, /* 10 */ }; static const uint8_t smartypants_cb_chars[UINT8_MAX+1] = { - 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, - 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, - 0, 0, 4, 0, 0, 0, 5, 3, 2, 0, 0, 0, 0, 1, 6, 0, - 0, 7, 0, 7, 0, 0, 0, 0, 0, 0, 0, 0, 8, 0, 0, 0, - 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, - 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 10, 0, 0, 0, - 9, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, - 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, - 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, - 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, - 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, - 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, - 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, - 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, - 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, - 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 4, 0, 0, 0, 5, 3, 2, 0, 0, 0, 0, 1, 6, 0, + 0, 7, 0, 7, 0, 0, 0, 0, 0, 0, 0, 0, 8, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 10, 0, 0, 0, + 9, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, }; static int word_boundary(uint8_t c) { - return c == 0 || isspace(c) || ispunct(c); + return c == 0 || isspace(c) || ispunct(c); } /* - If 'text' begins with any kind of single quote (e.g. "'" or "'" etc.), - returns the length of the sequence of characters that makes up the single- - quote. Otherwise, returns zero. + If 'text' begins with any kind of single quote (e.g. "'" or "'" etc.), + returns the length of the sequence of characters that makes up the single- + quote. Otherwise, returns zero. */ static size_t squote_len(const uint8_t *text, size_t size) { - static char* single_quote_list[] = { "'", "'", "'", "'", NULL }; - char** p; + static char* single_quote_list[] = { "'", "'", "'", "'", NULL }; + char** p; - for (p = single_quote_list; *p; ++p) { - size_t len = strlen(*p); - if (size >= len && memcmp(text, *p, len) == 0) { - return len; - } - } + for (p = single_quote_list; *p; ++p) { + size_t len = strlen(*p); + if (size >= len && memcmp(text, *p, len) == 0) { + return len; + } + } - return 0; + return 0; } /* Converts " or ' at very beginning or end of a word to left or right quote */ static int smartypants_quotes(hoedown_buffer *ob, uint8_t previous_char, uint8_t next_char, uint8_t quote, int *is_open) { - char ent[8]; + char ent[8]; - if (*is_open && !word_boundary(next_char)) - return 0; + if (*is_open && !word_boundary(next_char)) + return 0; - if (!(*is_open) && !word_boundary(previous_char)) - return 0; + if (!(*is_open) && !word_boundary(previous_char)) + return 0; - snprintf(ent, sizeof(ent), "&%c%cquo;", (*is_open) ? 'r' : 'l', quote); - *is_open = !(*is_open); - hoedown_buffer_puts(ob, ent); - return 1; + snprintf(ent, sizeof(ent), "&%c%cquo;", (*is_open) ? 'r' : 'l', quote); + *is_open = !(*is_open); + hoedown_buffer_puts(ob, ent); + return 1; } /* - Converts ' to left or right single quote; but the initial ' might be in - different forms, e.g. ' or ' or '. - 'squote_text' points to the original single quote, and 'squote_size' is its length. - 'text' points at the last character of the single-quote, e.g. ' or ; + Converts ' to left or right single quote; but the initial ' might be in + different forms, e.g. ' or ' or '. + 'squote_text' points to the original single quote, and 'squote_size' is its length. + 'text' points at the last character of the single-quote, e.g. ' or ; */ static size_t smartypants_squote(hoedown_buffer *ob, struct smartypants_data *smrt, uint8_t previous_char, const uint8_t *text, size_t size, - const uint8_t *squote_text, size_t squote_size) + const uint8_t *squote_text, size_t squote_size) { - if (size >= 2) { - uint8_t t1 = tolower(text[1]); - size_t next_squote_len = squote_len(text+1, size-1); - - /* convert '' to “ or ” */ - if (next_squote_len > 0) { - uint8_t next_char = (size > 1+next_squote_len) ? text[1+next_squote_len] : 0; - if (smartypants_quotes(ob, previous_char, next_char, 'd', &smrt->in_dquote)) - return next_squote_len; - } - - /* Tom's, isn't, I'm, I'd */ - if ((t1 == 's' || t1 == 't' || t1 == 'm' || t1 == 'd') && - (size == 3 || word_boundary(text[2]))) { - HOEDOWN_BUFPUTSL(ob, "’"); - return 0; - } - - /* you're, you'll, you've */ - if (size >= 3) { - uint8_t t2 = tolower(text[2]); - - if (((t1 == 'r' && t2 == 'e') || - (t1 == 'l' && t2 == 'l') || - (t1 == 'v' && t2 == 'e')) && - (size == 4 || word_boundary(text[3]))) { - HOEDOWN_BUFPUTSL(ob, "’"); - return 0; - } - } - } - - if (smartypants_quotes(ob, previous_char, size > 0 ? text[1] : 0, 's', &smrt->in_squote)) - return 0; - - hoedown_buffer_put(ob, squote_text, squote_size); - return 0; + if (size >= 2) { + uint8_t t1 = tolower(text[1]); + size_t next_squote_len = squote_len(text+1, size-1); + + /* convert '' to “ or ” */ + if (next_squote_len > 0) { + uint8_t next_char = (size > 1+next_squote_len) ? text[1+next_squote_len] : 0; + if (smartypants_quotes(ob, previous_char, next_char, 'd', &smrt->in_dquote)) + return next_squote_len; + } + + /* Tom's, isn't, I'm, I'd */ + if ((t1 == 's' || t1 == 't' || t1 == 'm' || t1 == 'd') && + (size == 3 || word_boundary(text[2]))) { + HOEDOWN_BUFPUTSL(ob, "’"); + return 0; + } + + /* you're, you'll, you've */ + if (size >= 3) { + uint8_t t2 = tolower(text[2]); + + if (((t1 == 'r' && t2 == 'e') || + (t1 == 'l' && t2 == 'l') || + (t1 == 'v' && t2 == 'e')) && + (size == 4 || word_boundary(text[3]))) { + HOEDOWN_BUFPUTSL(ob, "’"); + return 0; + } + } + } + + if (smartypants_quotes(ob, previous_char, size > 0 ? text[1] : 0, 's', &smrt->in_squote)) + return 0; + + hoedown_buffer_put(ob, squote_text, squote_size); + return 0; } /* Converts ' to left or right single quote. */ static size_t smartypants_cb__squote(hoedown_buffer *ob, struct smartypants_data *smrt, uint8_t previous_char, const uint8_t *text, size_t size) { - return smartypants_squote(ob, smrt, previous_char, text, size, text, 1); + return smartypants_squote(ob, smrt, previous_char, text, size, text, 1); } /* Converts (c), (r), (tm) */ static size_t smartypants_cb__parens(hoedown_buffer *ob, struct smartypants_data *smrt, uint8_t previous_char, const uint8_t *text, size_t size) { - if (size >= 3) { - uint8_t t1 = tolower(text[1]); - uint8_t t2 = tolower(text[2]); - - if (t1 == 'c' && t2 == ')') { - HOEDOWN_BUFPUTSL(ob, "©"); - return 2; - } - - if (t1 == 'r' && t2 == ')') { - HOEDOWN_BUFPUTSL(ob, "®"); - return 2; - } - - if (size >= 4 && t1 == 't' && t2 == 'm' && text[3] == ')') { - HOEDOWN_BUFPUTSL(ob, "™"); - return 3; - } - } - - hoedown_buffer_putc(ob, text[0]); - return 0; + if (size >= 3) { + uint8_t t1 = tolower(text[1]); + uint8_t t2 = tolower(text[2]); + + if (t1 == 'c' && t2 == ')') { + HOEDOWN_BUFPUTSL(ob, "©"); + return 2; + } + + if (t1 == 'r' && t2 == ')') { + HOEDOWN_BUFPUTSL(ob, "®"); + return 2; + } + + if (size >= 4 && t1 == 't' && t2 == 'm' && text[3] == ')') { + HOEDOWN_BUFPUTSL(ob, "™"); + return 3; + } + } + + hoedown_buffer_putc(ob, text[0]); + return 0; } /* Converts "--" to em-dash, etc. */ static size_t smartypants_cb__dash(hoedown_buffer *ob, struct smartypants_data *smrt, uint8_t previous_char, const uint8_t *text, size_t size) { - if (size >= 3 && text[1] == '-' && text[2] == '-') { - HOEDOWN_BUFPUTSL(ob, "—"); - return 2; - } - - if (size >= 2 && text[1] == '-') { - HOEDOWN_BUFPUTSL(ob, "–"); - return 1; - } - - hoedown_buffer_putc(ob, text[0]); - return 0; + if (size >= 3 && text[1] == '-' && text[2] == '-') { + HOEDOWN_BUFPUTSL(ob, "—"); + return 2; + } + + if (size >= 2 && text[1] == '-') { + HOEDOWN_BUFPUTSL(ob, "–"); + return 1; + } + + hoedown_buffer_putc(ob, text[0]); + return 0; } /* Converts " etc. */ static size_t smartypants_cb__amp(hoedown_buffer *ob, struct smartypants_data *smrt, uint8_t previous_char, const uint8_t *text, size_t size) { - size_t len; - if (size >= 6 && memcmp(text, """, 6) == 0) { - if (smartypants_quotes(ob, previous_char, size >= 7 ? text[6] : 0, 'd', &smrt->in_dquote)) - return 5; - } - - len = squote_len(text, size); - if (len > 0) { - return (len-1) + smartypants_squote(ob, smrt, previous_char, text+(len-1), size-(len-1), text, len); - } - - if (size >= 4 && memcmp(text, "�", 4) == 0) - return 3; - - hoedown_buffer_putc(ob, '&'); - return 0; + size_t len; + if (size >= 6 && memcmp(text, """, 6) == 0) { + if (smartypants_quotes(ob, previous_char, size >= 7 ? text[6] : 0, 'd', &smrt->in_dquote)) + return 5; + } + + len = squote_len(text, size); + if (len > 0) { + return (len-1) + smartypants_squote(ob, smrt, previous_char, text+(len-1), size-(len-1), text, len); + } + + if (size >= 4 && memcmp(text, "�", 4) == 0) + return 3; + + hoedown_buffer_putc(ob, '&'); + return 0; } /* Converts "..." to ellipsis */ static size_t smartypants_cb__period(hoedown_buffer *ob, struct smartypants_data *smrt, uint8_t previous_char, const uint8_t *text, size_t size) { - if (size >= 3 && text[1] == '.' && text[2] == '.') { - HOEDOWN_BUFPUTSL(ob, "…"); - return 2; - } - - if (size >= 5 && text[1] == ' ' && text[2] == '.' && text[3] == ' ' && text[4] == '.') { - HOEDOWN_BUFPUTSL(ob, "…"); - return 4; - } - - hoedown_buffer_putc(ob, text[0]); - return 0; + if (size >= 3 && text[1] == '.' && text[2] == '.') { + HOEDOWN_BUFPUTSL(ob, "…"); + return 2; + } + + if (size >= 5 && text[1] == ' ' && text[2] == '.' && text[3] == ' ' && text[4] == '.') { + HOEDOWN_BUFPUTSL(ob, "…"); + return 4; + } + + hoedown_buffer_putc(ob, text[0]); + return 0; } /* Converts `` to opening double quote */ static size_t smartypants_cb__backtick(hoedown_buffer *ob, struct smartypants_data *smrt, uint8_t previous_char, const uint8_t *text, size_t size) { - if (size >= 2 && text[1] == '`') { - if (smartypants_quotes(ob, previous_char, size >= 3 ? text[2] : 0, 'd', &smrt->in_dquote)) - return 1; - } + if (size >= 2 && text[1] == '`') { + if (smartypants_quotes(ob, previous_char, size >= 3 ? text[2] : 0, 'd', &smrt->in_dquote)) + return 1; + } - hoedown_buffer_putc(ob, text[0]); - return 0; + hoedown_buffer_putc(ob, text[0]); + return 0; } /* Converts 1/2, 1/4, 3/4 */ static size_t smartypants_cb__number(hoedown_buffer *ob, struct smartypants_data *smrt, uint8_t previous_char, const uint8_t *text, size_t size) { - if (word_boundary(previous_char) && size >= 3) { - if (text[0] == '1' && text[1] == '/' && text[2] == '2') { - if (size == 3 || word_boundary(text[3])) { - HOEDOWN_BUFPUTSL(ob, "½"); - return 2; - } - } - - if (text[0] == '1' && text[1] == '/' && text[2] == '4') { - if (size == 3 || word_boundary(text[3]) || - (size >= 5 && tolower(text[3]) == 't' && tolower(text[4]) == 'h')) { - HOEDOWN_BUFPUTSL(ob, "¼"); - return 2; - } - } - - if (text[0] == '3' && text[1] == '/' && text[2] == '4') { - if (size == 3 || word_boundary(text[3]) || - (size >= 6 && tolower(text[3]) == 't' && tolower(text[4]) == 'h' && tolower(text[5]) == 's')) { - HOEDOWN_BUFPUTSL(ob, "¾"); - return 2; - } - } - } - - hoedown_buffer_putc(ob, text[0]); - return 0; + if (word_boundary(previous_char) && size >= 3) { + if (text[0] == '1' && text[1] == '/' && text[2] == '2') { + if (size == 3 || word_boundary(text[3])) { + HOEDOWN_BUFPUTSL(ob, "½"); + return 2; + } + } + + if (text[0] == '1' && text[1] == '/' && text[2] == '4') { + if (size == 3 || word_boundary(text[3]) || + (size >= 5 && tolower(text[3]) == 't' && tolower(text[4]) == 'h')) { + HOEDOWN_BUFPUTSL(ob, "¼"); + return 2; + } + } + + if (text[0] == '3' && text[1] == '/' && text[2] == '4') { + if (size == 3 || word_boundary(text[3]) || + (size >= 6 && tolower(text[3]) == 't' && tolower(text[4]) == 'h' && tolower(text[5]) == 's')) { + HOEDOWN_BUFPUTSL(ob, "¾"); + return 2; + } + } + } + + hoedown_buffer_putc(ob, text[0]); + return 0; } /* Converts " to left or right double quote */ static size_t smartypants_cb__dquote(hoedown_buffer *ob, struct smartypants_data *smrt, uint8_t previous_char, const uint8_t *text, size_t size) { - if (!smartypants_quotes(ob, previous_char, size > 0 ? text[1] : 0, 'd', &smrt->in_dquote)) - HOEDOWN_BUFPUTSL(ob, """); + if (!smartypants_quotes(ob, previous_char, size > 0 ? text[1] : 0, 'd', &smrt->in_dquote)) + HOEDOWN_BUFPUTSL(ob, """); - return 0; + return 0; } static size_t smartypants_cb__ltag(hoedown_buffer *ob, struct smartypants_data *smrt, uint8_t previous_char, const uint8_t *text, size_t size) { - static const char *skip_tags[] = { - "pre", "code", "var", "samp", "kbd", "math", "script", "style" - }; - static const size_t skip_tags_count = 8; - - size_t tag, i = 0; - - /* This is a comment. Copy everything verbatim until --> or EOF is seen. */ - if (i + 4 < size && memcmp(text, "<!--", 4) == 0) { - i += 4; - while (i + 3 < size && memcmp(text + i, "-->", 3) != 0) - i++; - i += 3; - hoedown_buffer_put(ob, text, i + 1); - return i; - } - - while (i < size && text[i] != '>') - i++; - - for (tag = 0; tag < skip_tags_count; ++tag) { - if (hoedown_html_is_tag(text, size, skip_tags[tag]) == HOEDOWN_HTML_TAG_OPEN) - break; - } - - if (tag < skip_tags_count) { - for (;;) { - while (i < size && text[i] != '<') - i++; - - if (i == size) - break; - - if (hoedown_html_is_tag(text + i, size - i, skip_tags[tag]) == HOEDOWN_HTML_TAG_CLOSE) - break; - - i++; - } - - while (i < size && text[i] != '>') - i++; - } - - hoedown_buffer_put(ob, text, i + 1); - return i; + static const char *skip_tags[] = { + "pre", "code", "var", "samp", "kbd", "math", "script", "style" + }; + static const size_t skip_tags_count = 8; + + size_t tag, i = 0; + + /* This is a comment. Copy everything verbatim until --> or EOF is seen. */ + if (i + 4 < size && memcmp(text, "<!--", 4) == 0) { + i += 4; + while (i + 3 < size && memcmp(text + i, "-->", 3) != 0) + i++; + i += 3; + hoedown_buffer_put(ob, text, i + 1); + return i; + } + + while (i < size && text[i] != '>') + i++; + + for (tag = 0; tag < skip_tags_count; ++tag) { + if (hoedown_html_is_tag(text, size, skip_tags[tag]) == HOEDOWN_HTML_TAG_OPEN) + break; + } + + if (tag < skip_tags_count) { + for (;;) { + while (i < size && text[i] != '<') + i++; + + if (i == size) + break; + + if (hoedown_html_is_tag(text + i, size - i, skip_tags[tag]) == HOEDOWN_HTML_TAG_CLOSE) + break; + + i++; + } + + while (i < size && text[i] != '>') + i++; + } + + hoedown_buffer_put(ob, text, i + 1); + return i; } static size_t smartypants_cb__escape(hoedown_buffer *ob, struct smartypants_data *smrt, uint8_t previous_char, const uint8_t *text, size_t size) { - if (size < 2) - return 0; - - switch (text[1]) { - case '\\': - case '"': - case '\'': - case '.': - case '-': - case '`': - hoedown_buffer_putc(ob, text[1]); - return 1; - - default: - hoedown_buffer_putc(ob, '\\'); - return 0; - } + if (size < 2) + return 0; + + switch (text[1]) { + case '\\': + case '"': + case '\'': + case '.': + case '-': + case '`': + hoedown_buffer_putc(ob, text[1]); + return 1; + + default: + hoedown_buffer_putc(ob, '\\'); + return 0; + } } #if 0 @@ -408,28 +408,28 @@ static struct { void hoedown_html_smartypants(hoedown_buffer *ob, const uint8_t *text, size_t size) { - size_t i; - struct smartypants_data smrt = {0, 0}; + size_t i; + struct smartypants_data smrt = {0, 0}; - if (!text) - return; + if (!text) + return; - hoedown_buffer_grow(ob, size); + hoedown_buffer_grow(ob, size); - for (i = 0; i < size; ++i) { - size_t org; - uint8_t action = 0; + for (i = 0; i < size; ++i) { + size_t org; + uint8_t action = 0; - org = i; - while (i < size && (action = smartypants_cb_chars[text[i]]) == 0) - i++; + org = i; + while (i < size && (action = smartypants_cb_chars[text[i]]) == 0) + i++; - if (i > org) - hoedown_buffer_put(ob, text + org, i - org); + if (i > org) + hoedown_buffer_put(ob, text + org, i - org); - if (i < size) { - i += smartypants_cb_ptrs[(int)action] - (ob, &smrt, i ? text[i - 1] : 0, text + i, size - i); - } - } + if (i < size) { + i += smartypants_cb_ptrs[(int)action] + (ob, &smrt, i ? text[i - 1] : 0, text + i, size - i); + } + } } diff --git a/libraries/hoedown/src/stack.c b/libraries/hoedown/src/stack.c index 46ead232..0523c11b 100644 --- a/libraries/hoedown/src/stack.c +++ b/libraries/hoedown/src/stack.c @@ -9,71 +9,71 @@ void hoedown_stack_init(hoedown_stack *st, size_t initial_size) { - assert(st); + assert(st); - st->item = NULL; - st->size = st->asize = 0; + st->item = NULL; + st->size = st->asize = 0; - if (!initial_size) - initial_size = 8; + if (!initial_size) + initial_size = 8; - hoedown_stack_grow(st, initial_size); + hoedown_stack_grow(st, initial_size); } void hoedown_stack_uninit(hoedown_stack *st) { - assert(st); + assert(st); - free(st->item); + free(st->item); } void hoedown_stack_grow(hoedown_stack *st, size_t neosz) { - assert(st); + assert(st); - if (st->asize >= neosz) - return; + if (st->asize >= neosz) + return; - st->item = hoedown_realloc(st->item, neosz * sizeof(void *)); - memset(st->item + st->asize, 0x0, (neosz - st->asize) * sizeof(void *)); + st->item = hoedown_realloc(st->item, neosz * sizeof(void *)); + memset(st->item + st->asize, 0x0, (neosz - st->asize) * sizeof(void *)); - st->asize = neosz; + st->asize = neosz; - if (st->size > neosz) - st->size = neosz; + if (st->size > neosz) + st->size = neosz; } void hoedown_stack_push(hoedown_stack *st, void *item) { - assert(st); + assert(st); - if (st->size >= st->asize) - hoedown_stack_grow(st, st->size * 2); + if (st->size >= st->asize) + hoedown_stack_grow(st, st->size * 2); - st->item[st->size++] = item; + st->item[st->size++] = item; } void * hoedown_stack_pop(hoedown_stack *st) { - assert(st); + assert(st); - if (!st->size) - return NULL; + if (!st->size) + return NULL; - return st->item[--st->size]; + return st->item[--st->size]; } void * hoedown_stack_top(const hoedown_stack *st) { - assert(st); + assert(st); - if (!st->size) - return NULL; + if (!st->size) + return NULL; - return st->item[st->size - 1]; + return st->item[st->size - 1]; } diff --git a/libraries/hoedown/src/version.c b/libraries/hoedown/src/version.c index 625ed196..10d36cb9 100644 --- a/libraries/hoedown/src/version.c +++ b/libraries/hoedown/src/version.c @@ -3,7 +3,7 @@ void hoedown_version(int *major, int *minor, int *revision) { - *major = HOEDOWN_VERSION_MAJOR; - *minor = HOEDOWN_VERSION_MINOR; - *revision = HOEDOWN_VERSION_REVISION; + *major = HOEDOWN_VERSION_MAJOR; + *minor = HOEDOWN_VERSION_MINOR; + *revision = HOEDOWN_VERSION_REVISION; } diff --git a/libraries/iconfix/CMakeLists.txt b/libraries/iconfix/CMakeLists.txt index 93bfdd06..ccf0edea 100644 --- a/libraries/iconfix/CMakeLists.txt +++ b/libraries/iconfix/CMakeLists.txt @@ -15,14 +15,14 @@ internal/qiconloader_p.h add_library(MultiMC_iconfix SHARED ${ICONFIX_SOURCES}) target_include_directories(MultiMC_iconfix PUBLIC ${CMAKE_CURRENT_SOURCE_DIR} "${CMAKE_CURRENT_BINARY_DIR}" ) -qt5_use_modules(MultiMC_iconfix Core Widgets) +target_link_libraries(MultiMC_iconfix Qt5::Core Qt5::Widgets) set_target_properties(MultiMC_iconfix PROPERTIES CXX_VISIBILITY_PRESET hidden VISIBILITY_INLINES_HIDDEN 1) generate_export_header(MultiMC_iconfix) # Install it install( - TARGETS MultiMC_iconfix - RUNTIME DESTINATION ${LIBRARY_DEST_DIR} - LIBRARY DESTINATION ${LIBRARY_DEST_DIR} + TARGETS MultiMC_iconfix + RUNTIME DESTINATION ${LIBRARY_DEST_DIR} + LIBRARY DESTINATION ${LIBRARY_DEST_DIR} )
\ No newline at end of file diff --git a/libraries/iconfix/internal/qhexstring_p.h b/libraries/iconfix/internal/qhexstring_p.h index f01b4cdd..c81904e5 100644 --- a/libraries/iconfix/internal/qhexstring_p.h +++ b/libraries/iconfix/internal/qhexstring_p.h @@ -61,40 +61,40 @@ // internal helper. Converts an integer value to an unique string token template <typename T> struct HexString { - inline HexString(const T t) : val(t) - { - } + inline HexString(const T t) : val(t) + { + } - inline void write(QChar *&dest) const - { - const ushort hexChars[] = {'0', '1', '2', '3', '4', '5', '6', '7', - '8', '9', 'a', 'b', 'c', 'd', 'e', 'f'}; - const char *c = reinterpret_cast<const char *>(&val); - for (uint i = 0; i < sizeof(T); ++i) - { - *dest++ = hexChars[*c & 0xf]; - *dest++ = hexChars[(*c & 0xf0) >> 4]; - ++c; - } - } - const T val; + inline void write(QChar *&dest) const + { + const ushort hexChars[] = {'0', '1', '2', '3', '4', '5', '6', '7', + '8', '9', 'a', 'b', 'c', 'd', 'e', 'f'}; + const char *c = reinterpret_cast<const char *>(&val); + for (uint i = 0; i < sizeof(T); ++i) + { + *dest++ = hexChars[*c & 0xf]; + *dest++ = hexChars[(*c & 0xf0) >> 4]; + ++c; + } + } + const T val; }; // specialization to enable fast concatenating of our string tokens to a string template <typename T> struct QConcatenable<HexString<T>> { - typedef HexString<T> type; - enum - { - ExactSize = true - }; - static int size(const HexString<T> &) - { - return sizeof(T) * 2; - } - static inline void appendTo(const HexString<T> &str, QChar *&out) - { - str.write(out); - } - typedef QString ConvertTo; + typedef HexString<T> type; + enum + { + ExactSize = true + }; + static int size(const HexString<T> &) + { + return sizeof(T) * 2; + } + static inline void appendTo(const HexString<T> &str, QChar *&out) + { + str.write(out); + } + typedef QString ConvertTo; }; diff --git a/libraries/iconfix/internal/qiconloader.cpp b/libraries/iconfix/internal/qiconloader.cpp index b1195893..41cf3d50 100644 --- a/libraries/iconfix/internal/qiconloader.cpp +++ b/libraries/iconfix/internal/qiconloader.cpp @@ -55,7 +55,7 @@ Q_GLOBAL_STATIC(QIconLoader, iconLoaderInstance) static QString fallbackTheme() { - return QString("hicolor"); + return QString("hicolor"); } QIconLoader::QIconLoader() : m_themeKey(1), m_supportsSvg(false), m_initialized(false) @@ -67,403 +67,403 @@ QIconLoader::QIconLoader() : m_themeKey(1), m_supportsSvg(false), m_initialized( static inline QString systemThemeName() { - return QIcon::themeName(); + return QIcon::themeName(); } static inline QStringList systemIconSearchPaths() { - auto paths = QIcon::themeSearchPaths(); - paths.push_front(":/icons"); - return paths; + auto paths = QIcon::themeSearchPaths(); + paths.push_front(":/icons"); + return paths; } void QIconLoader::ensureInitialized() { - if (!m_initialized) - { - m_initialized = true; + if (!m_initialized) + { + m_initialized = true; - Q_ASSERT(qApp); + Q_ASSERT(qApp); - m_systemTheme = QIcon::themeName(); + m_systemTheme = QIcon::themeName(); - if (m_systemTheme.isEmpty()) - m_systemTheme = fallbackTheme(); - m_supportsSvg = true; - } + if (m_systemTheme.isEmpty()) + m_systemTheme = fallbackTheme(); + m_supportsSvg = true; + } } QIconLoader *QIconLoader::instance() { - iconLoaderInstance()->ensureInitialized(); - return iconLoaderInstance(); + iconLoaderInstance()->ensureInitialized(); + return iconLoaderInstance(); } // Queries the system theme and invalidates existing // icons if the theme has changed. void QIconLoader::updateSystemTheme() { - // Only change if this is not explicitly set by the user - if (m_userTheme.isEmpty()) - { - QString theme = systemThemeName(); - if (theme.isEmpty()) - theme = fallbackTheme(); - if (theme != m_systemTheme) - { - m_systemTheme = theme; - invalidateKey(); - } - } + // Only change if this is not explicitly set by the user + if (m_userTheme.isEmpty()) + { + QString theme = systemThemeName(); + if (theme.isEmpty()) + theme = fallbackTheme(); + if (theme != m_systemTheme) + { + m_systemTheme = theme; + invalidateKey(); + } + } } void QIconLoader::setThemeName(const QString &themeName) { - m_userTheme = themeName; - invalidateKey(); + m_userTheme = themeName; + invalidateKey(); } void QIconLoader::setThemeSearchPath(const QStringList &searchPaths) { - m_iconDirs = searchPaths; - themeList.clear(); - invalidateKey(); + m_iconDirs = searchPaths; + themeList.clear(); + invalidateKey(); } QStringList QIconLoader::themeSearchPaths() const { - if (m_iconDirs.isEmpty()) - { - m_iconDirs = systemIconSearchPaths(); - } - return m_iconDirs; + if (m_iconDirs.isEmpty()) + { + m_iconDirs = systemIconSearchPaths(); + } + return m_iconDirs; } QIconTheme::QIconTheme(const QString &themeName) : m_valid(false) { - QFile themeIndex; - - QStringList iconDirs = systemIconSearchPaths(); - for (int i = 0; i < iconDirs.size(); ++i) - { - QDir iconDir(iconDirs[i]); - QString themeDir = iconDir.path() + QLatin1Char('/') + themeName; - themeIndex.setFileName(themeDir + QLatin1String("/index.theme")); - if (themeIndex.exists()) - { - m_contentDir = themeDir; - m_valid = true; - - foreach (QString path, iconDirs) - { - if (QFileInfo(path).isDir()) - m_contentDirs.append(path + QLatin1Char('/') + themeName); - } - - break; - } - } - - // if there is no index file, abscond. - if (!themeIndex.exists()) - return; - - // otherwise continue reading index file - const QSettings indexReader(themeIndex.fileName(), QSettings::IniFormat); - QStringListIterator keyIterator(indexReader.allKeys()); - while (keyIterator.hasNext()) - { - const QString key = keyIterator.next(); - if (!key.endsWith(QLatin1String("/Size"))) - continue; - - // Note the QSettings ini-format does not accept - // slashes in key names, hence we have to cheat - int size = indexReader.value(key).toInt(); - if (!size) - continue; - - QString directoryKey = key.left(key.size() - 5); - QIconDirInfo dirInfo(directoryKey); - dirInfo.size = size; - QString type = - indexReader.value(directoryKey + QLatin1String("/Type")).toString(); - - if (type == QLatin1String("Fixed")) - dirInfo.type = QIconDirInfo::Fixed; - else if (type == QLatin1String("Scalable")) - dirInfo.type = QIconDirInfo::Scalable; - else - dirInfo.type = QIconDirInfo::Threshold; - - dirInfo.threshold = - indexReader.value(directoryKey + QLatin1String("/Threshold"), 2) - .toInt(); - - dirInfo.minSize = - indexReader.value(directoryKey + QLatin1String("/MinSize"), size) - .toInt(); - - dirInfo.maxSize = - indexReader.value(directoryKey + QLatin1String("/MaxSize"), size) - .toInt(); - m_keyList.append(dirInfo); - } - - // Parent themes provide fallbacks for missing icons - m_parents = indexReader.value(QLatin1String("Icon Theme/Inherits")).toStringList(); - m_parents.removeAll(QString()); - - // Ensure a default platform fallback for all themes - if (m_parents.isEmpty()) - { - const QString fallback = fallbackTheme(); - if (!fallback.isEmpty()) - m_parents.append(fallback); - } - - // Ensure that all themes fall back to hicolor - if (!m_parents.contains(QLatin1String("hicolor"))) - m_parents.append(QLatin1String("hicolor")); + QFile themeIndex; + + QStringList iconDirs = systemIconSearchPaths(); + for (int i = 0; i < iconDirs.size(); ++i) + { + QDir iconDir(iconDirs[i]); + QString themeDir = iconDir.path() + QLatin1Char('/') + themeName; + themeIndex.setFileName(themeDir + QLatin1String("/index.theme")); + if (themeIndex.exists()) + { + m_contentDir = themeDir; + m_valid = true; + + foreach (QString path, iconDirs) + { + if (QFileInfo(path).isDir()) + m_contentDirs.append(path + QLatin1Char('/') + themeName); + } + + break; + } + } + + // if there is no index file, abscond. + if (!themeIndex.exists()) + return; + + // otherwise continue reading index file + const QSettings indexReader(themeIndex.fileName(), QSettings::IniFormat); + QStringListIterator keyIterator(indexReader.allKeys()); + while (keyIterator.hasNext()) + { + const QString key = keyIterator.next(); + if (!key.endsWith(QLatin1String("/Size"))) + continue; + + // Note the QSettings ini-format does not accept + // slashes in key names, hence we have to cheat + int size = indexReader.value(key).toInt(); + if (!size) + continue; + + QString directoryKey = key.left(key.size() - 5); + QIconDirInfo dirInfo(directoryKey); + dirInfo.size = size; + QString type = + indexReader.value(directoryKey + QLatin1String("/Type")).toString(); + + if (type == QLatin1String("Fixed")) + dirInfo.type = QIconDirInfo::Fixed; + else if (type == QLatin1String("Scalable")) + dirInfo.type = QIconDirInfo::Scalable; + else + dirInfo.type = QIconDirInfo::Threshold; + + dirInfo.threshold = + indexReader.value(directoryKey + QLatin1String("/Threshold"), 2) + .toInt(); + + dirInfo.minSize = + indexReader.value(directoryKey + QLatin1String("/MinSize"), size) + .toInt(); + + dirInfo.maxSize = + indexReader.value(directoryKey + QLatin1String("/MaxSize"), size) + .toInt(); + m_keyList.append(dirInfo); + } + + // Parent themes provide fallbacks for missing icons + m_parents = indexReader.value(QLatin1String("Icon Theme/Inherits")).toStringList(); + m_parents.removeAll(QString()); + + // Ensure a default platform fallback for all themes + if (m_parents.isEmpty()) + { + const QString fallback = fallbackTheme(); + if (!fallback.isEmpty()) + m_parents.append(fallback); + } + + // Ensure that all themes fall back to hicolor + if (!m_parents.contains(QLatin1String("hicolor"))) + m_parents.append(QLatin1String("hicolor")); } QThemeIconEntries QIconLoader::findIconHelper(const QString &themeName, const QString &iconName, - QStringList &visited) const -{ - QThemeIconEntries entries; - Q_ASSERT(!themeName.isEmpty()); - - QPixmap pixmap; - - // Used to protect against potential recursions - visited << themeName; - - QIconTheme theme = themeList.value(themeName); - if (!theme.isValid()) - { - theme = QIconTheme(themeName); - if (!theme.isValid()) - theme = QIconTheme(fallbackTheme()); - - themeList.insert(themeName, theme); - } - - QStringList contentDirs = theme.contentDirs(); - const QVector<QIconDirInfo> subDirs = theme.keyList(); - - const QString svgext(QLatin1String(".svg")); - const QString pngext(QLatin1String(".png")); - const QString xpmext(QLatin1String(".xpm")); - - // Add all relevant files - for (int i = 0; i < subDirs.size(); ++i) - { - const QIconDirInfo &dirInfo = subDirs.at(i); - QString subdir = dirInfo.path; - - foreach (QString contentDir, contentDirs) - { - QDir currentDir(contentDir + '/' + subdir); - - if (currentDir.exists(iconName + pngext)) - { - PixmapEntry *iconEntry = new PixmapEntry; - iconEntry->dir = dirInfo; - iconEntry->filename = currentDir.filePath(iconName + pngext); - // Notice we ensure that pixmap entries always come before - // scalable to preserve search order afterwards - entries.prepend(iconEntry); - } - else if (m_supportsSvg && currentDir.exists(iconName + svgext)) - { - ScalableEntry *iconEntry = new ScalableEntry; - iconEntry->dir = dirInfo; - iconEntry->filename = currentDir.filePath(iconName + svgext); - entries.append(iconEntry); - break; - } - else if (currentDir.exists(iconName + xpmext)) - { - PixmapEntry *iconEntry = new PixmapEntry; - iconEntry->dir = dirInfo; - iconEntry->filename = currentDir.filePath(iconName + xpmext); - // Notice we ensure that pixmap entries always come before - // scalable to preserve search order afterwards - entries.append(iconEntry); - break; - } - } - } - - if (entries.isEmpty()) - { - const QStringList parents = theme.parents(); - // Search recursively through inherited themes - for (int i = 0; i < parents.size(); ++i) - { - - const QString parentTheme = parents.at(i).trimmed(); - - if (!visited.contains(parentTheme)) // guard against recursion - entries = findIconHelper(parentTheme, iconName, visited); - - if (!entries.isEmpty()) // success - break; - } - } + QStringList &visited) const +{ + QThemeIconEntries entries; + Q_ASSERT(!themeName.isEmpty()); + + QPixmap pixmap; + + // Used to protect against potential recursions + visited << themeName; + + QIconTheme theme = themeList.value(themeName); + if (!theme.isValid()) + { + theme = QIconTheme(themeName); + if (!theme.isValid()) + theme = QIconTheme(fallbackTheme()); + + themeList.insert(themeName, theme); + } + + QStringList contentDirs = theme.contentDirs(); + const QVector<QIconDirInfo> subDirs = theme.keyList(); + + const QString svgext(QLatin1String(".svg")); + const QString pngext(QLatin1String(".png")); + const QString xpmext(QLatin1String(".xpm")); + + // Add all relevant files + for (int i = 0; i < subDirs.size(); ++i) + { + const QIconDirInfo &dirInfo = subDirs.at(i); + QString subdir = dirInfo.path; + + foreach (QString contentDir, contentDirs) + { + QDir currentDir(contentDir + '/' + subdir); + + if (currentDir.exists(iconName + pngext)) + { + PixmapEntry *iconEntry = new PixmapEntry; + iconEntry->dir = dirInfo; + iconEntry->filename = currentDir.filePath(iconName + pngext); + // Notice we ensure that pixmap entries always come before + // scalable to preserve search order afterwards + entries.prepend(iconEntry); + } + else if (m_supportsSvg && currentDir.exists(iconName + svgext)) + { + ScalableEntry *iconEntry = new ScalableEntry; + iconEntry->dir = dirInfo; + iconEntry->filename = currentDir.filePath(iconName + svgext); + entries.append(iconEntry); + break; + } + else if (currentDir.exists(iconName + xpmext)) + { + PixmapEntry *iconEntry = new PixmapEntry; + iconEntry->dir = dirInfo; + iconEntry->filename = currentDir.filePath(iconName + xpmext); + // Notice we ensure that pixmap entries always come before + // scalable to preserve search order afterwards + entries.append(iconEntry); + break; + } + } + } + + if (entries.isEmpty()) + { + const QStringList parents = theme.parents(); + // Search recursively through inherited themes + for (int i = 0; i < parents.size(); ++i) + { + + const QString parentTheme = parents.at(i).trimmed(); + + if (!visited.contains(parentTheme)) // guard against recursion + entries = findIconHelper(parentTheme, iconName, visited); + + if (!entries.isEmpty()) // success + break; + } + } /********************************************************************* Author: Kaitlin Rupert <kaitlin.rupert@intel.com> Date: Aug 12, 2010 Description: Make it so that the QIcon loader honors /usr/share/pixmaps - directory. This is a valid directory per the Freedesktop.org - icon theme specification. + directory. This is a valid directory per the Freedesktop.org + icon theme specification. Bug: https://bugreports.qt.nokia.com/browse/QTBUG-12874 *********************************************************************/ #ifdef Q_OS_LINUX - /* Freedesktop standard says to look in /usr/share/pixmaps last */ - if (entries.isEmpty()) - { - const QString pixmaps(QLatin1String("/usr/share/pixmaps")); - - QDir currentDir(pixmaps); - QIconDirInfo dirInfo(pixmaps); - if (currentDir.exists(iconName + pngext)) - { - PixmapEntry *iconEntry = new PixmapEntry; - iconEntry->dir = dirInfo; - iconEntry->filename = currentDir.filePath(iconName + pngext); - // Notice we ensure that pixmap entries always come before - // scalable to preserve search order afterwards - entries.prepend(iconEntry); - } - else if (m_supportsSvg && currentDir.exists(iconName + svgext)) - { - ScalableEntry *iconEntry = new ScalableEntry; - iconEntry->dir = dirInfo; - iconEntry->filename = currentDir.filePath(iconName + svgext); - entries.append(iconEntry); - } - else if (currentDir.exists(iconName + xpmext)) - { - PixmapEntry *iconEntry = new PixmapEntry; - iconEntry->dir = dirInfo; - iconEntry->filename = currentDir.filePath(iconName + xpmext); - // Notice we ensure that pixmap entries always come before - // scalable to preserve search order afterwards - entries.append(iconEntry); - } - } + /* Freedesktop standard says to look in /usr/share/pixmaps last */ + if (entries.isEmpty()) + { + const QString pixmaps(QLatin1String("/usr/share/pixmaps")); + + QDir currentDir(pixmaps); + QIconDirInfo dirInfo(pixmaps); + if (currentDir.exists(iconName + pngext)) + { + PixmapEntry *iconEntry = new PixmapEntry; + iconEntry->dir = dirInfo; + iconEntry->filename = currentDir.filePath(iconName + pngext); + // Notice we ensure that pixmap entries always come before + // scalable to preserve search order afterwards + entries.prepend(iconEntry); + } + else if (m_supportsSvg && currentDir.exists(iconName + svgext)) + { + ScalableEntry *iconEntry = new ScalableEntry; + iconEntry->dir = dirInfo; + iconEntry->filename = currentDir.filePath(iconName + svgext); + entries.append(iconEntry); + } + else if (currentDir.exists(iconName + xpmext)) + { + PixmapEntry *iconEntry = new PixmapEntry; + iconEntry->dir = dirInfo; + iconEntry->filename = currentDir.filePath(iconName + xpmext); + // Notice we ensure that pixmap entries always come before + // scalable to preserve search order afterwards + entries.append(iconEntry); + } + } #endif - if (entries.isEmpty()) - { - // Search for unthemed icons in main dir of search paths - QStringList themeSearchPaths = QIcon::themeSearchPaths(); - foreach (QString contentDir, themeSearchPaths) - { - QDir currentDir(contentDir); - - if (currentDir.exists(iconName + pngext)) - { - PixmapEntry *iconEntry = new PixmapEntry; - iconEntry->filename = currentDir.filePath(iconName + pngext); - // Notice we ensure that pixmap entries always come before - // scalable to preserve search order afterwards - entries.prepend(iconEntry); - } - else if (m_supportsSvg && currentDir.exists(iconName + svgext)) - { - ScalableEntry *iconEntry = new ScalableEntry; - iconEntry->filename = currentDir.filePath(iconName + svgext); - entries.append(iconEntry); - break; - } - else if (currentDir.exists(iconName + xpmext)) - { - PixmapEntry *iconEntry = new PixmapEntry; - iconEntry->filename = currentDir.filePath(iconName + xpmext); - // Notice we ensure that pixmap entries always come before - // scalable to preserve search order afterwards - entries.append(iconEntry); - break; - } - } - } - return entries; + if (entries.isEmpty()) + { + // Search for unthemed icons in main dir of search paths + QStringList themeSearchPaths = QIcon::themeSearchPaths(); + foreach (QString contentDir, themeSearchPaths) + { + QDir currentDir(contentDir); + + if (currentDir.exists(iconName + pngext)) + { + PixmapEntry *iconEntry = new PixmapEntry; + iconEntry->filename = currentDir.filePath(iconName + pngext); + // Notice we ensure that pixmap entries always come before + // scalable to preserve search order afterwards + entries.prepend(iconEntry); + } + else if (m_supportsSvg && currentDir.exists(iconName + svgext)) + { + ScalableEntry *iconEntry = new ScalableEntry; + iconEntry->filename = currentDir.filePath(iconName + svgext); + entries.append(iconEntry); + break; + } + else if (currentDir.exists(iconName + xpmext)) + { + PixmapEntry *iconEntry = new PixmapEntry; + iconEntry->filename = currentDir.filePath(iconName + xpmext); + // Notice we ensure that pixmap entries always come before + // scalable to preserve search order afterwards + entries.append(iconEntry); + break; + } + } + } + return entries; } QThemeIconEntries QIconLoader::loadIcon(const QString &name) const { - if (!themeName().isEmpty()) - { - QStringList visited; - return findIconHelper(themeName(), name, visited); - } + if (!themeName().isEmpty()) + { + QStringList visited; + return findIconHelper(themeName(), name, visited); + } - return QThemeIconEntries(); + return QThemeIconEntries(); } // -------- Icon Loader Engine -------- // QIconLoaderEngineFixed::QIconLoaderEngineFixed(const QString &iconName) - : m_iconName(iconName), m_key(0) + : m_iconName(iconName), m_key(0) { } QIconLoaderEngineFixed::~QIconLoaderEngineFixed() { - qDeleteAll(m_entries); + qDeleteAll(m_entries); } QIconLoaderEngineFixed::QIconLoaderEngineFixed(const QIconLoaderEngineFixed &other) - : QIconEngine(other), m_iconName(other.m_iconName), m_key(0) + : QIconEngine(other), m_iconName(other.m_iconName), m_key(0) { } QIconEngine *QIconLoaderEngineFixed::clone() const { - return new QIconLoaderEngineFixed(*this); + return new QIconLoaderEngineFixed(*this); } bool QIconLoaderEngineFixed::read(QDataStream &in) { - in >> m_iconName; - return true; + in >> m_iconName; + return true; } bool QIconLoaderEngineFixed::write(QDataStream &out) const { - out << m_iconName; - return true; + out << m_iconName; + return true; } bool QIconLoaderEngineFixed::hasIcon() const { - return !(m_entries.isEmpty()); + return !(m_entries.isEmpty()); } // Lazily load the icon void QIconLoaderEngineFixed::ensureLoaded() { - if (!(QIconLoader::instance()->themeKey() == m_key)) - { + if (!(QIconLoader::instance()->themeKey() == m_key)) + { - qDeleteAll(m_entries); + qDeleteAll(m_entries); - m_entries = QIconLoader::instance()->loadIcon(m_iconName); - m_key = QIconLoader::instance()->themeKey(); - } + m_entries = QIconLoader::instance()->loadIcon(m_iconName); + m_key = QIconLoader::instance()->themeKey(); + } } void QIconLoaderEngineFixed::paint(QPainter *painter, const QRect &rect, QIcon::Mode mode, - QIcon::State state) + QIcon::State state) { - QSize pixmapSize = rect.size(); + QSize pixmapSize = rect.size(); #if defined(Q_WS_MAC) - pixmapSize *= qt_mac_get_scalefactor(); + pixmapSize *= qt_mac_get_scalefactor(); #endif - painter->drawPixmap(rect, pixmap(pixmapSize, mode, state)); + painter->drawPixmap(rect, pixmap(pixmapSize, mode, state)); } /* @@ -472,21 +472,21 @@ void QIconLoaderEngineFixed::paint(QPainter *painter, const QRect &rect, QIcon:: */ static bool directoryMatchesSize(const QIconDirInfo &dir, int iconsize) { - if (dir.type == QIconDirInfo::Fixed) - { - return dir.size == iconsize; - } - else if (dir.type == QIconDirInfo::Scalable) - { - return dir.size <= dir.maxSize && iconsize >= dir.minSize; - } - else if (dir.type == QIconDirInfo::Threshold) - { - return iconsize >= dir.size - dir.threshold && iconsize <= dir.size + dir.threshold; - } - - Q_ASSERT(1); // Not a valid value - return false; + if (dir.type == QIconDirInfo::Fixed) + { + return dir.size == iconsize; + } + else if (dir.type == QIconDirInfo::Scalable) + { + return dir.size <= dir.maxSize && iconsize >= dir.minSize; + } + else if (dir.type == QIconDirInfo::Threshold) + { + return iconsize >= dir.size - dir.threshold && iconsize <= dir.size + dir.threshold; + } + + Q_ASSERT(1); // Not a valid value + return false; } /* @@ -495,66 +495,66 @@ static bool directoryMatchesSize(const QIconDirInfo &dir, int iconsize) */ static int directorySizeDistance(const QIconDirInfo &dir, int iconsize) { - if (dir.type == QIconDirInfo::Fixed) - { - return qAbs(dir.size - iconsize); - } - else if (dir.type == QIconDirInfo::Scalable) - { - if (iconsize < dir.minSize) - return dir.minSize - iconsize; - else if (iconsize > dir.maxSize) - return iconsize - dir.maxSize; - else - return 0; - } - else if (dir.type == QIconDirInfo::Threshold) - { - if (iconsize < dir.size - dir.threshold) - return dir.minSize - iconsize; - else if (iconsize > dir.size + dir.threshold) - return iconsize - dir.maxSize; - else - return 0; - } - - Q_ASSERT(1); // Not a valid value - return INT_MAX; + if (dir.type == QIconDirInfo::Fixed) + { + return qAbs(dir.size - iconsize); + } + else if (dir.type == QIconDirInfo::Scalable) + { + if (iconsize < dir.minSize) + return dir.minSize - iconsize; + else if (iconsize > dir.maxSize) + return iconsize - dir.maxSize; + else + return 0; + } + else if (dir.type == QIconDirInfo::Threshold) + { + if (iconsize < dir.size - dir.threshold) + return dir.minSize - iconsize; + else if (iconsize > dir.size + dir.threshold) + return iconsize - dir.maxSize; + else + return 0; + } + + Q_ASSERT(1); // Not a valid value + return INT_MAX; } QIconLoaderEngineEntry *QIconLoaderEngineFixed::entryForSize(const QSize &size) { - int iconsize = qMin(size.width(), size.height()); - - // Note that m_entries are sorted so that png-files - // come first - - const int numEntries = m_entries.size(); - - // Search for exact matches first - for (int i = 0; i < numEntries; ++i) - { - QIconLoaderEngineEntry *entry = m_entries.at(i); - if (directoryMatchesSize(entry->dir, iconsize)) - { - return entry; - } - } - - // Find the minimum distance icon - int minimalSize = INT_MAX; - QIconLoaderEngineEntry *closestMatch = 0; - for (int i = 0; i < numEntries; ++i) - { - QIconLoaderEngineEntry *entry = m_entries.at(i); - int distance = directorySizeDistance(entry->dir, iconsize); - if (distance < minimalSize) - { - minimalSize = distance; - closestMatch = entry; - } - } - return closestMatch; + int iconsize = qMin(size.width(), size.height()); + + // Note that m_entries are sorted so that png-files + // come first + + const int numEntries = m_entries.size(); + + // Search for exact matches first + for (int i = 0; i < numEntries; ++i) + { + QIconLoaderEngineEntry *entry = m_entries.at(i); + if (directoryMatchesSize(entry->dir, iconsize)) + { + return entry; + } + } + + // Find the minimum distance icon + int minimalSize = INT_MAX; + QIconLoaderEngineEntry *closestMatch = 0; + for (int i = 0; i < numEntries; ++i) + { + QIconLoaderEngineEntry *entry = m_entries.at(i); + int distance = directorySizeDistance(entry->dir, iconsize); + if (distance < minimalSize) + { + minimalSize = distance; + closestMatch = entry; + } + } + return closestMatch; } /* @@ -564,125 +564,125 @@ QIconLoaderEngineEntry *QIconLoaderEngineFixed::entryForSize(const QSize &size) * */ QSize QIconLoaderEngineFixed::actualSize(const QSize &size, QIcon::Mode mode, - QIcon::State state) -{ - ensureLoaded(); - - QIconLoaderEngineEntry *entry = entryForSize(size); - if (entry) - { - const QIconDirInfo &dir = entry->dir; - if (dir.type == QIconDirInfo::Scalable) - return size; - else - { - int result = qMin<int>(dir.size, qMin(size.width(), size.height())); - return QSize(result, result); - } - } - return QIconEngine::actualSize(size, mode, state); + QIcon::State state) +{ + ensureLoaded(); + + QIconLoaderEngineEntry *entry = entryForSize(size); + if (entry) + { + const QIconDirInfo &dir = entry->dir; + if (dir.type == QIconDirInfo::Scalable) + return size; + else + { + int result = qMin<int>(dir.size, qMin(size.width(), size.height())); + return QSize(result, result); + } + } + return QIconEngine::actualSize(size, mode, state); } QPixmap PixmapEntry::pixmap(const QSize &size, QIcon::Mode mode, QIcon::State state) { - Q_UNUSED(state); - - // Ensure that basePixmap is lazily initialized before generating the - // key, otherwise the cache key is not unique - if (basePixmap.isNull()) - basePixmap.load(filename); - - QSize actualSize = basePixmap.size(); - if (!actualSize.isNull() && - (actualSize.width() > size.width() || actualSize.height() > size.height())) - actualSize.scale(size, Qt::KeepAspectRatio); - - QString key = QLatin1String("$qt_theme_") % HexString<qint64>(basePixmap.cacheKey()) % - HexString<int>(mode) % - HexString<qint64>(QGuiApplication::palette().cacheKey()) % - HexString<int>(actualSize.width()) % HexString<int>(actualSize.height()); - - QPixmap cachedPixmap; - if (QPixmapCache::find(key, &cachedPixmap)) - { - return cachedPixmap; - } - else - { - if (basePixmap.size() != actualSize) - { - cachedPixmap = basePixmap.scaled(actualSize, Qt::IgnoreAspectRatio, Qt::SmoothTransformation); - } - else - { - cachedPixmap = basePixmap; - } - QPixmapCache::insert(key, cachedPixmap); - } - return cachedPixmap; + Q_UNUSED(state); + + // Ensure that basePixmap is lazily initialized before generating the + // key, otherwise the cache key is not unique + if (basePixmap.isNull()) + basePixmap.load(filename); + + QSize actualSize = basePixmap.size(); + if (!actualSize.isNull() && + (actualSize.width() > size.width() || actualSize.height() > size.height())) + actualSize.scale(size, Qt::KeepAspectRatio); + + QString key = QLatin1String("$qt_theme_") % HexString<qint64>(basePixmap.cacheKey()) % + HexString<int>(mode) % + HexString<qint64>(QGuiApplication::palette().cacheKey()) % + HexString<int>(actualSize.width()) % HexString<int>(actualSize.height()); + + QPixmap cachedPixmap; + if (QPixmapCache::find(key, &cachedPixmap)) + { + return cachedPixmap; + } + else + { + if (basePixmap.size() != actualSize) + { + cachedPixmap = basePixmap.scaled(actualSize, Qt::IgnoreAspectRatio, Qt::SmoothTransformation); + } + else + { + cachedPixmap = basePixmap; + } + QPixmapCache::insert(key, cachedPixmap); + } + return cachedPixmap; } QPixmap ScalableEntry::pixmap(const QSize &size, QIcon::Mode mode, QIcon::State state) { - if (svgIcon.isNull()) - { - svgIcon = QIcon(filename); - } + if (svgIcon.isNull()) + { + svgIcon = QIcon(filename); + } - // Simply reuse svg icon engine - return svgIcon.pixmap(size, mode, state); + // Simply reuse svg icon engine + return svgIcon.pixmap(size, mode, state); } QPixmap QIconLoaderEngineFixed::pixmap(const QSize &size, QIcon::Mode mode, QIcon::State state) { - ensureLoaded(); + ensureLoaded(); - QIconLoaderEngineEntry *entry = entryForSize(size); - if (entry) - { - return entry->pixmap(size, mode, state); - } + QIconLoaderEngineEntry *entry = entryForSize(size); + if (entry) + { + return entry->pixmap(size, mode, state); + } - return QPixmap(); + return QPixmap(); } QString QIconLoaderEngineFixed::key() const { - return QLatin1String("QIconLoaderEngineFixed"); + return QLatin1String("QIconLoaderEngineFixed"); } void QIconLoaderEngineFixed::virtual_hook(int id, void *data) { - ensureLoaded(); - - switch (id) - { - case QIconEngine::AvailableSizesHook: - { - QIconEngine::AvailableSizesArgument &arg = - *reinterpret_cast<QIconEngine::AvailableSizesArgument *>(data); - const int N = m_entries.size(); - QList<QSize> sizes; - sizes.reserve(N); - - // Gets all sizes from the DirectoryInfo entries - for (int i = 0; i < N; ++i) - { - int size = m_entries.at(i)->dir.size; - sizes.append(QSize(size, size)); - } - arg.sizes.swap(sizes); // commit - } - break; - case QIconEngine::IconNameHook: - { - QString &name = *reinterpret_cast<QString *>(data); - name = m_iconName; - } - break; - default: - QIconEngine::virtual_hook(id, data); - } + ensureLoaded(); + + switch (id) + { + case QIconEngine::AvailableSizesHook: + { + QIconEngine::AvailableSizesArgument &arg = + *reinterpret_cast<QIconEngine::AvailableSizesArgument *>(data); + const int N = m_entries.size(); + QList<QSize> sizes; + sizes.reserve(N); + + // Gets all sizes from the DirectoryInfo entries + for (int i = 0; i < N; ++i) + { + int size = m_entries.at(i)->dir.size; + sizes.append(QSize(size, size)); + } + arg.sizes.swap(sizes); // commit + } + break; + case QIconEngine::IconNameHook: + { + QString &name = *reinterpret_cast<QString *>(data); + name = m_iconName; + } + break; + default: + QIconEngine::virtual_hook(id, data); + } } } // QtXdg diff --git a/libraries/iconfix/internal/qiconloader_p.h b/libraries/iconfix/internal/qiconloader_p.h index b71bdd83..e45a08d6 100644 --- a/libraries/iconfix/internal/qiconloader_p.h +++ b/libraries/iconfix/internal/qiconloader_p.h @@ -61,46 +61,46 @@ class QIconLoader; struct QIconDirInfo { - enum Type - { - Fixed, - Scalable, - Threshold - }; - QIconDirInfo(const QString &_path = QString()) - : path(_path), size(0), maxSize(0), minSize(0), threshold(0), type(Threshold) - { - } - QString path; - short size; - short maxSize; - short minSize; - short threshold; - Type type : 4; + enum Type + { + Fixed, + Scalable, + Threshold + }; + QIconDirInfo(const QString &_path = QString()) + : path(_path), size(0), maxSize(0), minSize(0), threshold(0), type(Threshold) + { + } + QString path; + short size; + short maxSize; + short minSize; + short threshold; + Type type : 4; }; class QIconLoaderEngineEntry { public: - virtual ~QIconLoaderEngineEntry() - { - } - virtual QPixmap pixmap(const QSize &size, QIcon::Mode mode, QIcon::State state) = 0; - QString filename; - QIconDirInfo dir; - static int count; + virtual ~QIconLoaderEngineEntry() + { + } + virtual QPixmap pixmap(const QSize &size, QIcon::Mode mode, QIcon::State state) = 0; + QString filename; + QIconDirInfo dir; + static int count; }; struct ScalableEntry : public QIconLoaderEngineEntry { - QPixmap pixmap(const QSize &size, QIcon::Mode mode, QIcon::State state) Q_DECL_OVERRIDE; - QIcon svgIcon; + QPixmap pixmap(const QSize &size, QIcon::Mode mode, QIcon::State state) Q_DECL_OVERRIDE; + QIcon svgIcon; }; struct PixmapEntry : public QIconLoaderEngineEntry { - QPixmap pixmap(const QSize &size, QIcon::Mode mode, QIcon::State state) Q_DECL_OVERRIDE; - QPixmap basePixmap; + QPixmap pixmap(const QSize &size, QIcon::Mode mode, QIcon::State state) Q_DECL_OVERRIDE; + QPixmap basePixmap; }; typedef QList<QIconLoaderEngineEntry *> QThemeIconEntries; @@ -109,107 +109,107 @@ typedef QList<QIconLoaderEngineEntry *> QThemeIconEntries; class QIconLoaderEngineFixed : public QIconEngine { public: - QIconLoaderEngineFixed(const QString &iconName = QString()); - ~QIconLoaderEngineFixed(); + QIconLoaderEngineFixed(const QString &iconName = QString()); + ~QIconLoaderEngineFixed(); - void paint(QPainter *painter, const QRect &rect, QIcon::Mode mode, QIcon::State state); - QPixmap pixmap(const QSize &size, QIcon::Mode mode, QIcon::State state); - QSize actualSize(const QSize &size, QIcon::Mode mode, QIcon::State state); - QIconEngine *clone() const; - bool read(QDataStream &in); - bool write(QDataStream &out) const; + void paint(QPainter *painter, const QRect &rect, QIcon::Mode mode, QIcon::State state); + QPixmap pixmap(const QSize &size, QIcon::Mode mode, QIcon::State state); + QSize actualSize(const QSize &size, QIcon::Mode mode, QIcon::State state); + QIconEngine *clone() const; + bool read(QDataStream &in); + bool write(QDataStream &out) const; private: - QString key() const; - bool hasIcon() const; - void ensureLoaded(); - void virtual_hook(int id, void *data); - QIconLoaderEngineEntry *entryForSize(const QSize &size); - QIconLoaderEngineFixed(const QIconLoaderEngineFixed &other); - QThemeIconEntries m_entries; - QString m_iconName; - uint m_key; - - friend class QIconLoader; + QString key() const; + bool hasIcon() const; + void ensureLoaded(); + void virtual_hook(int id, void *data); + QIconLoaderEngineEntry *entryForSize(const QSize &size); + QIconLoaderEngineFixed(const QIconLoaderEngineFixed &other); + QThemeIconEntries m_entries; + QString m_iconName; + uint m_key; + + friend class QIconLoader; }; class QIconTheme { public: - QIconTheme(const QString &name); - QIconTheme() : m_valid(false) - { - } - QStringList parents() - { - return m_parents; - } - QVector<QIconDirInfo> keyList() - { - return m_keyList; - } - QString contentDir() - { - return m_contentDir; - } - QStringList contentDirs() - { - return m_contentDirs; - } - bool isValid() - { - return m_valid; - } + QIconTheme(const QString &name); + QIconTheme() : m_valid(false) + { + } + QStringList parents() + { + return m_parents; + } + QVector<QIconDirInfo> keyList() + { + return m_keyList; + } + QString contentDir() + { + return m_contentDir; + } + QStringList contentDirs() + { + return m_contentDirs; + } + bool isValid() + { + return m_valid; + } private: - QString m_contentDir; - QStringList m_contentDirs; - QVector<QIconDirInfo> m_keyList; - QStringList m_parents; - bool m_valid; + QString m_contentDir; + QStringList m_contentDirs; + QVector<QIconDirInfo> m_keyList; + QStringList m_parents; + bool m_valid; }; class QIconLoader { public: - QIconLoader(); - QThemeIconEntries loadIcon(const QString &iconName) const; - uint themeKey() const - { - return m_themeKey; - } - - QString themeName() const - { - return m_userTheme.isEmpty() ? m_systemTheme : m_userTheme; - } - void setThemeName(const QString &themeName); - QIconTheme theme() - { - return themeList.value(themeName()); - } - void setThemeSearchPath(const QStringList &searchPaths); - QStringList themeSearchPaths() const; - QIconDirInfo dirInfo(int dirindex); - static QIconLoader *instance(); - void updateSystemTheme(); - void invalidateKey() - { - m_themeKey++; - } - void ensureInitialized(); + QIconLoader(); + QThemeIconEntries loadIcon(const QString &iconName) const; + uint themeKey() const + { + return m_themeKey; + } + + QString themeName() const + { + return m_userTheme.isEmpty() ? m_systemTheme : m_userTheme; + } + void setThemeName(const QString &themeName); + QIconTheme theme() + { + return themeList.value(themeName()); + } + void setThemeSearchPath(const QStringList &searchPaths); + QStringList themeSearchPaths() const; + QIconDirInfo dirInfo(int dirindex); + static QIconLoader *instance(); + void updateSystemTheme(); + void invalidateKey() + { + m_themeKey++; + } + void ensureInitialized(); private: - QThemeIconEntries findIconHelper(const QString &themeName, const QString &iconName, - QStringList &visited) const; - uint m_themeKey; - bool m_supportsSvg; - bool m_initialized; - - mutable QString m_userTheme; - mutable QString m_systemTheme; - mutable QStringList m_iconDirs; - mutable QHash<QString, QIconTheme> themeList; + QThemeIconEntries findIconHelper(const QString &themeName, const QString &iconName, + QStringList &visited) const; + uint m_themeKey; + bool m_supportsSvg; + bool m_initialized; + + mutable QString m_userTheme; + mutable QString m_systemTheme; + mutable QStringList m_iconDirs; + mutable QHash<QString, QIconTheme> themeList; }; } // QtXdg diff --git a/libraries/iconfix/xdgicon.cpp b/libraries/iconfix/xdgicon.cpp index a36d80a9..36fb7d42 100644 --- a/libraries/iconfix/xdgicon.cpp +++ b/libraries/iconfix/xdgicon.cpp @@ -46,17 +46,17 @@ namespace { struct QtIconCache : public IconCache { - QtIconCache() - { - qAddPostRoutine(qt_cleanup_icon_cache); - } + QtIconCache() + { + qAddPostRoutine(qt_cleanup_icon_cache); + } }; } -Q_GLOBAL_STATIC(IconCache, qtIconCache); +Q_GLOBAL_STATIC(IconCache, qtIconCache) static void qt_cleanup_icon_cache() { - qtIconCache()->clear(); + qtIconCache()->clear(); } /************************************************ @@ -78,7 +78,7 @@ XdgIcon::~XdgIcon() ************************************************/ QString XdgIcon::themeName() { - return QIcon::themeName(); + return QIcon::themeName(); } /************************************************ @@ -86,8 +86,8 @@ QString XdgIcon::themeName() ************************************************/ void XdgIcon::setThemeName(const QString &themeName) { - QIcon::setThemeName(themeName); - QtXdg::QIconLoader::instance()->updateSystemTheme(); + QIcon::setThemeName(themeName); + QtXdg::QIconLoader::instance()->updateSystemTheme(); } /************************************************ @@ -96,43 +96,43 @@ void XdgIcon::setThemeName(const QString &themeName) ************************************************/ QIcon XdgIcon::fromTheme(const QString &iconName, const QIcon &fallback) { - if (iconName.isEmpty()) - return fallback; - - bool isAbsolute = (iconName[0] == '/'); - - QString name = QFileInfo(iconName).fileName(); - if (name.endsWith(".png", Qt::CaseInsensitive) || - name.endsWith(".svg", Qt::CaseInsensitive) || - name.endsWith(".xpm", Qt::CaseInsensitive)) - { - name.truncate(name.length() - 4); - } - - QIcon icon; - - if (qtIconCache()->contains(name)) - { - icon = *qtIconCache()->object(name); - } - else - { - QIcon *cachedIcon; - if (!isAbsolute) - cachedIcon = new QIcon(new QtXdg::QIconLoaderEngineFixed(name)); - else - cachedIcon = new QIcon(iconName); - qtIconCache()->insert(name, cachedIcon); - icon = *cachedIcon; - } - - // Note the qapp check is to allow lazy loading of static icons - // Supporting fallbacks will not work for this case. - if (qApp && !isAbsolute && icon.availableSizes().isEmpty()) - { - return fallback; - } - return icon; + if (iconName.isEmpty()) + return fallback; + + bool isAbsolute = (iconName[0] == '/'); + + QString name = QFileInfo(iconName).fileName(); + if (name.endsWith(".png", Qt::CaseInsensitive) || + name.endsWith(".svg", Qt::CaseInsensitive) || + name.endsWith(".xpm", Qt::CaseInsensitive)) + { + name.truncate(name.length() - 4); + } + + QIcon icon; + + if (qtIconCache()->contains(name)) + { + icon = *qtIconCache()->object(name); + } + else + { + QIcon *cachedIcon; + if (!isAbsolute) + cachedIcon = new QIcon(new QtXdg::QIconLoaderEngineFixed(name)); + else + cachedIcon = new QIcon(iconName); + qtIconCache()->insert(name, cachedIcon); + icon = *cachedIcon; + } + + // Note the qapp check is to allow lazy loading of static icons + // Supporting fallbacks will not work for this case. + if (qApp && !isAbsolute && icon.availableSizes().isEmpty()) + { + return fallback; + } + return icon; } /************************************************ @@ -141,12 +141,12 @@ QIcon XdgIcon::fromTheme(const QString &iconName, const QIcon &fallback) ************************************************/ QIcon XdgIcon::fromTheme(const QStringList &iconNames, const QIcon &fallback) { - foreach (QString iconName, iconNames) - { - QIcon icon = fromTheme(iconName); - if (!icon.isNull()) - return icon; - } - - return fallback; + foreach (QString iconName, iconNames) + { + QIcon icon = fromTheme(iconName); + if (!icon.isNull()) + return icon; + } + + return fallback; } diff --git a/libraries/iconfix/xdgicon.h b/libraries/iconfix/xdgicon.h index df8f026e..1380607c 100644 --- a/libraries/iconfix/xdgicon.h +++ b/libraries/iconfix/xdgicon.h @@ -36,13 +36,13 @@ class MULTIMC_ICONFIX_EXPORT XdgIcon { public: - static QIcon fromTheme(const QString &iconName, const QIcon &fallback = QIcon()); - static QIcon fromTheme(const QStringList &iconNames, const QIcon &fallback = QIcon()); + static QIcon fromTheme(const QString &iconName, const QIcon &fallback = QIcon()); + static QIcon fromTheme(const QStringList &iconNames, const QIcon &fallback = QIcon()); - static QString themeName(); - static void setThemeName(const QString &themeName); + static QString themeName(); + static void setThemeName(const QString &themeName); protected: - explicit XdgIcon(); - virtual ~XdgIcon(); + explicit XdgIcon(); + virtual ~XdgIcon(); }; diff --git a/libraries/javacheck/CMakeLists.txt b/libraries/javacheck/CMakeLists.txt index 381efe08..dba5a1ae 100644 --- a/libraries/javacheck/CMakeLists.txt +++ b/libraries/javacheck/CMakeLists.txt @@ -7,7 +7,7 @@ set(CMAKE_JAVA_JAR_ENTRY_POINT JavaCheck) set(CMAKE_JAVA_COMPILE_FLAGS -target 1.6 -source 1.6 -Xlint:deprecation -Xlint:unchecked) set(SRC - JavaCheck.java + JavaCheck.java ) add_jar(JavaCheck ${SRC}) diff --git a/libraries/javacheck/JavaCheck.java b/libraries/javacheck/JavaCheck.java index 11420b86..69933040 100644 --- a/libraries/javacheck/JavaCheck.java +++ b/libraries/javacheck/JavaCheck.java @@ -2,23 +2,23 @@ import java.lang.Integer; public class JavaCheck { - private static final String[] keys = {"os.arch", "java.version"}; - public static void main (String [] args) - { - int ret = 0; - for(String key : keys) - { - String property = System.getProperty(key); - if(property != null) - { - System.out.println(key + "=" + property); - } - else - { - ret = 1; - } - } - - System.exit(ret); - } + private static final String[] keys = {"os.arch", "java.version"}; + public static void main (String [] args) + { + int ret = 0; + for(String key : keys) + { + String property = System.getProperty(key); + if(property != null) + { + System.out.println(key + "=" + property); + } + else + { + ret = 1; + } + } + + System.exit(ret); + } } diff --git a/libraries/launcher/CMakeLists.txt b/libraries/launcher/CMakeLists.txt index a4f52edb..a64c601d 100644 --- a/libraries/launcher/CMakeLists.txt +++ b/libraries/launcher/CMakeLists.txt @@ -7,15 +7,15 @@ set(CMAKE_JAVA_JAR_ENTRY_POINT org.multimc.EntryPoint) set(CMAKE_JAVA_COMPILE_FLAGS -target 1.6 -source 1.6 -Xlint:deprecation -Xlint:unchecked) set(SRC - org/multimc/EntryPoint.java - org/multimc/Launcher.java - org/multimc/LegacyFrame.java - org/multimc/NotFoundException.java - org/multimc/ParamBucket.java - org/multimc/ParseException.java - org/multimc/Utils.java - org/multimc/onesix/OneSixLauncher.java - net/minecraft/Launcher.java + org/multimc/EntryPoint.java + org/multimc/Launcher.java + org/multimc/LegacyFrame.java + org/multimc/NotFoundException.java + org/multimc/ParamBucket.java + org/multimc/ParseException.java + org/multimc/Utils.java + org/multimc/onesix/OneSixLauncher.java + net/minecraft/Launcher.java ) add_jar(NewLaunch ${SRC}) install_jar(NewLaunch "${JARS_DEST_DIR}") diff --git a/libraries/launcher/net/minecraft/Launcher.java b/libraries/launcher/net/minecraft/Launcher.java index 0c991cf5..dd704484 100644 --- a/libraries/launcher/net/minecraft/Launcher.java +++ b/libraries/launcher/net/minecraft/Launcher.java @@ -1,5 +1,5 @@ /* - * Copyright 2012-2018 MultiMC Contributors + * Copyright 2012-2019 MultiMC Contributors * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. @@ -28,138 +28,138 @@ import java.net.MalformedURLException; public class Launcher extends Applet implements AppletStub { - private Applet wrappedApplet; - private URL documentBase; - private boolean active = false; - private final Map<String, String> params; - - public Launcher(Applet applet, URL documentBase) - { - params = new TreeMap<String, String>(); - - this.setLayout(new BorderLayout()); - this.add(applet, "Center"); - this.wrappedApplet = applet; - this.documentBase = documentBase; - } - - public void setParameter(String name, String value) - { - params.put(name, value); - } - - public void replace(Applet applet) - { - this.wrappedApplet = applet; - - applet.setStub(this); - applet.setSize(getWidth(), getHeight()); - - this.setLayout(new BorderLayout()); - this.add(applet, "Center"); - - applet.init(); - active = true; - applet.start(); - validate(); - } - - @Override - public String getParameter(String name) - { - String param = params.get(name); - if (param != null) - return param; - try - { - return super.getParameter(name); - } catch (Exception ignore){} - return null; - } - - @Override - public boolean isActive() - { - return active; - } - - @Override - public void appletResize(int width, int height) - { - wrappedApplet.resize(width, height); - } - - @Override - public void resize(int width, int height) - { - wrappedApplet.resize(width, height); - } - - @Override - public void resize(Dimension d) - { - wrappedApplet.resize(d); - } - - @Override - public void init() - { - if (wrappedApplet != null) - { - wrappedApplet.init(); - } - } - - @Override - public void start() - { - wrappedApplet.start(); - active = true; - } - - @Override - public void stop() - { - wrappedApplet.stop(); - active = false; - } - - public void destroy() - { - wrappedApplet.destroy(); - } - - @Override - public URL getCodeBase() { - try { - return new URL("http://www.minecraft.net/game/"); - } catch (MalformedURLException e) { - e.printStackTrace(); - } - return null; - } - - @Override - public URL getDocumentBase() - { - try { - return new URL("http://www.minecraft.net/game/"); - } catch (MalformedURLException e) { - e.printStackTrace(); - } - return null; - } - - @Override - public void setVisible(boolean b) - { - super.setVisible(b); - wrappedApplet.setVisible(b); - } - public void update(Graphics paramGraphics) - { - } - public void paint(Graphics paramGraphics) - { - } + private Applet wrappedApplet; + private URL documentBase; + private boolean active = false; + private final Map<String, String> params; + + public Launcher(Applet applet, URL documentBase) + { + params = new TreeMap<String, String>(); + + this.setLayout(new BorderLayout()); + this.add(applet, "Center"); + this.wrappedApplet = applet; + this.documentBase = documentBase; + } + + public void setParameter(String name, String value) + { + params.put(name, value); + } + + public void replace(Applet applet) + { + this.wrappedApplet = applet; + + applet.setStub(this); + applet.setSize(getWidth(), getHeight()); + + this.setLayout(new BorderLayout()); + this.add(applet, "Center"); + + applet.init(); + active = true; + applet.start(); + validate(); + } + + @Override + public String getParameter(String name) + { + String param = params.get(name); + if (param != null) + return param; + try + { + return super.getParameter(name); + } catch (Exception ignore){} + return null; + } + + @Override + public boolean isActive() + { + return active; + } + + @Override + public void appletResize(int width, int height) + { + wrappedApplet.resize(width, height); + } + + @Override + public void resize(int width, int height) + { + wrappedApplet.resize(width, height); + } + + @Override + public void resize(Dimension d) + { + wrappedApplet.resize(d); + } + + @Override + public void init() + { + if (wrappedApplet != null) + { + wrappedApplet.init(); + } + } + + @Override + public void start() + { + wrappedApplet.start(); + active = true; + } + + @Override + public void stop() + { + wrappedApplet.stop(); + active = false; + } + + public void destroy() + { + wrappedApplet.destroy(); + } + + @Override + public URL getCodeBase() { + try { + return new URL("http://www.minecraft.net/game/"); + } catch (MalformedURLException e) { + e.printStackTrace(); + } + return null; + } + + @Override + public URL getDocumentBase() + { + try { + return new URL("http://www.minecraft.net/game/"); + } catch (MalformedURLException e) { + e.printStackTrace(); + } + return null; + } + + @Override + public void setVisible(boolean b) + { + super.setVisible(b); + wrappedApplet.setVisible(b); + } + public void update(Graphics paramGraphics) + { + } + public void paint(Graphics paramGraphics) + { + } }
\ No newline at end of file diff --git a/libraries/launcher/org/multimc/EntryPoint.java b/libraries/launcher/org/multimc/EntryPoint.java index 8c9b8074..12a494b9 100644 --- a/libraries/launcher/org/multimc/EntryPoint.java +++ b/libraries/launcher/org/multimc/EntryPoint.java @@ -1,5 +1,5 @@ package org.multimc;/* - * Copyright 2012-2018 MultiMC Contributors + * Copyright 2012-2019 MultiMC Contributors * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. @@ -21,131 +21,131 @@ import java.nio.charset.Charset; public class EntryPoint { - private enum Action - { - Proceed, - Launch, - Abort - } + private enum Action + { + Proceed, + Launch, + Abort + } - public static void main(String[] args) - { - EntryPoint listener = new EntryPoint(); - int retCode = listener.listen(); - if (retCode != 0) - { - System.out.println("Exiting with " + retCode); - System.exit(retCode); - } - } + public static void main(String[] args) + { + EntryPoint listener = new EntryPoint(); + int retCode = listener.listen(); + if (retCode != 0) + { + System.out.println("Exiting with " + retCode); + System.exit(retCode); + } + } - private Action parseLine(String inData) throws ParseException - { - String[] pair = inData.split(" ", 2); + private Action parseLine(String inData) throws ParseException + { + String[] pair = inData.split(" ", 2); - if(pair.length == 1) - { - String command = pair[0]; - if (pair[0].equals("launch")) - return Action.Launch; + if(pair.length == 1) + { + String command = pair[0]; + if (pair[0].equals("launch")) + return Action.Launch; - else if (pair[0].equals("abort")) - return Action.Abort; + else if (pair[0].equals("abort")) + return Action.Abort; - else throw new ParseException(); - } + else throw new ParseException("Error while parsing:" + pair[0]); + } - if(pair.length != 2) - throw new ParseException(); + if(pair.length != 2) + throw new ParseException("Pair length is not 2."); - String command = pair[0]; - String param = pair[1]; + String command = pair[0]; + String param = pair[1]; - if(command.equals("launcher")) - { - if(param.equals("onesix")) - { - m_launcher = new OneSixLauncher(); - Utils.log("Using onesix launcher."); - Utils.log(); - return Action.Proceed; - } - else - throw new ParseException(); - } + if(command.equals("launcher")) + { + if(param.equals("onesix")) + { + m_launcher = new OneSixLauncher(); + Utils.log("Using onesix launcher."); + Utils.log(); + return Action.Proceed; + } + else + throw new ParseException("Invalid launcher type: " + param); + } - m_params.add(command, param); - //System.out.println(command + " : " + param); - return Action.Proceed; - } + m_params.add(command, param); + //System.out.println(command + " : " + param); + return Action.Proceed; + } - public int listen() - { - BufferedReader buffer; - try - { - buffer = new BufferedReader(new InputStreamReader(System.in, "UTF-8")); - } catch (UnsupportedEncodingException e) - { - System.err.println("For some reason, your java does not support UTF-8. Consider living in the current century."); - e.printStackTrace(); - return 1; - } - boolean isListening = true; - boolean isAborted = false; - // Main loop - while (isListening) - { - String inData; - try - { - // Read from the pipe one line at a time - inData = buffer.readLine(); - if (inData != null) - { - Action a = parseLine(inData); - if(a == Action.Abort) - { - isListening = false; - isAborted = true; - } - if(a == Action.Launch) - { - isListening = false; - } - } - else - { - isListening = false; - isAborted = true; - } - } - catch (IOException e) - { - System.err.println("Launcher ABORT due to IO exception:"); - e.printStackTrace(); - return 1; - } - catch (ParseException e) - { - System.err.println("Launcher ABORT due to PARSE exception:"); - e.printStackTrace(); - return 1; - } - } - if(isAborted) - { - System.err.println("Launch aborted by MultiMC."); - return 1; - } - if(m_launcher != null) - { - return m_launcher.launch(m_params); - } - System.err.println("No valid launcher implementation specified."); - return 1; - } + public int listen() + { + BufferedReader buffer; + try + { + buffer = new BufferedReader(new InputStreamReader(System.in, "UTF-8")); + } catch (UnsupportedEncodingException e) + { + System.err.println("For some reason, your java does not support UTF-8. Consider living in the current century."); + e.printStackTrace(); + return 1; + } + boolean isListening = true; + boolean isAborted = false; + // Main loop + while (isListening) + { + String inData; + try + { + // Read from the pipe one line at a time + inData = buffer.readLine(); + if (inData != null) + { + Action a = parseLine(inData); + if(a == Action.Abort) + { + isListening = false; + isAborted = true; + } + if(a == Action.Launch) + { + isListening = false; + } + } + else + { + isListening = false; + isAborted = true; + } + } + catch (IOException e) + { + System.err.println("Launcher ABORT due to IO exception:"); + e.printStackTrace(); + return 1; + } + catch (ParseException e) + { + System.err.println("Launcher ABORT due to PARSE exception:"); + e.printStackTrace(); + return 1; + } + } + if(isAborted) + { + System.err.println("Launch aborted by MultiMC."); + return 1; + } + if(m_launcher != null) + { + return m_launcher.launch(m_params); + } + System.err.println("No valid launcher implementation specified."); + return 1; + } - private ParamBucket m_params = new ParamBucket(); - private org.multimc.Launcher m_launcher; + private ParamBucket m_params = new ParamBucket(); + private org.multimc.Launcher m_launcher; } diff --git a/libraries/launcher/org/multimc/Launcher.java b/libraries/launcher/org/multimc/Launcher.java index 2e851d18..8f9b043f 100644 --- a/libraries/launcher/org/multimc/Launcher.java +++ b/libraries/launcher/org/multimc/Launcher.java @@ -1,5 +1,5 @@ /* - * Copyright 2012-2018 MultiMC Contributors + * Copyright 2012-2019 MultiMC Contributors * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. @@ -18,5 +18,5 @@ package org.multimc; public interface Launcher { - abstract int launch(ParamBucket params); + abstract int launch(ParamBucket params); } diff --git a/libraries/launcher/org/multimc/LegacyFrame.java b/libraries/launcher/org/multimc/LegacyFrame.java index 9842eb0e..19cfdfb7 100644 --- a/libraries/launcher/org/multimc/LegacyFrame.java +++ b/libraries/launcher/org/multimc/LegacyFrame.java @@ -1,5 +1,5 @@ package org.multimc;/* - * Copyright 2012-2018 MultiMC Contributors + * Copyright 2012-2019 MultiMC Contributors * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. diff --git a/libraries/launcher/org/multimc/NotFoundException.java b/libraries/launcher/org/multimc/NotFoundException.java index 2c5da6de..c1084fe6 100644 --- a/libraries/launcher/org/multimc/NotFoundException.java +++ b/libraries/launcher/org/multimc/NotFoundException.java @@ -1,5 +1,5 @@ /* - * Copyright 2012-2018 MultiMC Contributors + * Copyright 2012-2019 MultiMC Contributors * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. diff --git a/libraries/launcher/org/multimc/ParamBucket.java b/libraries/launcher/org/multimc/ParamBucket.java index 5e9c3ff6..f5b40c40 100644 --- a/libraries/launcher/org/multimc/ParamBucket.java +++ b/libraries/launcher/org/multimc/ParamBucket.java @@ -1,5 +1,5 @@ /* - * Copyright 2012-2018 MultiMC Contributors + * Copyright 2012-2019 MultiMC Contributors * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. @@ -22,65 +22,65 @@ import java.util.List; public class ParamBucket { - public void add(String key, String value) - { - List<String> coll = null; - if(!m_params.containsKey(key)) - { - coll = new ArrayList<String>(); - m_params.put(key, coll); - } - else - { - coll = m_params.get(key); - } - coll.add(value); - } + public void add(String key, String value) + { + List<String> coll = null; + if(!m_params.containsKey(key)) + { + coll = new ArrayList<String>(); + m_params.put(key, coll); + } + else + { + coll = m_params.get(key); + } + coll.add(value); + } - public List<String> all(String key) throws NotFoundException - { - if(!m_params.containsKey(key)) - throw new NotFoundException(); - return m_params.get(key); - } + public List<String> all(String key) throws NotFoundException + { + if(!m_params.containsKey(key)) + throw new NotFoundException(); + return m_params.get(key); + } - public List<String> allSafe(String key, List<String> def) - { - if(!m_params.containsKey(key) || m_params.get(key).size() < 1) - { - return def; - } - return m_params.get(key); - } + public List<String> allSafe(String key, List<String> def) + { + if(!m_params.containsKey(key) || m_params.get(key).size() < 1) + { + return def; + } + return m_params.get(key); + } - public List<String> allSafe(String key) - { - return allSafe(key, new ArrayList<String>()); - } + public List<String> allSafe(String key) + { + return allSafe(key, new ArrayList<String>()); + } - public String first(String key) throws NotFoundException - { - List<String> list = all(key); - if(list.size() < 1) - { - throw new NotFoundException(); - } - return list.get(0); - } + public String first(String key) throws NotFoundException + { + List<String> list = all(key); + if(list.size() < 1) + { + throw new NotFoundException(); + } + return list.get(0); + } - public String firstSafe(String key, String def) - { - if(!m_params.containsKey(key) || m_params.get(key).size() < 1) - { - return def; - } - return m_params.get(key).get(0); - } + public String firstSafe(String key, String def) + { + if(!m_params.containsKey(key) || m_params.get(key).size() < 1) + { + return def; + } + return m_params.get(key).get(0); + } - public String firstSafe(String key) - { - return firstSafe(key, ""); - } + public String firstSafe(String key) + { + return firstSafe(key, ""); + } - private HashMap<String, List<String>> m_params = new HashMap<String, List<String>>(); + private HashMap<String, List<String>> m_params = new HashMap<String, List<String>>(); } diff --git a/libraries/launcher/org/multimc/ParseException.java b/libraries/launcher/org/multimc/ParseException.java index 9a8fe521..e9c84f6e 100644 --- a/libraries/launcher/org/multimc/ParseException.java +++ b/libraries/launcher/org/multimc/ParseException.java @@ -1,5 +1,5 @@ /* - * Copyright 2012-2018 MultiMC Contributors + * Copyright 2012-2019 MultiMC Contributors * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. @@ -18,5 +18,8 @@ package org.multimc; public class ParseException extends java.lang.Exception { - + public ParseException() { super(); } + public ParseException(String message) { + super(message); + } } diff --git a/libraries/launcher/org/multimc/Utils.java b/libraries/launcher/org/multimc/Utils.java index c5292eaf..b0a2d5b0 100644 --- a/libraries/launcher/org/multimc/Utils.java +++ b/libraries/launcher/org/multimc/Utils.java @@ -1,5 +1,5 @@ /* - * Copyright 2012-2018 MultiMC Contributors + * Copyright 2012-2019 MultiMC Contributors * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. @@ -34,86 +34,86 @@ import java.util.zip.ZipFile; public class Utils { - /** - * Combine two parts of a path. - * - * @param path1 - * @param path2 - * @return the paths, combined - */ - public static String combine(String path1, String path2) - { - File file1 = new File(path1); - File file2 = new File(file1, path2); - return file2.getPath(); - } + /** + * Combine two parts of a path. + * + * @param path1 + * @param path2 + * @return the paths, combined + */ + public static String combine(String path1, String path2) + { + File file1 = new File(path1); + File file2 = new File(file1, path2); + return file2.getPath(); + } - /** - * Join a list of strings into a string using a separator! - * - * @param strings the string list to join - * @param separator the glue - * @return the result. - */ - public static String join(List<String> strings, String separator) - { - StringBuilder sb = new StringBuilder(); - String sep = ""; - for (String s : strings) - { - sb.append(sep).append(s); - sep = separator; - } - return sb.toString(); - } + /** + * Join a list of strings into a string using a separator! + * + * @param strings the string list to join + * @param separator the glue + * @return the result. + */ + public static String join(List<String> strings, String separator) + { + StringBuilder sb = new StringBuilder(); + String sep = ""; + for (String s : strings) + { + sb.append(sep).append(s); + sep = separator; + } + return sb.toString(); + } - /** - * Finds a field that looks like a Minecraft base folder in a supplied class - * - * @param mc the class to scan - */ - public static Field getMCPathField(Class<?> mc) - { - Field[] fields = mc.getDeclaredFields(); + /** + * Finds a field that looks like a Minecraft base folder in a supplied class + * + * @param mc the class to scan + */ + public static Field getMCPathField(Class<?> mc) + { + Field[] fields = mc.getDeclaredFields(); - for (Field f : fields) - { - if (f.getType() != File.class) - { - // Has to be File - continue; - } - if (f.getModifiers() != (Modifier.PRIVATE + Modifier.STATIC)) - { - // And Private Static. - continue; - } - return f; - } - return null; - } + for (Field f : fields) + { + if (f.getType() != File.class) + { + // Has to be File + continue; + } + if (f.getModifiers() != (Modifier.PRIVATE + Modifier.STATIC)) + { + // And Private Static. + continue; + } + return f; + } + return null; + } - /** - * Log to the MultiMC console - * - * @param message A String containing the message - * @param level A String containing the level name. See MinecraftLauncher::getLevel() - */ - public static void log(String message, String level) - { - // Kinda dirty - String tag = "!![" + level + "]!"; - System.out.println(tag + message.replace("\n", "\n" + tag)); - } + /** + * Log to the MultiMC console + * + * @param message A String containing the message + * @param level A String containing the level name. See MinecraftLauncher::getLevel() + */ + public static void log(String message, String level) + { + // Kinda dirty + String tag = "!![" + level + "]!"; + System.out.println(tag + message.replace("\n", "\n" + tag)); + } - public static void log(String message) - { - log(message, "MultiMC"); - } + public static void log(String message) + { + log(message, "MultiMC"); + } - public static void log() - { - System.out.println(); - } + public static void log() + { + System.out.println(); + } } diff --git a/libraries/launcher/org/multimc/onesix/OneSixLauncher.java b/libraries/launcher/org/multimc/onesix/OneSixLauncher.java index 9667297d..ec688ee3 100644 --- a/libraries/launcher/org/multimc/onesix/OneSixLauncher.java +++ b/libraries/launcher/org/multimc/onesix/OneSixLauncher.java @@ -1,4 +1,4 @@ -/* Copyright 2012-2018 MultiMC Contributors +/* Copyright 2012-2019 MultiMC Contributors * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. @@ -27,208 +27,208 @@ import java.util.List; public class OneSixLauncher implements Launcher { - // parameters, separated from ParamBucket - private List<String> libraries; - private List<String> mcparams; - private List<String> mods; - private List<String> jarmods; - private List<String> coremods; - private List<String> traits; - private String appletClass; - private String mainClass; - private String nativePath; - private String userName, sessionId; - private String windowTitle; - private String windowParams; - - // secondary parameters - private int winSizeW; - private int winSizeH; - private boolean maximize; - private String cwd; - - // the much abused system classloader, for convenience (for further abuse) - private ClassLoader cl; - - private void processParams(ParamBucket params) throws NotFoundException - { - libraries = params.all("cp"); - mcparams = params.allSafe("param", new ArrayList<String>() ); - mainClass = params.firstSafe("mainClass", "net.minecraft.client.Minecraft"); - appletClass = params.firstSafe("appletClass", "net.minecraft.client.MinecraftApplet"); - traits = params.allSafe("traits", new ArrayList<String>()); - nativePath = params.first("natives"); - - userName = params.first("userName"); - sessionId = params.first("sessionId"); - windowTitle = params.firstSafe("windowTitle", "Minecraft"); - windowParams = params.firstSafe("windowParams", "854x480"); - - cwd = System.getProperty("user.dir"); - - winSizeW = 854; - winSizeH = 480; - maximize = false; - - String[] dimStrings = windowParams.split("x"); - - if (windowParams.equalsIgnoreCase("max")) - { - maximize = true; - } - else if (dimStrings.length == 2) - { - try - { - winSizeW = Integer.parseInt(dimStrings[0]); - winSizeH = Integer.parseInt(dimStrings[1]); - } catch (NumberFormatException ignored) {} - } - } - - int legacyLaunch() - { - // Get the Minecraft Class and set the base folder - Class<?> mc; - try - { - mc = cl.loadClass(mainClass); - - Field f = Utils.getMCPathField(mc); - - if (f == null) - { - System.err.println("Could not find Minecraft path field."); - } - else - { - f.setAccessible(true); - f.set(null, new File(cwd)); - } - } catch (Exception e) - { - System.err.println("Could not set base folder. Failed to find/access Minecraft main class:"); - e.printStackTrace(System.err); - return -1; - } - - System.setProperty("minecraft.applet.TargetDirectory", cwd); - - if(!traits.contains("noapplet")) - { - Utils.log("Launching with applet wrapper..."); - try - { - Class<?> MCAppletClass = cl.loadClass(appletClass); - Applet mcappl = (Applet) MCAppletClass.newInstance(); - LegacyFrame mcWindow = new LegacyFrame(windowTitle); - mcWindow.start(mcappl, userName, sessionId, winSizeW, winSizeH, maximize); - return 0; - } catch (Exception e) - { - Utils.log("Applet wrapper failed:", "Error"); - e.printStackTrace(System.err); - Utils.log(); - Utils.log("Falling back to using main class."); - } - } - - // init params for the main method to chomp on. - String[] paramsArray = mcparams.toArray(new String[mcparams.size()]); - try - { - mc.getMethod("main", String[].class).invoke(null, (Object) paramsArray); - return 0; - } catch (Exception e) - { - Utils.log("Failed to invoke the Minecraft main class:", "Fatal"); - e.printStackTrace(System.err); - return -1; - } - } - - int launchWithMainClass() - { - // window size, title and state, onesix - if (maximize) - { - // FIXME: there is no good way to maximize the minecraft window in onesix. - // the following often breaks linux screen setups - // mcparams.add("--fullscreen"); - } - else - { - mcparams.add("--width"); - mcparams.add(Integer.toString(winSizeW)); - mcparams.add("--height"); - mcparams.add(Integer.toString(winSizeH)); - } - - // Get the Minecraft Class. - Class<?> mc; - try - { - mc = cl.loadClass(mainClass); - } catch (ClassNotFoundException e) - { - System.err.println("Failed to find Minecraft main class:"); - e.printStackTrace(System.err); - return -1; - } - - // get the main method. - Method meth; - try - { - meth = mc.getMethod("main", String[].class); - } catch (NoSuchMethodException e) - { - System.err.println("Failed to acquire the main method:"); - e.printStackTrace(System.err); - return -1; - } - - // init params for the main method to chomp on. - String[] paramsArray = mcparams.toArray(new String[mcparams.size()]); - try - { - // static method doesn't have an instance - meth.invoke(null, (Object) paramsArray); - } catch (Exception e) - { - System.err.println("Failed to start Minecraft:"); - e.printStackTrace(System.err); - return -1; - } - return 0; - } - - @Override - public int launch(ParamBucket params) - { - // get and process the launch script params - try - { - processParams(params); - } catch (NotFoundException e) - { - System.err.println("Not enough arguments."); - e.printStackTrace(System.err); - return -1; - } - - // grab the system classloader and ... - cl = ClassLoader.getSystemClassLoader(); - - if (traits.contains("legacyLaunch") || traits.contains("alphaLaunch") ) - { - // legacy launch uses the applet wrapper - return legacyLaunch(); - } - else - { - // normal launch just calls main() - return launchWithMainClass(); - } - } + // parameters, separated from ParamBucket + private List<String> libraries; + private List<String> mcparams; + private List<String> mods; + private List<String> jarmods; + private List<String> coremods; + private List<String> traits; + private String appletClass; + private String mainClass; + private String nativePath; + private String userName, sessionId; + private String windowTitle; + private String windowParams; + + // secondary parameters + private int winSizeW; + private int winSizeH; + private boolean maximize; + private String cwd; + + // the much abused system classloader, for convenience (for further abuse) + private ClassLoader cl; + + private void processParams(ParamBucket params) throws NotFoundException + { + libraries = params.all("cp"); + mcparams = params.allSafe("param", new ArrayList<String>() ); + mainClass = params.firstSafe("mainClass", "net.minecraft.client.Minecraft"); + appletClass = params.firstSafe("appletClass", "net.minecraft.client.MinecraftApplet"); + traits = params.allSafe("traits", new ArrayList<String>()); + nativePath = params.first("natives"); + + userName = params.first("userName"); + sessionId = params.first("sessionId"); + windowTitle = params.firstSafe("windowTitle", "Minecraft"); + windowParams = params.firstSafe("windowParams", "854x480"); + + cwd = System.getProperty("user.dir"); + + winSizeW = 854; + winSizeH = 480; + maximize = false; + + String[] dimStrings = windowParams.split("x"); + + if (windowParams.equalsIgnoreCase("max")) + { + maximize = true; + } + else if (dimStrings.length == 2) + { + try + { + winSizeW = Integer.parseInt(dimStrings[0]); + winSizeH = Integer.parseInt(dimStrings[1]); + } catch (NumberFormatException ignored) {} + } + } + + int legacyLaunch() + { + // Get the Minecraft Class and set the base folder + Class<?> mc; + try + { + mc = cl.loadClass(mainClass); + + Field f = Utils.getMCPathField(mc); + + if (f == null) + { + System.err.println("Could not find Minecraft path field."); + } + else + { + f.setAccessible(true); + f.set(null, new File(cwd)); + } + } catch (Exception e) + { + System.err.println("Could not set base folder. Failed to find/access Minecraft main class:"); + e.printStackTrace(System.err); + return -1; + } + + System.setProperty("minecraft.applet.TargetDirectory", cwd); + + if(!traits.contains("noapplet")) + { + Utils.log("Launching with applet wrapper..."); + try + { + Class<?> MCAppletClass = cl.loadClass(appletClass); + Applet mcappl = (Applet) MCAppletClass.newInstance(); + LegacyFrame mcWindow = new LegacyFrame(windowTitle); + mcWindow.start(mcappl, userName, sessionId, winSizeW, winSizeH, maximize); + return 0; + } catch (Exception e) + { + Utils.log("Applet wrapper failed:", "Error"); + e.printStackTrace(System.err); + Utils.log(); + Utils.log("Falling back to using main class."); + } + } + + // init params for the main method to chomp on. + String[] paramsArray = mcparams.toArray(new String[mcparams.size()]); + try + { + mc.getMethod("main", String[].class).invoke(null, (Object) paramsArray); + return 0; + } catch (Exception e) + { + Utils.log("Failed to invoke the Minecraft main class:", "Fatal"); + e.printStackTrace(System.err); + return -1; + } + } + + int launchWithMainClass() + { + // window size, title and state, onesix + if (maximize) + { + // FIXME: there is no good way to maximize the minecraft window in onesix. + // the following often breaks linux screen setups + // mcparams.add("--fullscreen"); + } + else + { + mcparams.add("--width"); + mcparams.add(Integer.toString(winSizeW)); + mcparams.add("--height"); + mcparams.add(Integer.toString(winSizeH)); + } + + // Get the Minecraft Class. + Class<?> mc; + try + { + mc = cl.loadClass(mainClass); + } catch (ClassNotFoundException e) + { + System.err.println("Failed to find Minecraft main class:"); + e.printStackTrace(System.err); + return -1; + } + + // get the main method. + Method meth; + try + { + meth = mc.getMethod("main", String[].class); + } catch (NoSuchMethodException e) + { + System.err.println("Failed to acquire the main method:"); + e.printStackTrace(System.err); + return -1; + } + + // init params for the main method to chomp on. + String[] paramsArray = mcparams.toArray(new String[mcparams.size()]); + try + { + // static method doesn't have an instance + meth.invoke(null, (Object) paramsArray); + } catch (Exception e) + { + System.err.println("Failed to start Minecraft:"); + e.printStackTrace(System.err); + return -1; + } + return 0; + } + + @Override + public int launch(ParamBucket params) + { + // get and process the launch script params + try + { + processParams(params); + } catch (NotFoundException e) + { + System.err.println("Not enough arguments."); + e.printStackTrace(System.err); + return -1; + } + + // grab the system classloader and ... + cl = ClassLoader.getSystemClassLoader(); + + if (traits.contains("legacyLaunch") || traits.contains("alphaLaunch") ) + { + // legacy launch uses the applet wrapper + return legacyLaunch(); + } + else + { + // normal launch just calls main() + return launchWithMainClass(); + } + } } diff --git a/libraries/pack200/CMakeLists.txt b/libraries/pack200/CMakeLists.txt index b568e506..31eb0f73 100644 --- a/libraries/pack200/CMakeLists.txt +++ b/libraries/pack200/CMakeLists.txt @@ -8,22 +8,22 @@ option(PACK200_BUILD_BINARY "Build a tiny utility that decompresses pack200 stre find_package(ZLIB REQUIRED) set(PACK200_SRC - include/unpack200.h - src/bands.cpp - src/bands.h - src/bytes.cpp - src/bytes.h - src/coding.cpp - src/coding.h - src/constants.h - src/defines.h - src/unpack200.cpp - src/unpack.cpp - src/unpack.h - src/utils.cpp - src/utils.h - src/zip.cpp - src/zip.h + include/unpack200.h + src/bands.cpp + src/bands.h + src/bytes.cpp + src/bytes.h + src/coding.cpp + src/coding.h + src/constants.h + src/defines.h + src/unpack200.cpp + src/unpack.cpp + src/unpack.h + src/utils.cpp + src/utils.h + src/zip.cpp + src/zip.h ) if (Qt5_POSITION_INDEPENDENT_CODE) @@ -39,12 +39,12 @@ generate_export_header(MultiMC_unpack200) # Install it install( - TARGETS MultiMC_unpack200 - RUNTIME DESTINATION ${LIBRARY_DEST_DIR} - LIBRARY DESTINATION ${LIBRARY_DEST_DIR} + TARGETS MultiMC_unpack200 + RUNTIME DESTINATION ${LIBRARY_DEST_DIR} + LIBRARY DESTINATION ${LIBRARY_DEST_DIR} ) if(PACK200_BUILD_BINARY) - add_executable(anti200 anti200.cpp) - target_link_libraries(anti200 MultiMC_unpack200) + add_executable(anti200 anti200.cpp) + target_link_libraries(anti200 MultiMC_unpack200) endif() diff --git a/libraries/pack200/anti200.cpp b/libraries/pack200/anti200.cpp index 944e80e5..1e672847 100644 --- a/libraries/pack200/anti200.cpp +++ b/libraries/pack200/anti200.cpp @@ -8,36 +8,36 @@ int main(int argc, char **argv) { - if (argc != 3) - { - std::cerr << "Simple pack200 unpacker!" << std::endl << "Run like this:" << std::endl - << " " << argv[0] << " input.jar.lzma output.jar" << std::endl; - return EXIT_FAILURE; - } + if (argc != 3) + { + std::cerr << "Simple pack200 unpacker!" << std::endl << "Run like this:" << std::endl + << " " << argv[0] << " input.jar.lzma output.jar" << std::endl; + return EXIT_FAILURE; + } - FILE *input = fopen(argv[1], "rb"); - if (!input) - { - std::cerr << "Can't open input file"; - return EXIT_FAILURE; - } - FILE *output = fopen(argv[2], "wb"); - if (!output) - { - fclose(input); - std::cerr << "Can't open output file"; - return EXIT_FAILURE; - } - try - { - unpack_200(input, output); - } - catch (std::runtime_error &e) - { - std::cerr << "Bad things happened: " << e.what() << std::endl; - fclose(input); - fclose(output); - return EXIT_FAILURE; - } - return EXIT_SUCCESS; + FILE *input = fopen(argv[1], "rb"); + if (!input) + { + std::cerr << "Can't open input file"; + return EXIT_FAILURE; + } + FILE *output = fopen(argv[2], "wb"); + if (!output) + { + fclose(input); + std::cerr << "Can't open output file"; + return EXIT_FAILURE; + } + try + { + unpack_200(input, output); + } + catch (const std::runtime_error &e) + { + std::cerr << "Bad things happened: " << e.what() << std::endl; + fclose(input); + fclose(output); + return EXIT_FAILURE; + } + return EXIT_SUCCESS; } diff --git a/libraries/pack200/src/bands.cpp b/libraries/pack200/src/bands.cpp index 1608d838..e82613b5 100644 --- a/libraries/pack200/src/bands.cpp +++ b/libraries/pack200/src/bands.cpp @@ -47,188 +47,188 @@ void band::readData(int expectedLength) { - assert(expectedLength >= 0); - assert(vs[0].cmk == cmk_ERROR); - if (expectedLength != 0) - { - assert(length == 0); - length = expectedLength; - } - if (length == 0) - { - assert((rplimit = cm.vs0.rp = u->rp) != nullptr); - return; - } - assert(length > 0); + assert(expectedLength >= 0); + assert(vs[0].cmk == cmk_ERROR); + if (expectedLength != 0) + { + assert(length == 0); + length = expectedLength; + } + if (length == 0) + { + assert((rplimit = cm.vs0.rp = u->rp) != nullptr); + return; + } + assert(length > 0); - bool is_BYTE1 = (defc->spec == BYTE1_spec); + bool is_BYTE1 = (defc->spec == BYTE1_spec); - if (is_BYTE1) - { - // No possibility of coding change. Sizing is exact. - u->ensure_input(length); - } - else - { - // Make a conservatively generous estimate of band size in bytes. - // Assume B == 5 everywhere. - // Assume awkward pop with all {U} values (2*5 per value) - int64_t generous = (int64_t)length * (B_MAX * 3 + 1) + C_SLOP; - u->ensure_input(generous); - } + if (is_BYTE1) + { + // No possibility of coding change. Sizing is exact. + u->ensure_input(length); + } + else + { + // Make a conservatively generous estimate of band size in bytes. + // Assume B == 5 everywhere. + // Assume awkward pop with all {U} values (2*5 per value) + int64_t generous = (int64_t)length * (B_MAX * 3 + 1) + C_SLOP; + u->ensure_input(generous); + } - // Read one value to see what it might be. - int XB = _meta_default; - if (!is_BYTE1) - { - // must be a variable-length coding - assert(defc->B() > 1 && defc->L() > 0); + // Read one value to see what it might be. + int XB = _meta_default; + if (!is_BYTE1) + { + // must be a variable-length coding + assert(defc->B() > 1 && defc->L() > 0); - value_stream xvs; - coding *valc = defc; - if (valc->D() != 0) - { - valc = coding::findBySpec(defc->B(), defc->H(), defc->S()); - assert(!valc->isMalloc); - } - xvs.init(u->rp, u->rplimit, valc); - int X = xvs.getInt(); - if (valc->S() != 0) - { - assert(valc->min <= -256); - XB = -1 - X; - } - else - { - int L = valc->L(); - assert(valc->max >= L + 255); - XB = X - L; - } - if (0 <= XB && XB < 256) - { - // Skip over the escape value. - u->rp = xvs.rp; - } - else - { - // No, it's still default. - XB = _meta_default; - } - } + value_stream xvs; + coding *valc = defc; + if (valc->D() != 0) + { + valc = coding::findBySpec(defc->B(), defc->H(), defc->S()); + assert(!valc->isMalloc); + } + xvs.init(u->rp, u->rplimit, valc); + int X = xvs.getInt(); + if (valc->S() != 0) + { + assert(valc->min <= -256); + XB = -1 - X; + } + else + { + int L = valc->L(); + assert(valc->max >= L + 255); + XB = X - L; + } + if (0 <= XB && XB < 256) + { + // Skip over the escape value. + u->rp = xvs.rp; + } + else + { + // No, it's still default. + XB = _meta_default; + } + } - if (XB <= _meta_canon_max) - { - byte XB_byte = (byte)XB; - byte *XB_ptr = &XB_byte; - cm.init(u->rp, u->rplimit, XB_ptr, 0, defc, length, nullptr); - } - else - { - assert(u->meta_rp != nullptr); - // Scribble the initial byte onto the band. - byte *save_meta_rp = --u->meta_rp; - byte save_meta_xb = (*save_meta_rp); - (*save_meta_rp) = (byte)XB; - cm.init(u->rp, u->rplimit, u->meta_rp, 0, defc, length, nullptr); - (*save_meta_rp) = save_meta_xb; // put it back, just to be tidy - } - rplimit = u->rp; + if (XB <= _meta_canon_max) + { + byte XB_byte = (byte)XB; + byte *XB_ptr = &XB_byte; + cm.init(u->rp, u->rplimit, XB_ptr, 0, defc, length, nullptr); + } + else + { + assert(u->meta_rp != nullptr); + // Scribble the initial byte onto the band. + byte *save_meta_rp = --u->meta_rp; + byte save_meta_xb = (*save_meta_rp); + (*save_meta_rp) = (byte)XB; + cm.init(u->rp, u->rplimit, u->meta_rp, 0, defc, length, nullptr); + (*save_meta_rp) = save_meta_xb; // put it back, just to be tidy + } + rplimit = u->rp; - rewind(); + rewind(); } void band::setIndex(cpindex *ix_) { - assert(ix_ == nullptr || ixTag == ix_->ixTag); - ix = ix_; + assert(ix_ == nullptr || ixTag == ix_->ixTag); + ix = ix_; } void band::setIndexByTag(byte tag) { - setIndex(u->cp.getIndex(tag)); + setIndex(u->cp.getIndex(tag)); } entry *band::getRefCommon(cpindex *ix_, bool nullOKwithCaller) { - assert(ix_->ixTag == ixTag || - (ixTag == CONSTANT_Literal && ix_->ixTag >= CONSTANT_Integer && - ix_->ixTag <= CONSTANT_String)); - int n = vs[0].getInt() - nullOK; - // Note: band-local nullOK means nullptr encodes as 0. - // But nullOKwithCaller means caller is willing to tolerate a nullptr. - entry *ref = ix_->get(n); - if (ref == nullptr && !(nullOKwithCaller && n == -1)) - unpack_abort(n == -1 ? "nullptr ref" : "bad ref"); - return ref; + assert(ix_->ixTag == ixTag || + (ixTag == CONSTANT_Literal && ix_->ixTag >= CONSTANT_Integer && + ix_->ixTag <= CONSTANT_String)); + int n = vs[0].getInt() - nullOK; + // Note: band-local nullOK means nullptr encodes as 0. + // But nullOKwithCaller means caller is willing to tolerate a nullptr. + entry *ref = ix_->get(n); + if (ref == nullptr && !(nullOKwithCaller && n == -1)) + unpack_abort(n == -1 ? "nullptr ref" : "bad ref"); + return ref; } int64_t band::getLong(band &lo_band, bool have_hi) { - band &hi_band = (*this); - assert(lo_band.bn == hi_band.bn + 1); - uint32_t lo = lo_band.getInt(); - if (!have_hi) - { - assert(hi_band.length == 0); - return makeLong(0, lo); - } - uint32_t hi = hi_band.getInt(); - return makeLong(hi, lo); + band &hi_band = (*this); + assert(lo_band.bn == hi_band.bn + 1); + uint32_t lo = lo_band.getInt(); + if (!have_hi) + { + assert(hi_band.length == 0); + return makeLong(0, lo); + } + uint32_t hi = hi_band.getInt(); + return makeLong(hi, lo); } int band::getIntTotal() { - if (length == 0) - return 0; - if (total_memo > 0) - return total_memo - 1; - int total = getInt(); - // overflow checks require that none of the addends are <0, - // and that the partial sums never overflow (wrap negative) - if (total < 0) - { - unpack_abort("overflow detected"); - } - for (int k = length - 1; k > 0; k--) - { - int prev_total = total; - total += vs[0].getInt(); - if (total < prev_total) - { - unpack_abort("overflow detected"); - } - } - rewind(); - total_memo = total + 1; - return total; + if (length == 0) + return 0; + if (total_memo > 0) + return total_memo - 1; + int total = getInt(); + // overflow checks require that none of the addends are <0, + // and that the partial sums never overflow (wrap negative) + if (total < 0) + { + unpack_abort("overflow detected"); + } + for (int k = length - 1; k > 0; k--) + { + int prev_total = total; + total += vs[0].getInt(); + if (total < prev_total) + { + unpack_abort("overflow detected"); + } + } + rewind(); + total_memo = total + 1; + return total; } int band::getIntCount(int tag) { - if (length == 0) - return 0; - if (tag >= HIST0_MIN && tag <= HIST0_MAX) - { - if (hist0 == nullptr) - { - // Lazily calculate an approximate histogram. - hist0 = U_NEW(int, (HIST0_MAX - HIST0_MIN) + 1); - for (int k = length; k > 0; k--) - { - int x = vs[0].getInt(); - if (x >= HIST0_MIN && x <= HIST0_MAX) - hist0[x - HIST0_MIN] += 1; - } - rewind(); - } - return hist0[tag - HIST0_MIN]; - } - int total = 0; - for (int k = length; k > 0; k--) - { - total += (vs[0].getInt() == tag) ? 1 : 0; - } - rewind(); - return total; + if (length == 0) + return 0; + if (tag >= HIST0_MIN && tag <= HIST0_MAX) + { + if (hist0 == nullptr) + { + // Lazily calculate an approximate histogram. + hist0 = U_NEW(int, (HIST0_MAX - HIST0_MIN) + 1); + for (int k = length; k > 0; k--) + { + int x = vs[0].getInt(); + if (x >= HIST0_MIN && x <= HIST0_MAX) + hist0[x - HIST0_MIN] += 1; + } + rewind(); + } + return hist0[tag - HIST0_MIN]; + } + int total = 0; + for (int k = length; k > 0; k--) + { + total += (vs[0].getInt() == tag) ? 1 : 0; + } + rewind(); + return total; } #define INDEX_INIT(tag, nullOK, subindex) ((tag) + (subindex) * SUBINDEX_BIT + (nullOK) * 256) @@ -240,184 +240,184 @@ int band::getIntCount(int tag) struct band_init { - int defc; - int index; + int defc; + int index; }; #define BAND_INIT(name, cspec, ix) \ - { \ - cspec, ix \ - } + { \ + cspec, ix \ + } const band_init all_band_inits[] = - { - // BAND_INIT(archive_magic, BYTE1_spec, 0), - // BAND_INIT(archive_header, UNSIGNED5_spec, 0), - // BAND_INIT(band_headers, BYTE1_spec, 0), - BAND_INIT(cp_Utf8_prefix, DELTA5_spec, 0), BAND_INIT(cp_Utf8_suffix, UNSIGNED5_spec, 0), - BAND_INIT(cp_Utf8_chars, CHAR3_spec, 0), BAND_INIT(cp_Utf8_big_suffix, DELTA5_spec, 0), - BAND_INIT(cp_Utf8_big_chars, DELTA5_spec, 0), BAND_INIT(cp_Int, UDELTA5_spec, 0), - BAND_INIT(cp_Float, UDELTA5_spec, 0), BAND_INIT(cp_Long_hi, UDELTA5_spec, 0), - BAND_INIT(cp_Long_lo, DELTA5_spec, 0), BAND_INIT(cp_Double_hi, UDELTA5_spec, 0), - BAND_INIT(cp_Double_lo, DELTA5_spec, 0), - BAND_INIT(cp_String, UDELTA5_spec, INDEX(CONSTANT_Utf8)), - BAND_INIT(cp_Class, UDELTA5_spec, INDEX(CONSTANT_Utf8)), - BAND_INIT(cp_Signature_form, DELTA5_spec, INDEX(CONSTANT_Utf8)), - BAND_INIT(cp_Signature_classes, UDELTA5_spec, INDEX(CONSTANT_Class)), - BAND_INIT(cp_Descr_name, DELTA5_spec, INDEX(CONSTANT_Utf8)), - BAND_INIT(cp_Descr_type, UDELTA5_spec, INDEX(CONSTANT_Signature)), - BAND_INIT(cp_Field_class, DELTA5_spec, INDEX(CONSTANT_Class)), - BAND_INIT(cp_Field_desc, UDELTA5_spec, INDEX(CONSTANT_NameandType)), - BAND_INIT(cp_Method_class, DELTA5_spec, INDEX(CONSTANT_Class)), - BAND_INIT(cp_Method_desc, UDELTA5_spec, INDEX(CONSTANT_NameandType)), - BAND_INIT(cp_Imethod_class, DELTA5_spec, INDEX(CONSTANT_Class)), - BAND_INIT(cp_Imethod_desc, UDELTA5_spec, INDEX(CONSTANT_NameandType)), - BAND_INIT(attr_definition_headers, BYTE1_spec, 0), - BAND_INIT(attr_definition_name, UNSIGNED5_spec, INDEX(CONSTANT_Utf8)), - BAND_INIT(attr_definition_layout, UNSIGNED5_spec, INDEX(CONSTANT_Utf8)), - BAND_INIT(ic_this_class, UDELTA5_spec, INDEX(CONSTANT_Class)), - BAND_INIT(ic_flags, UNSIGNED5_spec, 0), - BAND_INIT(ic_outer_class, DELTA5_spec, NULL_OR_INDEX(CONSTANT_Class)), - BAND_INIT(ic_name, DELTA5_spec, NULL_OR_INDEX(CONSTANT_Utf8)), - BAND_INIT(class_this, DELTA5_spec, INDEX(CONSTANT_Class)), - BAND_INIT(class_super, DELTA5_spec, INDEX(CONSTANT_Class)), - BAND_INIT(class_interface_count, DELTA5_spec, 0), - BAND_INIT(class_interface, DELTA5_spec, INDEX(CONSTANT_Class)), - BAND_INIT(class_field_count, DELTA5_spec, 0), - BAND_INIT(class_method_count, DELTA5_spec, 0), - BAND_INIT(field_descr, DELTA5_spec, INDEX(CONSTANT_NameandType)), - BAND_INIT(field_flags_hi, UNSIGNED5_spec, 0), - BAND_INIT(field_flags_lo, UNSIGNED5_spec, 0), - BAND_INIT(field_attr_count, UNSIGNED5_spec, 0), - BAND_INIT(field_attr_indexes, UNSIGNED5_spec, 0), - BAND_INIT(field_attr_calls, UNSIGNED5_spec, 0), - BAND_INIT(field_ConstantValue_KQ, UNSIGNED5_spec, INDEX(CONSTANT_Literal)), - BAND_INIT(field_Signature_RS, UNSIGNED5_spec, INDEX(CONSTANT_Signature)), - BAND_INIT(field_metadata_bands, -1, -1), BAND_INIT(field_attr_bands, -1, -1), - BAND_INIT(method_descr, MDELTA5_spec, INDEX(CONSTANT_NameandType)), - BAND_INIT(method_flags_hi, UNSIGNED5_spec, 0), - BAND_INIT(method_flags_lo, UNSIGNED5_spec, 0), - BAND_INIT(method_attr_count, UNSIGNED5_spec, 0), - BAND_INIT(method_attr_indexes, UNSIGNED5_spec, 0), - BAND_INIT(method_attr_calls, UNSIGNED5_spec, 0), - BAND_INIT(method_Exceptions_N, UNSIGNED5_spec, 0), - BAND_INIT(method_Exceptions_RC, UNSIGNED5_spec, INDEX(CONSTANT_Class)), - BAND_INIT(method_Signature_RS, UNSIGNED5_spec, INDEX(CONSTANT_Signature)), - BAND_INIT(method_metadata_bands, -1, -1), BAND_INIT(method_attr_bands, -1, -1), - BAND_INIT(class_flags_hi, UNSIGNED5_spec, 0), - BAND_INIT(class_flags_lo, UNSIGNED5_spec, 0), - BAND_INIT(class_attr_count, UNSIGNED5_spec, 0), - BAND_INIT(class_attr_indexes, UNSIGNED5_spec, 0), - BAND_INIT(class_attr_calls, UNSIGNED5_spec, 0), - BAND_INIT(class_SourceFile_RUN, UNSIGNED5_spec, NULL_OR_INDEX(CONSTANT_Utf8)), - BAND_INIT(class_EnclosingMethod_RC, UNSIGNED5_spec, INDEX(CONSTANT_Class)), - BAND_INIT(class_EnclosingMethod_RDN, UNSIGNED5_spec, - NULL_OR_INDEX(CONSTANT_NameandType)), - BAND_INIT(class_Signature_RS, UNSIGNED5_spec, INDEX(CONSTANT_Signature)), - BAND_INIT(class_metadata_bands, -1, -1), - BAND_INIT(class_InnerClasses_N, UNSIGNED5_spec, 0), - BAND_INIT(class_InnerClasses_RC, UNSIGNED5_spec, INDEX(CONSTANT_Class)), - BAND_INIT(class_InnerClasses_F, UNSIGNED5_spec, 0), - BAND_INIT(class_InnerClasses_outer_RCN, UNSIGNED5_spec, NULL_OR_INDEX(CONSTANT_Class)), - BAND_INIT(class_InnerClasses_name_RUN, UNSIGNED5_spec, NULL_OR_INDEX(CONSTANT_Utf8)), - BAND_INIT(class_ClassFile_version_minor_H, UNSIGNED5_spec, 0), - BAND_INIT(class_ClassFile_version_major_H, UNSIGNED5_spec, 0), - BAND_INIT(class_attr_bands, -1, -1), BAND_INIT(code_headers, BYTE1_spec, 0), - BAND_INIT(code_max_stack, UNSIGNED5_spec, 0), - BAND_INIT(code_max_na_locals, UNSIGNED5_spec, 0), - BAND_INIT(code_handler_count, UNSIGNED5_spec, 0), - BAND_INIT(code_handler_start_P, BCI5_spec, 0), - BAND_INIT(code_handler_end_PO, BRANCH5_spec, 0), - BAND_INIT(code_handler_catch_PO, BRANCH5_spec, 0), - BAND_INIT(code_handler_class_RCN, UNSIGNED5_spec, NULL_OR_INDEX(CONSTANT_Class)), - BAND_INIT(code_flags_hi, UNSIGNED5_spec, 0), - BAND_INIT(code_flags_lo, UNSIGNED5_spec, 0), - BAND_INIT(code_attr_count, UNSIGNED5_spec, 0), - BAND_INIT(code_attr_indexes, UNSIGNED5_spec, 0), - BAND_INIT(code_attr_calls, UNSIGNED5_spec, 0), - BAND_INIT(code_StackMapTable_N, UNSIGNED5_spec, 0), - BAND_INIT(code_StackMapTable_frame_T, BYTE1_spec, 0), - BAND_INIT(code_StackMapTable_local_N, UNSIGNED5_spec, 0), - BAND_INIT(code_StackMapTable_stack_N, UNSIGNED5_spec, 0), - BAND_INIT(code_StackMapTable_offset, UNSIGNED5_spec, 0), - BAND_INIT(code_StackMapTable_T, BYTE1_spec, 0), - BAND_INIT(code_StackMapTable_RC, UNSIGNED5_spec, INDEX(CONSTANT_Class)), - BAND_INIT(code_StackMapTable_P, BCI5_spec, 0), - BAND_INIT(code_LineNumberTable_N, UNSIGNED5_spec, 0), - BAND_INIT(code_LineNumberTable_bci_P, BCI5_spec, 0), - BAND_INIT(code_LineNumberTable_line, UNSIGNED5_spec, 0), - BAND_INIT(code_LocalVariableTable_N, UNSIGNED5_spec, 0), - BAND_INIT(code_LocalVariableTable_bci_P, BCI5_spec, 0), - BAND_INIT(code_LocalVariableTable_span_O, BRANCH5_spec, 0), - BAND_INIT(code_LocalVariableTable_name_RU, UNSIGNED5_spec, INDEX(CONSTANT_Utf8)), - BAND_INIT(code_LocalVariableTable_type_RS, UNSIGNED5_spec, INDEX(CONSTANT_Signature)), - BAND_INIT(code_LocalVariableTable_slot, UNSIGNED5_spec, 0), - BAND_INIT(code_LocalVariableTypeTable_N, UNSIGNED5_spec, 0), - BAND_INIT(code_LocalVariableTypeTable_bci_P, BCI5_spec, 0), - BAND_INIT(code_LocalVariableTypeTable_span_O, BRANCH5_spec, 0), - BAND_INIT(code_LocalVariableTypeTable_name_RU, UNSIGNED5_spec, INDEX(CONSTANT_Utf8)), - BAND_INIT(code_LocalVariableTypeTable_type_RS, UNSIGNED5_spec, - INDEX(CONSTANT_Signature)), - BAND_INIT(code_LocalVariableTypeTable_slot, UNSIGNED5_spec, 0), - BAND_INIT(code_attr_bands, -1, -1), BAND_INIT(bc_codes, BYTE1_spec, 0), - BAND_INIT(bc_case_count, UNSIGNED5_spec, 0), BAND_INIT(bc_case_value, DELTA5_spec, 0), - BAND_INIT(bc_byte, BYTE1_spec, 0), BAND_INIT(bc_short, DELTA5_spec, 0), - BAND_INIT(bc_local, UNSIGNED5_spec, 0), BAND_INIT(bc_label, BRANCH5_spec, 0), - BAND_INIT(bc_intref, DELTA5_spec, INDEX(CONSTANT_Integer)), - BAND_INIT(bc_floatref, DELTA5_spec, INDEX(CONSTANT_Float)), - BAND_INIT(bc_longref, DELTA5_spec, INDEX(CONSTANT_Long)), - BAND_INIT(bc_doubleref, DELTA5_spec, INDEX(CONSTANT_Double)), - BAND_INIT(bc_stringref, DELTA5_spec, INDEX(CONSTANT_String)), - BAND_INIT(bc_classref, UNSIGNED5_spec, NULL_OR_INDEX(CONSTANT_Class)), - BAND_INIT(bc_fieldref, DELTA5_spec, INDEX(CONSTANT_Fieldref)), - BAND_INIT(bc_methodref, UNSIGNED5_spec, INDEX(CONSTANT_Methodref)), - BAND_INIT(bc_imethodref, DELTA5_spec, INDEX(CONSTANT_InterfaceMethodref)), - BAND_INIT(bc_thisfield, UNSIGNED5_spec, SUB_INDEX(CONSTANT_Fieldref)), - BAND_INIT(bc_superfield, UNSIGNED5_spec, SUB_INDEX(CONSTANT_Fieldref)), - BAND_INIT(bc_thismethod, UNSIGNED5_spec, SUB_INDEX(CONSTANT_Methodref)), - BAND_INIT(bc_supermethod, UNSIGNED5_spec, SUB_INDEX(CONSTANT_Methodref)), - BAND_INIT(bc_initref, UNSIGNED5_spec, SUB_INDEX(CONSTANT_Methodref)), - BAND_INIT(bc_escref, UNSIGNED5_spec, INDEX(CONSTANT_All)), - BAND_INIT(bc_escrefsize, UNSIGNED5_spec, 0), BAND_INIT(bc_escsize, UNSIGNED5_spec, 0), - BAND_INIT(bc_escbyte, BYTE1_spec, 0), - BAND_INIT(file_name, UNSIGNED5_spec, INDEX(CONSTANT_Utf8)), - BAND_INIT(file_size_hi, UNSIGNED5_spec, 0), BAND_INIT(file_size_lo, UNSIGNED5_spec, 0), - BAND_INIT(file_modtime, DELTA5_spec, 0), BAND_INIT(file_options, UNSIGNED5_spec, 0), - // BAND_INIT(file_bits, BYTE1_spec, 0), - {0, 0}}; + { + // BAND_INIT(archive_magic, BYTE1_spec, 0), + // BAND_INIT(archive_header, UNSIGNED5_spec, 0), + // BAND_INIT(band_headers, BYTE1_spec, 0), + BAND_INIT(cp_Utf8_prefix, DELTA5_spec, 0), BAND_INIT(cp_Utf8_suffix, UNSIGNED5_spec, 0), + BAND_INIT(cp_Utf8_chars, CHAR3_spec, 0), BAND_INIT(cp_Utf8_big_suffix, DELTA5_spec, 0), + BAND_INIT(cp_Utf8_big_chars, DELTA5_spec, 0), BAND_INIT(cp_Int, UDELTA5_spec, 0), + BAND_INIT(cp_Float, UDELTA5_spec, 0), BAND_INIT(cp_Long_hi, UDELTA5_spec, 0), + BAND_INIT(cp_Long_lo, DELTA5_spec, 0), BAND_INIT(cp_Double_hi, UDELTA5_spec, 0), + BAND_INIT(cp_Double_lo, DELTA5_spec, 0), + BAND_INIT(cp_String, UDELTA5_spec, INDEX(CONSTANT_Utf8)), + BAND_INIT(cp_Class, UDELTA5_spec, INDEX(CONSTANT_Utf8)), + BAND_INIT(cp_Signature_form, DELTA5_spec, INDEX(CONSTANT_Utf8)), + BAND_INIT(cp_Signature_classes, UDELTA5_spec, INDEX(CONSTANT_Class)), + BAND_INIT(cp_Descr_name, DELTA5_spec, INDEX(CONSTANT_Utf8)), + BAND_INIT(cp_Descr_type, UDELTA5_spec, INDEX(CONSTANT_Signature)), + BAND_INIT(cp_Field_class, DELTA5_spec, INDEX(CONSTANT_Class)), + BAND_INIT(cp_Field_desc, UDELTA5_spec, INDEX(CONSTANT_NameandType)), + BAND_INIT(cp_Method_class, DELTA5_spec, INDEX(CONSTANT_Class)), + BAND_INIT(cp_Method_desc, UDELTA5_spec, INDEX(CONSTANT_NameandType)), + BAND_INIT(cp_Imethod_class, DELTA5_spec, INDEX(CONSTANT_Class)), + BAND_INIT(cp_Imethod_desc, UDELTA5_spec, INDEX(CONSTANT_NameandType)), + BAND_INIT(attr_definition_headers, BYTE1_spec, 0), + BAND_INIT(attr_definition_name, UNSIGNED5_spec, INDEX(CONSTANT_Utf8)), + BAND_INIT(attr_definition_layout, UNSIGNED5_spec, INDEX(CONSTANT_Utf8)), + BAND_INIT(ic_this_class, UDELTA5_spec, INDEX(CONSTANT_Class)), + BAND_INIT(ic_flags, UNSIGNED5_spec, 0), + BAND_INIT(ic_outer_class, DELTA5_spec, NULL_OR_INDEX(CONSTANT_Class)), + BAND_INIT(ic_name, DELTA5_spec, NULL_OR_INDEX(CONSTANT_Utf8)), + BAND_INIT(class_this, DELTA5_spec, INDEX(CONSTANT_Class)), + BAND_INIT(class_super, DELTA5_spec, INDEX(CONSTANT_Class)), + BAND_INIT(class_interface_count, DELTA5_spec, 0), + BAND_INIT(class_interface, DELTA5_spec, INDEX(CONSTANT_Class)), + BAND_INIT(class_field_count, DELTA5_spec, 0), + BAND_INIT(class_method_count, DELTA5_spec, 0), + BAND_INIT(field_descr, DELTA5_spec, INDEX(CONSTANT_NameandType)), + BAND_INIT(field_flags_hi, UNSIGNED5_spec, 0), + BAND_INIT(field_flags_lo, UNSIGNED5_spec, 0), + BAND_INIT(field_attr_count, UNSIGNED5_spec, 0), + BAND_INIT(field_attr_indexes, UNSIGNED5_spec, 0), + BAND_INIT(field_attr_calls, UNSIGNED5_spec, 0), + BAND_INIT(field_ConstantValue_KQ, UNSIGNED5_spec, INDEX(CONSTANT_Literal)), + BAND_INIT(field_Signature_RS, UNSIGNED5_spec, INDEX(CONSTANT_Signature)), + BAND_INIT(field_metadata_bands, -1, -1), BAND_INIT(field_attr_bands, -1, -1), + BAND_INIT(method_descr, MDELTA5_spec, INDEX(CONSTANT_NameandType)), + BAND_INIT(method_flags_hi, UNSIGNED5_spec, 0), + BAND_INIT(method_flags_lo, UNSIGNED5_spec, 0), + BAND_INIT(method_attr_count, UNSIGNED5_spec, 0), + BAND_INIT(method_attr_indexes, UNSIGNED5_spec, 0), + BAND_INIT(method_attr_calls, UNSIGNED5_spec, 0), + BAND_INIT(method_Exceptions_N, UNSIGNED5_spec, 0), + BAND_INIT(method_Exceptions_RC, UNSIGNED5_spec, INDEX(CONSTANT_Class)), + BAND_INIT(method_Signature_RS, UNSIGNED5_spec, INDEX(CONSTANT_Signature)), + BAND_INIT(method_metadata_bands, -1, -1), BAND_INIT(method_attr_bands, -1, -1), + BAND_INIT(class_flags_hi, UNSIGNED5_spec, 0), + BAND_INIT(class_flags_lo, UNSIGNED5_spec, 0), + BAND_INIT(class_attr_count, UNSIGNED5_spec, 0), + BAND_INIT(class_attr_indexes, UNSIGNED5_spec, 0), + BAND_INIT(class_attr_calls, UNSIGNED5_spec, 0), + BAND_INIT(class_SourceFile_RUN, UNSIGNED5_spec, NULL_OR_INDEX(CONSTANT_Utf8)), + BAND_INIT(class_EnclosingMethod_RC, UNSIGNED5_spec, INDEX(CONSTANT_Class)), + BAND_INIT(class_EnclosingMethod_RDN, UNSIGNED5_spec, + NULL_OR_INDEX(CONSTANT_NameandType)), + BAND_INIT(class_Signature_RS, UNSIGNED5_spec, INDEX(CONSTANT_Signature)), + BAND_INIT(class_metadata_bands, -1, -1), + BAND_INIT(class_InnerClasses_N, UNSIGNED5_spec, 0), + BAND_INIT(class_InnerClasses_RC, UNSIGNED5_spec, INDEX(CONSTANT_Class)), + BAND_INIT(class_InnerClasses_F, UNSIGNED5_spec, 0), + BAND_INIT(class_InnerClasses_outer_RCN, UNSIGNED5_spec, NULL_OR_INDEX(CONSTANT_Class)), + BAND_INIT(class_InnerClasses_name_RUN, UNSIGNED5_spec, NULL_OR_INDEX(CONSTANT_Utf8)), + BAND_INIT(class_ClassFile_version_minor_H, UNSIGNED5_spec, 0), + BAND_INIT(class_ClassFile_version_major_H, UNSIGNED5_spec, 0), + BAND_INIT(class_attr_bands, -1, -1), BAND_INIT(code_headers, BYTE1_spec, 0), + BAND_INIT(code_max_stack, UNSIGNED5_spec, 0), + BAND_INIT(code_max_na_locals, UNSIGNED5_spec, 0), + BAND_INIT(code_handler_count, UNSIGNED5_spec, 0), + BAND_INIT(code_handler_start_P, BCI5_spec, 0), + BAND_INIT(code_handler_end_PO, BRANCH5_spec, 0), + BAND_INIT(code_handler_catch_PO, BRANCH5_spec, 0), + BAND_INIT(code_handler_class_RCN, UNSIGNED5_spec, NULL_OR_INDEX(CONSTANT_Class)), + BAND_INIT(code_flags_hi, UNSIGNED5_spec, 0), + BAND_INIT(code_flags_lo, UNSIGNED5_spec, 0), + BAND_INIT(code_attr_count, UNSIGNED5_spec, 0), + BAND_INIT(code_attr_indexes, UNSIGNED5_spec, 0), + BAND_INIT(code_attr_calls, UNSIGNED5_spec, 0), + BAND_INIT(code_StackMapTable_N, UNSIGNED5_spec, 0), + BAND_INIT(code_StackMapTable_frame_T, BYTE1_spec, 0), + BAND_INIT(code_StackMapTable_local_N, UNSIGNED5_spec, 0), + BAND_INIT(code_StackMapTable_stack_N, UNSIGNED5_spec, 0), + BAND_INIT(code_StackMapTable_offset, UNSIGNED5_spec, 0), + BAND_INIT(code_StackMapTable_T, BYTE1_spec, 0), + BAND_INIT(code_StackMapTable_RC, UNSIGNED5_spec, INDEX(CONSTANT_Class)), + BAND_INIT(code_StackMapTable_P, BCI5_spec, 0), + BAND_INIT(code_LineNumberTable_N, UNSIGNED5_spec, 0), + BAND_INIT(code_LineNumberTable_bci_P, BCI5_spec, 0), + BAND_INIT(code_LineNumberTable_line, UNSIGNED5_spec, 0), + BAND_INIT(code_LocalVariableTable_N, UNSIGNED5_spec, 0), + BAND_INIT(code_LocalVariableTable_bci_P, BCI5_spec, 0), + BAND_INIT(code_LocalVariableTable_span_O, BRANCH5_spec, 0), + BAND_INIT(code_LocalVariableTable_name_RU, UNSIGNED5_spec, INDEX(CONSTANT_Utf8)), + BAND_INIT(code_LocalVariableTable_type_RS, UNSIGNED5_spec, INDEX(CONSTANT_Signature)), + BAND_INIT(code_LocalVariableTable_slot, UNSIGNED5_spec, 0), + BAND_INIT(code_LocalVariableTypeTable_N, UNSIGNED5_spec, 0), + BAND_INIT(code_LocalVariableTypeTable_bci_P, BCI5_spec, 0), + BAND_INIT(code_LocalVariableTypeTable_span_O, BRANCH5_spec, 0), + BAND_INIT(code_LocalVariableTypeTable_name_RU, UNSIGNED5_spec, INDEX(CONSTANT_Utf8)), + BAND_INIT(code_LocalVariableTypeTable_type_RS, UNSIGNED5_spec, + INDEX(CONSTANT_Signature)), + BAND_INIT(code_LocalVariableTypeTable_slot, UNSIGNED5_spec, 0), + BAND_INIT(code_attr_bands, -1, -1), BAND_INIT(bc_codes, BYTE1_spec, 0), + BAND_INIT(bc_case_count, UNSIGNED5_spec, 0), BAND_INIT(bc_case_value, DELTA5_spec, 0), + BAND_INIT(bc_byte, BYTE1_spec, 0), BAND_INIT(bc_short, DELTA5_spec, 0), + BAND_INIT(bc_local, UNSIGNED5_spec, 0), BAND_INIT(bc_label, BRANCH5_spec, 0), + BAND_INIT(bc_intref, DELTA5_spec, INDEX(CONSTANT_Integer)), + BAND_INIT(bc_floatref, DELTA5_spec, INDEX(CONSTANT_Float)), + BAND_INIT(bc_longref, DELTA5_spec, INDEX(CONSTANT_Long)), + BAND_INIT(bc_doubleref, DELTA5_spec, INDEX(CONSTANT_Double)), + BAND_INIT(bc_stringref, DELTA5_spec, INDEX(CONSTANT_String)), + BAND_INIT(bc_classref, UNSIGNED5_spec, NULL_OR_INDEX(CONSTANT_Class)), + BAND_INIT(bc_fieldref, DELTA5_spec, INDEX(CONSTANT_Fieldref)), + BAND_INIT(bc_methodref, UNSIGNED5_spec, INDEX(CONSTANT_Methodref)), + BAND_INIT(bc_imethodref, DELTA5_spec, INDEX(CONSTANT_InterfaceMethodref)), + BAND_INIT(bc_thisfield, UNSIGNED5_spec, SUB_INDEX(CONSTANT_Fieldref)), + BAND_INIT(bc_superfield, UNSIGNED5_spec, SUB_INDEX(CONSTANT_Fieldref)), + BAND_INIT(bc_thismethod, UNSIGNED5_spec, SUB_INDEX(CONSTANT_Methodref)), + BAND_INIT(bc_supermethod, UNSIGNED5_spec, SUB_INDEX(CONSTANT_Methodref)), + BAND_INIT(bc_initref, UNSIGNED5_spec, SUB_INDEX(CONSTANT_Methodref)), + BAND_INIT(bc_escref, UNSIGNED5_spec, INDEX(CONSTANT_All)), + BAND_INIT(bc_escrefsize, UNSIGNED5_spec, 0), BAND_INIT(bc_escsize, UNSIGNED5_spec, 0), + BAND_INIT(bc_escbyte, BYTE1_spec, 0), + BAND_INIT(file_name, UNSIGNED5_spec, INDEX(CONSTANT_Utf8)), + BAND_INIT(file_size_hi, UNSIGNED5_spec, 0), BAND_INIT(file_size_lo, UNSIGNED5_spec, 0), + BAND_INIT(file_modtime, DELTA5_spec, 0), BAND_INIT(file_options, UNSIGNED5_spec, 0), + // BAND_INIT(file_bits, BYTE1_spec, 0), + {0, 0}}; band *band::makeBands(unpacker *u) { - band *tmp_all_bands = U_NEW(band, BAND_LIMIT); - for (int i = 0; i < BAND_LIMIT; i++) - { - assert((byte *)&all_band_inits[i + 1] < - (byte *)all_band_inits + sizeof(all_band_inits)); - const band_init &bi = all_band_inits[i]; - band &b = tmp_all_bands[i]; - coding *defc = coding::findBySpec(bi.defc); - assert((defc == nullptr) == (bi.defc == -1)); // no garbage, please - assert(defc == nullptr || !defc->isMalloc); - b.init(u, i, defc); - if (bi.index > 0) - { - b.nullOK = ((bi.index >> 8) & 1); - b.ixTag = (bi.index & 0xFF); - } - } - return tmp_all_bands; + band *tmp_all_bands = U_NEW(band, BAND_LIMIT); + for (int i = 0; i < BAND_LIMIT; i++) + { + assert((byte *)&all_band_inits[i + 1] < + (byte *)all_band_inits + sizeof(all_band_inits)); + const band_init &bi = all_band_inits[i]; + band &b = tmp_all_bands[i]; + coding *defc = coding::findBySpec(bi.defc); + assert((defc == nullptr) == (bi.defc == -1)); // no garbage, please + assert(defc == nullptr || !defc->isMalloc); + b.init(u, i, defc); + if (bi.index > 0) + { + b.nullOK = ((bi.index >> 8) & 1); + b.ixTag = (bi.index & 0xFF); + } + } + return tmp_all_bands; } void band::initIndexes(unpacker *u) { - band *tmp_all_bands = u->all_bands; - for (int i = 0; i < BAND_LIMIT; i++) - { - band *scan = &tmp_all_bands[i]; - uint32_t tag = scan->ixTag; // Cf. #define INDEX(tag) above - if (tag != 0 && tag != CONSTANT_Literal && (tag & SUBINDEX_BIT) == 0) - { - scan->setIndex(u->cp.getIndex(tag)); - } - } + band *tmp_all_bands = u->all_bands; + for (int i = 0; i < BAND_LIMIT; i++) + { + band *scan = &tmp_all_bands[i]; + uint32_t tag = scan->ixTag; // Cf. #define INDEX(tag) above + if (tag != 0 && tag != CONSTANT_Literal && (tag & SUBINDEX_BIT) == 0) + { + scan->setIndex(u->cp.getIndex(tag)); + } + } } diff --git a/libraries/pack200/src/bands.h b/libraries/pack200/src/bands.h index a56cd7d5..66c5aec4 100644 --- a/libraries/pack200/src/bands.h +++ b/libraries/pack200/src/bands.h @@ -30,138 +30,138 @@ struct unpacker; struct band { - int bn; // band_number of this band - coding *defc; // default coding method - cpindex *ix; // CP entry mapping, if CPRefBand - byte ixTag; // 0 or 1; nullptr is coded as (nullOK?0:-1) - byte nullOK; // 0 or 1; nullptr is coded as (nullOK?0:-1) - int length; // expected # values - unpacker *u; // back pointer - - value_stream vs[2]; // source of values - coding_method cm; // method used for initial state of vs[0] - byte *rplimit; // end of band (encoded, transmitted) - - int total_memo; // cached value of getIntTotal, or -1 - int *hist0; // approximate. histogram - enum - { - HIST0_MIN = 0, - HIST0_MAX = 255 - }; // catches the usual cases - - // properties for attribute layout elements: - byte le_kind; // EK_XXX - byte le_bci; // 0,EK_BCI,EK_BCD,EK_BCO - byte le_back; // ==EF_BACK - byte le_len; // 0,1,2,4 (size in classfile), or call addr - band **le_body; // body of repl, union, call (nullptr-terminated) + int bn; // band_number of this band + coding *defc; // default coding method + cpindex *ix; // CP entry mapping, if CPRefBand + byte ixTag; // 0 or 1; nullptr is coded as (nullOK?0:-1) + byte nullOK; // 0 or 1; nullptr is coded as (nullOK?0:-1) + int length; // expected # values + unpacker *u; // back pointer + + value_stream vs[2]; // source of values + coding_method cm; // method used for initial state of vs[0] + byte *rplimit; // end of band (encoded, transmitted) + + int total_memo; // cached value of getIntTotal, or -1 + int *hist0; // approximate. histogram + enum + { + HIST0_MIN = 0, + HIST0_MAX = 255 + }; // catches the usual cases + + // properties for attribute layout elements: + byte le_kind; // EK_XXX + byte le_bci; // 0,EK_BCI,EK_BCD,EK_BCO + byte le_back; // ==EF_BACK + byte le_len; // 0,1,2,4 (size in classfile), or call addr + band **le_body; // body of repl, union, call (nullptr-terminated) // Note: EK_CASE elements use hist0 to record union tags. #define le_casetags hist0 - band &nextBand() - { - return this[1]; - } - band &prevBand() - { - return this[-1]; - } - - void init(unpacker *u_, int bn_, coding *defc_) - { - u = u_; - cm.u = u_; - bn = bn_; - defc = defc_; - } - void init(unpacker *u_, int bn_, int defcSpec) - { - init(u_, bn_, coding::findBySpec(defcSpec)); - } - void initRef(int ixTag_ = 0, bool nullOK_ = false) - { - ixTag = ixTag_; - nullOK = nullOK_; - setIndexByTag(ixTag); - } - - void expectMoreLength(int l) - { - assert(length >= 0); // able to accept a length - assert((int)l >= 0); // no overflow - assert(rplimit == nullptr); // readData not yet called - length += l; - assert(length >= l); // no overflow - } - - void setIndex(cpindex *ix_); - void setIndexByTag(byte tag); - - // Parse the band and its meta-coding header. - void readData(int expectedLength = 0); - - // Reset the band for another pass (Cf. Java Band.resetForSecondPass.) - void rewind() - { - cm.reset(&vs[0]); - } - - byte *&curRP() - { - return vs[0].rp; - } - byte *minRP() - { - return cm.vs0.rp; - } - byte *maxRP() - { - return rplimit; - } - size_t size() - { - return maxRP() - minRP(); - } - - int getByte() - { - assert(ix == nullptr); - return vs[0].getByte(); - } - int getInt() - { - assert(ix == nullptr); - return vs[0].getInt(); - } - entry *getRefN() - { - assert(ix != nullptr); - return getRefCommon(ix, true); - } - entry *getRef() - { - assert(ix != nullptr); - return getRefCommon(ix, false); - } - entry *getRefUsing(cpindex *ix2) - { - assert(ix == nullptr); - return getRefCommon(ix2, true); - } - entry *getRefCommon(cpindex *ix, bool nullOK); - int64_t getLong(band &lo_band, bool have_hi); - - static int64_t makeLong(uint32_t hi, uint32_t lo) - { - return ((uint64_t)hi << 32) + (((uint64_t)lo << 32) >> 32); - } - - int getIntTotal(); - int getIntCount(int tag); - - static band *makeBands(unpacker *u); - static void initIndexes(unpacker *u); + band &nextBand() + { + return this[1]; + } + band &prevBand() + { + return this[-1]; + } + + void init(unpacker *u_, int bn_, coding *defc_) + { + u = u_; + cm.u = u_; + bn = bn_; + defc = defc_; + } + void init(unpacker *u_, int bn_, int defcSpec) + { + init(u_, bn_, coding::findBySpec(defcSpec)); + } + void initRef(int ixTag_ = 0, bool nullOK_ = false) + { + ixTag = ixTag_; + nullOK = nullOK_; + setIndexByTag(ixTag); + } + + void expectMoreLength(int l) + { + assert(length >= 0); // able to accept a length + assert((int)l >= 0); // no overflow + assert(rplimit == nullptr); // readData not yet called + length += l; + assert(length >= l); // no overflow + } + + void setIndex(cpindex *ix_); + void setIndexByTag(byte tag); + + // Parse the band and its meta-coding header. + void readData(int expectedLength = 0); + + // Reset the band for another pass (Cf. Java Band.resetForSecondPass.) + void rewind() + { + cm.reset(&vs[0]); + } + + byte *&curRP() + { + return vs[0].rp; + } + byte *minRP() + { + return cm.vs0.rp; + } + byte *maxRP() + { + return rplimit; + } + size_t size() + { + return maxRP() - minRP(); + } + + int getByte() + { + assert(ix == nullptr); + return vs[0].getByte(); + } + int getInt() + { + assert(ix == nullptr); + return vs[0].getInt(); + } + entry *getRefN() + { + assert(ix != nullptr); + return getRefCommon(ix, true); + } + entry *getRef() + { + assert(ix != nullptr); + return getRefCommon(ix, false); + } + entry *getRefUsing(cpindex *ix2) + { + assert(ix == nullptr); + return getRefCommon(ix2, true); + } + entry *getRefCommon(cpindex *ix, bool nullOK); + int64_t getLong(band &lo_band, bool have_hi); + + static int64_t makeLong(uint32_t hi, uint32_t lo) + { + return ((uint64_t)hi << 32) + (((uint64_t)lo << 32) >> 32); + } + + int getIntTotal(); + int getIntCount(int tag); + + static band *makeBands(unpacker *u); + static void initIndexes(unpacker *u); }; extern band all_bands[]; @@ -173,179 +173,179 @@ extern band all_bands[]; // Band schema: enum band_number { - // e_archive_magic, - // e_archive_header, - // e_band_headers, - - // constant pool contents - e_cp_Utf8_prefix, - e_cp_Utf8_suffix, - e_cp_Utf8_chars, - e_cp_Utf8_big_suffix, - e_cp_Utf8_big_chars, - e_cp_Int, - e_cp_Float, - e_cp_Long_hi, - e_cp_Long_lo, - e_cp_Double_hi, - e_cp_Double_lo, - e_cp_String, - e_cp_Class, - e_cp_Signature_form, - e_cp_Signature_classes, - e_cp_Descr_name, - e_cp_Descr_type, - e_cp_Field_class, - e_cp_Field_desc, - e_cp_Method_class, - e_cp_Method_desc, - e_cp_Imethod_class, - e_cp_Imethod_desc, - - // bands which define transmission of attributes - e_attr_definition_headers, - e_attr_definition_name, - e_attr_definition_layout, - - // band for hardwired InnerClasses attribute (shared across the package) - e_ic_this_class, - e_ic_flags, - // These bands contain data only where flags sets ACC_IC_LONG_FORM: - e_ic_outer_class, - e_ic_name, - - // bands for carrying class schema information: - e_class_this, - e_class_super, - e_class_interface_count, - e_class_interface, - - // bands for class members - e_class_field_count, - e_class_method_count, - e_field_descr, - e_field_flags_hi, - e_field_flags_lo, - e_field_attr_count, - e_field_attr_indexes, - e_field_attr_calls, - e_field_ConstantValue_KQ, - e_field_Signature_RS, - e_field_metadata_bands, - e_field_attr_bands, - e_method_descr, - e_method_flags_hi, - e_method_flags_lo, - e_method_attr_count, - e_method_attr_indexes, - e_method_attr_calls, - e_method_Exceptions_N, - e_method_Exceptions_RC, - e_method_Signature_RS, - e_method_metadata_bands, - e_method_attr_bands, - e_class_flags_hi, - e_class_flags_lo, - e_class_attr_count, - e_class_attr_indexes, - e_class_attr_calls, - e_class_SourceFile_RUN, - e_class_EnclosingMethod_RC, - e_class_EnclosingMethod_RDN, - e_class_Signature_RS, - e_class_metadata_bands, - e_class_InnerClasses_N, - e_class_InnerClasses_RC, - e_class_InnerClasses_F, - e_class_InnerClasses_outer_RCN, - e_class_InnerClasses_name_RUN, - e_class_ClassFile_version_minor_H, - e_class_ClassFile_version_major_H, - e_class_attr_bands, - e_code_headers, - e_code_max_stack, - e_code_max_na_locals, - e_code_handler_count, - e_code_handler_start_P, - e_code_handler_end_PO, - e_code_handler_catch_PO, - e_code_handler_class_RCN, - - // code attributes - e_code_flags_hi, - e_code_flags_lo, - e_code_attr_count, - e_code_attr_indexes, - e_code_attr_calls, - e_code_StackMapTable_N, - e_code_StackMapTable_frame_T, - e_code_StackMapTable_local_N, - e_code_StackMapTable_stack_N, - e_code_StackMapTable_offset, - e_code_StackMapTable_T, - e_code_StackMapTable_RC, - e_code_StackMapTable_P, - e_code_LineNumberTable_N, - e_code_LineNumberTable_bci_P, - e_code_LineNumberTable_line, - e_code_LocalVariableTable_N, - e_code_LocalVariableTable_bci_P, - e_code_LocalVariableTable_span_O, - e_code_LocalVariableTable_name_RU, - e_code_LocalVariableTable_type_RS, - e_code_LocalVariableTable_slot, - e_code_LocalVariableTypeTable_N, - e_code_LocalVariableTypeTable_bci_P, - e_code_LocalVariableTypeTable_span_O, - e_code_LocalVariableTypeTable_name_RU, - e_code_LocalVariableTypeTable_type_RS, - e_code_LocalVariableTypeTable_slot, - e_code_attr_bands, - - // bands for bytecodes - e_bc_codes, - // remaining bands provide typed opcode fields required by the bc_codes - e_bc_case_count, - e_bc_case_value, - e_bc_byte, - e_bc_short, - e_bc_local, - e_bc_label, - - // ldc* operands: - e_bc_intref, - e_bc_floatref, - e_bc_longref, - e_bc_doubleref, - e_bc_stringref, - e_bc_classref, - e_bc_fieldref, - e_bc_methodref, - e_bc_imethodref, - - // _self_linker_op family - e_bc_thisfield, - e_bc_superfield, - e_bc_thismethod, - e_bc_supermethod, - - // bc_invokeinit family: - e_bc_initref, - - // bytecode escape sequences - e_bc_escref, - e_bc_escrefsize, - e_bc_escsize, - e_bc_escbyte, - - // file attributes and contents - e_file_name, - e_file_size_hi, - e_file_size_lo, - e_file_modtime, - e_file_options, - // e_file_bits, // handled specially as an appendix - BAND_LIMIT + // e_archive_magic, + // e_archive_header, + // e_band_headers, + + // constant pool contents + e_cp_Utf8_prefix, + e_cp_Utf8_suffix, + e_cp_Utf8_chars, + e_cp_Utf8_big_suffix, + e_cp_Utf8_big_chars, + e_cp_Int, + e_cp_Float, + e_cp_Long_hi, + e_cp_Long_lo, + e_cp_Double_hi, + e_cp_Double_lo, + e_cp_String, + e_cp_Class, + e_cp_Signature_form, + e_cp_Signature_classes, + e_cp_Descr_name, + e_cp_Descr_type, + e_cp_Field_class, + e_cp_Field_desc, + e_cp_Method_class, + e_cp_Method_desc, + e_cp_Imethod_class, + e_cp_Imethod_desc, + + // bands which define transmission of attributes + e_attr_definition_headers, + e_attr_definition_name, + e_attr_definition_layout, + + // band for hardwired InnerClasses attribute (shared across the package) + e_ic_this_class, + e_ic_flags, + // These bands contain data only where flags sets ACC_IC_LONG_FORM: + e_ic_outer_class, + e_ic_name, + + // bands for carrying class schema information: + e_class_this, + e_class_super, + e_class_interface_count, + e_class_interface, + + // bands for class members + e_class_field_count, + e_class_method_count, + e_field_descr, + e_field_flags_hi, + e_field_flags_lo, + e_field_attr_count, + e_field_attr_indexes, + e_field_attr_calls, + e_field_ConstantValue_KQ, + e_field_Signature_RS, + e_field_metadata_bands, + e_field_attr_bands, + e_method_descr, + e_method_flags_hi, + e_method_flags_lo, + e_method_attr_count, + e_method_attr_indexes, + e_method_attr_calls, + e_method_Exceptions_N, + e_method_Exceptions_RC, + e_method_Signature_RS, + e_method_metadata_bands, + e_method_attr_bands, + e_class_flags_hi, + e_class_flags_lo, + e_class_attr_count, + e_class_attr_indexes, + e_class_attr_calls, + e_class_SourceFile_RUN, + e_class_EnclosingMethod_RC, + e_class_EnclosingMethod_RDN, + e_class_Signature_RS, + e_class_metadata_bands, + e_class_InnerClasses_N, + e_class_InnerClasses_RC, + e_class_InnerClasses_F, + e_class_InnerClasses_outer_RCN, + e_class_InnerClasses_name_RUN, + e_class_ClassFile_version_minor_H, + e_class_ClassFile_version_major_H, + e_class_attr_bands, + e_code_headers, + e_code_max_stack, + e_code_max_na_locals, + e_code_handler_count, + e_code_handler_start_P, + e_code_handler_end_PO, + e_code_handler_catch_PO, + e_code_handler_class_RCN, + + // code attributes + e_code_flags_hi, + e_code_flags_lo, + e_code_attr_count, + e_code_attr_indexes, + e_code_attr_calls, + e_code_StackMapTable_N, + e_code_StackMapTable_frame_T, + e_code_StackMapTable_local_N, + e_code_StackMapTable_stack_N, + e_code_StackMapTable_offset, + e_code_StackMapTable_T, + e_code_StackMapTable_RC, + e_code_StackMapTable_P, + e_code_LineNumberTable_N, + e_code_LineNumberTable_bci_P, + e_code_LineNumberTable_line, + e_code_LocalVariableTable_N, + e_code_LocalVariableTable_bci_P, + e_code_LocalVariableTable_span_O, + e_code_LocalVariableTable_name_RU, + e_code_LocalVariableTable_type_RS, + e_code_LocalVariableTable_slot, + e_code_LocalVariableTypeTable_N, + e_code_LocalVariableTypeTable_bci_P, + e_code_LocalVariableTypeTable_span_O, + e_code_LocalVariableTypeTable_name_RU, + e_code_LocalVariableTypeTable_type_RS, + e_code_LocalVariableTypeTable_slot, + e_code_attr_bands, + + // bands for bytecodes + e_bc_codes, + // remaining bands provide typed opcode fields required by the bc_codes + e_bc_case_count, + e_bc_case_value, + e_bc_byte, + e_bc_short, + e_bc_local, + e_bc_label, + + // ldc* operands: + e_bc_intref, + e_bc_floatref, + e_bc_longref, + e_bc_doubleref, + e_bc_stringref, + e_bc_classref, + e_bc_fieldref, + e_bc_methodref, + e_bc_imethodref, + + // _self_linker_op family + e_bc_thisfield, + e_bc_superfield, + e_bc_thismethod, + e_bc_supermethod, + + // bc_invokeinit family: + e_bc_initref, + + // bytecode escape sequences + e_bc_escref, + e_bc_escrefsize, + e_bc_escsize, + e_bc_escbyte, + + // file attributes and contents + e_file_name, + e_file_size_hi, + e_file_size_lo, + e_file_modtime, + e_file_options, + // e_file_bits, // handled specially as an appendix + BAND_LIMIT }; // Symbolic names for bands, as if in a giant global struct: diff --git a/libraries/pack200/src/bytes.cpp b/libraries/pack200/src/bytes.cpp index d3808afa..767fe0a5 100644 --- a/libraries/pack200/src/bytes.cpp +++ b/libraries/pack200/src/bytes.cpp @@ -36,182 +36,182 @@ static byte dummy[1 << 10]; bool bytes::inBounds(const void *p) { - return p >= ptr && p < limit(); + return p >= ptr && p < limit(); } void bytes::malloc(size_t len_) { - len = len_; - ptr = NEW(byte, add_size(len_, 1)); // add trailing zero byte always - if (ptr == nullptr) - { - // set ptr to some victim memory, to ease escape - set(dummy, sizeof(dummy) - 1); - unpack_abort(ERROR_ENOMEM); - } + len = len_; + ptr = NEW(byte, add_size(len_, 1)); // add trailing zero byte always + if (ptr == nullptr) + { + // set ptr to some victim memory, to ease escape + set(dummy, sizeof(dummy) - 1); + unpack_abort(ERROR_ENOMEM); + } } void bytes::realloc(size_t len_) { - if (len == len_) - return; // nothing to do - if (ptr == dummy) - return; // escaping from an error - if (ptr == nullptr) - { - malloc(len_); - return; - } - byte *oldptr = ptr; - ptr = (len_ >= PSIZE_MAX) ? nullptr : (byte *)::realloc(ptr, add_size(len_, 1)); - if (ptr != nullptr) - { - if (len < len_) - memset(ptr + len, 0, len_ - len); - ptr[len_] = 0; - len = len_; - } - else - { - ptr = oldptr; // ease our escape - unpack_abort(ERROR_ENOMEM); - } + if (len == len_) + return; // nothing to do + if (ptr == dummy) + return; // escaping from an error + if (ptr == nullptr) + { + malloc(len_); + return; + } + byte *oldptr = ptr; + ptr = (len_ >= PSIZE_MAX) ? nullptr : (byte *)::realloc(ptr, add_size(len_, 1)); + if (ptr != nullptr) + { + if (len < len_) + memset(ptr + len, 0, len_ - len); + ptr[len_] = 0; + len = len_; + } + else + { + ptr = oldptr; // ease our escape + unpack_abort(ERROR_ENOMEM); + } } void bytes::free() { - if (ptr == dummy) - return; // escaping from an error - if (ptr != nullptr) - { - ::free(ptr); - } - len = 0; - ptr = 0; + if (ptr == dummy) + return; // escaping from an error + if (ptr != nullptr) + { + ::free(ptr); + } + len = 0; + ptr = 0; } int bytes::indexOf(byte c) { - byte *p = (byte *)memchr(ptr, c, len); - return (p == 0) ? -1 : (int)(p - ptr); + byte *p = (byte *)memchr(ptr, c, len); + return (p == 0) ? -1 : (int)(p - ptr); } byte *bytes::writeTo(byte *bp) { - memcpy(bp, ptr, len); - return bp + len; + memcpy(bp, ptr, len); + return bp + len; } int bytes::compareTo(bytes &other) { - size_t l1 = len; - size_t l2 = other.len; - int cmp = memcmp(ptr, other.ptr, (l1 < l2) ? l1 : l2); - if (cmp != 0) - return cmp; - return (l1 < l2) ? -1 : (l1 > l2) ? 1 : 0; + size_t l1 = len; + size_t l2 = other.len; + int cmp = memcmp(ptr, other.ptr, (l1 < l2) ? l1 : l2); + if (cmp != 0) + return cmp; + return (l1 < l2) ? -1 : (l1 > l2) ? 1 : 0; } void bytes::saveFrom(const void *ptr_, size_t len_) { - malloc(len_); - // Save as much as possible. - if (len_ > len) - { - assert(ptr == dummy); // error recovery - len_ = len; - } - copyFrom(ptr_, len_); + malloc(len_); + // Save as much as possible. + if (len_ > len) + { + assert(ptr == dummy); // error recovery + len_ = len; + } + copyFrom(ptr_, len_); } //#TODO: Need to fix for exception handling void bytes::copyFrom(const void *ptr_, size_t len_, size_t offset) { - assert(len_ == 0 || inBounds(ptr + offset)); - assert(len_ == 0 || inBounds(ptr + offset + len_ - 1)); - memcpy(ptr + offset, ptr_, len_); + assert(len_ == 0 || inBounds(ptr + offset)); + assert(len_ == 0 || inBounds(ptr + offset + len_ - 1)); + memcpy(ptr + offset, ptr_, len_); } // Make sure there are 'o' bytes beyond the fill pointer, // advance the fill pointer, and return the old fill pointer. byte *fillbytes::grow(size_t s) { - size_t nlen = add_size(b.len, s); - if (nlen <= allocated) - { - b.len = nlen; - return limit() - s; - } - size_t maxlen = nlen; - if (maxlen < 128) - maxlen = 128; - if (maxlen < allocated * 2) - maxlen = allocated * 2; - if (allocated == 0) - { - // Initial buffer was not malloced. Do not reallocate it. - bytes old = b; - b.malloc(maxlen); - if (b.len == maxlen) - old.writeTo(b.ptr); - } - else - { - b.realloc(maxlen); - } - allocated = b.len; - if (allocated != maxlen) - { - b.len = nlen - s; // back up - return dummy; // scribble during error recov. - } - // after realloc, recompute pointers - b.len = nlen; - assert(b.len <= allocated); - return limit() - s; + size_t nlen = add_size(b.len, s); + if (nlen <= allocated) + { + b.len = nlen; + return limit() - s; + } + size_t maxlen = nlen; + if (maxlen < 128) + maxlen = 128; + if (maxlen < allocated * 2) + maxlen = allocated * 2; + if (allocated == 0) + { + // Initial buffer was not malloced. Do not reallocate it. + bytes old = b; + b.malloc(maxlen); + if (b.len == maxlen) + old.writeTo(b.ptr); + } + else + { + b.realloc(maxlen); + } + allocated = b.len; + if (allocated != maxlen) + { + b.len = nlen - s; // back up + return dummy; // scribble during error recov. + } + // after realloc, recompute pointers + b.len = nlen; + assert(b.len <= allocated); + return limit() - s; } void fillbytes::ensureSize(size_t s) { - if (allocated >= s) - return; - size_t len0 = b.len; - grow(s - size()); - b.len = len0; // put it back + if (allocated >= s) + return; + size_t len0 = b.len; + grow(s - size()); + b.len = len0; // put it back } int ptrlist::indexOf(const void *x) { - int len = length(); - for (int i = 0; i < len; i++) - { - if (get(i) == x) - return i; - } - return -1; + int len = length(); + for (int i = 0; i < len; i++) + { + if (get(i) == x) + return i; + } + return -1; } void ptrlist::freeAll() { - int len = length(); - for (int i = 0; i < len; i++) - { - void *p = (void *)get(i); - if (p != nullptr) - { - ::free(p); - } - } - free(); + int len = length(); + for (int i = 0; i < len; i++) + { + void *p = (void *)get(i); + if (p != nullptr) + { + ::free(p); + } + } + free(); } int intlist::indexOf(int x) { - int len = length(); - for (int i = 0; i < len; i++) - { - if (get(i) == x) - return i; - } - return -1; + int len = length(); + for (int i = 0; i < len; i++) + { + if (get(i) == x) + return i; + } + return -1; } diff --git a/libraries/pack200/src/bytes.h b/libraries/pack200/src/bytes.h index b116efda..2ce1f7f4 100644 --- a/libraries/pack200/src/bytes.h +++ b/libraries/pack200/src/bytes.h @@ -27,225 +27,225 @@ struct bytes { - int8_t *ptr; - size_t len; - int8_t *limit() - { - return ptr + len; - } + int8_t *ptr; + size_t len; + int8_t *limit() + { + return ptr + len; + } - void set(int8_t *ptr_, size_t len_) - { - ptr = ptr_; - len = len_; - } - void set(const char *str) - { - ptr = (int8_t *)str; - len = strlen(str); - } - bool inBounds(const void *p); // p in [ptr, limit) - void malloc(size_t len_); - void realloc(size_t len_); - void free(); - void copyFrom(const void *ptr_, size_t len_, size_t offset = 0); - void saveFrom(const void *ptr_, size_t len_); - void saveFrom(const char *str) - { - saveFrom(str, strlen(str)); - } - void copyFrom(bytes &other, size_t offset = 0) - { - copyFrom(other.ptr, other.len, offset); - } - void saveFrom(bytes &other) - { - saveFrom(other.ptr, other.len); - } - void clear(int fill_byte = 0) - { - memset(ptr, fill_byte, len); - } - int8_t *writeTo(int8_t *bp); - bool equals(bytes &other) - { - return 0 == compareTo(other); - } - int compareTo(bytes &other); - bool contains(int8_t c) - { - return indexOf(c) >= 0; - } - int indexOf(int8_t c); - // substrings: - static bytes of(int8_t *ptr, size_t len) - { - bytes res; - res.set(ptr, len); - return res; - } - bytes slice(size_t beg, size_t end) - { - bytes res; - res.ptr = ptr + beg; - res.len = end - beg; - assert(res.len == 0 ||(inBounds(res.ptr) && inBounds(res.limit() - 1))); - return res; - } - // building C strings inside byte buffers: - bytes &strcat(const char *str) - { - ::strcat((char *)ptr, str); - return *this; - } - bytes &strcat(bytes &other) - { - ::strncat((char *)ptr, (char *)other.ptr, other.len); - return *this; - } - char *strval() - { - assert(strlen((char *)ptr) == len); - return (char *)ptr; - } + void set(int8_t *ptr_, size_t len_) + { + ptr = ptr_; + len = len_; + } + void set(const char *str) + { + ptr = (int8_t *)str; + len = strlen(str); + } + bool inBounds(const void *p); // p in [ptr, limit) + void malloc(size_t len_); + void realloc(size_t len_); + void free(); + void copyFrom(const void *ptr_, size_t len_, size_t offset = 0); + void saveFrom(const void *ptr_, size_t len_); + void saveFrom(const char *str) + { + saveFrom(str, strlen(str)); + } + void copyFrom(bytes &other, size_t offset = 0) + { + copyFrom(other.ptr, other.len, offset); + } + void saveFrom(bytes &other) + { + saveFrom(other.ptr, other.len); + } + void clear(int fill_byte = 0) + { + memset(ptr, fill_byte, len); + } + int8_t *writeTo(int8_t *bp); + bool equals(bytes &other) + { + return 0 == compareTo(other); + } + int compareTo(bytes &other); + bool contains(int8_t c) + { + return indexOf(c) >= 0; + } + int indexOf(int8_t c); + // substrings: + static bytes of(int8_t *ptr, size_t len) + { + bytes res; + res.set(ptr, len); + return res; + } + bytes slice(size_t beg, size_t end) + { + bytes res; + res.ptr = ptr + beg; + res.len = end - beg; + assert(res.len == 0 ||(inBounds(res.ptr) && inBounds(res.limit() - 1))); + return res; + } + // building C strings inside byte buffers: + bytes &strcat(const char *str) + { + ::strcat((char *)ptr, str); + return *this; + } + bytes &strcat(bytes &other) + { + ::strncat((char *)ptr, (char *)other.ptr, other.len); + return *this; + } + char *strval() + { + assert(strlen((char *)ptr) == len); + return (char *)ptr; + } }; #define BYTES_OF(var) (bytes::of((int8_t *)&(var), sizeof(var))) struct fillbytes { - bytes b; - size_t allocated; + bytes b; + size_t allocated; - int8_t *base() - { - return b.ptr; - } - size_t size() - { - return b.len; - } - int8_t *limit() - { - return b.limit(); - } // logical limit - void setLimit(int8_t *lp) - { - assert(isAllocated(lp)); - b.len = lp - b.ptr; - } - int8_t *end() - { - return b.ptr + allocated; - } // physical limit - int8_t *loc(size_t o) - { - assert(o < b.len); - return b.ptr + o; - } - void init() - { - allocated = 0; - b.set(nullptr, 0); - } - void init(size_t s) - { - init(); - ensureSize(s); - } - void free() - { - if (allocated != 0) - b.free(); - allocated = 0; - } - void empty() - { - b.len = 0; - } - int8_t *grow(size_t s); // grow so that limit() += s - int getByte(uint32_t i) - { - return *loc(i) & 0xFF; - } - void addByte(int8_t x) - { - *grow(1) = x; - } - void ensureSize(size_t s); // make sure allocated >= s - void trimToSize() - { - if (allocated > size()) - b.realloc(allocated = size()); - } - bool canAppend(size_t s) - { - return allocated > b.len + s; - } - bool isAllocated(int8_t *p) - { - return p >= base() && p <= end(); - } // asserts - void set(bytes &src) - { - set(src.ptr, src.len); - } + int8_t *base() + { + return b.ptr; + } + size_t size() + { + return b.len; + } + int8_t *limit() + { + return b.limit(); + } // logical limit + void setLimit(int8_t *lp) + { + assert(isAllocated(lp)); + b.len = lp - b.ptr; + } + int8_t *end() + { + return b.ptr + allocated; + } // physical limit + int8_t *loc(size_t o) + { + assert(o < b.len); + return b.ptr + o; + } + void init() + { + allocated = 0; + b.set(nullptr, 0); + } + void init(size_t s) + { + init(); + ensureSize(s); + } + void free() + { + if (allocated != 0) + b.free(); + allocated = 0; + } + void empty() + { + b.len = 0; + } + int8_t *grow(size_t s); // grow so that limit() += s + int getByte(uint32_t i) + { + return *loc(i) & 0xFF; + } + void addByte(int8_t x) + { + *grow(1) = x; + } + void ensureSize(size_t s); // make sure allocated >= s + void trimToSize() + { + if (allocated > size()) + b.realloc(allocated = size()); + } + bool canAppend(size_t s) + { + return allocated > b.len + s; + } + bool isAllocated(int8_t *p) + { + return p >= base() && p <= end(); + } // asserts + void set(bytes &src) + { + set(src.ptr, src.len); + } - void set(int8_t *ptr, size_t len) - { - b.set(ptr, len); - allocated = 0; // mark as not reallocatable - } + void set(int8_t *ptr, size_t len) + { + b.set(ptr, len); + allocated = 0; // mark as not reallocatable + } - // block operations on resizing byte buffer: - fillbytes &append(const void *ptr_, size_t len_) - { - memcpy(grow(len_), ptr_, len_); - return (*this); - } - fillbytes &append(bytes &other) - { - return append(other.ptr, other.len); - } - fillbytes &append(const char *str) - { - return append(str, strlen(str)); - } + // block operations on resizing byte buffer: + fillbytes &append(const void *ptr_, size_t len_) + { + memcpy(grow(len_), ptr_, len_); + return (*this); + } + fillbytes &append(bytes &other) + { + return append(other.ptr, other.len); + } + fillbytes &append(const char *str) + { + return append(str, strlen(str)); + } }; struct ptrlist : fillbytes { - typedef const void *cvptr; - int length() - { - return (int)(size() / sizeof(cvptr)); - } - cvptr *base() - { - return (cvptr *)fillbytes::base(); - } - cvptr &get(int i) - { - return *(cvptr *)loc(i * sizeof(cvptr)); - } - cvptr *limit() - { - return (cvptr *)fillbytes::limit(); - } - void add(cvptr x) - { - *(cvptr *)grow(sizeof(x)) = x; - } - void popTo(int l) - { - assert(l <= length()); - b.len = l * sizeof(cvptr); - } - int indexOf(cvptr x); - bool contains(cvptr x) - { - return indexOf(x) >= 0; - } - void freeAll(); // frees every ptr on the list, plus the list itself + typedef const void *cvptr; + int length() + { + return (int)(size() / sizeof(cvptr)); + } + cvptr *base() + { + return (cvptr *)fillbytes::base(); + } + cvptr &get(int i) + { + return *(cvptr *)loc(i * sizeof(cvptr)); + } + cvptr *limit() + { + return (cvptr *)fillbytes::limit(); + } + void add(cvptr x) + { + *(cvptr *)grow(sizeof(x)) = x; + } + void popTo(int l) + { + assert(l <= length()); + b.len = l * sizeof(cvptr); + } + int indexOf(cvptr x); + bool contains(cvptr x) + { + return indexOf(x) >= 0; + } + void freeAll(); // frees every ptr on the list, plus the list itself }; // Use a macro rather than mess with subtle mismatches // between member and non-member function pointers. @@ -253,34 +253,34 @@ struct ptrlist : fillbytes struct intlist : fillbytes { - int length() - { - return (int)(size() / sizeof(int)); - } - int *base() - { - return (int *)fillbytes::base(); - } - int &get(int i) - { - return *(int *)loc(i * sizeof(int)); - } - int *limit() - { - return (int *)fillbytes::limit(); - } - void add(int x) - { - *(int *)grow(sizeof(x)) = x; - } - void popTo(int l) - { - assert(l <= length()); - b.len = l * sizeof(int); - } - int indexOf(int x); - bool contains(int x) - { - return indexOf(x) >= 0; - } + int length() + { + return (int)(size() / sizeof(int)); + } + int *base() + { + return (int *)fillbytes::base(); + } + int &get(int i) + { + return *(int *)loc(i * sizeof(int)); + } + int *limit() + { + return (int *)fillbytes::limit(); + } + void add(int x) + { + *(int *)grow(sizeof(x)) = x; + } + void popTo(int l) + { + assert(l <= length()); + b.len = l * sizeof(int); + } + int indexOf(int x); + bool contains(int x) + { + return indexOf(x) >= 0; + } }; diff --git a/libraries/pack200/src/coding.cpp b/libraries/pack200/src/coding.cpp index 6bd17a3c..8e872013 100644 --- a/libraries/pack200/src/coding.cpp +++ b/libraries/pack200/src/coding.cpp @@ -48,12 +48,12 @@ extern coding basic_codings[]; #pragma GCC diagnostic ignored "-Wunused-variable" #define CODING_PRIVATE(spec) \ - int spec_ = spec; \ - int B = CODING_B(spec_); \ - int H = CODING_H(spec_); \ - int L = 256 - H; \ - int S = CODING_S(spec_); \ - int D = CODING_D(spec_) + int spec_ = spec; \ + int B = CODING_B(spec_); \ + int H = CODING_H(spec_); \ + int L = 256 - H; \ + int S = CODING_S(spec_); \ + int D = CODING_D(spec_) #define IS_NEG_CODE(S, codeVal) ((((int)(codeVal) + 1) & ((1 << S) - 1)) == 0) @@ -61,568 +61,568 @@ extern coding basic_codings[]; static int decode_sign(int S, uint32_t ux) { // == Coding.decodeSign32 - assert(S > 0); - uint32_t sigbits = (ux >> S); - if (IS_NEG_CODE(S, ux)) - return (int)(~sigbits); - else - return (int)(ux - sigbits); - // Note that (int)(ux-sigbits) can be negative, if ux is large enough. + assert(S > 0); + uint32_t sigbits = (ux >> S); + if (IS_NEG_CODE(S, ux)) + return (int)(~sigbits); + else + return (int)(ux - sigbits); + // Note that (int)(ux-sigbits) can be negative, if ux is large enough. } coding *coding::init() { - if (umax > 0) - return this; // already done - assert(spec != 0); // sanity - - // fill in derived fields - CODING_PRIVATE(spec); - - // Return nullptr if 'arb(BHSD)' parameter constraints are not met: - if (B < 1 || B > B_MAX) - return nullptr; - if (H < 1 || H > 256) - return nullptr; - if (S < 0 || S > 2) - return nullptr; - if (D < 0 || D > 1) - return nullptr; - if (B == 1 && H != 256) - return nullptr; // 1-byte coding must be fixed-size - if (B >= 5 && H == 256) - return nullptr; // no 5-byte fixed-size coding - - // first compute the range of the coding, in 64 bits - int64_t range = 0; - { - int64_t H_i = 1; - for (int i = 0; i < B; i++) - { - range += H_i; - H_i *= H; - } - range *= L; - range += H_i; - } - assert(range > 0); // no useless codings, please - - int this_umax; - - // now, compute min and max - if (range >= ((int64_t)1 << 32)) - { - this_umax = INT_MAX_VALUE; - this->umin = INT_MIN_VALUE; - this->max = INT_MAX_VALUE; - this->min = INT_MIN_VALUE; - } - else - { - this_umax = (range > INT_MAX_VALUE) ? INT_MAX_VALUE : (int)range - 1; - this->max = this_umax; - this->min = this->umin = 0; - if (S != 0 && range != 0) - { - int64_t maxPosCode = range - 1; - int64_t maxNegCode = range - 1; - while (IS_NEG_CODE(S, maxPosCode)) - --maxPosCode; - while (!IS_NEG_CODE(S, maxNegCode)) - --maxNegCode; - int maxPos = decode_sign(S, (uint32_t)maxPosCode); - if (maxPos < 0) - this->max = INT_MAX_VALUE; // 32-bit wraparound - else - this->max = maxPos; - if (maxNegCode < 0) - this->min = 0; // No negative codings at all. - else - this->min = decode_sign(S, (uint32_t)maxNegCode); - } - } - - assert(!(isFullRange | isSigned | isSubrange)); // init - if (min < 0) - this->isSigned = true; - if (max < INT_MAX_VALUE && range <= INT_MAX_VALUE) - this->isSubrange = true; - if (max == INT_MAX_VALUE && min == INT_MIN_VALUE) - this->isFullRange = true; - - // do this last, to reduce MT exposure (should have a membar too) - this->umax = this_umax; - - return this; + if (umax > 0) + return this; // already done + assert(spec != 0); // sanity + + // fill in derived fields + CODING_PRIVATE(spec); + + // Return nullptr if 'arb(BHSD)' parameter constraints are not met: + if (B < 1 || B > B_MAX) + return nullptr; + if (H < 1 || H > 256) + return nullptr; + if (S < 0 || S > 2) + return nullptr; + if (D < 0 || D > 1) + return nullptr; + if (B == 1 && H != 256) + return nullptr; // 1-byte coding must be fixed-size + if (B >= 5 && H == 256) + return nullptr; // no 5-byte fixed-size coding + + // first compute the range of the coding, in 64 bits + int64_t range = 0; + { + int64_t H_i = 1; + for (int i = 0; i < B; i++) + { + range += H_i; + H_i *= H; + } + range *= L; + range += H_i; + } + assert(range > 0); // no useless codings, please + + int this_umax; + + // now, compute min and max + if (range >= ((int64_t)1 << 32)) + { + this_umax = INT_MAX_VALUE; + this->umin = INT_MIN_VALUE; + this->max = INT_MAX_VALUE; + this->min = INT_MIN_VALUE; + } + else + { + this_umax = (range > INT_MAX_VALUE) ? INT_MAX_VALUE : (int)range - 1; + this->max = this_umax; + this->min = this->umin = 0; + if (S != 0 && range != 0) + { + int64_t maxPosCode = range - 1; + int64_t maxNegCode = range - 1; + while (IS_NEG_CODE(S, maxPosCode)) + --maxPosCode; + while (!IS_NEG_CODE(S, maxNegCode)) + --maxNegCode; + int maxPos = decode_sign(S, (uint32_t)maxPosCode); + if (maxPos < 0) + this->max = INT_MAX_VALUE; // 32-bit wraparound + else + this->max = maxPos; + if (maxNegCode < 0) + this->min = 0; // No negative codings at all. + else + this->min = decode_sign(S, (uint32_t)maxNegCode); + } + } + + assert(!(isFullRange | isSigned | isSubrange)); // init + if (min < 0) + this->isSigned = true; + if (max < INT_MAX_VALUE && range <= INT_MAX_VALUE) + this->isSubrange = true; + if (max == INT_MAX_VALUE && min == INT_MIN_VALUE) + this->isFullRange = true; + + // do this last, to reduce MT exposure (should have a membar too) + this->umax = this_umax; + + return this; } coding *coding::findBySpec(int spec) { - for (coding *scan = &basic_codings[0];; scan++) - { - if (scan->spec == spec) - return scan->init(); - if (scan->spec == 0) - break; - } - coding *ptr = NEW(coding, 1); - if (!ptr) - return nullptr; - coding *c = ptr->initFrom(spec); - if (c == nullptr) - { - ::free(ptr); - } - else - // else caller should free it... - c->isMalloc = true; - return c; + for (coding *scan = &basic_codings[0];; scan++) + { + if (scan->spec == spec) + return scan->init(); + if (scan->spec == 0) + break; + } + coding *ptr = NEW(coding, 1); + if (!ptr) + return nullptr; + coding *c = ptr->initFrom(spec); + if (c == nullptr) + { + ::free(ptr); + } + else + // else caller should free it... + c->isMalloc = true; + return c; } coding *coding::findBySpec(int B, int H, int S, int D) { - if (B < 1 || B > B_MAX) - return nullptr; - if (H < 1 || H > 256) - return nullptr; - if (S < 0 || S > 2) - return nullptr; - if (D < 0 || D > 1) - return nullptr; - return findBySpec(CODING_SPEC(B, H, S, D)); + if (B < 1 || B > B_MAX) + return nullptr; + if (H < 1 || H > 256) + return nullptr; + if (S < 0 || S > 2) + return nullptr; + if (D < 0 || D > 1) + return nullptr; + return findBySpec(CODING_SPEC(B, H, S, D)); } void coding::free() { - if (isMalloc) - { - ::free(this); - } + if (isMalloc) + { + ::free(this); + } } void coding_method::reset(value_stream *state) { - assert(state->rp == state->rplimit); // not in mid-stream, please - // assert(this == vs0.cm); - state[0] = vs0; - if (uValues != nullptr) - { - uValues->reset(state->helper()); - } + assert(state->rp == state->rplimit); // not in mid-stream, please + // assert(this == vs0.cm); + state[0] = vs0; + if (uValues != nullptr) + { + uValues->reset(state->helper()); + } } uint32_t coding::parse(byte *&rp, int B, int H) { - int L = 256 - H; - byte *ptr = rp; - // hand peel the i==0 part of the loop: - uint32_t b_i = *ptr++ & 0xFF; - if (B == 1 || b_i < (uint32_t)L) - { - rp = ptr; - return b_i; - } - uint32_t sum = b_i; - uint32_t H_i = H; - assert(B <= B_MAX); - for (int i = 2; i <= B_MAX; i++) - { // easy for compilers to unroll if desired - b_i = *ptr++ & 0xFF; - sum += b_i * H_i; - if (i == B || b_i < (uint32_t)L) - { - rp = ptr; - return sum; - } - H_i *= H; - } - assert(false); - return 0; + int L = 256 - H; + byte *ptr = rp; + // hand peel the i==0 part of the loop: + uint32_t b_i = *ptr++ & 0xFF; + if (B == 1 || b_i < (uint32_t)L) + { + rp = ptr; + return b_i; + } + uint32_t sum = b_i; + uint32_t H_i = H; + assert(B <= B_MAX); + for (int i = 2; i <= B_MAX; i++) + { // easy for compilers to unroll if desired + b_i = *ptr++ & 0xFF; + sum += b_i * H_i; + if (i == B || b_i < (uint32_t)L) + { + rp = ptr; + return sum; + } + H_i *= H; + } + assert(false); + return 0; } uint32_t coding::parse_lgH(byte *&rp, int B, int H, int lgH) { - assert(H == (1 << lgH)); - int L = 256 - (1 << lgH); - byte *ptr = rp; - // hand peel the i==0 part of the loop: - uint32_t b_i = *ptr++ & 0xFF; - if (B == 1 || b_i < (uint32_t)L) - { - rp = ptr; - return b_i; - } - uint32_t sum = b_i; - uint32_t lg_H_i = lgH; - assert(B <= B_MAX); - for (int i = 2; i <= B_MAX; i++) - { // easy for compilers to unroll if desired - b_i = *ptr++ & 0xFF; - sum += b_i << lg_H_i; - if (i == B || b_i < (uint32_t)L) - { - rp = ptr; - return sum; - } - lg_H_i += lgH; - } - assert(false); - return 0; + assert(H == (1 << lgH)); + int L = 256 - (1 << lgH); + byte *ptr = rp; + // hand peel the i==0 part of the loop: + uint32_t b_i = *ptr++ & 0xFF; + if (B == 1 || b_i < (uint32_t)L) + { + rp = ptr; + return b_i; + } + uint32_t sum = b_i; + uint32_t lg_H_i = lgH; + assert(B <= B_MAX); + for (int i = 2; i <= B_MAX; i++) + { // easy for compilers to unroll if desired + b_i = *ptr++ & 0xFF; + sum += b_i << lg_H_i; + if (i == B || b_i < (uint32_t)L) + { + rp = ptr; + return sum; + } + lg_H_i += lgH; + } + assert(false); + return 0; } static const char ERB[] = "EOF reading band"; void coding::parseMultiple(byte *&rp, int N, byte *limit, int B, int H) { - if (N < 0) - { - unpack_abort("bad value count"); - return; - } - byte *ptr = rp; - if (B == 1 || H == 256) - { - size_t len = (size_t)N * B; - if (len / B != (size_t)N || ptr + len > limit) - { - unpack_abort(ERB); - return; - } - rp = ptr + len; - return; - } - // Note: We assume rp has enough zero-padding. - int L = 256 - H; - int n = B; - while (N > 0) - { - ptr += 1; - if (--n == 0) - { - // end of encoding at B bytes, regardless of byte value - } - else - { - int b = (ptr[-1] & 0xFF); - if (b >= L) - { - // keep going, unless we find a byte < L - continue; - } - } - // found the last byte - N -= 1; - n = B; // reset length counter - // do an error check here - if (ptr > limit) - { - unpack_abort(ERB); - return; - } - } - rp = ptr; - return; + if (N < 0) + { + unpack_abort("bad value count"); + return; + } + byte *ptr = rp; + if (B == 1 || H == 256) + { + size_t len = (size_t)N * B; + if (len / B != (size_t)N || ptr + len > limit) + { + unpack_abort(ERB); + return; + } + rp = ptr + len; + return; + } + // Note: We assume rp has enough zero-padding. + int L = 256 - H; + int n = B; + while (N > 0) + { + ptr += 1; + if (--n == 0) + { + // end of encoding at B bytes, regardless of byte value + } + else + { + int b = (ptr[-1] & 0xFF); + if (b >= L) + { + // keep going, unless we find a byte < L + continue; + } + } + // found the last byte + N -= 1; + n = B; // reset length counter + // do an error check here + if (ptr > limit) + { + unpack_abort(ERB); + return; + } + } + rp = ptr; + return; } bool value_stream::hasHelper() { - // If my coding method is a pop-style method, - // then I need a second value stream to transmit - // unfavored values. - // This can be determined by examining fValues. - return cm->fValues != nullptr; + // If my coding method is a pop-style method, + // then I need a second value stream to transmit + // unfavored values. + // This can be determined by examining fValues. + return cm->fValues != nullptr; } void value_stream::init(byte *rp_, byte *rplimit_, coding *defc) { - rp = rp_; - rplimit = rplimit_; - sum = 0; - cm = nullptr; // no need in the simple case - setCoding(defc); + rp = rp_; + rplimit = rplimit_; + sum = 0; + cm = nullptr; // no need in the simple case + setCoding(defc); } void value_stream::setCoding(coding *defc) { - if (defc == nullptr) - { - unpack_abort("bad coding"); - defc = coding::findByIndex(_meta_canon_min); // random pick for recovery - } - - c = (*defc); - - // choose cmk - cmk = cmk_ERROR; - switch (c.spec) - { - case BYTE1_spec: - cmk = cmk_BYTE1; - break; - case CHAR3_spec: - cmk = cmk_CHAR3; - break; - case UNSIGNED5_spec: - cmk = cmk_UNSIGNED5; - break; - case DELTA5_spec: - cmk = cmk_DELTA5; - break; - case BCI5_spec: - cmk = cmk_BCI5; - break; - case BRANCH5_spec: - cmk = cmk_BRANCH5; - break; - default: - if (c.D() == 0) - { - switch (c.S()) - { - case 0: - cmk = cmk_BHS0; - break; - case 1: - cmk = cmk_BHS1; - break; - default: - cmk = cmk_BHS; - break; - } - } - else - { - if (c.S() == 1) - { - if (c.isFullRange) - cmk = cmk_BHS1D1full; - if (c.isSubrange) - cmk = cmk_BHS1D1sub; - } - if (cmk == cmk_ERROR) - cmk = cmk_BHSD1; - } - } + if (defc == nullptr) + { + unpack_abort("bad coding"); + defc = coding::findByIndex(_meta_canon_min); // random pick for recovery + } + + c = (*defc); + + // choose cmk + cmk = cmk_ERROR; + switch (c.spec) + { + case BYTE1_spec: + cmk = cmk_BYTE1; + break; + case CHAR3_spec: + cmk = cmk_CHAR3; + break; + case UNSIGNED5_spec: + cmk = cmk_UNSIGNED5; + break; + case DELTA5_spec: + cmk = cmk_DELTA5; + break; + case BCI5_spec: + cmk = cmk_BCI5; + break; + case BRANCH5_spec: + cmk = cmk_BRANCH5; + break; + default: + if (c.D() == 0) + { + switch (c.S()) + { + case 0: + cmk = cmk_BHS0; + break; + case 1: + cmk = cmk_BHS1; + break; + default: + cmk = cmk_BHS; + break; + } + } + else + { + if (c.S() == 1) + { + if (c.isFullRange) + cmk = cmk_BHS1D1full; + if (c.isSubrange) + cmk = cmk_BHS1D1sub; + } + if (cmk == cmk_ERROR) + cmk = cmk_BHSD1; + } + } } static int getPopValue(value_stream *self, uint32_t uval) { - if (uval > 0) - { - // note that the initial parse performed a range check - assert(uval <= (uint32_t)self->cm->fVlength); - return self->cm->fValues[uval - 1]; - } - else - { - // take an unfavored value - return self->helper()->getInt(); - } + if (uval > 0) + { + // note that the initial parse performed a range check + assert(uval <= (uint32_t)self->cm->fVlength); + return self->cm->fValues[uval - 1]; + } + else + { + // take an unfavored value + return self->helper()->getInt(); + } } int coding::sumInUnsignedRange(int x, int y) { - assert(isSubrange); - int range = (int)(umax + 1); - assert(range > 0); - x += y; - if (x != (int)((int64_t)(x - y) + (int64_t)y)) - { - // 32-bit overflow interferes with range reduction. - // Back off from the overflow by adding a multiple of range: - if (x < 0) - { - x -= range; - assert(x >= 0); - } - else - { - x += range; - assert(x < 0); - } - } - if (x < 0) - { - x += range; - if (x >= 0) - return x; - } - else if (x >= range) - { - x -= range; - if (x < range) - return x; - } - else - { - // in range - return x; - } - // do it the hard way - x %= range; - if (x < 0) - x += range; - return x; + assert(isSubrange); + int range = (int)(umax + 1); + assert(range > 0); + x += y; + if (x != (int)((int64_t)(x - y) + (int64_t)y)) + { + // 32-bit overflow interferes with range reduction. + // Back off from the overflow by adding a multiple of range: + if (x < 0) + { + x -= range; + assert(x >= 0); + } + else + { + x += range; + assert(x < 0); + } + } + if (x < 0) + { + x += range; + if (x >= 0) + return x; + } + else if (x >= range) + { + x -= range; + if (x < range) + return x; + } + else + { + // in range + return x; + } + // do it the hard way + x %= range; + if (x < 0) + x += range; + return x; } static int getDeltaValue(value_stream *self, uint32_t uval, bool isSubrange) { - assert((uint32_t)(self->c.isSubrange) == (uint32_t)isSubrange); - assert(self->c.isSubrange | self->c.isFullRange); - if (isSubrange) - return self->sum = self->c.sumInUnsignedRange(self->sum, (int)uval); - else - return self->sum += (int)uval; + assert((uint32_t)(self->c.isSubrange) == (uint32_t)isSubrange); + assert(self->c.isSubrange | self->c.isFullRange); + if (isSubrange) + return self->sum = self->c.sumInUnsignedRange(self->sum, (int)uval); + else + return self->sum += (int)uval; } bool value_stream::hasValue() { - if (rp < rplimit) - return true; - if (cm == nullptr) - return false; - if (cm->next == nullptr) - return false; - cm->next->reset(this); - return hasValue(); + if (rp < rplimit) + return true; + if (cm == nullptr) + return false; + if (cm->next == nullptr) + return false; + cm->next->reset(this); + return hasValue(); } int value_stream::getInt() { - if (rp >= rplimit) - { - // Advance to next coding segment. - if (rp > rplimit || cm == nullptr || cm->next == nullptr) - { - // Must perform this check and throw an exception on bad input. - unpack_abort(ERB); - return 0; - } - cm->next->reset(this); - return getInt(); - } - - CODING_PRIVATE(c.spec); - uint32_t uval; - enum - { - B5 = 5, - B3 = 3, - H128 = 128, - H64 = 64, - H4 = 4 - }; - switch (cmk) - { - case cmk_BHS: - assert(D == 0); - uval = coding::parse(rp, B, H); - if (S == 0) - return (int)uval; - return decode_sign(S, uval); - - case cmk_BHS0: - assert(S == 0 && D == 0); - uval = coding::parse(rp, B, H); - return (int)uval; - - case cmk_BHS1: - assert(S == 1 && D == 0); - uval = coding::parse(rp, B, H); - return DECODE_SIGN_S1(uval); - - case cmk_BYTE1: - assert(c.spec == BYTE1_spec); - assert(B == 1 && H == 256 && S == 0 && D == 0); - return *rp++ & 0xFF; - - case cmk_CHAR3: - assert(c.spec == CHAR3_spec); - assert(B == B3 && H == H128 && S == 0 && D == 0); - return coding::parse_lgH(rp, B3, H128, 7); - - case cmk_UNSIGNED5: - assert(c.spec == UNSIGNED5_spec); - assert(B == B5 && H == H64 && S == 0 && D == 0); - return coding::parse_lgH(rp, B5, H64, 6); - - case cmk_BHSD1: - assert(D == 1); - uval = coding::parse(rp, B, H); - if (S != 0) - uval = (uint32_t)decode_sign(S, uval); - return getDeltaValue(this, uval, (bool)c.isSubrange); - - case cmk_BHS1D1full: - assert(S == 1 && D == 1 && c.isFullRange); - uval = coding::parse(rp, B, H); - uval = (uint32_t)DECODE_SIGN_S1(uval); - return getDeltaValue(this, uval, false); - - case cmk_BHS1D1sub: - assert(S == 1 && D == 1 && c.isSubrange); - uval = coding::parse(rp, B, H); - uval = (uint32_t)DECODE_SIGN_S1(uval); - return getDeltaValue(this, uval, true); - - case cmk_DELTA5: - assert(c.spec == DELTA5_spec); - assert(B == B5 && H == H64 && S == 1 && D == 1 && c.isFullRange); - uval = coding::parse_lgH(rp, B5, H64, 6); - sum += DECODE_SIGN_S1(uval); - return sum; - - case cmk_BCI5: - assert(c.spec == BCI5_spec); - assert(B == B5 && H == H4 && S == 0 && D == 0); - return coding::parse_lgH(rp, B5, H4, 2); - - case cmk_BRANCH5: - assert(c.spec == BRANCH5_spec); - assert(B == B5 && H == H4 && S == 2 && D == 0); - uval = coding::parse_lgH(rp, B5, H4, 2); - return decode_sign(S, uval); - - case cmk_pop: - uval = coding::parse(rp, B, H); - if (S != 0) - { - uval = (uint32_t)decode_sign(S, uval); - } - if (D != 0) - { - assert(c.isSubrange | c.isFullRange); - if (c.isSubrange) - sum = c.sumInUnsignedRange(sum, (int)uval); - else - sum += (int)uval; - uval = (uint32_t)sum; - } - return getPopValue(this, uval); - - case cmk_pop_BHS0: - assert(S == 0 && D == 0); - uval = coding::parse(rp, B, H); - return getPopValue(this, uval); - - case cmk_pop_BYTE1: - assert(c.spec == BYTE1_spec); - assert(B == 1 && H == 256 && S == 0 && D == 0); - return getPopValue(this, *rp++ & 0xFF); - - default: - break; - } - assert(false); - return 0; + if (rp >= rplimit) + { + // Advance to next coding segment. + if (rp > rplimit || cm == nullptr || cm->next == nullptr) + { + // Must perform this check and throw an exception on bad input. + unpack_abort(ERB); + return 0; + } + cm->next->reset(this); + return getInt(); + } + + CODING_PRIVATE(c.spec); + uint32_t uval; + enum + { + B5 = 5, + B3 = 3, + H128 = 128, + H64 = 64, + H4 = 4 + }; + switch (cmk) + { + case cmk_BHS: + assert(D == 0); + uval = coding::parse(rp, B, H); + if (S == 0) + return (int)uval; + return decode_sign(S, uval); + + case cmk_BHS0: + assert(S == 0 && D == 0); + uval = coding::parse(rp, B, H); + return (int)uval; + + case cmk_BHS1: + assert(S == 1 && D == 0); + uval = coding::parse(rp, B, H); + return DECODE_SIGN_S1(uval); + + case cmk_BYTE1: + assert(c.spec == BYTE1_spec); + assert(B == 1 && H == 256 && S == 0 && D == 0); + return *rp++ & 0xFF; + + case cmk_CHAR3: + assert(c.spec == CHAR3_spec); + assert(B == B3 && H == H128 && S == 0 && D == 0); + return coding::parse_lgH(rp, B3, H128, 7); + + case cmk_UNSIGNED5: + assert(c.spec == UNSIGNED5_spec); + assert(B == B5 && H == H64 && S == 0 && D == 0); + return coding::parse_lgH(rp, B5, H64, 6); + + case cmk_BHSD1: + assert(D == 1); + uval = coding::parse(rp, B, H); + if (S != 0) + uval = (uint32_t)decode_sign(S, uval); + return getDeltaValue(this, uval, (bool)c.isSubrange); + + case cmk_BHS1D1full: + assert(S == 1 && D == 1 && c.isFullRange); + uval = coding::parse(rp, B, H); + uval = (uint32_t)DECODE_SIGN_S1(uval); + return getDeltaValue(this, uval, false); + + case cmk_BHS1D1sub: + assert(S == 1 && D == 1 && c.isSubrange); + uval = coding::parse(rp, B, H); + uval = (uint32_t)DECODE_SIGN_S1(uval); + return getDeltaValue(this, uval, true); + + case cmk_DELTA5: + assert(c.spec == DELTA5_spec); + assert(B == B5 && H == H64 && S == 1 && D == 1 && c.isFullRange); + uval = coding::parse_lgH(rp, B5, H64, 6); + sum += DECODE_SIGN_S1(uval); + return sum; + + case cmk_BCI5: + assert(c.spec == BCI5_spec); + assert(B == B5 && H == H4 && S == 0 && D == 0); + return coding::parse_lgH(rp, B5, H4, 2); + + case cmk_BRANCH5: + assert(c.spec == BRANCH5_spec); + assert(B == B5 && H == H4 && S == 2 && D == 0); + uval = coding::parse_lgH(rp, B5, H4, 2); + return decode_sign(S, uval); + + case cmk_pop: + uval = coding::parse(rp, B, H); + if (S != 0) + { + uval = (uint32_t)decode_sign(S, uval); + } + if (D != 0) + { + assert(c.isSubrange | c.isFullRange); + if (c.isSubrange) + sum = c.sumInUnsignedRange(sum, (int)uval); + else + sum += (int)uval; + uval = (uint32_t)sum; + } + return getPopValue(this, uval); + + case cmk_pop_BHS0: + assert(S == 0 && D == 0); + uval = coding::parse(rp, B, H); + return getPopValue(this, uval); + + case cmk_pop_BYTE1: + assert(c.spec == BYTE1_spec); + assert(B == 1 && H == 256 && S == 0 && D == 0); + return getPopValue(this, *rp++ & 0xFF); + + default: + break; + } + assert(false); + return 0; } static int moreCentral(int x, int y) { // used to find end of Pop.{F} - // Suggested implementation from the Pack200 specification: - uint32_t kx = (x >> 31) ^ (x << 1); - uint32_t ky = (y >> 31) ^ (y << 1); - return (kx < ky ? x : y); + // Suggested implementation from the Pack200 specification: + uint32_t kx = (x >> 31) ^ (x << 1); + uint32_t ky = (y >> 31) ^ (y << 1); + return (kx < ky ? x : y); } // static maybe_inline // int moreCentral2(int x, int y, int min) { @@ -641,7 +641,7 @@ static const byte *no_meta[] = {nullptr}; #define NO_META (*(byte **)no_meta) enum { - POP_FAVORED_N = -2 + POP_FAVORED_N = -2 }; // mode bits @@ -650,395 +650,395 @@ enum // This function knows all about meta-coding. void coding_method::init(byte *&band_rp, byte *band_limit, byte *&meta_rp, int mode, - coding *defc, int N, intlist *valueSink) + coding *defc, int N, intlist *valueSink) { - assert(N != 0); - - assert(u != nullptr); // must be pre-initialized - // if (u == nullptr) u = unpacker::current(); // expensive - - int op = (meta_rp == nullptr) ? _meta_default : (*meta_rp++ & 0xFF); - coding *foundc = nullptr; - coding *to_free = nullptr; - - if (op == _meta_default) - { - foundc = defc; - // and fall through - } - else if (op >= _meta_canon_min && op <= _meta_canon_max) - { - foundc = coding::findByIndex(op); - // and fall through - } - else if (op == _meta_arb) - { - int args = (*meta_rp++ & 0xFF); - // args = (D:[0..1] + 2*S[0..2] + 8*(B:[1..5]-1)) - int D = ((args >> 0) & 1); - int S = ((args >> 1) & 3); - int B = ((args >> 3) & -1) + 1; - // & (H[1..256]-1) - int H = (*meta_rp++ & 0xFF) + 1; - foundc = coding::findBySpec(B, H, S, D); - to_free = foundc; // findBySpec may dynamically allocate - if (foundc == nullptr) - { - unpack_abort("illegal arbitrary coding"); - return; - } - // and fall through - } - else if (op >= _meta_run && op < _meta_pop) - { - int args = (op - _meta_run); - // args: KX:[0..3] + 4*(KBFlag:[0..1]) + 8*(ABDef:[0..2]) - int KX = ((args >> 0) & 3); - int KBFlag = ((args >> 2) & 1); - int ABDef = ((args >> 3) & -1); - assert(ABDef <= 2); - // & KB: one of [0..255] if KBFlag=1 - int KB = (!KBFlag ? 3 : (*meta_rp++ & 0xFF)); - int K = (KB + 1) << (KX * 4); - int N2 = (N >= 0) ? N - K : N; - if (N == 0 || (N2 <= 0 && N2 != N)) - { - unpack_abort("illegal run encoding"); - } - if ((mode & DISABLE_RUN) != 0) - { - unpack_abort("illegal nested run encoding"); - } - - // & Enc{ ACode } if ADef=0 (ABDef != 1) - // No direct nesting of 'run' in ACode, but in BCode it's OK. - int disRun = mode | DISABLE_RUN; - if (ABDef == 1) - { - this->init(band_rp, band_limit, NO_META, disRun, defc, K, valueSink); - } - else - { - this->init(band_rp, band_limit, meta_rp, disRun, defc, K, valueSink); - } - - // & Enc{ BCode } if BDef=0 (ABDef != 2) - coding_method *tail = U_NEW(coding_method, 1); - if (!tail) - return; - tail->u = u; - - // The 'run' codings may be nested indirectly via 'pop' codings. - // This means that this->next may already be filled in, if - // ACode was of type 'pop' with a 'run' token coding. - // No problem: Just chain the upcoming BCode onto the end. - for (coding_method *self = this;; self = self->next) - { - if (self->next == nullptr) - { - self->next = tail; - break; - } - } - - if (ABDef == 2) - { - tail->init(band_rp, band_limit, NO_META, mode, defc, N2, valueSink); - } - else - { - tail->init(band_rp, band_limit, meta_rp, mode, defc, N2, valueSink); - } - // Note: The preceding calls to init should be tail-recursive. - - return; // done; no falling through - } - else if (op >= _meta_pop && op < _meta_limit) - { - int args = (op - _meta_pop); - // args: (FDef:[0..1]) + 2*UDef:[0..1] + 4*(TDefL:[0..11]) - int FDef = ((args >> 0) & 1); - int UDef = ((args >> 1) & 1); - int TDefL = ((args >> 2) & -1); - assert(TDefL <= 11); - int TDef = (TDefL > 0); - int TL = (TDefL <= 6) ? (2 << TDefL) : (256 - (4 << (11 - TDefL))); - int TH = (256 - TL); - if (N <= 0) - { - unpack_abort("illegal pop encoding"); - } - if ((mode & DISABLE_POP) != 0) - { - unpack_abort("illegal nested pop encoding"); - } - - // No indirect nesting of 'pop', but 'run' is OK. - int disPop = DISABLE_POP; - - // & Enc{ FCode } if FDef=0 - int FN = POP_FAVORED_N; - assert(valueSink == nullptr); - intlist fValueSink; - fValueSink.init(); - coding_method fval; - BYTES_OF(fval).clear(); - fval.u = u; - if (FDef != 0) - { - fval.init(band_rp, band_limit, NO_META, disPop, defc, FN, &fValueSink); - } - else - { - fval.init(band_rp, band_limit, meta_rp, disPop, defc, FN, &fValueSink); - } - bytes fvbuf; - fValues = (u->saveTo(fvbuf, fValueSink.b), (int *)fvbuf.ptr); - fVlength = fValueSink.length(); // i.e., the parameter K - fValueSink.free(); - - // Skip the first {F} run in all subsequent passes. - // The next call to this->init(...) will set vs0.rp to point after the {F}. - - // & Enc{ TCode } if TDef=0 (TDefL==0) - if (TDef != 0) - { - coding *tcode = coding::findBySpec(1, 256); // BYTE1 - // find the most narrowly sufficient code: - for (int B = 2; B <= B_MAX; B++) - { - if (fVlength <= tcode->umax) - break; // found it - tcode->free(); - tcode = coding::findBySpec(B, TH); - if (!tcode) - return; - } - if (!(fVlength <= tcode->umax)) - { - unpack_abort("pop.L value too small"); - } - this->init(band_rp, band_limit, NO_META, disPop, tcode, N, nullptr); - tcode->free(); - } - else - { - this->init(band_rp, band_limit, meta_rp, disPop, defc, N, nullptr); - } - - // Count the number of zero tokens right now. - // Also verify that they are in bounds. - int UN = 0; // one {U} for each zero in {T} - value_stream vs = vs0; - for (int i = 0; i < N; i++) - { - uint32_t val = vs.getInt(); - if (val == 0) - UN += 1; - if (!(val <= (uint32_t)fVlength)) - { - unpack_abort("pop token out of range"); - } - } - vs.done(); - - // & Enc{ UCode } if UDef=0 - if (UN != 0) - { - uValues = U_NEW(coding_method, 1); - if (uValues == nullptr) - return; - uValues->u = u; - if (UDef != 0) - { - uValues->init(band_rp, band_limit, NO_META, disPop, defc, UN, nullptr); - } - else - { - uValues->init(band_rp, band_limit, meta_rp, disPop, defc, UN, nullptr); - } - } - else - { - if (UDef == 0) - { - int uop = (*meta_rp++ & 0xFF); - if (uop > _meta_canon_max) - // %%% Spec. requires the more strict (uop != _meta_default). - unpack_abort("bad meta-coding for empty pop/U"); - } - } - - // Bug fix for 6259542 - // Last of all, adjust vs0.cmk to the 'pop' flavor - for (coding_method *self = this; self != nullptr; self = self->next) - { - coding_method_kind cmk2 = cmk_pop; - switch (self->vs0.cmk) - { - case cmk_BHS0: - cmk2 = cmk_pop_BHS0; - break; - case cmk_BYTE1: - cmk2 = cmk_pop_BYTE1; - break; - default: - break; - } - self->vs0.cmk = cmk2; - if (self != this) - { - assert(self->fValues == nullptr); // no double init - self->fValues = this->fValues; - self->fVlength = this->fVlength; - assert(self->uValues == nullptr); // must stay nullptr - } - } - - return; // done; no falling through - } - else - { - unpack_abort("bad meta-coding"); - } - - // Common code here skips a series of values with one coding. - assert(foundc != nullptr); - - assert(vs0.cmk == cmk_ERROR); // no garbage, please - assert(vs0.rp == nullptr); // no garbage, please - assert(vs0.rplimit == nullptr); // no garbage, please - assert(vs0.sum == 0); // no garbage, please - - vs0.init(band_rp, band_limit, foundc); - - // Done with foundc. Free if necessary. - if (to_free != nullptr) - { - to_free->free(); - to_free = nullptr; - } - foundc = nullptr; - - coding &c = vs0.c; - CODING_PRIVATE(c.spec); - // assert sane N - assert((uint32_t)N < INT_MAX_VALUE || N == POP_FAVORED_N); - - // Look at the values, or at least skip over them quickly. - if (valueSink == nullptr) - { - // Skip and ignore values in the first pass. - c.parseMultiple(band_rp, N, band_limit, B, H); - } - else if (N >= 0) - { - // Pop coding, {F} sequence, initial run of values... - assert((mode & DISABLE_POP) != 0); - value_stream vs = vs0; - for (int n = 0; n < N; n++) - { - int val = vs.getInt(); - valueSink->add(val); - } - band_rp = vs.rp; - } - else - { - // Pop coding, {F} sequence, final run of values... - assert((mode & DISABLE_POP) != 0); - assert(N == POP_FAVORED_N); - int min = INT_MIN_VALUE; // farthest from the center - // min2 is based on the buggy specification of centrality in version 150.7 - // no known implementations transmit this value, but just in case... - // int min2 = INT_MIN_VALUE; - int last = 0; - // if there were initial runs, find the potential sentinels in them: - for (int i = 0; i < valueSink->length(); i++) - { - last = valueSink->get(i); - min = moreCentral(min, last); - // min2 = moreCentral2(min2, last, min); - } - value_stream vs = vs0; - for (;;) - { - int val = vs.getInt(); - if (valueSink->length() > 0 && (val == last || val == min)) //|| val == min2 - break; - valueSink->add(val); - last = val; - min = moreCentral(min, last); - // min2 = moreCentral2(min2, last, min); - } - band_rp = vs.rp; - } - - // Get an accurate upper limit now. - vs0.rplimit = band_rp; - vs0.cm = this; - - return; // success + assert(N != 0); + + assert(u != nullptr); // must be pre-initialized + // if (u == nullptr) u = unpacker::current(); // expensive + + int op = (meta_rp == nullptr) ? _meta_default : (*meta_rp++ & 0xFF); + coding *foundc = nullptr; + coding *to_free = nullptr; + + if (op == _meta_default) + { + foundc = defc; + // and fall through + } + else if (op >= _meta_canon_min && op <= _meta_canon_max) + { + foundc = coding::findByIndex(op); + // and fall through + } + else if (op == _meta_arb) + { + int args = (*meta_rp++ & 0xFF); + // args = (D:[0..1] + 2*S[0..2] + 8*(B:[1..5]-1)) + int D = ((args >> 0) & 1); + int S = ((args >> 1) & 3); + int B = ((args >> 3) & -1) + 1; + // & (H[1..256]-1) + int H = (*meta_rp++ & 0xFF) + 1; + foundc = coding::findBySpec(B, H, S, D); + to_free = foundc; // findBySpec may dynamically allocate + if (foundc == nullptr) + { + unpack_abort("illegal arbitrary coding"); + return; + } + // and fall through + } + else if (op >= _meta_run && op < _meta_pop) + { + int args = (op - _meta_run); + // args: KX:[0..3] + 4*(KBFlag:[0..1]) + 8*(ABDef:[0..2]) + int KX = ((args >> 0) & 3); + int KBFlag = ((args >> 2) & 1); + int ABDef = ((args >> 3) & -1); + assert(ABDef <= 2); + // & KB: one of [0..255] if KBFlag=1 + int KB = (!KBFlag ? 3 : (*meta_rp++ & 0xFF)); + int K = (KB + 1) << (KX * 4); + int N2 = (N >= 0) ? N - K : N; + if (N == 0 || (N2 <= 0 && N2 != N)) + { + unpack_abort("illegal run encoding"); + } + if ((mode & DISABLE_RUN) != 0) + { + unpack_abort("illegal nested run encoding"); + } + + // & Enc{ ACode } if ADef=0 (ABDef != 1) + // No direct nesting of 'run' in ACode, but in BCode it's OK. + int disRun = mode | DISABLE_RUN; + if (ABDef == 1) + { + this->init(band_rp, band_limit, NO_META, disRun, defc, K, valueSink); + } + else + { + this->init(band_rp, band_limit, meta_rp, disRun, defc, K, valueSink); + } + + // & Enc{ BCode } if BDef=0 (ABDef != 2) + coding_method *tail = U_NEW(coding_method, 1); + if (!tail) + return; + tail->u = u; + + // The 'run' codings may be nested indirectly via 'pop' codings. + // This means that this->next may already be filled in, if + // ACode was of type 'pop' with a 'run' token coding. + // No problem: Just chain the upcoming BCode onto the end. + for (coding_method *self = this;; self = self->next) + { + if (self->next == nullptr) + { + self->next = tail; + break; + } + } + + if (ABDef == 2) + { + tail->init(band_rp, band_limit, NO_META, mode, defc, N2, valueSink); + } + else + { + tail->init(band_rp, band_limit, meta_rp, mode, defc, N2, valueSink); + } + // Note: The preceding calls to init should be tail-recursive. + + return; // done; no falling through + } + else if (op >= _meta_pop && op < _meta_limit) + { + int args = (op - _meta_pop); + // args: (FDef:[0..1]) + 2*UDef:[0..1] + 4*(TDefL:[0..11]) + int FDef = ((args >> 0) & 1); + int UDef = ((args >> 1) & 1); + int TDefL = ((args >> 2) & -1); + assert(TDefL <= 11); + int TDef = (TDefL > 0); + int TL = (TDefL <= 6) ? (2 << TDefL) : (256 - (4 << (11 - TDefL))); + int TH = (256 - TL); + if (N <= 0) + { + unpack_abort("illegal pop encoding"); + } + if ((mode & DISABLE_POP) != 0) + { + unpack_abort("illegal nested pop encoding"); + } + + // No indirect nesting of 'pop', but 'run' is OK. + int disPop = DISABLE_POP; + + // & Enc{ FCode } if FDef=0 + int FN = POP_FAVORED_N; + assert(valueSink == nullptr); + intlist fValueSink; + fValueSink.init(); + coding_method fval; + BYTES_OF(fval).clear(); + fval.u = u; + if (FDef != 0) + { + fval.init(band_rp, band_limit, NO_META, disPop, defc, FN, &fValueSink); + } + else + { + fval.init(band_rp, band_limit, meta_rp, disPop, defc, FN, &fValueSink); + } + bytes fvbuf; + fValues = (u->saveTo(fvbuf, fValueSink.b), (int *)fvbuf.ptr); + fVlength = fValueSink.length(); // i.e., the parameter K + fValueSink.free(); + + // Skip the first {F} run in all subsequent passes. + // The next call to this->init(...) will set vs0.rp to point after the {F}. + + // & Enc{ TCode } if TDef=0 (TDefL==0) + if (TDef != 0) + { + coding *tcode = coding::findBySpec(1, 256); // BYTE1 + // find the most narrowly sufficient code: + for (int B = 2; B <= B_MAX; B++) + { + if (fVlength <= tcode->umax) + break; // found it + tcode->free(); + tcode = coding::findBySpec(B, TH); + if (!tcode) + return; + } + if (!(fVlength <= tcode->umax)) + { + unpack_abort("pop.L value too small"); + } + this->init(band_rp, band_limit, NO_META, disPop, tcode, N, nullptr); + tcode->free(); + } + else + { + this->init(band_rp, band_limit, meta_rp, disPop, defc, N, nullptr); + } + + // Count the number of zero tokens right now. + // Also verify that they are in bounds. + int UN = 0; // one {U} for each zero in {T} + value_stream vs = vs0; + for (int i = 0; i < N; i++) + { + uint32_t val = vs.getInt(); + if (val == 0) + UN += 1; + if (!(val <= (uint32_t)fVlength)) + { + unpack_abort("pop token out of range"); + } + } + vs.done(); + + // & Enc{ UCode } if UDef=0 + if (UN != 0) + { + uValues = U_NEW(coding_method, 1); + if (uValues == nullptr) + return; + uValues->u = u; + if (UDef != 0) + { + uValues->init(band_rp, band_limit, NO_META, disPop, defc, UN, nullptr); + } + else + { + uValues->init(band_rp, band_limit, meta_rp, disPop, defc, UN, nullptr); + } + } + else + { + if (UDef == 0) + { + int uop = (*meta_rp++ & 0xFF); + if (uop > _meta_canon_max) + // %%% Spec. requires the more strict (uop != _meta_default). + unpack_abort("bad meta-coding for empty pop/U"); + } + } + + // Bug fix for 6259542 + // Last of all, adjust vs0.cmk to the 'pop' flavor + for (coding_method *self = this; self != nullptr; self = self->next) + { + coding_method_kind cmk2 = cmk_pop; + switch (self->vs0.cmk) + { + case cmk_BHS0: + cmk2 = cmk_pop_BHS0; + break; + case cmk_BYTE1: + cmk2 = cmk_pop_BYTE1; + break; + default: + break; + } + self->vs0.cmk = cmk2; + if (self != this) + { + assert(self->fValues == nullptr); // no double init + self->fValues = this->fValues; + self->fVlength = this->fVlength; + assert(self->uValues == nullptr); // must stay nullptr + } + } + + return; // done; no falling through + } + else + { + unpack_abort("bad meta-coding"); + } + + // Common code here skips a series of values with one coding. + assert(foundc != nullptr); + + assert(vs0.cmk == cmk_ERROR); // no garbage, please + assert(vs0.rp == nullptr); // no garbage, please + assert(vs0.rplimit == nullptr); // no garbage, please + assert(vs0.sum == 0); // no garbage, please + + vs0.init(band_rp, band_limit, foundc); + + // Done with foundc. Free if necessary. + if (to_free != nullptr) + { + to_free->free(); + to_free = nullptr; + } + foundc = nullptr; + + coding &c = vs0.c; + CODING_PRIVATE(c.spec); + // assert sane N + assert((uint32_t)N < INT_MAX_VALUE || N == POP_FAVORED_N); + + // Look at the values, or at least skip over them quickly. + if (valueSink == nullptr) + { + // Skip and ignore values in the first pass. + c.parseMultiple(band_rp, N, band_limit, B, H); + } + else if (N >= 0) + { + // Pop coding, {F} sequence, initial run of values... + assert((mode & DISABLE_POP) != 0); + value_stream vs = vs0; + for (int n = 0; n < N; n++) + { + int val = vs.getInt(); + valueSink->add(val); + } + band_rp = vs.rp; + } + else + { + // Pop coding, {F} sequence, final run of values... + assert((mode & DISABLE_POP) != 0); + assert(N == POP_FAVORED_N); + int min = INT_MIN_VALUE; // farthest from the center + // min2 is based on the buggy specification of centrality in version 150.7 + // no known implementations transmit this value, but just in case... + // int min2 = INT_MIN_VALUE; + int last = 0; + // if there were initial runs, find the potential sentinels in them: + for (int i = 0; i < valueSink->length(); i++) + { + last = valueSink->get(i); + min = moreCentral(min, last); + // min2 = moreCentral2(min2, last, min); + } + value_stream vs = vs0; + for (;;) + { + int val = vs.getInt(); + if (valueSink->length() > 0 && (val == last || val == min)) //|| val == min2 + break; + valueSink->add(val); + last = val; + min = moreCentral(min, last); + // min2 = moreCentral2(min2, last, min); + } + band_rp = vs.rp; + } + + // Get an accurate upper limit now. + vs0.rplimit = band_rp; + vs0.cm = this; + + return; // success } coding basic_codings[] = { - // This one is not a usable irregular coding, but is used by cp_Utf8_chars. - CODING_INIT(3, 128, 0, 0), - - // Fixed-length codings: - CODING_INIT(1, 256, 0, 0), CODING_INIT(1, 256, 1, 0), CODING_INIT(1, 256, 0, 1), - CODING_INIT(1, 256, 1, 1), CODING_INIT(2, 256, 0, 0), CODING_INIT(2, 256, 1, 0), - CODING_INIT(2, 256, 0, 1), CODING_INIT(2, 256, 1, 1), CODING_INIT(3, 256, 0, 0), - CODING_INIT(3, 256, 1, 0), CODING_INIT(3, 256, 0, 1), CODING_INIT(3, 256, 1, 1), - CODING_INIT(4, 256, 0, 0), CODING_INIT(4, 256, 1, 0), CODING_INIT(4, 256, 0, 1), - CODING_INIT(4, 256, 1, 1), - - // Full-range variable-length codings: - CODING_INIT(5, 4, 0, 0), CODING_INIT(5, 4, 1, 0), CODING_INIT(5, 4, 2, 0), - CODING_INIT(5, 16, 0, 0), CODING_INIT(5, 16, 1, 0), CODING_INIT(5, 16, 2, 0), - CODING_INIT(5, 32, 0, 0), CODING_INIT(5, 32, 1, 0), CODING_INIT(5, 32, 2, 0), - CODING_INIT(5, 64, 0, 0), CODING_INIT(5, 64, 1, 0), CODING_INIT(5, 64, 2, 0), - CODING_INIT(5, 128, 0, 0), CODING_INIT(5, 128, 1, 0), CODING_INIT(5, 128, 2, 0), - CODING_INIT(5, 4, 0, 1), CODING_INIT(5, 4, 1, 1), CODING_INIT(5, 4, 2, 1), - CODING_INIT(5, 16, 0, 1), CODING_INIT(5, 16, 1, 1), CODING_INIT(5, 16, 2, 1), - CODING_INIT(5, 32, 0, 1), CODING_INIT(5, 32, 1, 1), CODING_INIT(5, 32, 2, 1), - CODING_INIT(5, 64, 0, 1), CODING_INIT(5, 64, 1, 1), CODING_INIT(5, 64, 2, 1), - CODING_INIT(5, 128, 0, 1), CODING_INIT(5, 128, 1, 1), CODING_INIT(5, 128, 2, 1), - - // Variable length subrange codings: - CODING_INIT(2, 192, 0, 0), CODING_INIT(2, 224, 0, 0), CODING_INIT(2, 240, 0, 0), - CODING_INIT(2, 248, 0, 0), CODING_INIT(2, 252, 0, 0), CODING_INIT(2, 8, 0, 1), - CODING_INIT(2, 8, 1, 1), CODING_INIT(2, 16, 0, 1), CODING_INIT(2, 16, 1, 1), - CODING_INIT(2, 32, 0, 1), CODING_INIT(2, 32, 1, 1), CODING_INIT(2, 64, 0, 1), - CODING_INIT(2, 64, 1, 1), CODING_INIT(2, 128, 0, 1), CODING_INIT(2, 128, 1, 1), - CODING_INIT(2, 192, 0, 1), CODING_INIT(2, 192, 1, 1), CODING_INIT(2, 224, 0, 1), - CODING_INIT(2, 224, 1, 1), CODING_INIT(2, 240, 0, 1), CODING_INIT(2, 240, 1, 1), - CODING_INIT(2, 248, 0, 1), CODING_INIT(2, 248, 1, 1), CODING_INIT(3, 192, 0, 0), - CODING_INIT(3, 224, 0, 0), CODING_INIT(3, 240, 0, 0), CODING_INIT(3, 248, 0, 0), - CODING_INIT(3, 252, 0, 0), CODING_INIT(3, 8, 0, 1), CODING_INIT(3, 8, 1, 1), - CODING_INIT(3, 16, 0, 1), CODING_INIT(3, 16, 1, 1), CODING_INIT(3, 32, 0, 1), - CODING_INIT(3, 32, 1, 1), CODING_INIT(3, 64, 0, 1), CODING_INIT(3, 64, 1, 1), - CODING_INIT(3, 128, 0, 1), CODING_INIT(3, 128, 1, 1), CODING_INIT(3, 192, 0, 1), - CODING_INIT(3, 192, 1, 1), CODING_INIT(3, 224, 0, 1), CODING_INIT(3, 224, 1, 1), - CODING_INIT(3, 240, 0, 1), CODING_INIT(3, 240, 1, 1), CODING_INIT(3, 248, 0, 1), - CODING_INIT(3, 248, 1, 1), CODING_INIT(4, 192, 0, 0), CODING_INIT(4, 224, 0, 0), - CODING_INIT(4, 240, 0, 0), CODING_INIT(4, 248, 0, 0), CODING_INIT(4, 252, 0, 0), - CODING_INIT(4, 8, 0, 1), CODING_INIT(4, 8, 1, 1), CODING_INIT(4, 16, 0, 1), - CODING_INIT(4, 16, 1, 1), CODING_INIT(4, 32, 0, 1), CODING_INIT(4, 32, 1, 1), - CODING_INIT(4, 64, 0, 1), CODING_INIT(4, 64, 1, 1), CODING_INIT(4, 128, 0, 1), - CODING_INIT(4, 128, 1, 1), CODING_INIT(4, 192, 0, 1), CODING_INIT(4, 192, 1, 1), - CODING_INIT(4, 224, 0, 1), CODING_INIT(4, 224, 1, 1), CODING_INIT(4, 240, 0, 1), - CODING_INIT(4, 240, 1, 1), CODING_INIT(4, 248, 0, 1), CODING_INIT(4, 248, 1, 1), - CODING_INIT(0, 0, 0, 0)}; + // This one is not a usable irregular coding, but is used by cp_Utf8_chars. + CODING_INIT(3, 128, 0, 0), + + // Fixed-length codings: + CODING_INIT(1, 256, 0, 0), CODING_INIT(1, 256, 1, 0), CODING_INIT(1, 256, 0, 1), + CODING_INIT(1, 256, 1, 1), CODING_INIT(2, 256, 0, 0), CODING_INIT(2, 256, 1, 0), + CODING_INIT(2, 256, 0, 1), CODING_INIT(2, 256, 1, 1), CODING_INIT(3, 256, 0, 0), + CODING_INIT(3, 256, 1, 0), CODING_INIT(3, 256, 0, 1), CODING_INIT(3, 256, 1, 1), + CODING_INIT(4, 256, 0, 0), CODING_INIT(4, 256, 1, 0), CODING_INIT(4, 256, 0, 1), + CODING_INIT(4, 256, 1, 1), + + // Full-range variable-length codings: + CODING_INIT(5, 4, 0, 0), CODING_INIT(5, 4, 1, 0), CODING_INIT(5, 4, 2, 0), + CODING_INIT(5, 16, 0, 0), CODING_INIT(5, 16, 1, 0), CODING_INIT(5, 16, 2, 0), + CODING_INIT(5, 32, 0, 0), CODING_INIT(5, 32, 1, 0), CODING_INIT(5, 32, 2, 0), + CODING_INIT(5, 64, 0, 0), CODING_INIT(5, 64, 1, 0), CODING_INIT(5, 64, 2, 0), + CODING_INIT(5, 128, 0, 0), CODING_INIT(5, 128, 1, 0), CODING_INIT(5, 128, 2, 0), + CODING_INIT(5, 4, 0, 1), CODING_INIT(5, 4, 1, 1), CODING_INIT(5, 4, 2, 1), + CODING_INIT(5, 16, 0, 1), CODING_INIT(5, 16, 1, 1), CODING_INIT(5, 16, 2, 1), + CODING_INIT(5, 32, 0, 1), CODING_INIT(5, 32, 1, 1), CODING_INIT(5, 32, 2, 1), + CODING_INIT(5, 64, 0, 1), CODING_INIT(5, 64, 1, 1), CODING_INIT(5, 64, 2, 1), + CODING_INIT(5, 128, 0, 1), CODING_INIT(5, 128, 1, 1), CODING_INIT(5, 128, 2, 1), + + // Variable length subrange codings: + CODING_INIT(2, 192, 0, 0), CODING_INIT(2, 224, 0, 0), CODING_INIT(2, 240, 0, 0), + CODING_INIT(2, 248, 0, 0), CODING_INIT(2, 252, 0, 0), CODING_INIT(2, 8, 0, 1), + CODING_INIT(2, 8, 1, 1), CODING_INIT(2, 16, 0, 1), CODING_INIT(2, 16, 1, 1), + CODING_INIT(2, 32, 0, 1), CODING_INIT(2, 32, 1, 1), CODING_INIT(2, 64, 0, 1), + CODING_INIT(2, 64, 1, 1), CODING_INIT(2, 128, 0, 1), CODING_INIT(2, 128, 1, 1), + CODING_INIT(2, 192, 0, 1), CODING_INIT(2, 192, 1, 1), CODING_INIT(2, 224, 0, 1), + CODING_INIT(2, 224, 1, 1), CODING_INIT(2, 240, 0, 1), CODING_INIT(2, 240, 1, 1), + CODING_INIT(2, 248, 0, 1), CODING_INIT(2, 248, 1, 1), CODING_INIT(3, 192, 0, 0), + CODING_INIT(3, 224, 0, 0), CODING_INIT(3, 240, 0, 0), CODING_INIT(3, 248, 0, 0), + CODING_INIT(3, 252, 0, 0), CODING_INIT(3, 8, 0, 1), CODING_INIT(3, 8, 1, 1), + CODING_INIT(3, 16, 0, 1), CODING_INIT(3, 16, 1, 1), CODING_INIT(3, 32, 0, 1), + CODING_INIT(3, 32, 1, 1), CODING_INIT(3, 64, 0, 1), CODING_INIT(3, 64, 1, 1), + CODING_INIT(3, 128, 0, 1), CODING_INIT(3, 128, 1, 1), CODING_INIT(3, 192, 0, 1), + CODING_INIT(3, 192, 1, 1), CODING_INIT(3, 224, 0, 1), CODING_INIT(3, 224, 1, 1), + CODING_INIT(3, 240, 0, 1), CODING_INIT(3, 240, 1, 1), CODING_INIT(3, 248, 0, 1), + CODING_INIT(3, 248, 1, 1), CODING_INIT(4, 192, 0, 0), CODING_INIT(4, 224, 0, 0), + CODING_INIT(4, 240, 0, 0), CODING_INIT(4, 248, 0, 0), CODING_INIT(4, 252, 0, 0), + CODING_INIT(4, 8, 0, 1), CODING_INIT(4, 8, 1, 1), CODING_INIT(4, 16, 0, 1), + CODING_INIT(4, 16, 1, 1), CODING_INIT(4, 32, 0, 1), CODING_INIT(4, 32, 1, 1), + CODING_INIT(4, 64, 0, 1), CODING_INIT(4, 64, 1, 1), CODING_INIT(4, 128, 0, 1), + CODING_INIT(4, 128, 1, 1), CODING_INIT(4, 192, 0, 1), CODING_INIT(4, 192, 1, 1), + CODING_INIT(4, 224, 0, 1), CODING_INIT(4, 224, 1, 1), CODING_INIT(4, 240, 0, 1), + CODING_INIT(4, 240, 1, 1), CODING_INIT(4, 248, 0, 1), CODING_INIT(4, 248, 1, 1), + CODING_INIT(0, 0, 0, 0)}; #define BASIC_INDEX_LIMIT (int)(sizeof(basic_codings) / sizeof(basic_codings[0]) - 1) coding *coding::findByIndex(int idx) { - int index_limit = BASIC_INDEX_LIMIT; - assert(_meta_canon_min == 1 && _meta_canon_max + 1 == index_limit); + int index_limit = BASIC_INDEX_LIMIT; + assert(_meta_canon_min == 1 && _meta_canon_max + 1 == index_limit); - if (idx >= _meta_canon_min && idx <= _meta_canon_max) - return basic_codings[idx].init(); - else - return nullptr; + if (idx >= _meta_canon_min && idx <= _meta_canon_max) + return basic_codings[idx].init(); + else + return nullptr; } diff --git a/libraries/pack200/src/coding.h b/libraries/pack200/src/coding.h index f9bd6ca2..bfdd252e 100644 --- a/libraries/pack200/src/coding.h +++ b/libraries/pack200/src/coding.h @@ -35,9 +35,9 @@ struct unpacker; #define CODING_D(x) ((x) >> 0 & 0xF) #define CODING_INIT(B, H, S, D) \ - { \ - CODING_SPEC(B, H, S, D), 0, 0, 0, 0, 0, 0, 0, 0 \ - } + { \ + CODING_SPEC(B, H, S, D), 0, 0, 0, 0, 0, 0, 0, 0 \ + } // For debugging purposes, some compilers do not like this and will complain. // #define long do_not_use_C_long_types_use_jlong_or_int @@ -45,126 +45,126 @@ struct unpacker; struct coding { - int spec; // B,H,S,D - - // Handy values derived from the spec: - int B() - { - return CODING_B(spec); - } - int H() - { - return CODING_H(spec); - } - int S() - { - return CODING_S(spec); - } - int D() - { - return CODING_D(spec); - } - int L() - { - return 256 - CODING_H(spec); - } - int min, max; - int umin, umax; - char isSigned, isSubrange, isFullRange, isMalloc; - - coding *init(); // returns self or nullptr if error - coding *initFrom(int spec_) - { - assert(this->spec == 0); - this->spec = spec_; - return init(); - } - - static coding *findBySpec(int spec); - static coding *findBySpec(int B, int H, int S = 0, int D = 0); - static coding *findByIndex(int irregularCodingIndex); - - static uint32_t parse(byte *&rp, int B, int H); - static uint32_t parse_lgH(byte *&rp, int B, int H, int lgH); - static void parseMultiple(byte *&rp, int N, byte *limit, int B, int H); - - uint32_t parse(byte *&rp) - { - return parse(rp, CODING_B(spec), CODING_H(spec)); - } - void parseMultiple(byte *&rp, int N, byte *limit) - { - parseMultiple(rp, N, limit, CODING_B(spec), CODING_H(spec)); - } - - bool canRepresent(int x) - { - return (x >= min && x <= max); - } - bool canRepresentUnsigned(int x) - { - return (x >= umin && x <= umax); - } - - int sumInUnsignedRange(int x, int y); - - int readFrom(byte *&rpVar, int *dbase); - void readArrayFrom(byte *&rpVar, int *dbase, int length, int *values); - void skipArrayFrom(byte *&rpVar, int length) - { - readArrayFrom(rpVar, (int *)NULL, length, (int *)NULL); - } - - void free(); // free self if isMalloc + int spec; // B,H,S,D + + // Handy values derived from the spec: + int B() + { + return CODING_B(spec); + } + int H() + { + return CODING_H(spec); + } + int S() + { + return CODING_S(spec); + } + int D() + { + return CODING_D(spec); + } + int L() + { + return 256 - CODING_H(spec); + } + int min, max; + int umin, umax; + char isSigned, isSubrange, isFullRange, isMalloc; + + coding *init(); // returns self or nullptr if error + coding *initFrom(int spec_) + { + assert(this->spec == 0); + this->spec = spec_; + return init(); + } + + static coding *findBySpec(int spec); + static coding *findBySpec(int B, int H, int S = 0, int D = 0); + static coding *findByIndex(int irregularCodingIndex); + + static uint32_t parse(byte *&rp, int B, int H); + static uint32_t parse_lgH(byte *&rp, int B, int H, int lgH); + static void parseMultiple(byte *&rp, int N, byte *limit, int B, int H); + + uint32_t parse(byte *&rp) + { + return parse(rp, CODING_B(spec), CODING_H(spec)); + } + void parseMultiple(byte *&rp, int N, byte *limit) + { + parseMultiple(rp, N, limit, CODING_B(spec), CODING_H(spec)); + } + + bool canRepresent(int x) + { + return (x >= min && x <= max); + } + bool canRepresentUnsigned(int x) + { + return (x >= umin && x <= umax); + } + + int sumInUnsignedRange(int x, int y); + + int readFrom(byte *&rpVar, int *dbase); + void readArrayFrom(byte *&rpVar, int *dbase, int length, int *values); + void skipArrayFrom(byte *&rpVar, int length) + { + readArrayFrom(rpVar, (int *)NULL, length, (int *)NULL); + } + + void free(); // free self if isMalloc }; enum coding_method_kind { - cmk_ERROR, - cmk_BHS, - cmk_BHS0, - cmk_BHS1, - cmk_BHSD1, - cmk_BHS1D1full, // isFullRange - cmk_BHS1D1sub, // isSubRange - - // special cases hand-optimized (~50% of all decoded values) - cmk_BYTE1, //(1,256) 6% - cmk_CHAR3, //(3,128) 7% - cmk_UNSIGNED5, //(5,64) 13% - cmk_DELTA5, //(5,64,1,1) 5% - cmk_BCI5, //(5,4) 18% - cmk_BRANCH5, //(5,4,2) 4% - // cmk_UNSIGNED5H16, //(5,16) 5% - // cmk_UNSIGNED2H4, //(2,4) 6% - // cmk_DELTA4H8, //(4,8,1,1) 10% - // cmk_DELTA3H16, //(3,16,1,1) 9% - cmk_BHS_LIMIT, - cmk_pop, - cmk_pop_BHS0, - cmk_pop_BYTE1, - cmk_pop_LIMIT, - cmk_LIMIT + cmk_ERROR, + cmk_BHS, + cmk_BHS0, + cmk_BHS1, + cmk_BHSD1, + cmk_BHS1D1full, // isFullRange + cmk_BHS1D1sub, // isSubRange + + // special cases hand-optimized (~50% of all decoded values) + cmk_BYTE1, //(1,256) 6% + cmk_CHAR3, //(3,128) 7% + cmk_UNSIGNED5, //(5,64) 13% + cmk_DELTA5, //(5,64,1,1) 5% + cmk_BCI5, //(5,4) 18% + cmk_BRANCH5, //(5,4,2) 4% + // cmk_UNSIGNED5H16, //(5,16) 5% + // cmk_UNSIGNED2H4, //(2,4) 6% + // cmk_DELTA4H8, //(4,8,1,1) 10% + // cmk_DELTA3H16, //(3,16,1,1) 9% + cmk_BHS_LIMIT, + cmk_pop, + cmk_pop_BHS0, + cmk_pop_BYTE1, + cmk_pop_LIMIT, + cmk_LIMIT }; enum { - BYTE1_spec = CODING_SPEC(1, 256, 0, 0), - CHAR3_spec = CODING_SPEC(3, 128, 0, 0), - UNSIGNED4_spec = CODING_SPEC(4, 256, 0, 0), - UNSIGNED5_spec = CODING_SPEC(5, 64, 0, 0), - SIGNED5_spec = CODING_SPEC(5, 64, 1, 0), - DELTA5_spec = CODING_SPEC(5, 64, 1, 1), - UDELTA5_spec = CODING_SPEC(5, 64, 0, 1), - MDELTA5_spec = CODING_SPEC(5, 64, 2, 1), - BCI5_spec = CODING_SPEC(5, 4, 0, 0), - BRANCH5_spec = CODING_SPEC(5, 4, 2, 0) + BYTE1_spec = CODING_SPEC(1, 256, 0, 0), + CHAR3_spec = CODING_SPEC(3, 128, 0, 0), + UNSIGNED4_spec = CODING_SPEC(4, 256, 0, 0), + UNSIGNED5_spec = CODING_SPEC(5, 64, 0, 0), + SIGNED5_spec = CODING_SPEC(5, 64, 1, 0), + DELTA5_spec = CODING_SPEC(5, 64, 1, 1), + UDELTA5_spec = CODING_SPEC(5, 64, 0, 1), + MDELTA5_spec = CODING_SPEC(5, 64, 2, 1), + BCI5_spec = CODING_SPEC(5, 4, 0, 0), + BRANCH5_spec = CODING_SPEC(5, 4, 2, 0) }; enum { - B_MAX = 5, - C_SLOP = B_MAX * 10 + B_MAX = 5, + C_SLOP = B_MAX * 10 }; struct coding_method; @@ -172,76 +172,76 @@ struct coding_method; // iterator under the control of a meta-coding struct value_stream { - // current coding of values or values - coding c; // B,H,S,D,etc. - coding_method_kind cmk; // type of decoding needed - byte *rp; // read pointer - byte *rplimit; // final value of read pointer - int sum; // partial sum of all values so far (D=1 only) - coding_method *cm; // coding method that defines this stream - - void init(byte *band_rp, byte *band_limit, coding *defc); - void init(byte *band_rp, byte *band_limit, int spec) - { - init(band_rp, band_limit, coding::findBySpec(spec)); - } - - void setCoding(coding *c); - void setCoding(int spec) - { - setCoding(coding::findBySpec(spec)); - } - - // Parse and decode a single value. - int getInt(); - - // Parse and decode a single byte, with no error checks. - int getByte() - { - assert(cmk == cmk_BYTE1); - assert(rp < rplimit); - return *rp++ & 0xFF; - } - - // Used only for asserts. - bool hasValue(); - - void done() - { - assert(!hasValue()); - } - - // Sometimes a value stream has an auxiliary (but there are never two). - value_stream *helper() - { - assert(hasHelper()); - return this + 1; - } - bool hasHelper(); + // current coding of values or values + coding c; // B,H,S,D,etc. + coding_method_kind cmk; // type of decoding needed + byte *rp; // read pointer + byte *rplimit; // final value of read pointer + int sum; // partial sum of all values so far (D=1 only) + coding_method *cm; // coding method that defines this stream + + void init(byte *band_rp, byte *band_limit, coding *defc); + void init(byte *band_rp, byte *band_limit, int spec) + { + init(band_rp, band_limit, coding::findBySpec(spec)); + } + + void setCoding(coding *c); + void setCoding(int spec) + { + setCoding(coding::findBySpec(spec)); + } + + // Parse and decode a single value. + int getInt(); + + // Parse and decode a single byte, with no error checks. + int getByte() + { + assert(cmk == cmk_BYTE1); + assert(rp < rplimit); + return *rp++ & 0xFF; + } + + // Used only for asserts. + bool hasValue(); + + void done() + { + assert(!hasValue()); + } + + // Sometimes a value stream has an auxiliary (but there are never two). + value_stream *helper() + { + assert(hasHelper()); + return this + 1; + } + bool hasHelper(); }; struct coding_method { - value_stream vs0; // initial state snapshot (vs.meta==this) + value_stream vs0; // initial state snapshot (vs.meta==this) - coding_method *next; // what to do when we run out of bytes + coding_method *next; // what to do when we run out of bytes - // these fields are used for pop codes only: - int *fValues; // favored value array - int fVlength; // maximum favored value token - coding_method *uValues; // unfavored value stream + // these fields are used for pop codes only: + int *fValues; // favored value array + int fVlength; // maximum favored value token + coding_method *uValues; // unfavored value stream - // pointer to outer unpacker, for error checks etc. - unpacker *u; + // pointer to outer unpacker, for error checks etc. + unpacker *u; - // Initialize a value stream. - void reset(value_stream *state); + // Initialize a value stream. + void reset(value_stream *state); - // Parse a band header, size a band, and initialize for further action. - // band_rp advances (but not past band_limit), and meta_rp advances. - // The mode gives context, such as "inside a pop". - // The defc and N are the incoming parameters to a meta-coding. - // The value sink is used to collect output values, when desired. - void init(byte *&band_rp, byte *band_limit, byte *&meta_rp, int mode, coding *defc, int N, - intlist *valueSink); + // Parse a band header, size a band, and initialize for further action. + // band_rp advances (but not past band_limit), and meta_rp advances. + // The mode gives context, such as "inside a pop". + // The defc and N are the incoming parameters to a meta-coding. + // The value sink is used to collect output values, when desired. + void init(byte *&band_rp, byte *band_limit, byte *&meta_rp, int mode, coding *defc, int N, + intlist *valueSink); }; diff --git a/libraries/pack200/src/constants.h b/libraries/pack200/src/constants.h index 2cc14b7d..f1baf42a 100644 --- a/libraries/pack200/src/constants.h +++ b/libraries/pack200/src/constants.h @@ -55,388 +55,388 @@ enum { - CONSTANT_None, - CONSTANT_Utf8, - CONSTANT_unused2, /* unused, was Unicode */ - CONSTANT_Integer, - CONSTANT_Float, - CONSTANT_Long, - CONSTANT_Double, - CONSTANT_Class, - CONSTANT_String, - CONSTANT_Fieldref, - CONSTANT_Methodref, - CONSTANT_InterfaceMethodref, - CONSTANT_NameandType, - CONSTANT_Signature = 13, - CONSTANT_All = 14, - CONSTANT_Limit = 15, - CONSTANT_NONE = 0, - CONSTANT_Literal = 20, // pseudo-tag for debugging - CONSTANT_Member = 21, // pseudo-tag for debugging - SUBINDEX_BIT = 64, // combined with CONSTANT_xxx for ixTag - ACC_STATIC = 0x0008, - ACC_IC_LONG_FORM = (1 << 16), // for ic_flags - CLASS_ATTR_SourceFile = 17, - CLASS_ATTR_EnclosingMethod = 18, - CLASS_ATTR_InnerClasses = 23, - CLASS_ATTR_ClassFile_version = 24, - FIELD_ATTR_ConstantValue = 17, - METHOD_ATTR_Code = 17, - METHOD_ATTR_Exceptions = 18, - METHOD_ATTR_RuntimeVisibleParameterAnnotations = 23, - METHOD_ATTR_RuntimeInvisibleParameterAnnotations = 24, - METHOD_ATTR_AnnotationDefault = 25, - CODE_ATTR_StackMapTable = 0, - CODE_ATTR_LineNumberTable = 1, - CODE_ATTR_LocalVariableTable = 2, - CODE_ATTR_LocalVariableTypeTable = 3, - // X_ATTR_Synthetic = 12, // ACC_SYNTHETIC; not predefined - X_ATTR_Signature = 19, - X_ATTR_Deprecated = 20, - X_ATTR_RuntimeVisibleAnnotations = 21, - X_ATTR_RuntimeInvisibleAnnotations = 22, - X_ATTR_OVERFLOW = 16, - X_ATTR_LIMIT_NO_FLAGS_HI = 32, - X_ATTR_LIMIT_FLAGS_HI = 63, + CONSTANT_None, + CONSTANT_Utf8, + CONSTANT_unused2, /* unused, was Unicode */ + CONSTANT_Integer, + CONSTANT_Float, + CONSTANT_Long, + CONSTANT_Double, + CONSTANT_Class, + CONSTANT_String, + CONSTANT_Fieldref, + CONSTANT_Methodref, + CONSTANT_InterfaceMethodref, + CONSTANT_NameandType, + CONSTANT_Signature = 13, + CONSTANT_All = 14, + CONSTANT_Limit = 15, + CONSTANT_NONE = 0, + CONSTANT_Literal = 20, // pseudo-tag for debugging + CONSTANT_Member = 21, // pseudo-tag for debugging + SUBINDEX_BIT = 64, // combined with CONSTANT_xxx for ixTag + ACC_STATIC = 0x0008, + ACC_IC_LONG_FORM = (1 << 16), // for ic_flags + CLASS_ATTR_SourceFile = 17, + CLASS_ATTR_EnclosingMethod = 18, + CLASS_ATTR_InnerClasses = 23, + CLASS_ATTR_ClassFile_version = 24, + FIELD_ATTR_ConstantValue = 17, + METHOD_ATTR_Code = 17, + METHOD_ATTR_Exceptions = 18, + METHOD_ATTR_RuntimeVisibleParameterAnnotations = 23, + METHOD_ATTR_RuntimeInvisibleParameterAnnotations = 24, + METHOD_ATTR_AnnotationDefault = 25, + CODE_ATTR_StackMapTable = 0, + CODE_ATTR_LineNumberTable = 1, + CODE_ATTR_LocalVariableTable = 2, + CODE_ATTR_LocalVariableTypeTable = 3, + // X_ATTR_Synthetic = 12, // ACC_SYNTHETIC; not predefined + X_ATTR_Signature = 19, + X_ATTR_Deprecated = 20, + X_ATTR_RuntimeVisibleAnnotations = 21, + X_ATTR_RuntimeInvisibleAnnotations = 22, + X_ATTR_OVERFLOW = 16, + X_ATTR_LIMIT_NO_FLAGS_HI = 32, + X_ATTR_LIMIT_FLAGS_HI = 63, #define O_ATTR_DO(F) \ - F(X_ATTR_OVERFLOW, 01) \ - /*(end)*/ + F(X_ATTR_OVERFLOW, 01) \ + /*(end)*/ #define X_ATTR_DO(F) \ - O_ATTR_DO(F) F(X_ATTR_Signature, Signature) F(X_ATTR_Deprecated, Deprecated) \ - F(X_ATTR_RuntimeVisibleAnnotations, RuntimeVisibleAnnotations) \ - F(X_ATTR_RuntimeInvisibleAnnotations, RuntimeInvisibleAnnotations) \ - /*F(X_ATTR_Synthetic,Synthetic)*/ \ - /*(end)*/ + O_ATTR_DO(F) F(X_ATTR_Signature, Signature) F(X_ATTR_Deprecated, Deprecated) \ + F(X_ATTR_RuntimeVisibleAnnotations, RuntimeVisibleAnnotations) \ + F(X_ATTR_RuntimeInvisibleAnnotations, RuntimeInvisibleAnnotations) \ + /*F(X_ATTR_Synthetic,Synthetic)*/ \ + /*(end)*/ #define CLASS_ATTR_DO(F) \ - F(CLASS_ATTR_SourceFile, SourceFile) F(CLASS_ATTR_InnerClasses, InnerClasses) \ - F(CLASS_ATTR_EnclosingMethod, EnclosingMethod) F(CLASS_ATTR_ClassFile_version, 02) \ - /*(end)*/ + F(CLASS_ATTR_SourceFile, SourceFile) F(CLASS_ATTR_InnerClasses, InnerClasses) \ + F(CLASS_ATTR_EnclosingMethod, EnclosingMethod) F(CLASS_ATTR_ClassFile_version, 02) \ + /*(end)*/ #define FIELD_ATTR_DO(F) \ - F(FIELD_ATTR_ConstantValue, ConstantValue) \ - /*(end)*/ + F(FIELD_ATTR_ConstantValue, ConstantValue) \ + /*(end)*/ #define METHOD_ATTR_DO(F) \ - F(METHOD_ATTR_Code, Code) F(METHOD_ATTR_Exceptions, Exceptions) \ - F(METHOD_ATTR_RuntimeVisibleParameterAnnotations, RuntimeVisibleParameterAnnotations) \ - F(METHOD_ATTR_RuntimeInvisibleParameterAnnotations, \ - RuntimeInvisibleParameterAnnotations) \ - F(METHOD_ATTR_AnnotationDefault, AnnotationDefault) \ - /*(end)*/ + F(METHOD_ATTR_Code, Code) F(METHOD_ATTR_Exceptions, Exceptions) \ + F(METHOD_ATTR_RuntimeVisibleParameterAnnotations, RuntimeVisibleParameterAnnotations) \ + F(METHOD_ATTR_RuntimeInvisibleParameterAnnotations, \ + RuntimeInvisibleParameterAnnotations) \ + F(METHOD_ATTR_AnnotationDefault, AnnotationDefault) \ + /*(end)*/ #define CODE_ATTR_DO(F) \ - F(CODE_ATTR_StackMapTable, StackMapTable) F(CODE_ATTR_LineNumberTable, LineNumberTable) \ - F(CODE_ATTR_LocalVariableTable, LocalVariableTable) \ - F(CODE_ATTR_LocalVariableTypeTable, LocalVariableTypeTable) \ - /*(end)*/ + F(CODE_ATTR_StackMapTable, StackMapTable) F(CODE_ATTR_LineNumberTable, LineNumberTable) \ + F(CODE_ATTR_LocalVariableTable, LocalVariableTable) \ + F(CODE_ATTR_LocalVariableTypeTable, LocalVariableTypeTable) \ + /*(end)*/ #define ALL_ATTR_DO(F) \ - X_ATTR_DO(F) CLASS_ATTR_DO(F) FIELD_ATTR_DO(F) METHOD_ATTR_DO(F) CODE_ATTR_DO(F) \ - /*(end)*/ + X_ATTR_DO(F) CLASS_ATTR_DO(F) FIELD_ATTR_DO(F) METHOD_ATTR_DO(F) CODE_ATTR_DO(F) \ + /*(end)*/ - // attribute "context types" - ATTR_CONTEXT_CLASS = 0, - ATTR_CONTEXT_FIELD = 1, - ATTR_CONTEXT_METHOD = 2, - ATTR_CONTEXT_CODE = 3, - ATTR_CONTEXT_LIMIT = 4, + // attribute "context types" + ATTR_CONTEXT_CLASS = 0, + ATTR_CONTEXT_FIELD = 1, + ATTR_CONTEXT_METHOD = 2, + ATTR_CONTEXT_CODE = 3, + ATTR_CONTEXT_LIMIT = 4, - // constants for parsed layouts (stored in band::le_kind) - EK_NONE = 0, // not a layout element - EK_INT = 'I', // B H I SH etc., also FH etc. - EK_BCI = 'P', // PH etc. - EK_BCID = 'Q', // POH etc. - EK_BCO = 'O', // OH etc. - EK_REPL = 'N', // NH[...] etc. - EK_REF = 'R', // RUH, RUNH, KQH, etc. - EK_UN = 'T', // TB(...)[...] etc. - EK_CASE = 'K', // (...)[...] etc. - EK_CALL = '(', // (0), (1), etc. - EK_CBLE = '[', // [...][...] etc. - NO_BAND_INDEX = -1, + // constants for parsed layouts (stored in band::le_kind) + EK_NONE = 0, // not a layout element + EK_INT = 'I', // B H I SH etc., also FH etc. + EK_BCI = 'P', // PH etc. + EK_BCID = 'Q', // POH etc. + EK_BCO = 'O', // OH etc. + EK_REPL = 'N', // NH[...] etc. + EK_REF = 'R', // RUH, RUNH, KQH, etc. + EK_UN = 'T', // TB(...)[...] etc. + EK_CASE = 'K', // (...)[...] etc. + EK_CALL = '(', // (0), (1), etc. + EK_CBLE = '[', // [...][...] etc. + NO_BAND_INDEX = -1, - // File option bits, from LSB in ascending bit position. - FO_DEFLATE_HINT = 1 << 0, - FO_IS_CLASS_STUB = 1 << 1, + // File option bits, from LSB in ascending bit position. + FO_DEFLATE_HINT = 1 << 0, + FO_IS_CLASS_STUB = 1 << 1, - // Archive option bits, from LSB in ascending bit position: - AO_HAVE_SPECIAL_FORMATS = 1 << 0, - AO_HAVE_CP_NUMBERS = 1 << 1, - AO_HAVE_ALL_CODE_FLAGS = 1 << 2, - AO_3_UNUSED_MBZ = 1 << 3, - AO_HAVE_FILE_HEADERS = 1 << 4, - AO_DEFLATE_HINT = 1 << 5, - AO_HAVE_FILE_MODTIME = 1 << 6, - AO_HAVE_FILE_OPTIONS = 1 << 7, - AO_HAVE_FILE_SIZE_HI = 1 << 8, - AO_HAVE_CLASS_FLAGS_HI = 1 << 9, - AO_HAVE_FIELD_FLAGS_HI = 1 << 10, - AO_HAVE_METHOD_FLAGS_HI = 1 << 11, - AO_HAVE_CODE_FLAGS_HI = 1 << 12, + // Archive option bits, from LSB in ascending bit position: + AO_HAVE_SPECIAL_FORMATS = 1 << 0, + AO_HAVE_CP_NUMBERS = 1 << 1, + AO_HAVE_ALL_CODE_FLAGS = 1 << 2, + AO_3_UNUSED_MBZ = 1 << 3, + AO_HAVE_FILE_HEADERS = 1 << 4, + AO_DEFLATE_HINT = 1 << 5, + AO_HAVE_FILE_MODTIME = 1 << 6, + AO_HAVE_FILE_OPTIONS = 1 << 7, + AO_HAVE_FILE_SIZE_HI = 1 << 8, + AO_HAVE_CLASS_FLAGS_HI = 1 << 9, + AO_HAVE_FIELD_FLAGS_HI = 1 << 10, + AO_HAVE_METHOD_FLAGS_HI = 1 << 11, + AO_HAVE_CODE_FLAGS_HI = 1 << 12, #define ARCHIVE_BIT_DO(F) \ - F(AO_HAVE_SPECIAL_FORMATS) F(AO_HAVE_CP_NUMBERS) F(AO_HAVE_ALL_CODE_FLAGS) \ - /*F(AO_3_UNUSED_MBZ)*/ \ - F(AO_HAVE_FILE_HEADERS) F(AO_DEFLATE_HINT) F(AO_HAVE_FILE_MODTIME) \ - F(AO_HAVE_FILE_OPTIONS) F(AO_HAVE_FILE_SIZE_HI) F(AO_HAVE_CLASS_FLAGS_HI) \ - F(AO_HAVE_FIELD_FLAGS_HI) F(AO_HAVE_METHOD_FLAGS_HI) F(AO_HAVE_CODE_FLAGS_HI) \ - /*(end)*/ + F(AO_HAVE_SPECIAL_FORMATS) F(AO_HAVE_CP_NUMBERS) F(AO_HAVE_ALL_CODE_FLAGS) \ + /*F(AO_3_UNUSED_MBZ)*/ \ + F(AO_HAVE_FILE_HEADERS) F(AO_DEFLATE_HINT) F(AO_HAVE_FILE_MODTIME) \ + F(AO_HAVE_FILE_OPTIONS) F(AO_HAVE_FILE_SIZE_HI) F(AO_HAVE_CLASS_FLAGS_HI) \ + F(AO_HAVE_FIELD_FLAGS_HI) F(AO_HAVE_METHOD_FLAGS_HI) F(AO_HAVE_CODE_FLAGS_HI) \ + /*(end)*/ - // Constants for decoding attribute definition header bytes. - ADH_CONTEXT_MASK = 0x3, // (hdr & ADH_CONTEXT_MASK) - ADH_BIT_SHIFT = 0x2, // (hdr >> ADH_BIT_SHIFT) - ADH_BIT_IS_LSB = 1, // (hdr >> ADH_BIT_SHIFT) - ADH_BIT_IS_LSB + // Constants for decoding attribute definition header bytes. + ADH_CONTEXT_MASK = 0x3, // (hdr & ADH_CONTEXT_MASK) + ADH_BIT_SHIFT = 0x2, // (hdr >> ADH_BIT_SHIFT) + ADH_BIT_IS_LSB = 1, // (hdr >> ADH_BIT_SHIFT) - ADH_BIT_IS_LSB #define ADH_BYTE(context, index) ((((index) + ADH_BIT_IS_LSB) << ADH_BIT_SHIFT) + (context)) #define ADH_BYTE_CONTEXT(adhb) ((adhb) & ADH_CONTEXT_MASK) #define ADH_BYTE_INDEX(adhb) (((adhb) >> ADH_BIT_SHIFT) - ADH_BIT_IS_LSB) - NO_MODTIME = 0, // nullptr modtime value + NO_MODTIME = 0, // nullptr modtime value - // meta-coding - _meta_default = 0, - _meta_canon_min = 1, - _meta_canon_max = 115, - _meta_arb = 116, - _meta_run = 117, - _meta_pop = 141, - _meta_limit = 189, - _meta_error = 255, - _xxx_1_end + // meta-coding + _meta_default = 0, + _meta_canon_min = 1, + _meta_canon_max = 115, + _meta_arb = 116, + _meta_run = 117, + _meta_pop = 141, + _meta_limit = 189, + _meta_error = 255, + _xxx_1_end }; // Bytecodes. enum { - bc_nop = 0, // 0x00 - bc_aconst_null = 1, // 0x01 - bc_iconst_m1 = 2, // 0x02 - bc_iconst_0 = 3, // 0x03 - bc_iconst_1 = 4, // 0x04 - bc_iconst_2 = 5, // 0x05 - bc_iconst_3 = 6, // 0x06 - bc_iconst_4 = 7, // 0x07 - bc_iconst_5 = 8, // 0x08 - bc_lconst_0 = 9, // 0x09 - bc_lconst_1 = 10, // 0x0a - bc_fconst_0 = 11, // 0x0b - bc_fconst_1 = 12, // 0x0c - bc_fconst_2 = 13, // 0x0d - bc_dconst_0 = 14, // 0x0e - bc_dconst_1 = 15, // 0x0f - bc_bipush = 16, // 0x10 - bc_sipush = 17, // 0x11 - bc_ldc = 18, // 0x12 - bc_ldc_w = 19, // 0x13 - bc_ldc2_w = 20, // 0x14 - bc_iload = 21, // 0x15 - bc_lload = 22, // 0x16 - bc_fload = 23, // 0x17 - bc_dload = 24, // 0x18 - bc_aload = 25, // 0x19 - bc_iload_0 = 26, // 0x1a - bc_iload_1 = 27, // 0x1b - bc_iload_2 = 28, // 0x1c - bc_iload_3 = 29, // 0x1d - bc_lload_0 = 30, // 0x1e - bc_lload_1 = 31, // 0x1f - bc_lload_2 = 32, // 0x20 - bc_lload_3 = 33, // 0x21 - bc_fload_0 = 34, // 0x22 - bc_fload_1 = 35, // 0x23 - bc_fload_2 = 36, // 0x24 - bc_fload_3 = 37, // 0x25 - bc_dload_0 = 38, // 0x26 - bc_dload_1 = 39, // 0x27 - bc_dload_2 = 40, // 0x28 - bc_dload_3 = 41, // 0x29 - bc_aload_0 = 42, // 0x2a - bc_aload_1 = 43, // 0x2b - bc_aload_2 = 44, // 0x2c - bc_aload_3 = 45, // 0x2d - bc_iaload = 46, // 0x2e - bc_laload = 47, // 0x2f - bc_faload = 48, // 0x30 - bc_daload = 49, // 0x31 - bc_aaload = 50, // 0x32 - bc_baload = 51, // 0x33 - bc_caload = 52, // 0x34 - bc_saload = 53, // 0x35 - bc_istore = 54, // 0x36 - bc_lstore = 55, // 0x37 - bc_fstore = 56, // 0x38 - bc_dstore = 57, // 0x39 - bc_astore = 58, // 0x3a - bc_istore_0 = 59, // 0x3b - bc_istore_1 = 60, // 0x3c - bc_istore_2 = 61, // 0x3d - bc_istore_3 = 62, // 0x3e - bc_lstore_0 = 63, // 0x3f - bc_lstore_1 = 64, // 0x40 - bc_lstore_2 = 65, // 0x41 - bc_lstore_3 = 66, // 0x42 - bc_fstore_0 = 67, // 0x43 - bc_fstore_1 = 68, // 0x44 - bc_fstore_2 = 69, // 0x45 - bc_fstore_3 = 70, // 0x46 - bc_dstore_0 = 71, // 0x47 - bc_dstore_1 = 72, // 0x48 - bc_dstore_2 = 73, // 0x49 - bc_dstore_3 = 74, // 0x4a - bc_astore_0 = 75, // 0x4b - bc_astore_1 = 76, // 0x4c - bc_astore_2 = 77, // 0x4d - bc_astore_3 = 78, // 0x4e - bc_iastore = 79, // 0x4f - bc_lastore = 80, // 0x50 - bc_fastore = 81, // 0x51 - bc_dastore = 82, // 0x52 - bc_aastore = 83, // 0x53 - bc_bastore = 84, // 0x54 - bc_castore = 85, // 0x55 - bc_sastore = 86, // 0x56 - bc_pop = 87, // 0x57 - bc_pop2 = 88, // 0x58 - bc_dup = 89, // 0x59 - bc_dup_x1 = 90, // 0x5a - bc_dup_x2 = 91, // 0x5b - bc_dup2 = 92, // 0x5c - bc_dup2_x1 = 93, // 0x5d - bc_dup2_x2 = 94, // 0x5e - bc_swap = 95, // 0x5f - bc_iadd = 96, // 0x60 - bc_ladd = 97, // 0x61 - bc_fadd = 98, // 0x62 - bc_dadd = 99, // 0x63 - bc_isub = 100, // 0x64 - bc_lsub = 101, // 0x65 - bc_fsub = 102, // 0x66 - bc_dsub = 103, // 0x67 - bc_imul = 104, // 0x68 - bc_lmul = 105, // 0x69 - bc_fmul = 106, // 0x6a - bc_dmul = 107, // 0x6b - bc_idiv = 108, // 0x6c - bc_ldiv = 109, // 0x6d - bc_fdiv = 110, // 0x6e - bc_ddiv = 111, // 0x6f - bc_irem = 112, // 0x70 - bc_lrem = 113, // 0x71 - bc_frem = 114, // 0x72 - bc_drem = 115, // 0x73 - bc_ineg = 116, // 0x74 - bc_lneg = 117, // 0x75 - bc_fneg = 118, // 0x76 - bc_dneg = 119, // 0x77 - bc_ishl = 120, // 0x78 - bc_lshl = 121, // 0x79 - bc_ishr = 122, // 0x7a - bc_lshr = 123, // 0x7b - bc_iushr = 124, // 0x7c - bc_lushr = 125, // 0x7d - bc_iand = 126, // 0x7e - bc_land = 127, // 0x7f - bc_ior = 128, // 0x80 - bc_lor = 129, // 0x81 - bc_ixor = 130, // 0x82 - bc_lxor = 131, // 0x83 - bc_iinc = 132, // 0x84 - bc_i2l = 133, // 0x85 - bc_i2f = 134, // 0x86 - bc_i2d = 135, // 0x87 - bc_l2i = 136, // 0x88 - bc_l2f = 137, // 0x89 - bc_l2d = 138, // 0x8a - bc_f2i = 139, // 0x8b - bc_f2l = 140, // 0x8c - bc_f2d = 141, // 0x8d - bc_d2i = 142, // 0x8e - bc_d2l = 143, // 0x8f - bc_d2f = 144, // 0x90 - bc_i2b = 145, // 0x91 - bc_i2c = 146, // 0x92 - bc_i2s = 147, // 0x93 - bc_lcmp = 148, // 0x94 - bc_fcmpl = 149, // 0x95 - bc_fcmpg = 150, // 0x96 - bc_dcmpl = 151, // 0x97 - bc_dcmpg = 152, // 0x98 - bc_ifeq = 153, // 0x99 - bc_ifne = 154, // 0x9a - bc_iflt = 155, // 0x9b - bc_ifge = 156, // 0x9c - bc_ifgt = 157, // 0x9d - bc_ifle = 158, // 0x9e - bc_if_icmpeq = 159, // 0x9f - bc_if_icmpne = 160, // 0xa0 - bc_if_icmplt = 161, // 0xa1 - bc_if_icmpge = 162, // 0xa2 - bc_if_icmpgt = 163, // 0xa3 - bc_if_icmple = 164, // 0xa4 - bc_if_acmpeq = 165, // 0xa5 - bc_if_acmpne = 166, // 0xa6 - bc_goto = 167, // 0xa7 - bc_jsr = 168, // 0xa8 - bc_ret = 169, // 0xa9 - bc_tableswitch = 170, // 0xaa - bc_lookupswitch = 171, // 0xab - bc_ireturn = 172, // 0xac - bc_lreturn = 173, // 0xad - bc_freturn = 174, // 0xae - bc_dreturn = 175, // 0xaf - bc_areturn = 176, // 0xb0 - bc_return = 177, // 0xb1 - bc_getstatic = 178, // 0xb2 - bc_putstatic = 179, // 0xb3 - bc_getfield = 180, // 0xb4 - bc_putfield = 181, // 0xb5 - bc_invokevirtual = 182, // 0xb6 - bc_invokespecial = 183, // 0xb7 - bc_invokestatic = 184, // 0xb8 - bc_invokeinterface = 185, // 0xb9 - bc_xxxunusedxxx = 186, // 0xba - bc_new = 187, // 0xbb - bc_newarray = 188, // 0xbc - bc_anewarray = 189, // 0xbd - bc_arraylength = 190, // 0xbe - bc_athrow = 191, // 0xbf - bc_checkcast = 192, // 0xc0 - bc_instanceof = 193, // 0xc1 - bc_monitorenter = 194, // 0xc2 - bc_monitorexit = 195, // 0xc3 - bc_wide = 196, // 0xc4 - bc_multianewarray = 197, // 0xc5 - bc_ifnull = 198, // 0xc6 - bc_ifnonnull = 199, // 0xc7 - bc_goto_w = 200, // 0xc8 - bc_jsr_w = 201, // 0xc9 - bc_bytecode_limit = 202 // 0xca + bc_nop = 0, // 0x00 + bc_aconst_null = 1, // 0x01 + bc_iconst_m1 = 2, // 0x02 + bc_iconst_0 = 3, // 0x03 + bc_iconst_1 = 4, // 0x04 + bc_iconst_2 = 5, // 0x05 + bc_iconst_3 = 6, // 0x06 + bc_iconst_4 = 7, // 0x07 + bc_iconst_5 = 8, // 0x08 + bc_lconst_0 = 9, // 0x09 + bc_lconst_1 = 10, // 0x0a + bc_fconst_0 = 11, // 0x0b + bc_fconst_1 = 12, // 0x0c + bc_fconst_2 = 13, // 0x0d + bc_dconst_0 = 14, // 0x0e + bc_dconst_1 = 15, // 0x0f + bc_bipush = 16, // 0x10 + bc_sipush = 17, // 0x11 + bc_ldc = 18, // 0x12 + bc_ldc_w = 19, // 0x13 + bc_ldc2_w = 20, // 0x14 + bc_iload = 21, // 0x15 + bc_lload = 22, // 0x16 + bc_fload = 23, // 0x17 + bc_dload = 24, // 0x18 + bc_aload = 25, // 0x19 + bc_iload_0 = 26, // 0x1a + bc_iload_1 = 27, // 0x1b + bc_iload_2 = 28, // 0x1c + bc_iload_3 = 29, // 0x1d + bc_lload_0 = 30, // 0x1e + bc_lload_1 = 31, // 0x1f + bc_lload_2 = 32, // 0x20 + bc_lload_3 = 33, // 0x21 + bc_fload_0 = 34, // 0x22 + bc_fload_1 = 35, // 0x23 + bc_fload_2 = 36, // 0x24 + bc_fload_3 = 37, // 0x25 + bc_dload_0 = 38, // 0x26 + bc_dload_1 = 39, // 0x27 + bc_dload_2 = 40, // 0x28 + bc_dload_3 = 41, // 0x29 + bc_aload_0 = 42, // 0x2a + bc_aload_1 = 43, // 0x2b + bc_aload_2 = 44, // 0x2c + bc_aload_3 = 45, // 0x2d + bc_iaload = 46, // 0x2e + bc_laload = 47, // 0x2f + bc_faload = 48, // 0x30 + bc_daload = 49, // 0x31 + bc_aaload = 50, // 0x32 + bc_baload = 51, // 0x33 + bc_caload = 52, // 0x34 + bc_saload = 53, // 0x35 + bc_istore = 54, // 0x36 + bc_lstore = 55, // 0x37 + bc_fstore = 56, // 0x38 + bc_dstore = 57, // 0x39 + bc_astore = 58, // 0x3a + bc_istore_0 = 59, // 0x3b + bc_istore_1 = 60, // 0x3c + bc_istore_2 = 61, // 0x3d + bc_istore_3 = 62, // 0x3e + bc_lstore_0 = 63, // 0x3f + bc_lstore_1 = 64, // 0x40 + bc_lstore_2 = 65, // 0x41 + bc_lstore_3 = 66, // 0x42 + bc_fstore_0 = 67, // 0x43 + bc_fstore_1 = 68, // 0x44 + bc_fstore_2 = 69, // 0x45 + bc_fstore_3 = 70, // 0x46 + bc_dstore_0 = 71, // 0x47 + bc_dstore_1 = 72, // 0x48 + bc_dstore_2 = 73, // 0x49 + bc_dstore_3 = 74, // 0x4a + bc_astore_0 = 75, // 0x4b + bc_astore_1 = 76, // 0x4c + bc_astore_2 = 77, // 0x4d + bc_astore_3 = 78, // 0x4e + bc_iastore = 79, // 0x4f + bc_lastore = 80, // 0x50 + bc_fastore = 81, // 0x51 + bc_dastore = 82, // 0x52 + bc_aastore = 83, // 0x53 + bc_bastore = 84, // 0x54 + bc_castore = 85, // 0x55 + bc_sastore = 86, // 0x56 + bc_pop = 87, // 0x57 + bc_pop2 = 88, // 0x58 + bc_dup = 89, // 0x59 + bc_dup_x1 = 90, // 0x5a + bc_dup_x2 = 91, // 0x5b + bc_dup2 = 92, // 0x5c + bc_dup2_x1 = 93, // 0x5d + bc_dup2_x2 = 94, // 0x5e + bc_swap = 95, // 0x5f + bc_iadd = 96, // 0x60 + bc_ladd = 97, // 0x61 + bc_fadd = 98, // 0x62 + bc_dadd = 99, // 0x63 + bc_isub = 100, // 0x64 + bc_lsub = 101, // 0x65 + bc_fsub = 102, // 0x66 + bc_dsub = 103, // 0x67 + bc_imul = 104, // 0x68 + bc_lmul = 105, // 0x69 + bc_fmul = 106, // 0x6a + bc_dmul = 107, // 0x6b + bc_idiv = 108, // 0x6c + bc_ldiv = 109, // 0x6d + bc_fdiv = 110, // 0x6e + bc_ddiv = 111, // 0x6f + bc_irem = 112, // 0x70 + bc_lrem = 113, // 0x71 + bc_frem = 114, // 0x72 + bc_drem = 115, // 0x73 + bc_ineg = 116, // 0x74 + bc_lneg = 117, // 0x75 + bc_fneg = 118, // 0x76 + bc_dneg = 119, // 0x77 + bc_ishl = 120, // 0x78 + bc_lshl = 121, // 0x79 + bc_ishr = 122, // 0x7a + bc_lshr = 123, // 0x7b + bc_iushr = 124, // 0x7c + bc_lushr = 125, // 0x7d + bc_iand = 126, // 0x7e + bc_land = 127, // 0x7f + bc_ior = 128, // 0x80 + bc_lor = 129, // 0x81 + bc_ixor = 130, // 0x82 + bc_lxor = 131, // 0x83 + bc_iinc = 132, // 0x84 + bc_i2l = 133, // 0x85 + bc_i2f = 134, // 0x86 + bc_i2d = 135, // 0x87 + bc_l2i = 136, // 0x88 + bc_l2f = 137, // 0x89 + bc_l2d = 138, // 0x8a + bc_f2i = 139, // 0x8b + bc_f2l = 140, // 0x8c + bc_f2d = 141, // 0x8d + bc_d2i = 142, // 0x8e + bc_d2l = 143, // 0x8f + bc_d2f = 144, // 0x90 + bc_i2b = 145, // 0x91 + bc_i2c = 146, // 0x92 + bc_i2s = 147, // 0x93 + bc_lcmp = 148, // 0x94 + bc_fcmpl = 149, // 0x95 + bc_fcmpg = 150, // 0x96 + bc_dcmpl = 151, // 0x97 + bc_dcmpg = 152, // 0x98 + bc_ifeq = 153, // 0x99 + bc_ifne = 154, // 0x9a + bc_iflt = 155, // 0x9b + bc_ifge = 156, // 0x9c + bc_ifgt = 157, // 0x9d + bc_ifle = 158, // 0x9e + bc_if_icmpeq = 159, // 0x9f + bc_if_icmpne = 160, // 0xa0 + bc_if_icmplt = 161, // 0xa1 + bc_if_icmpge = 162, // 0xa2 + bc_if_icmpgt = 163, // 0xa3 + bc_if_icmple = 164, // 0xa4 + bc_if_acmpeq = 165, // 0xa5 + bc_if_acmpne = 166, // 0xa6 + bc_goto = 167, // 0xa7 + bc_jsr = 168, // 0xa8 + bc_ret = 169, // 0xa9 + bc_tableswitch = 170, // 0xaa + bc_lookupswitch = 171, // 0xab + bc_ireturn = 172, // 0xac + bc_lreturn = 173, // 0xad + bc_freturn = 174, // 0xae + bc_dreturn = 175, // 0xaf + bc_areturn = 176, // 0xb0 + bc_return = 177, // 0xb1 + bc_getstatic = 178, // 0xb2 + bc_putstatic = 179, // 0xb3 + bc_getfield = 180, // 0xb4 + bc_putfield = 181, // 0xb5 + bc_invokevirtual = 182, // 0xb6 + bc_invokespecial = 183, // 0xb7 + bc_invokestatic = 184, // 0xb8 + bc_invokeinterface = 185, // 0xb9 + bc_xxxunusedxxx = 186, // 0xba + bc_new = 187, // 0xbb + bc_newarray = 188, // 0xbc + bc_anewarray = 189, // 0xbd + bc_arraylength = 190, // 0xbe + bc_athrow = 191, // 0xbf + bc_checkcast = 192, // 0xc0 + bc_instanceof = 193, // 0xc1 + bc_monitorenter = 194, // 0xc2 + bc_monitorexit = 195, // 0xc3 + bc_wide = 196, // 0xc4 + bc_multianewarray = 197, // 0xc5 + bc_ifnull = 198, // 0xc6 + bc_ifnonnull = 199, // 0xc7 + bc_goto_w = 200, // 0xc8 + bc_jsr_w = 201, // 0xc9 + bc_bytecode_limit = 202 // 0xca }; enum { - bc_end_marker = 255, - bc_byte_escape = 254, - bc_ref_escape = 253, - _first_linker_op = bc_getstatic, - _last_linker_op = bc_invokestatic, - _num_linker_ops = (_last_linker_op - _first_linker_op) + 1, - _self_linker_op = bc_bytecode_limit, - _self_linker_aload_flag = 1 * _num_linker_ops, - _self_linker_super_flag = 2 * _num_linker_ops, - _self_linker_limit = _self_linker_op + 4 * _num_linker_ops, - _invokeinit_op = _self_linker_limit, - _invokeinit_self_option = 0, - _invokeinit_super_option = 1, - _invokeinit_new_option = 2, - _invokeinit_limit = _invokeinit_op + 3, - _xldc_op = _invokeinit_limit, - bc_aldc = bc_ldc, - bc_cldc = _xldc_op + 0, - bc_ildc = _xldc_op + 1, - bc_fldc = _xldc_op + 2, - bc_aldc_w = bc_ldc_w, - bc_cldc_w = _xldc_op + 3, - bc_ildc_w = _xldc_op + 4, - bc_fldc_w = _xldc_op + 5, - bc_lldc2_w = bc_ldc2_w, - bc_dldc2_w = _xldc_op + 6, - _xldc_limit = _xldc_op + 7, - _xxx_3_end + bc_end_marker = 255, + bc_byte_escape = 254, + bc_ref_escape = 253, + _first_linker_op = bc_getstatic, + _last_linker_op = bc_invokestatic, + _num_linker_ops = (_last_linker_op - _first_linker_op) + 1, + _self_linker_op = bc_bytecode_limit, + _self_linker_aload_flag = 1 * _num_linker_ops, + _self_linker_super_flag = 2 * _num_linker_ops, + _self_linker_limit = _self_linker_op + 4 * _num_linker_ops, + _invokeinit_op = _self_linker_limit, + _invokeinit_self_option = 0, + _invokeinit_super_option = 1, + _invokeinit_new_option = 2, + _invokeinit_limit = _invokeinit_op + 3, + _xldc_op = _invokeinit_limit, + bc_aldc = bc_ldc, + bc_cldc = _xldc_op + 0, + bc_ildc = _xldc_op + 1, + bc_fldc = _xldc_op + 2, + bc_aldc_w = bc_ldc_w, + bc_cldc_w = _xldc_op + 3, + bc_ildc_w = _xldc_op + 4, + bc_fldc_w = _xldc_op + 5, + bc_lldc2_w = bc_ldc2_w, + bc_dldc2_w = _xldc_op + 6, + _xldc_limit = _xldc_op + 7, + _xxx_3_end }; diff --git a/libraries/pack200/src/unpack.cpp b/libraries/pack200/src/unpack.cpp index 55d253b2..9c4c633c 100644 --- a/libraries/pack200/src/unpack.cpp +++ b/libraries/pack200/src/unpack.cpp @@ -34,13 +34,7 @@ * _LP64 can be explicitly set (used on Linux). * Solaris compilers will define __sparcv9 or __x86_64 on 64bit compilations. */ -#if defined(_LP64) || defined(__sparcv9) || defined(__x86_64) -#define LONG_LONG_FORMAT "%ld" -#define LONG_LONG_HEX_FORMAT "%lx" -#else -#define LONG_LONG_FORMAT "%lld" -#define LONG_LONG_HEX_FORMAT "%016llx" -#endif +#include <cinttypes> #include <sys/types.h> @@ -67,248 +61,248 @@ // tags, in canonical order: static const byte TAGS_IN_ORDER[] = { - CONSTANT_Utf8, CONSTANT_Integer, CONSTANT_Float, CONSTANT_Long, - CONSTANT_Double, CONSTANT_String, CONSTANT_Class, CONSTANT_Signature, - CONSTANT_NameandType, CONSTANT_Fieldref, CONSTANT_Methodref, CONSTANT_InterfaceMethodref}; + CONSTANT_Utf8, CONSTANT_Integer, CONSTANT_Float, CONSTANT_Long, + CONSTANT_Double, CONSTANT_String, CONSTANT_Class, CONSTANT_Signature, + CONSTANT_NameandType, CONSTANT_Fieldref, CONSTANT_Methodref, CONSTANT_InterfaceMethodref}; #define N_TAGS_IN_ORDER (sizeof TAGS_IN_ORDER) // REQUESTED must be -2 for u2 and REQUESTED_LDC must be -1 for u1 enum { - NOT_REQUESTED = 0, - REQUESTED = -2, - REQUESTED_LDC = -1 + NOT_REQUESTED = 0, + REQUESTED = -2, + REQUESTED_LDC = -1 }; #define NO_INORD ((uint32_t) - 1) struct entry { - byte tag; - unsigned short nrefs; // pack w/ tag - - int outputIndex; - uint32_t inord; // &cp.entries[cp.tag_base[this->tag]+this->inord] == this - - entry **refs; - - // put last to pack best - union - { - bytes b; - int i; - int64_t l; - } value; - - void requestOutputIndex(constant_pool &cp, int req = REQUESTED); - int getOutputIndex() - { - assert(outputIndex > NOT_REQUESTED); - return outputIndex; - } - - entry *ref(int refnum) - { - assert((uint32_t)refnum < nrefs); - return refs[refnum]; - } - - const char *utf8String() - { - assert(tagMatches(CONSTANT_Utf8)); - assert(value.b.len == strlen((const char *)value.b.ptr)); - return (const char *)value.b.ptr; - } - - entry *className() - { - assert(tagMatches(CONSTANT_Class)); - return ref(0); - } - - entry *memberClass() - { - assert(tagMatches(CONSTANT_Member)); - return ref(0); - } - - entry *memberDescr() - { - assert(tagMatches(CONSTANT_Member)); - return ref(1); - } - - entry *descrName() - { - assert(tagMatches(CONSTANT_NameandType)); - return ref(0); - } - - entry *descrType() - { - assert(tagMatches(CONSTANT_NameandType)); - return ref(1); - } - - int typeSize(); - - bytes &asUtf8(); - int asInteger() - { - assert(tag == CONSTANT_Integer); - return value.i; - } - - bool isUtf8(bytes &b) - { - return tagMatches(CONSTANT_Utf8) && value.b.equals(b); - } - - bool isDoubleWord() - { - return tag == CONSTANT_Double || tag == CONSTANT_Long; - } - - bool tagMatches(byte tag2) - { - return (tag2 == tag) || (tag2 == CONSTANT_Utf8 && tag == CONSTANT_Signature) || - (tag2 == CONSTANT_Literal && tag >= CONSTANT_Integer && tag <= CONSTANT_String && - tag != CONSTANT_Class) || - (tag2 == CONSTANT_Member && tag >= CONSTANT_Fieldref && - tag <= CONSTANT_InterfaceMethodref); - } + byte tag; + unsigned short nrefs; // pack w/ tag + + int outputIndex; + uint32_t inord; // &cp.entries[cp.tag_base[this->tag]+this->inord] == this + + entry **refs; + + // put last to pack best + union + { + bytes b; + int i; + int64_t l; + } value; + + void requestOutputIndex(constant_pool &cp, int req = REQUESTED); + int getOutputIndex() + { + assert(outputIndex > NOT_REQUESTED); + return outputIndex; + } + + entry *ref(int refnum) + { + assert((uint32_t)refnum < nrefs); + return refs[refnum]; + } + + const char *utf8String() + { + assert(tagMatches(CONSTANT_Utf8)); + assert(value.b.len == strlen((const char *)value.b.ptr)); + return (const char *)value.b.ptr; + } + + entry *className() + { + assert(tagMatches(CONSTANT_Class)); + return ref(0); + } + + entry *memberClass() + { + assert(tagMatches(CONSTANT_Member)); + return ref(0); + } + + entry *memberDescr() + { + assert(tagMatches(CONSTANT_Member)); + return ref(1); + } + + entry *descrName() + { + assert(tagMatches(CONSTANT_NameandType)); + return ref(0); + } + + entry *descrType() + { + assert(tagMatches(CONSTANT_NameandType)); + return ref(1); + } + + int typeSize(); + + bytes &asUtf8(); + int asInteger() + { + assert(tag == CONSTANT_Integer); + return value.i; + } + + bool isUtf8(bytes &b) + { + return tagMatches(CONSTANT_Utf8) && value.b.equals(b); + } + + bool isDoubleWord() + { + return tag == CONSTANT_Double || tag == CONSTANT_Long; + } + + bool tagMatches(byte tag2) + { + return (tag2 == tag) || (tag2 == CONSTANT_Utf8 && tag == CONSTANT_Signature) || + (tag2 == CONSTANT_Literal && tag >= CONSTANT_Integer && tag <= CONSTANT_String && + tag != CONSTANT_Class) || + (tag2 == CONSTANT_Member && tag >= CONSTANT_Fieldref && + tag <= CONSTANT_InterfaceMethodref); + } }; entry *cpindex::get(uint32_t i) { - if (i >= len) - return nullptr; - else if (base1 != nullptr) - // primary index - return &base1[i]; - else - // secondary index - return base2[i]; + if (i >= len) + return nullptr; + else if (base1 != nullptr) + // primary index + return &base1[i]; + else + // secondary index + return base2[i]; } inline bytes &entry::asUtf8() { - assert(tagMatches(CONSTANT_Utf8)); - return value.b; + assert(tagMatches(CONSTANT_Utf8)); + return value.b; } int entry::typeSize() { - assert(tagMatches(CONSTANT_Utf8)); - const char *sigp = (char *)value.b.ptr; - switch (*sigp) - { - case '(': - sigp++; - break; // skip opening '(' - case 'D': - case 'J': - return 2; // double field - default: - return 1; // field - } - int siglen = 0; - for (;;) - { - int ch = *sigp++; - switch (ch) - { - case 'D': - case 'J': - siglen += 1; - break; - case '[': - // Skip rest of array info. - while (ch == '[') - { - ch = *sigp++; - } - if (ch != 'L') - break; - // else fall through - case 'L': - sigp = strchr(sigp, ';'); - if (sigp == nullptr) - { - unpack_abort("bad data"); - return 0; - } - sigp += 1; - break; - case ')': // closing ')' - return siglen; - } - siglen += 1; - } + assert(tagMatches(CONSTANT_Utf8)); + const char *sigp = (char *)value.b.ptr; + switch (*sigp) + { + case '(': + sigp++; + break; // skip opening '(' + case 'D': + case 'J': + return 2; // double field + default: + return 1; // field + } + int siglen = 0; + for (;;) + { + int ch = *sigp++; + switch (ch) + { + case 'D': + case 'J': + siglen += 1; + break; + case '[': + // Skip rest of array info. + while (ch == '[') + { + ch = *sigp++; + } + if (ch != 'L') + break; + // else fall through + case 'L': + sigp = strchr(sigp, ';'); + if (sigp == nullptr) + { + unpack_abort("bad data"); + return 0; + } + sigp += 1; + break; + case ')': // closing ')' + return siglen; + } + siglen += 1; + } } inline cpindex *constant_pool::getFieldIndex(entry *classRef) { - assert(classRef->tagMatches(CONSTANT_Class)); - assert((uint32_t)classRef->inord < (uint32_t)tag_count[CONSTANT_Class]); - return &member_indexes[classRef->inord * 2 + 0]; + assert(classRef->tagMatches(CONSTANT_Class)); + assert((uint32_t)classRef->inord < (uint32_t)tag_count[CONSTANT_Class]); + return &member_indexes[classRef->inord * 2 + 0]; } inline cpindex *constant_pool::getMethodIndex(entry *classRef) { - assert(classRef->tagMatches(CONSTANT_Class)); - assert((uint32_t)classRef->inord < (uint32_t)tag_count[CONSTANT_Class]); - return &member_indexes[classRef->inord * 2 + 1]; + assert(classRef->tagMatches(CONSTANT_Class)); + assert((uint32_t)classRef->inord < (uint32_t)tag_count[CONSTANT_Class]); + return &member_indexes[classRef->inord * 2 + 1]; } struct inner_class { - entry *inner; - entry *outer; - entry *name; - int flags; - inner_class *next_sibling; - bool requested; + entry *inner; + entry *outer; + entry *name; + int flags; + inner_class *next_sibling; + bool requested; }; // Here is where everything gets deallocated: void unpacker::free() { - int i; - if (jarout != nullptr) - jarout->reset(); - if (gzin != nullptr) - { - gzin->free(); - gzin = nullptr; - } - if (free_input) - input.free(); - /* - * free everybody ever allocated with U_NEW or (recently) with T_NEW - */ - assert(smallbuf.base() == nullptr || mallocs.contains(smallbuf.base())); - assert(tsmallbuf.base() == nullptr || tmallocs.contains(tsmallbuf.base())); - mallocs.freeAll(); - tmallocs.freeAll(); - smallbuf.init(); - tsmallbuf.init(); - bcimap.free(); - class_fixup_type.free(); - class_fixup_offset.free(); - class_fixup_ref.free(); - code_fixup_type.free(); - code_fixup_offset.free(); - code_fixup_source.free(); - requested_ics.free(); - cur_classfile_head.free(); - cur_classfile_tail.free(); - for (i = 0; i < ATTR_CONTEXT_LIMIT; i++) - attr_defs[i].free(); - - // free CP state - cp.outputEntries.free(); - for (i = 0; i < CONSTANT_Limit; i++) - cp.tag_extras[i].free(); + int i; + if (jarout != nullptr) + jarout->reset(); + if (gzin != nullptr) + { + gzin->free(); + gzin = nullptr; + } + if (free_input) + input.free(); + /* + * free everybody ever allocated with U_NEW or (recently) with T_NEW + */ + assert(smallbuf.base() == nullptr || mallocs.contains(smallbuf.base())); + assert(tsmallbuf.base() == nullptr || tmallocs.contains(tsmallbuf.base())); + mallocs.freeAll(); + tmallocs.freeAll(); + smallbuf.init(); + tsmallbuf.init(); + bcimap.free(); + class_fixup_type.free(); + class_fixup_offset.free(); + class_fixup_ref.free(); + code_fixup_type.free(); + code_fixup_offset.free(); + code_fixup_source.free(); + requested_ics.free(); + cur_classfile_head.free(); + cur_classfile_tail.free(); + for (i = 0; i < ATTR_CONTEXT_LIMIT; i++) + attr_defs[i].free(); + + // free CP state + cp.outputEntries.free(); + for (i = 0; i < CONSTANT_Limit; i++) + cp.tag_extras[i].free(); } // input handling @@ -318,1933 +312,1933 @@ void unpacker::free() // unless rplimit hits input.limit(). bool unpacker::ensure_input(int64_t more) { - uint64_t want = more - input_remaining(); - if ((int64_t)want <= 0) - return true; // it's already in the buffer - if (rplimit == input.limit()) - return true; // not expecting any more - - if (read_input_fn == nullptr) - { - // assume it is already all there - bytes_read += input.limit() - rplimit; - rplimit = input.limit(); - return true; - } - - uint64_t remaining = (input.limit() - rplimit); // how much left to read? - byte *rpgoal = (want >= remaining) ? input.limit() : rplimit + (size_t)want; - enum - { - CHUNK_SIZE = (1 << 14) - }; - uint64_t fetch = want; - if (fetch < CHUNK_SIZE) - fetch = CHUNK_SIZE; - if (fetch > remaining * 3 / 4) - fetch = remaining; - // Try to fetch at least "more" bytes. - while ((int64_t)fetch > 0) - { - int64_t nr = (*read_input_fn)(this, rplimit, fetch, remaining); - if (nr <= 0) - { - return (rplimit >= rpgoal); - } - remaining -= nr; - rplimit += nr; - fetch -= nr; - bytes_read += nr; - assert(remaining == (uint64_t)(input.limit() - rplimit)); - } - return true; + uint64_t want = more - input_remaining(); + if ((int64_t)want <= 0) + return true; // it's already in the buffer + if (rplimit == input.limit()) + return true; // not expecting any more + + if (read_input_fn == nullptr) + { + // assume it is already all there + bytes_read += input.limit() - rplimit; + rplimit = input.limit(); + return true; + } + + uint64_t remaining = (input.limit() - rplimit); // how much left to read? + byte *rpgoal = (want >= remaining) ? input.limit() : rplimit + (size_t)want; + enum + { + CHUNK_SIZE = (1 << 14) + }; + uint64_t fetch = want; + if (fetch < CHUNK_SIZE) + fetch = CHUNK_SIZE; + if (fetch > remaining * 3 / 4) + fetch = remaining; + // Try to fetch at least "more" bytes. + while ((int64_t)fetch > 0) + { + int64_t nr = (*read_input_fn)(this, rplimit, fetch, remaining); + if (nr <= 0) + { + return (rplimit >= rpgoal); + } + remaining -= nr; + rplimit += nr; + fetch -= nr; + bytes_read += nr; + assert(remaining == (uint64_t)(input.limit() - rplimit)); + } + return true; } // output handling fillbytes *unpacker::close_output(fillbytes *which) { - assert(wp != nullptr); - if (which == nullptr) - { - if (wpbase == cur_classfile_head.base()) - { - which = &cur_classfile_head; - } - else - { - which = &cur_classfile_tail; - } - } - assert(wpbase == which->base()); - assert(wplimit == which->end()); - which->setLimit(wp); - wp = nullptr; - wplimit = nullptr; - // wpbase = nullptr; - return which; + assert(wp != nullptr); + if (which == nullptr) + { + if (wpbase == cur_classfile_head.base()) + { + which = &cur_classfile_head; + } + else + { + which = &cur_classfile_tail; + } + } + assert(wpbase == which->base()); + assert(wplimit == which->end()); + which->setLimit(wp); + wp = nullptr; + wplimit = nullptr; + // wpbase = nullptr; + return which; } // maybe_inline void unpacker::ensure_put_space(size_t size) { - if (wp + size <= wplimit) - return; - // Determine which segment needs expanding. - fillbytes *which = close_output(); - byte *wp0 = which->grow(size); - wpbase = which->base(); - wplimit = which->end(); - wp = wp0; + if (wp + size <= wplimit) + return; + // Determine which segment needs expanding. + fillbytes *which = close_output(); + byte *wp0 = which->grow(size); + wpbase = which->base(); + wplimit = which->end(); + wp = wp0; } byte *unpacker::put_space(size_t size) { - byte *wp0 = wp; - byte *wp1 = wp0 + size; - if (wp1 > wplimit) - { - ensure_put_space(size); - wp0 = wp; - wp1 = wp0 + size; - } - wp = wp1; - return wp0; + byte *wp0 = wp; + byte *wp1 = wp0 + size; + if (wp1 > wplimit) + { + ensure_put_space(size); + wp0 = wp; + wp1 = wp0 + size; + } + wp = wp1; + return wp0; } void unpacker::putu2_at(byte *wp, int n) { - if (n != (unsigned short)n) - { - unpack_abort(ERROR_OVERFLOW); - return; - } - wp[0] = (n) >> 8; - wp[1] = (n) >> 0; + if (n != (unsigned short)n) + { + unpack_abort(ERROR_OVERFLOW); + return; + } + wp[0] = (n) >> 8; + wp[1] = (n) >> 0; } void unpacker::putu4_at(byte *wp, int n) { - wp[0] = (n) >> 24; - wp[1] = (n) >> 16; - wp[2] = (n) >> 8; - wp[3] = (n) >> 0; + wp[0] = (n) >> 24; + wp[1] = (n) >> 16; + wp[2] = (n) >> 8; + wp[3] = (n) >> 0; } void unpacker::putu8_at(byte *wp, int64_t n) { - putu4_at(wp + 0, (int)((uint64_t)n >> 32)); - putu4_at(wp + 4, (int)((uint64_t)n >> 0)); + putu4_at(wp + 0, (int)((uint64_t)n >> 32)); + putu4_at(wp + 4, (int)((uint64_t)n >> 0)); } void unpacker::putu2(int n) { - putu2_at(put_space(2), n); + putu2_at(put_space(2), n); } void unpacker::putu4(int n) { - putu4_at(put_space(4), n); + putu4_at(put_space(4), n); } void unpacker::putu8(int64_t n) { - putu8_at(put_space(8), n); + putu8_at(put_space(8), n); } int unpacker::putref_index(entry *e, int size) { - if (e == nullptr) - return 0; - else if (e->outputIndex > NOT_REQUESTED) - return e->outputIndex; - else if (e->tag == CONSTANT_Signature) - return putref_index(e->ref(0), size); - else - { - e->requestOutputIndex(cp, -size); - // Later on we'll fix the bits. - class_fixup_type.addByte(size); - class_fixup_offset.add((int)wpoffset()); - class_fixup_ref.add(e); - return 0; - } + if (e == nullptr) + return 0; + else if (e->outputIndex > NOT_REQUESTED) + return e->outputIndex; + else if (e->tag == CONSTANT_Signature) + return putref_index(e->ref(0), size); + else + { + e->requestOutputIndex(cp, -size); + // Later on we'll fix the bits. + class_fixup_type.addByte(size); + class_fixup_offset.add((int)wpoffset()); + class_fixup_ref.add(e); + return 0; + } } void unpacker::putref(entry *e) { - int oidx = putref_index(e, 2); - putu2_at(put_space(2), oidx); + int oidx = putref_index(e, 2); + putu2_at(put_space(2), oidx); } void unpacker::putu1ref(entry *e) { - int oidx = putref_index(e, 1); - putu1_at(put_space(1), oidx); + int oidx = putref_index(e, 1); + putu1_at(put_space(1), oidx); } // Allocation of small and large blocks. enum { - CHUNK = (1 << 14), - SMALL = (1 << 9) + CHUNK = (1 << 14), + SMALL = (1 << 9) }; // Call malloc. Try to combine small blocks and free much later. void *unpacker::alloc_heap(size_t size, bool smallOK, bool temp) { - if (!smallOK || size > SMALL) - { - void *res = must_malloc((int)size); - (temp ? &tmallocs : &mallocs)->add(res); - return res; - } - fillbytes &xsmallbuf = *(temp ? &tsmallbuf : &smallbuf); - if (!xsmallbuf.canAppend(size + 1)) - { - xsmallbuf.init(CHUNK); - (temp ? &tmallocs : &mallocs)->add(xsmallbuf.base()); - } - int growBy = (int)size; - growBy += -growBy & 7; // round up mod 8 - return xsmallbuf.grow(growBy); + if (!smallOK || size > SMALL) + { + void *res = must_malloc((int)size); + (temp ? &tmallocs : &mallocs)->add(res); + return res; + } + fillbytes &xsmallbuf = *(temp ? &tsmallbuf : &smallbuf); + if (!xsmallbuf.canAppend(size + 1)) + { + xsmallbuf.init(CHUNK); + (temp ? &tmallocs : &mallocs)->add(xsmallbuf.base()); + } + int growBy = (int)size; + growBy += -growBy & 7; // round up mod 8 + return xsmallbuf.grow(growBy); } void unpacker::saveTo(bytes &b, byte *ptr, size_t len) { - b.ptr = U_NEW(byte, add_size(len, 1)); - b.len = len; - b.copyFrom(ptr, len); + b.ptr = U_NEW(byte, add_size(len, 1)); + b.len = len; + b.copyFrom(ptr, len); } // Read up through band_headers. // Do the archive_size dance to set the size of the input mega-buffer. void unpacker::read_file_header() { - // Read file header to determine file type and total size. - enum - { - MAGIC_BYTES = 4, - AH_LENGTH_0 = 3, // minver, majver, options are outside of archive_size - AH_LENGTH_0_MAX = AH_LENGTH_0 + 1, // options might have 2 bytes - AH_LENGTH = 26, // maximum archive header length (w/ all fields) - // Length contributions from optional header fields: - AH_FILE_HEADER_LEN = 5, // sizehi/lo/next/modtime/files - AH_ARCHIVE_SIZE_LEN = 2, // sizehi/lo only; part of AH_FILE_HEADER_LEN - AH_CP_NUMBER_LEN = 4, // int/float/long/double - AH_SPECIAL_FORMAT_LEN = 2, // layouts/band-headers - AH_LENGTH_MIN = - AH_LENGTH - (AH_FILE_HEADER_LEN + AH_SPECIAL_FORMAT_LEN + AH_CP_NUMBER_LEN), - ARCHIVE_SIZE_MIN = AH_LENGTH_MIN - (AH_LENGTH_0 + AH_ARCHIVE_SIZE_LEN), - FIRST_READ = MAGIC_BYTES + AH_LENGTH_MIN - }; - - assert(AH_LENGTH_MIN == 15); // # of UNSIGNED5 fields required after archive_magic - assert(ARCHIVE_SIZE_MIN == 10); // # of UNSIGNED5 fields required after archive_size - // An absolute minimum nullptr archive is magic[4], {minver,majver,options}[3], - // archive_size[0], cp_counts[8], class_counts[4], for a total of 19 bytes. - // (Note that archive_size is optional; it may be 0..10 bytes in length.) - // The first read must capture everything up through the options field. - // This happens to work even if {minver,majver,options} is a pathological - // 15 bytes long. Legal pack files limit those three fields to 1+1+2 bytes. - assert(FIRST_READ >= MAGIC_BYTES + AH_LENGTH_0 * B_MAX); - - // Up through archive_size, the largest possible archive header is - // magic[4], {minver,majver,options}[4], archive_size[10]. - // (Note only the low 12 bits of options are allowed to be non-zero.) - // In order to parse archive_size, we need at least this many bytes - // in the first read. Of course, if archive_size_hi is more than - // a byte, we probably will fail to allocate the buffer, since it - // will be many gigabytes long. This is a practical, not an - // architectural limit to Pack200 archive sizes. - assert(FIRST_READ >= MAGIC_BYTES + AH_LENGTH_0_MAX + 2 * B_MAX); - - bool foreign_buf = (read_input_fn == nullptr); - byte initbuf[(int)FIRST_READ + (int)C_SLOP + 200]; // 200 is for JAR I/O - if (foreign_buf) - { - // inbytes is all there is - input.set(inbytes); - rp = input.base(); - rplimit = input.limit(); - } - else - { - // inbytes, if not empty, contains some read-ahead we must use first - // ensure_input will take care of copying it into initbuf, - // then querying read_input_fn for any additional data needed. - // However, the caller must assume that we use up all of inbytes. - // There is no way to tell the caller that we used only part of them. - // Therefore, the caller must use only a bare minimum of read-ahead. - if (inbytes.len > FIRST_READ) - { - unpack_abort("too much read-ahead"); - } - input.set(initbuf, sizeof(initbuf)); - input.b.clear(); - input.b.copyFrom(inbytes); - rplimit = rp = input.base(); - rplimit += inbytes.len; - bytes_read += inbytes.len; - } - // Read only 19 bytes, which is certain to contain #archive_options fields, - // but is certain not to overflow past the archive_header. - input.b.len = FIRST_READ; - if (!ensure_input(FIRST_READ)) - unpack_abort("EOF reading archive magic number"); - - if (rp[0] == 'P' && rp[1] == 'K') - { - // In the Unix-style program, we simply simulate a copy command. - // Copy until EOF; assume the JAR file is the last segment. - fprintf(stderr, "Copy-mode.\n"); - for (;;) - { - jarout->write_data(rp, (int)input_remaining()); - if (foreign_buf) - break; // one-time use of a passed in buffer - if (input.size() < CHUNK) - { - // Get some breathing room. - input.set(U_NEW(byte, (size_t)CHUNK + C_SLOP), (size_t)CHUNK); - } - rp = rplimit = input.base(); - if (!ensure_input(1)) - break; - } - jarout->closeJarFile(false); - return; - } - - // Read the magic number. - magic = 0; - for (int i1 = 0; i1 < (int)sizeof(magic); i1++) - { - magic <<= 8; - magic += (*rp++ & 0xFF); - } - - // Read the first 3 values from the header. - value_stream hdr; - int hdrVals = 0; - int hdrValsSkipped = 0; // debug only - hdr.init(rp, rplimit, UNSIGNED5_spec); - minver = hdr.getInt(); - majver = hdr.getInt(); - hdrVals += 2; - - if (magic != (int)JAVA_PACKAGE_MAGIC || - (majver != JAVA5_PACKAGE_MAJOR_VERSION && majver != JAVA6_PACKAGE_MAJOR_VERSION) || - (minver != JAVA5_PACKAGE_MINOR_VERSION && minver != JAVA6_PACKAGE_MINOR_VERSION)) - { - char message[200]; - sprintf(message, "@" ERROR_FORMAT ": magic/ver = " - "%08X/%d.%d should be %08X/%d.%d OR %08X/%d.%d\n", - magic, majver, minver, JAVA_PACKAGE_MAGIC, JAVA5_PACKAGE_MAJOR_VERSION, - JAVA5_PACKAGE_MINOR_VERSION, JAVA_PACKAGE_MAGIC, JAVA6_PACKAGE_MAJOR_VERSION, - JAVA6_PACKAGE_MINOR_VERSION); - unpack_abort(message); - } - - archive_options = hdr.getInt(); - hdrVals += 1; - assert(hdrVals == AH_LENGTH_0); // first three fields only + // Read file header to determine file type and total size. + enum + { + MAGIC_BYTES = 4, + AH_LENGTH_0 = 3, // minver, majver, options are outside of archive_size + AH_LENGTH_0_MAX = AH_LENGTH_0 + 1, // options might have 2 bytes + AH_LENGTH = 26, // maximum archive header length (w/ all fields) + // Length contributions from optional header fields: + AH_FILE_HEADER_LEN = 5, // sizehi/lo/next/modtime/files + AH_ARCHIVE_SIZE_LEN = 2, // sizehi/lo only; part of AH_FILE_HEADER_LEN + AH_CP_NUMBER_LEN = 4, // int/float/long/double + AH_SPECIAL_FORMAT_LEN = 2, // layouts/band-headers + AH_LENGTH_MIN = + AH_LENGTH - (AH_FILE_HEADER_LEN + AH_SPECIAL_FORMAT_LEN + AH_CP_NUMBER_LEN), + ARCHIVE_SIZE_MIN = AH_LENGTH_MIN - (AH_LENGTH_0 + AH_ARCHIVE_SIZE_LEN), + FIRST_READ = MAGIC_BYTES + AH_LENGTH_MIN + }; + + assert(AH_LENGTH_MIN == 15); // # of UNSIGNED5 fields required after archive_magic + assert(ARCHIVE_SIZE_MIN == 10); // # of UNSIGNED5 fields required after archive_size + // An absolute minimum nullptr archive is magic[4], {minver,majver,options}[3], + // archive_size[0], cp_counts[8], class_counts[4], for a total of 19 bytes. + // (Note that archive_size is optional; it may be 0..10 bytes in length.) + // The first read must capture everything up through the options field. + // This happens to work even if {minver,majver,options} is a pathological + // 15 bytes long. Legal pack files limit those three fields to 1+1+2 bytes. + assert(FIRST_READ >= MAGIC_BYTES + AH_LENGTH_0 * B_MAX); + + // Up through archive_size, the largest possible archive header is + // magic[4], {minver,majver,options}[4], archive_size[10]. + // (Note only the low 12 bits of options are allowed to be non-zero.) + // In order to parse archive_size, we need at least this many bytes + // in the first read. Of course, if archive_size_hi is more than + // a byte, we probably will fail to allocate the buffer, since it + // will be many gigabytes long. This is a practical, not an + // architectural limit to Pack200 archive sizes. + assert(FIRST_READ >= MAGIC_BYTES + AH_LENGTH_0_MAX + 2 * B_MAX); + + bool foreign_buf = (read_input_fn == nullptr); + byte initbuf[(int)FIRST_READ + (int)C_SLOP + 200]; // 200 is for JAR I/O + if (foreign_buf) + { + // inbytes is all there is + input.set(inbytes); + rp = input.base(); + rplimit = input.limit(); + } + else + { + // inbytes, if not empty, contains some read-ahead we must use first + // ensure_input will take care of copying it into initbuf, + // then querying read_input_fn for any additional data needed. + // However, the caller must assume that we use up all of inbytes. + // There is no way to tell the caller that we used only part of them. + // Therefore, the caller must use only a bare minimum of read-ahead. + if (inbytes.len > FIRST_READ) + { + unpack_abort("too much read-ahead"); + } + input.set(initbuf, sizeof(initbuf)); + input.b.clear(); + input.b.copyFrom(inbytes); + rplimit = rp = input.base(); + rplimit += inbytes.len; + bytes_read += inbytes.len; + } + // Read only 19 bytes, which is certain to contain #archive_options fields, + // but is certain not to overflow past the archive_header. + input.b.len = FIRST_READ; + if (!ensure_input(FIRST_READ)) + unpack_abort("EOF reading archive magic number"); + + if (rp[0] == 'P' && rp[1] == 'K') + { + // In the Unix-style program, we simply simulate a copy command. + // Copy until EOF; assume the JAR file is the last segment. + fprintf(stderr, "Copy-mode.\n"); + for (;;) + { + jarout->write_data(rp, (int)input_remaining()); + if (foreign_buf) + break; // one-time use of a passed in buffer + if (input.size() < CHUNK) + { + // Get some breathing room. + input.set(U_NEW(byte, (size_t)CHUNK + C_SLOP), (size_t)CHUNK); + } + rp = rplimit = input.base(); + if (!ensure_input(1)) + break; + } + jarout->closeJarFile(false); + return; + } + + // Read the magic number. + magic = 0; + for (int i1 = 0; i1 < (int)sizeof(magic); i1++) + { + magic <<= 8; + magic += (*rp++ & 0xFF); + } + + // Read the first 3 values from the header. + value_stream hdr; + int hdrVals = 0; + int hdrValsSkipped = 0; // debug only + hdr.init(rp, rplimit, UNSIGNED5_spec); + minver = hdr.getInt(); + majver = hdr.getInt(); + hdrVals += 2; + + if (magic != (int)JAVA_PACKAGE_MAGIC || + (majver != JAVA5_PACKAGE_MAJOR_VERSION && majver != JAVA6_PACKAGE_MAJOR_VERSION) || + (minver != JAVA5_PACKAGE_MINOR_VERSION && minver != JAVA6_PACKAGE_MINOR_VERSION)) + { + char message[200]; + sprintf(message, "@" ERROR_FORMAT ": magic/ver = " + "%08X/%d.%d should be %08X/%d.%d OR %08X/%d.%d\n", + magic, majver, minver, JAVA_PACKAGE_MAGIC, JAVA5_PACKAGE_MAJOR_VERSION, + JAVA5_PACKAGE_MINOR_VERSION, JAVA_PACKAGE_MAGIC, JAVA6_PACKAGE_MAJOR_VERSION, + JAVA6_PACKAGE_MINOR_VERSION); + unpack_abort(message); + } + + archive_options = hdr.getInt(); + hdrVals += 1; + assert(hdrVals == AH_LENGTH_0); // first three fields only #define ORBIT(bit) | (bit) - int OPTION_LIMIT = (0 ARCHIVE_BIT_DO(ORBIT)); + int OPTION_LIMIT = (0 ARCHIVE_BIT_DO(ORBIT)); #undef ORBIT - if ((archive_options & ~OPTION_LIMIT) != 0) - { - fprintf(stderr, "Warning: Illegal archive options 0x%x\n", archive_options); - unpack_abort("illegal archive options"); - return; - } - - if ((archive_options & AO_HAVE_FILE_HEADERS) != 0) - { - uint32_t hi = hdr.getInt(); - uint32_t lo = hdr.getInt(); - uint64_t x = band::makeLong(hi, lo); - archive_size = (size_t)x; - if (archive_size != x) - { - // Silly size specified; force overflow. - archive_size = PSIZE_MAX + 1; - } - hdrVals += 2; - } - else - { - hdrValsSkipped += 2; - } - - // Now we can size the whole archive. - // Read everything else into a mega-buffer. - rp = hdr.rp; - int header_size_0 = (int)(rp - input.base()); // used-up header (4byte + 3int) - int header_size_1 = (int)(rplimit - rp); // buffered unused initial fragment - int header_size = header_size_0 + header_size_1; - unsized_bytes_read = header_size_0; - if (foreign_buf) - { - if (archive_size > (size_t)header_size_1) - { - unpack_abort("EOF reading fixed input buffer"); - return; - } - } - else if (archive_size != 0) - { - if (archive_size < ARCHIVE_SIZE_MIN) - { - unpack_abort("impossible archive size"); // bad input data - return; - } - if (archive_size < (size_t)header_size_1) - { - unpack_abort("too much read-ahead"); // somehow we pre-fetched too much? - return; - } - input.set(U_NEW(byte, add_size(header_size_0, archive_size, C_SLOP)), - (size_t)header_size_0 + archive_size); - assert(input.limit()[0] == 0); - // Move all the bytes we read initially into the real buffer. - input.b.copyFrom(initbuf, header_size); - rp = input.b.ptr + header_size_0; - rplimit = input.b.ptr + header_size; - } - else - { - // It's more complicated and painful. - // A zero archive_size means that we must read until EOF. - input.init(CHUNK * 2); - input.b.len = input.allocated; - rp = rplimit = input.base(); - // Set up input buffer as if we already read the header: - input.b.copyFrom(initbuf, header_size); - rplimit += header_size; - while (ensure_input(input.limit() - rp)) - { - size_t dataSoFar = input_remaining(); - size_t nextSize = add_size(dataSoFar, CHUNK); - input.ensureSize(nextSize); - input.b.len = input.allocated; - rp = rplimit = input.base(); - rplimit += dataSoFar; - } - size_t dataSize = (rplimit - input.base()); - input.b.len = dataSize; - input.grow(C_SLOP); - free_input = true; // free it later - input.b.len = dataSize; - assert(input.limit()[0] == 0); - rp = rplimit = input.base(); - rplimit += dataSize; - rp += header_size_0; // already scanned these bytes... - } - live_input = true; // mark as "do not reuse" - - // read the rest of the header fields - ensure_input((AH_LENGTH - AH_LENGTH_0) * B_MAX); - hdr.rp = rp; - hdr.rplimit = rplimit; - - if ((archive_options & AO_HAVE_FILE_HEADERS) != 0) - { - archive_next_count = hdr.getInt(); - if (archive_next_count < 0) - unpack_abort("bad archive_next_count"); - archive_modtime = hdr.getInt(); - file_count = hdr.getInt(); - if (file_count < 0) - unpack_abort("bad file_count"); - hdrVals += 3; - } - else - { - hdrValsSkipped += 3; - } - - if ((archive_options & AO_HAVE_SPECIAL_FORMATS) != 0) - { - band_headers_size = hdr.getInt(); - if (band_headers_size < 0) - unpack_abort("bad band_headers_size"); - attr_definition_count = hdr.getInt(); - if (attr_definition_count < 0) - unpack_abort("bad attr_definition_count"); - hdrVals += 2; - } - else - { - hdrValsSkipped += 2; - } - - int cp_counts[N_TAGS_IN_ORDER]; - for (int k = 0; k < (int)N_TAGS_IN_ORDER; k++) - { - if (!(archive_options & AO_HAVE_CP_NUMBERS)) - { - switch (TAGS_IN_ORDER[k]) - { - case CONSTANT_Integer: - case CONSTANT_Float: - case CONSTANT_Long: - case CONSTANT_Double: - cp_counts[k] = 0; - hdrValsSkipped += 1; - continue; - } - } - cp_counts[k] = hdr.getInt(); - if (cp_counts[k] < 0) - unpack_abort("bad cp_counts"); - hdrVals += 1; - } - - ic_count = hdr.getInt(); - if (ic_count < 0) - unpack_abort("bad ic_count"); - - default_class_minver = hdr.getInt(); - default_class_majver = hdr.getInt(); - - class_count = hdr.getInt(); - if (class_count < 0) - unpack_abort("bad class_count"); - - hdrVals += 4; - - // done with archive_header - hdrVals += hdrValsSkipped; - assert(hdrVals == AH_LENGTH); - - rp = hdr.rp; - if (rp > rplimit) - unpack_abort("EOF reading archive header"); - - // Now size the CP. - cp.init(this, cp_counts); - - default_file_modtime = archive_modtime; - if (default_file_modtime == 0 && !(archive_options & AO_HAVE_FILE_MODTIME)) - default_file_modtime = DEFAULT_ARCHIVE_MODTIME; // taken from driver - if ((archive_options & AO_DEFLATE_HINT) != 0) - default_file_options |= FO_DEFLATE_HINT; - - // meta-bytes, if any, immediately follow archive header - // band_headers.readData(band_headers_size); - ensure_input(band_headers_size); - if (input_remaining() < (size_t)band_headers_size) - { - unpack_abort("EOF reading band headers"); - return; - } - bytes band_headers; - // The "1+" allows an initial byte to be pushed on the front. - band_headers.set(1 + U_NEW(byte, 1 + band_headers_size + C_SLOP), band_headers_size); - - // Start scanning band headers here: - band_headers.copyFrom(rp, band_headers.len); - rp += band_headers.len; - assert(rp <= rplimit); - meta_rp = band_headers.ptr; - // Put evil meta-codes at the end of the band headers, - // so we are sure to throw an error if we run off the end. - bytes::of(band_headers.limit(), C_SLOP).clear(_meta_error); + if ((archive_options & ~OPTION_LIMIT) != 0) + { + fprintf(stderr, "Warning: Illegal archive options 0x%x\n", archive_options); + unpack_abort("illegal archive options"); + return; + } + + if ((archive_options & AO_HAVE_FILE_HEADERS) != 0) + { + uint32_t hi = hdr.getInt(); + uint32_t lo = hdr.getInt(); + uint64_t x = band::makeLong(hi, lo); + archive_size = (size_t)x; + if (archive_size != x) + { + // Silly size specified; force overflow. + archive_size = PSIZE_MAX + 1; + } + hdrVals += 2; + } + else + { + hdrValsSkipped += 2; + } + + // Now we can size the whole archive. + // Read everything else into a mega-buffer. + rp = hdr.rp; + int header_size_0 = (int)(rp - input.base()); // used-up header (4byte + 3int) + int header_size_1 = (int)(rplimit - rp); // buffered unused initial fragment + int header_size = header_size_0 + header_size_1; + unsized_bytes_read = header_size_0; + if (foreign_buf) + { + if (archive_size > (size_t)header_size_1) + { + unpack_abort("EOF reading fixed input buffer"); + return; + } + } + else if (archive_size != 0) + { + if (archive_size < ARCHIVE_SIZE_MIN) + { + unpack_abort("impossible archive size"); // bad input data + return; + } + if (archive_size < (size_t)header_size_1) + { + unpack_abort("too much read-ahead"); // somehow we pre-fetched too much? + return; + } + input.set(U_NEW(byte, add_size(header_size_0, archive_size, C_SLOP)), + (size_t)header_size_0 + archive_size); + assert(input.limit()[0] == 0); + // Move all the bytes we read initially into the real buffer. + input.b.copyFrom(initbuf, header_size); + rp = input.b.ptr + header_size_0; + rplimit = input.b.ptr + header_size; + } + else + { + // It's more complicated and painful. + // A zero archive_size means that we must read until EOF. + input.init(CHUNK * 2); + input.b.len = input.allocated; + rp = rplimit = input.base(); + // Set up input buffer as if we already read the header: + input.b.copyFrom(initbuf, header_size); + rplimit += header_size; + while (ensure_input(input.limit() - rp)) + { + size_t dataSoFar = input_remaining(); + size_t nextSize = add_size(dataSoFar, CHUNK); + input.ensureSize(nextSize); + input.b.len = input.allocated; + rp = rplimit = input.base(); + rplimit += dataSoFar; + } + size_t dataSize = (rplimit - input.base()); + input.b.len = dataSize; + input.grow(C_SLOP); + free_input = true; // free it later + input.b.len = dataSize; + assert(input.limit()[0] == 0); + rp = rplimit = input.base(); + rplimit += dataSize; + rp += header_size_0; // already scanned these bytes... + } + live_input = true; // mark as "do not reuse" + + // read the rest of the header fields + ensure_input((AH_LENGTH - AH_LENGTH_0) * B_MAX); + hdr.rp = rp; + hdr.rplimit = rplimit; + + if ((archive_options & AO_HAVE_FILE_HEADERS) != 0) + { + archive_next_count = hdr.getInt(); + if (archive_next_count < 0) + unpack_abort("bad archive_next_count"); + archive_modtime = hdr.getInt(); + file_count = hdr.getInt(); + if (file_count < 0) + unpack_abort("bad file_count"); + hdrVals += 3; + } + else + { + hdrValsSkipped += 3; + } + + if ((archive_options & AO_HAVE_SPECIAL_FORMATS) != 0) + { + band_headers_size = hdr.getInt(); + if (band_headers_size < 0) + unpack_abort("bad band_headers_size"); + attr_definition_count = hdr.getInt(); + if (attr_definition_count < 0) + unpack_abort("bad attr_definition_count"); + hdrVals += 2; + } + else + { + hdrValsSkipped += 2; + } + + int cp_counts[N_TAGS_IN_ORDER]; + for (int k = 0; k < (int)N_TAGS_IN_ORDER; k++) + { + if (!(archive_options & AO_HAVE_CP_NUMBERS)) + { + switch (TAGS_IN_ORDER[k]) + { + case CONSTANT_Integer: + case CONSTANT_Float: + case CONSTANT_Long: + case CONSTANT_Double: + cp_counts[k] = 0; + hdrValsSkipped += 1; + continue; + } + } + cp_counts[k] = hdr.getInt(); + if (cp_counts[k] < 0) + unpack_abort("bad cp_counts"); + hdrVals += 1; + } + + ic_count = hdr.getInt(); + if (ic_count < 0) + unpack_abort("bad ic_count"); + + default_class_minver = hdr.getInt(); + default_class_majver = hdr.getInt(); + + class_count = hdr.getInt(); + if (class_count < 0) + unpack_abort("bad class_count"); + + hdrVals += 4; + + // done with archive_header + hdrVals += hdrValsSkipped; + assert(hdrVals == AH_LENGTH); + + rp = hdr.rp; + if (rp > rplimit) + unpack_abort("EOF reading archive header"); + + // Now size the CP. + cp.init(this, cp_counts); + + default_file_modtime = archive_modtime; + if (default_file_modtime == 0 && !(archive_options & AO_HAVE_FILE_MODTIME)) + default_file_modtime = DEFAULT_ARCHIVE_MODTIME; // taken from driver + if ((archive_options & AO_DEFLATE_HINT) != 0) + default_file_options |= FO_DEFLATE_HINT; + + // meta-bytes, if any, immediately follow archive header + // band_headers.readData(band_headers_size); + ensure_input(band_headers_size); + if (input_remaining() < (size_t)band_headers_size) + { + unpack_abort("EOF reading band headers"); + return; + } + bytes band_headers; + // The "1+" allows an initial byte to be pushed on the front. + band_headers.set(1 + U_NEW(byte, 1 + band_headers_size + C_SLOP), band_headers_size); + + // Start scanning band headers here: + band_headers.copyFrom(rp, band_headers.len); + rp += band_headers.len; + assert(rp <= rplimit); + meta_rp = band_headers.ptr; + // Put evil meta-codes at the end of the band headers, + // so we are sure to throw an error if we run off the end. + bytes::of(band_headers.limit(), C_SLOP).clear(_meta_error); } void unpacker::finish() { - if (verbose >= 1) - { - fprintf(stderr, "A total of " LONG_LONG_FORMAT " bytes were read in %d segment(s).\n", - (bytes_read_before_reset + bytes_read), segments_read_before_reset + 1); - fprintf(stderr, "A total of " LONG_LONG_FORMAT " file content bytes were written.\n", - (bytes_written_before_reset + bytes_written)); - fprintf(stderr, - "A total of %d files (of which %d are classes) were written to output.\n", - files_written_before_reset + files_written, - classes_written_before_reset + classes_written); - } - if (jarout != nullptr) - jarout->closeJarFile(true); + if (verbose >= 1) + { + fprintf(stderr, "A total of %" PRIu64 " bytes were read in %d segment(s).\n", + (bytes_read_before_reset + bytes_read), segments_read_before_reset + 1); + fprintf(stderr, "A total of %" PRIu64 " file content bytes were written.\n", + (bytes_written_before_reset + bytes_written)); + fprintf(stderr, + "A total of %d files (of which %d are classes) were written to output.\n", + files_written_before_reset + files_written, + classes_written_before_reset + classes_written); + } + if (jarout != nullptr) + jarout->closeJarFile(true); } // Cf. PackageReader.readConstantPoolCounts void constant_pool::init(unpacker *u_, int counts[NUM_COUNTS]) { - this->u = u_; - - // Fill-pointer for CP. - int next_entry = 0; - - // Size the constant pool: - for (int k = 0; k < (int)N_TAGS_IN_ORDER; k++) - { - byte tag = TAGS_IN_ORDER[k]; - int len = counts[k]; - tag_count[tag] = len; - tag_base[tag] = next_entry; - next_entry += len; - // Detect and defend against constant pool size overflow. - // (Pack200 forbids the sum of CP counts to exceed 2^29-1.) - enum - { - CP_SIZE_LIMIT = (1 << 29), - IMPLICIT_ENTRY_COUNT = 1 // empty Utf8 string - }; - if (len >= (1 << 29) || len < 0 || next_entry >= CP_SIZE_LIMIT + IMPLICIT_ENTRY_COUNT) - { - unpack_abort("archive too large: constant pool limit exceeded"); - } - } - - // Close off the end of the CP: - nentries = next_entry; - - // place a limit on future CP growth: - int generous = 0; - generous = add_size(generous, u->ic_count); // implicit name - generous = add_size(generous, u->ic_count); // outer - generous = add_size(generous, u->ic_count); // outer.utf8 - generous = add_size(generous, 40); // WKUs, misc - generous = add_size(generous, u->class_count); // implicit SourceFile strings - maxentries = add_size(nentries, generous); - - // Note that this CP does not include "empty" entries - // for longs and doubles. Those are introduced when - // the entries are renumbered for classfile output. - - entries = U_NEW(entry, maxentries); - - first_extra_entry = &entries[nentries]; - - // Initialize the standard indexes. - tag_count[CONSTANT_All] = nentries; - tag_base[CONSTANT_All] = 0; - for (int tag = 0; tag < CONSTANT_Limit; tag++) - { - entry *cpMap = &entries[tag_base[tag]]; - tag_index[tag].init(tag_count[tag], cpMap, tag); - } - - // Initialize hashTab to a generous power-of-two size. - uint32_t pow2 = 1; - uint32_t target = maxentries + maxentries / 2; // 60% full - while (pow2 < target) - pow2 <<= 1; - hashTab = U_NEW(entry *, hashTabLength = pow2); + this->u = u_; + + // Fill-pointer for CP. + int next_entry = 0; + + // Size the constant pool: + for (int k = 0; k < (int)N_TAGS_IN_ORDER; k++) + { + byte tag = TAGS_IN_ORDER[k]; + int len = counts[k]; + tag_count[tag] = len; + tag_base[tag] = next_entry; + next_entry += len; + // Detect and defend against constant pool size overflow. + // (Pack200 forbids the sum of CP counts to exceed 2^29-1.) + enum + { + CP_SIZE_LIMIT = (1 << 29), + IMPLICIT_ENTRY_COUNT = 1 // empty Utf8 string + }; + if (len >= (1 << 29) || len < 0 || next_entry >= CP_SIZE_LIMIT + IMPLICIT_ENTRY_COUNT) + { + unpack_abort("archive too large: constant pool limit exceeded"); + } + } + + // Close off the end of the CP: + nentries = next_entry; + + // place a limit on future CP growth: + int generous = 0; + generous = add_size(generous, u->ic_count); // implicit name + generous = add_size(generous, u->ic_count); // outer + generous = add_size(generous, u->ic_count); // outer.utf8 + generous = add_size(generous, 40); // WKUs, misc + generous = add_size(generous, u->class_count); // implicit SourceFile strings + maxentries = add_size(nentries, generous); + + // Note that this CP does not include "empty" entries + // for longs and doubles. Those are introduced when + // the entries are renumbered for classfile output. + + entries = U_NEW(entry, maxentries); + + first_extra_entry = &entries[nentries]; + + // Initialize the standard indexes. + tag_count[CONSTANT_All] = nentries; + tag_base[CONSTANT_All] = 0; + for (int tag = 0; tag < CONSTANT_Limit; tag++) + { + entry *cpMap = &entries[tag_base[tag]]; + tag_index[tag].init(tag_count[tag], cpMap, tag); + } + + // Initialize hashTab to a generous power-of-two size. + uint32_t pow2 = 1; + uint32_t target = maxentries + maxentries / 2; // 60% full + while (pow2 < target) + pow2 <<= 1; + hashTab = U_NEW(entry *, hashTabLength = pow2); } static byte *store_Utf8_char(byte *cp, unsigned short ch) { - if (ch >= 0x001 && ch <= 0x007F) - { - *cp++ = (byte)ch; - } - else if (ch <= 0x07FF) - { - *cp++ = (byte)(0xC0 | ((ch >> 6) & 0x1F)); - *cp++ = (byte)(0x80 | ((ch >> 0) & 0x3F)); - } - else - { - *cp++ = (byte)(0xE0 | ((ch >> 12) & 0x0F)); - *cp++ = (byte)(0x80 | ((ch >> 6) & 0x3F)); - *cp++ = (byte)(0x80 | ((ch >> 0) & 0x3F)); - } - return cp; + if (ch >= 0x001 && ch <= 0x007F) + { + *cp++ = (byte)ch; + } + else if (ch <= 0x07FF) + { + *cp++ = (byte)(0xC0 | ((ch >> 6) & 0x1F)); + *cp++ = (byte)(0x80 | ((ch >> 0) & 0x3F)); + } + else + { + *cp++ = (byte)(0xE0 | ((ch >> 12) & 0x0F)); + *cp++ = (byte)(0x80 | ((ch >> 6) & 0x3F)); + *cp++ = (byte)(0x80 | ((ch >> 0) & 0x3F)); + } + return cp; } static byte *skip_Utf8_chars(byte *cp, int len) { - for (;; cp++) - { - int ch = *cp & 0xFF; - if ((ch & 0xC0) != 0x80) - { - if (len-- == 0) - return cp; - if (ch < 0x80 && len == 0) - return cp + 1; - } - } + for (;; cp++) + { + int ch = *cp & 0xFF; + if ((ch & 0xC0) != 0x80) + { + if (len-- == 0) + return cp; + if (ch < 0x80 && len == 0) + return cp + 1; + } + } } static int compare_Utf8_chars(bytes &b1, bytes &b2) { - int l1 = (int)b1.len; - int l2 = (int)b2.len; - int l0 = (l1 < l2) ? l1 : l2; - byte *p1 = b1.ptr; - byte *p2 = b2.ptr; - int c0 = 0; - for (int i = 0; i < l0; i++) - { - int c1 = p1[i] & 0xFF; - int c2 = p2[i] & 0xFF; - if (c1 != c2) - { - // Before returning the obvious answer, - // check to see if c1 or c2 is part of a 0x0000, - // which encodes as {0xC0,0x80}. The 0x0000 is the - // lowest-sorting Java char value, and yet it encodes - // as if it were the first char after 0x7F, which causes - // strings containing nulls to sort too high. All other - // comparisons are consistent between Utf8 and Java chars. - if (c1 == 0xC0 && (p1[i + 1] & 0xFF) == 0x80) - c1 = 0; - if (c2 == 0xC0 && (p2[i + 1] & 0xFF) == 0x80) - c2 = 0; - if (c0 == 0xC0) - { - assert(((c1 | c2) & 0xC0) == 0x80); // c1 & c2 are extension chars - if (c1 == 0x80) - c1 = 0; // will sort below c2 - if (c2 == 0x80) - c2 = 0; // will sort below c1 - } - return c1 - c2; - } - c0 = c1; // save away previous char - } - // common prefix is identical; return length difference if any - return l1 - l2; + int l1 = (int)b1.len; + int l2 = (int)b2.len; + int l0 = (l1 < l2) ? l1 : l2; + byte *p1 = b1.ptr; + byte *p2 = b2.ptr; + int c0 = 0; + for (int i = 0; i < l0; i++) + { + int c1 = p1[i] & 0xFF; + int c2 = p2[i] & 0xFF; + if (c1 != c2) + { + // Before returning the obvious answer, + // check to see if c1 or c2 is part of a 0x0000, + // which encodes as {0xC0,0x80}. The 0x0000 is the + // lowest-sorting Java char value, and yet it encodes + // as if it were the first char after 0x7F, which causes + // strings containing nulls to sort too high. All other + // comparisons are consistent between Utf8 and Java chars. + if (c1 == 0xC0 && (p1[i + 1] & 0xFF) == 0x80) + c1 = 0; + if (c2 == 0xC0 && (p2[i + 1] & 0xFF) == 0x80) + c2 = 0; + if (c0 == 0xC0) + { + assert(((c1 | c2) & 0xC0) == 0x80); // c1 & c2 are extension chars + if (c1 == 0x80) + c1 = 0; // will sort below c2 + if (c2 == 0x80) + c2 = 0; // will sort below c1 + } + return c1 - c2; + } + c0 = c1; // save away previous char + } + // common prefix is identical; return length difference if any + return l1 - l2; } // Cf. PackageReader.readUtf8Bands void unpacker::read_Utf8_values(entry *cpMap, int len) { - // Implicit first Utf8 string is the empty string. - enum - { - // certain bands begin with implicit zeroes - PREFIX_SKIP_2 = 2, - SUFFIX_SKIP_1 = 1 - }; - - int i; - - // First band: Read lengths of shared prefixes. - if (len > PREFIX_SKIP_2) - cp_Utf8_prefix.readData(len - PREFIX_SKIP_2); - - // Second band: Read lengths of unshared suffixes: - if (len > SUFFIX_SKIP_1) - cp_Utf8_suffix.readData(len - SUFFIX_SKIP_1); - - bytes *allsuffixes = T_NEW(bytes, len); - - int nbigsuf = 0; - fillbytes charbuf; // buffer to allocate small strings - charbuf.init(); - - // Third band: Read the char values in the unshared suffixes: - cp_Utf8_chars.readData(cp_Utf8_suffix.getIntTotal()); - for (i = 0; i < len; i++) - { - int suffix = (i < SUFFIX_SKIP_1) ? 0 : cp_Utf8_suffix.getInt(); - if (suffix < 0) - { - unpack_abort("bad utf8 suffix"); - } - if (suffix == 0 && i >= SUFFIX_SKIP_1) - { - // chars are packed in cp_Utf8_big_chars - nbigsuf += 1; - continue; - } - bytes &chars = allsuffixes[i]; - uint32_t size3 = suffix * 3; // max Utf8 length - bool isMalloc = (suffix > SMALL); - if (isMalloc) - { - chars.malloc(size3); - } - else - { - if (!charbuf.canAppend(size3 + 1)) - { - assert(charbuf.allocated == 0 || tmallocs.contains(charbuf.base())); - charbuf.init(CHUNK); // Reset to new buffer. - tmallocs.add(charbuf.base()); - } - chars.set(charbuf.grow(size3 + 1), size3); - } - - byte *chp = chars.ptr; - for (int j = 0; j < suffix; j++) - { - unsigned short ch = cp_Utf8_chars.getInt(); - chp = store_Utf8_char(chp, ch); - } - // shrink to fit: - if (isMalloc) - { - chars.realloc(chp - chars.ptr); - tmallocs.add(chars.ptr); // free it later - } - else - { - int shrink = (int)(chars.limit() - chp); - chars.len -= shrink; - charbuf.b.len -= shrink; // ungrow to reclaim buffer space - // Note that we did not reclaim the final '\0'. - assert(chars.limit() == charbuf.limit() - 1); - assert(strlen((char *)chars.ptr) == chars.len); - } - } - // cp_Utf8_chars.done(); - - // Fourth band: Go back and size the specially packed strings. - int maxlen = 0; - cp_Utf8_big_suffix.readData(nbigsuf); - cp_Utf8_suffix.rewind(); - for (i = 0; i < len; i++) - { - int suffix = (i < SUFFIX_SKIP_1) ? 0 : cp_Utf8_suffix.getInt(); - int prefix = (i < PREFIX_SKIP_2) ? 0 : cp_Utf8_prefix.getInt(); - if (prefix < 0 || prefix + suffix < 0) - { - unpack_abort("bad utf8 prefix"); - } - bytes &chars = allsuffixes[i]; - if (suffix == 0 && i >= SUFFIX_SKIP_1) - { - suffix = cp_Utf8_big_suffix.getInt(); - assert(chars.ptr == nullptr); - chars.len = suffix; // just a momentary hack - } - else - { - assert(chars.ptr != nullptr); - } - if (maxlen < prefix + suffix) - { - maxlen = prefix + suffix; - } - } - // cp_Utf8_suffix.done(); // will use allsuffixes[i].len (ptr!=nullptr) - // cp_Utf8_big_suffix.done(); // will use allsuffixes[i].len - - // Fifth band(s): Get the specially packed characters. - cp_Utf8_big_suffix.rewind(); - for (i = 0; i < len; i++) - { - bytes &chars = allsuffixes[i]; - if (chars.ptr != nullptr) - continue; // already input - int suffix = (int)chars.len; // pick up the hack - uint32_t size3 = suffix * 3; - if (suffix == 0) - continue; // done with empty string - chars.malloc(size3); - byte *chp = chars.ptr; - band saved_band = cp_Utf8_big_chars; - cp_Utf8_big_chars.readData(suffix); - for (int j = 0; j < suffix; j++) - { - unsigned short ch = cp_Utf8_big_chars.getInt(); - chp = store_Utf8_char(chp, ch); - } - chars.realloc(chp - chars.ptr); - tmallocs.add(chars.ptr); // free it later - // cp_Utf8_big_chars.done(); - cp_Utf8_big_chars = saved_band; // reset the band for the next string - } - cp_Utf8_big_chars.readData(0); // zero chars - // cp_Utf8_big_chars.done(); - - // Finally, sew together all the prefixes and suffixes. - bytes bigbuf; - bigbuf.malloc(maxlen * 3 + 1); // max Utf8 length, plus slop for nullptr - int prevlen = 0; // previous string length (in chars) - tmallocs.add(bigbuf.ptr); // free after this block - cp_Utf8_prefix.rewind(); - for (i = 0; i < len; i++) - { - bytes &chars = allsuffixes[i]; - int prefix = (i < PREFIX_SKIP_2) ? 0 : cp_Utf8_prefix.getInt(); - int suffix = (int)chars.len; - byte *fillp; - // by induction, the buffer is already filled with the prefix - // make sure the prefix value is not corrupted, though: - if (prefix > prevlen) - { - unpack_abort("utf8 prefix overflow"); - return; - } - fillp = skip_Utf8_chars(bigbuf.ptr, prefix); - // copy the suffix into the same buffer: - fillp = chars.writeTo(fillp); - assert(bigbuf.inBounds(fillp)); - *fillp = 0; // bigbuf must contain a well-formed Utf8 string - int length = (int)(fillp - bigbuf.ptr); - bytes &value = cpMap[i].value.b; - value.set(U_NEW(byte, add_size(length, 1)), length); - value.copyFrom(bigbuf.ptr, length); - // Index all Utf8 strings - entry *&htref = cp.hashTabRef(CONSTANT_Utf8, value); - if (htref == nullptr) - { - // Note that if two identical strings are transmitted, - // the first is taken to be the canonical one. - htref = &cpMap[i]; - } - prevlen = prefix + suffix; - } - // cp_Utf8_prefix.done(); - - // Free intermediate buffers. - free_temps(); + // Implicit first Utf8 string is the empty string. + enum + { + // certain bands begin with implicit zeroes + PREFIX_SKIP_2 = 2, + SUFFIX_SKIP_1 = 1 + }; + + int i; + + // First band: Read lengths of shared prefixes. + if (len > PREFIX_SKIP_2) + cp_Utf8_prefix.readData(len - PREFIX_SKIP_2); + + // Second band: Read lengths of unshared suffixes: + if (len > SUFFIX_SKIP_1) + cp_Utf8_suffix.readData(len - SUFFIX_SKIP_1); + + bytes *allsuffixes = T_NEW(bytes, len); + + int nbigsuf = 0; + fillbytes charbuf; // buffer to allocate small strings + charbuf.init(); + + // Third band: Read the char values in the unshared suffixes: + cp_Utf8_chars.readData(cp_Utf8_suffix.getIntTotal()); + for (i = 0; i < len; i++) + { + int suffix = (i < SUFFIX_SKIP_1) ? 0 : cp_Utf8_suffix.getInt(); + if (suffix < 0) + { + unpack_abort("bad utf8 suffix"); + } + if (suffix == 0 && i >= SUFFIX_SKIP_1) + { + // chars are packed in cp_Utf8_big_chars + nbigsuf += 1; + continue; + } + bytes &chars = allsuffixes[i]; + uint32_t size3 = suffix * 3; // max Utf8 length + bool isMalloc = (suffix > SMALL); + if (isMalloc) + { + chars.malloc(size3); + } + else + { + if (!charbuf.canAppend(size3 + 1)) + { + assert(charbuf.allocated == 0 || tmallocs.contains(charbuf.base())); + charbuf.init(CHUNK); // Reset to new buffer. + tmallocs.add(charbuf.base()); + } + chars.set(charbuf.grow(size3 + 1), size3); + } + + byte *chp = chars.ptr; + for (int j = 0; j < suffix; j++) + { + unsigned short ch = cp_Utf8_chars.getInt(); + chp = store_Utf8_char(chp, ch); + } + // shrink to fit: + if (isMalloc) + { + chars.realloc(chp - chars.ptr); + tmallocs.add(chars.ptr); // free it later + } + else + { + int shrink = (int)(chars.limit() - chp); + chars.len -= shrink; + charbuf.b.len -= shrink; // ungrow to reclaim buffer space + // Note that we did not reclaim the final '\0'. + assert(chars.limit() == charbuf.limit() - 1); + assert(strlen((char *)chars.ptr) == chars.len); + } + } + // cp_Utf8_chars.done(); + + // Fourth band: Go back and size the specially packed strings. + int maxlen = 0; + cp_Utf8_big_suffix.readData(nbigsuf); + cp_Utf8_suffix.rewind(); + for (i = 0; i < len; i++) + { + int suffix = (i < SUFFIX_SKIP_1) ? 0 : cp_Utf8_suffix.getInt(); + int prefix = (i < PREFIX_SKIP_2) ? 0 : cp_Utf8_prefix.getInt(); + if (prefix < 0 || prefix + suffix < 0) + { + unpack_abort("bad utf8 prefix"); + } + bytes &chars = allsuffixes[i]; + if (suffix == 0 && i >= SUFFIX_SKIP_1) + { + suffix = cp_Utf8_big_suffix.getInt(); + assert(chars.ptr == nullptr); + chars.len = suffix; // just a momentary hack + } + else + { + assert(chars.ptr != nullptr); + } + if (maxlen < prefix + suffix) + { + maxlen = prefix + suffix; + } + } + // cp_Utf8_suffix.done(); // will use allsuffixes[i].len (ptr!=nullptr) + // cp_Utf8_big_suffix.done(); // will use allsuffixes[i].len + + // Fifth band(s): Get the specially packed characters. + cp_Utf8_big_suffix.rewind(); + for (i = 0; i < len; i++) + { + bytes &chars = allsuffixes[i]; + if (chars.ptr != nullptr) + continue; // already input + int suffix = (int)chars.len; // pick up the hack + uint32_t size3 = suffix * 3; + if (suffix == 0) + continue; // done with empty string + chars.malloc(size3); + byte *chp = chars.ptr; + band saved_band = cp_Utf8_big_chars; + cp_Utf8_big_chars.readData(suffix); + for (int j = 0; j < suffix; j++) + { + unsigned short ch = cp_Utf8_big_chars.getInt(); + chp = store_Utf8_char(chp, ch); + } + chars.realloc(chp - chars.ptr); + tmallocs.add(chars.ptr); // free it later + // cp_Utf8_big_chars.done(); + cp_Utf8_big_chars = saved_band; // reset the band for the next string + } + cp_Utf8_big_chars.readData(0); // zero chars + // cp_Utf8_big_chars.done(); + + // Finally, sew together all the prefixes and suffixes. + bytes bigbuf; + bigbuf.malloc(maxlen * 3 + 1); // max Utf8 length, plus slop for nullptr + int prevlen = 0; // previous string length (in chars) + tmallocs.add(bigbuf.ptr); // free after this block + cp_Utf8_prefix.rewind(); + for (i = 0; i < len; i++) + { + bytes &chars = allsuffixes[i]; + int prefix = (i < PREFIX_SKIP_2) ? 0 : cp_Utf8_prefix.getInt(); + int suffix = (int)chars.len; + byte *fillp; + // by induction, the buffer is already filled with the prefix + // make sure the prefix value is not corrupted, though: + if (prefix > prevlen) + { + unpack_abort("utf8 prefix overflow"); + return; + } + fillp = skip_Utf8_chars(bigbuf.ptr, prefix); + // copy the suffix into the same buffer: + fillp = chars.writeTo(fillp); + assert(bigbuf.inBounds(fillp)); + *fillp = 0; // bigbuf must contain a well-formed Utf8 string + int length = (int)(fillp - bigbuf.ptr); + bytes &value = cpMap[i].value.b; + value.set(U_NEW(byte, add_size(length, 1)), length); + value.copyFrom(bigbuf.ptr, length); + // Index all Utf8 strings + entry *&htref = cp.hashTabRef(CONSTANT_Utf8, value); + if (htref == nullptr) + { + // Note that if two identical strings are transmitted, + // the first is taken to be the canonical one. + htref = &cpMap[i]; + } + prevlen = prefix + suffix; + } + // cp_Utf8_prefix.done(); + + // Free intermediate buffers. + free_temps(); } void unpacker::read_single_words(band &cp_band, entry *cpMap, int len) { - cp_band.readData(len); - for (int i = 0; i < len; i++) - { - cpMap[i].value.i = cp_band.getInt(); // coding handles signs OK - } + cp_band.readData(len); + for (int i = 0; i < len; i++) + { + cpMap[i].value.i = cp_band.getInt(); // coding handles signs OK + } } void unpacker::read_double_words(band &cp_bands, entry *cpMap, int len) { - band &cp_band_hi = cp_bands; - band &cp_band_lo = cp_bands.nextBand(); - cp_band_hi.readData(len); - cp_band_lo.readData(len); - for (int i = 0; i < len; i++) - { - cpMap[i].value.l = cp_band_hi.getLong(cp_band_lo, true); - } - // cp_band_hi.done(); - // cp_band_lo.done(); + band &cp_band_hi = cp_bands; + band &cp_band_lo = cp_bands.nextBand(); + cp_band_hi.readData(len); + cp_band_lo.readData(len); + for (int i = 0; i < len; i++) + { + cpMap[i].value.l = cp_band_hi.getLong(cp_band_lo, true); + } + // cp_band_hi.done(); + // cp_band_lo.done(); } void unpacker::read_single_refs(band &cp_band, byte refTag, entry *cpMap, int len) { - assert(refTag == CONSTANT_Utf8); - cp_band.setIndexByTag(refTag); - cp_band.readData(len); - int indexTag = (cp_band.bn == e_cp_Class) ? CONSTANT_Class : 0; - for (int i = 0; i < len; i++) - { - entry &e = cpMap[i]; - e.refs = U_NEW(entry *, e.nrefs = 1); - entry *utf = cp_band.getRef(); - e.refs[0] = utf; - e.value.b = utf->value.b; // copy value of Utf8 string to self - if (indexTag != 0) - { - // Maintain cross-reference: - entry *&htref = cp.hashTabRef(indexTag, e.value.b); - if (htref == nullptr) - { - // Note that if two identical classes are transmitted, - // the first is taken to be the canonical one. - htref = &e; - } - } - } - // cp_band.done(); + assert(refTag == CONSTANT_Utf8); + cp_band.setIndexByTag(refTag); + cp_band.readData(len); + int indexTag = (cp_band.bn == e_cp_Class) ? CONSTANT_Class : 0; + for (int i = 0; i < len; i++) + { + entry &e = cpMap[i]; + e.refs = U_NEW(entry *, e.nrefs = 1); + entry *utf = cp_band.getRef(); + e.refs[0] = utf; + e.value.b = utf->value.b; // copy value of Utf8 string to self + if (indexTag != 0) + { + // Maintain cross-reference: + entry *&htref = cp.hashTabRef(indexTag, e.value.b); + if (htref == nullptr) + { + // Note that if two identical classes are transmitted, + // the first is taken to be the canonical one. + htref = &e; + } + } + } + // cp_band.done(); } void unpacker::read_double_refs(band &cp_band, byte ref1Tag, byte ref2Tag, entry *cpMap, - int len) + int len) { - band &cp_band1 = cp_band; - band &cp_band2 = cp_band.nextBand(); - cp_band1.setIndexByTag(ref1Tag); - cp_band2.setIndexByTag(ref2Tag); - cp_band1.readData(len); - cp_band2.readData(len); - for (int i = 0; i < len; i++) - { - entry &e = cpMap[i]; - e.refs = U_NEW(entry *, e.nrefs = 2); - e.refs[0] = cp_band1.getRef(); - e.refs[1] = cp_band2.getRef(); - } - // cp_band1.done(); - // cp_band2.done(); + band &cp_band1 = cp_band; + band &cp_band2 = cp_band.nextBand(); + cp_band1.setIndexByTag(ref1Tag); + cp_band2.setIndexByTag(ref2Tag); + cp_band1.readData(len); + cp_band2.readData(len); + for (int i = 0; i < len; i++) + { + entry &e = cpMap[i]; + e.refs = U_NEW(entry *, e.nrefs = 2); + e.refs[0] = cp_band1.getRef(); + e.refs[1] = cp_band2.getRef(); + } + // cp_band1.done(); + // cp_band2.done(); } // Cf. PackageReader.readSignatureBands void unpacker::read_signature_values(entry *cpMap, int len) { - cp_Signature_form.setIndexByTag(CONSTANT_Utf8); - cp_Signature_form.readData(len); - int ncTotal = 0; - int i; - for (i = 0; i < len; i++) - { - entry &e = cpMap[i]; - entry &form = *cp_Signature_form.getRef(); - int nc = 0; - - for (const char *ncp = form.utf8String(); *ncp; ncp++) - { - if (*ncp == 'L') - nc++; - } - - ncTotal += nc; - e.refs = U_NEW(entry *, cpMap[i].nrefs = 1 + nc); - e.refs[0] = &form; - } - // cp_Signature_form.done(); - cp_Signature_classes.setIndexByTag(CONSTANT_Class); - cp_Signature_classes.readData(ncTotal); - for (i = 0; i < len; i++) - { - entry &e = cpMap[i]; - for (int j = 1; j < e.nrefs; j++) - { - e.refs[j] = cp_Signature_classes.getRef(); - } - } - // cp_Signature_classes.done(); + cp_Signature_form.setIndexByTag(CONSTANT_Utf8); + cp_Signature_form.readData(len); + int ncTotal = 0; + int i; + for (i = 0; i < len; i++) + { + entry &e = cpMap[i]; + entry &form = *cp_Signature_form.getRef(); + int nc = 0; + + for (const char *ncp = form.utf8String(); *ncp; ncp++) + { + if (*ncp == 'L') + nc++; + } + + ncTotal += nc; + e.refs = U_NEW(entry *, cpMap[i].nrefs = 1 + nc); + e.refs[0] = &form; + } + // cp_Signature_form.done(); + cp_Signature_classes.setIndexByTag(CONSTANT_Class); + cp_Signature_classes.readData(ncTotal); + for (i = 0; i < len; i++) + { + entry &e = cpMap[i]; + for (int j = 1; j < e.nrefs; j++) + { + e.refs[j] = cp_Signature_classes.getRef(); + } + } + // cp_Signature_classes.done(); } // Cf. PackageReader.readConstantPool void unpacker::read_cp() { - int i; - - for (int k = 0; k < (int)N_TAGS_IN_ORDER; k++) - { - byte tag = TAGS_IN_ORDER[k]; - int len = cp.tag_count[tag]; - int base = cp.tag_base[tag]; - - entry *cpMap = &cp.entries[base]; - for (i = 0; i < len; i++) - { - cpMap[i].tag = tag; - cpMap[i].inord = i; - } - - switch (tag) - { - case CONSTANT_Utf8: - read_Utf8_values(cpMap, len); - break; - case CONSTANT_Integer: - read_single_words(cp_Int, cpMap, len); - break; - case CONSTANT_Float: - read_single_words(cp_Float, cpMap, len); - break; - case CONSTANT_Long: - read_double_words(cp_Long_hi /*& cp_Long_lo*/, cpMap, len); - break; - case CONSTANT_Double: - read_double_words(cp_Double_hi /*& cp_Double_lo*/, cpMap, len); - break; - case CONSTANT_String: - read_single_refs(cp_String, CONSTANT_Utf8, cpMap, len); - break; - case CONSTANT_Class: - read_single_refs(cp_Class, CONSTANT_Utf8, cpMap, len); - break; - case CONSTANT_Signature: - read_signature_values(cpMap, len); - break; - case CONSTANT_NameandType: - read_double_refs(cp_Descr_name /*& cp_Descr_type*/, CONSTANT_Utf8, - CONSTANT_Signature, cpMap, len); - break; - case CONSTANT_Fieldref: - read_double_refs(cp_Field_class /*& cp_Field_desc*/, CONSTANT_Class, - CONSTANT_NameandType, cpMap, len); - break; - case CONSTANT_Methodref: - read_double_refs(cp_Method_class /*& cp_Method_desc*/, CONSTANT_Class, - CONSTANT_NameandType, cpMap, len); - break; - case CONSTANT_InterfaceMethodref: - read_double_refs(cp_Imethod_class /*& cp_Imethod_desc*/, CONSTANT_Class, - CONSTANT_NameandType, cpMap, len); - break; - default: - assert(false); - break; - } - } - - cp.expandSignatures(); - cp.initMemberIndexes(); + int i; + + for (int k = 0; k < (int)N_TAGS_IN_ORDER; k++) + { + byte tag = TAGS_IN_ORDER[k]; + int len = cp.tag_count[tag]; + int base = cp.tag_base[tag]; + + entry *cpMap = &cp.entries[base]; + for (i = 0; i < len; i++) + { + cpMap[i].tag = tag; + cpMap[i].inord = i; + } + + switch (tag) + { + case CONSTANT_Utf8: + read_Utf8_values(cpMap, len); + break; + case CONSTANT_Integer: + read_single_words(cp_Int, cpMap, len); + break; + case CONSTANT_Float: + read_single_words(cp_Float, cpMap, len); + break; + case CONSTANT_Long: + read_double_words(cp_Long_hi /*& cp_Long_lo*/, cpMap, len); + break; + case CONSTANT_Double: + read_double_words(cp_Double_hi /*& cp_Double_lo*/, cpMap, len); + break; + case CONSTANT_String: + read_single_refs(cp_String, CONSTANT_Utf8, cpMap, len); + break; + case CONSTANT_Class: + read_single_refs(cp_Class, CONSTANT_Utf8, cpMap, len); + break; + case CONSTANT_Signature: + read_signature_values(cpMap, len); + break; + case CONSTANT_NameandType: + read_double_refs(cp_Descr_name /*& cp_Descr_type*/, CONSTANT_Utf8, + CONSTANT_Signature, cpMap, len); + break; + case CONSTANT_Fieldref: + read_double_refs(cp_Field_class /*& cp_Field_desc*/, CONSTANT_Class, + CONSTANT_NameandType, cpMap, len); + break; + case CONSTANT_Methodref: + read_double_refs(cp_Method_class /*& cp_Method_desc*/, CONSTANT_Class, + CONSTANT_NameandType, cpMap, len); + break; + case CONSTANT_InterfaceMethodref: + read_double_refs(cp_Imethod_class /*& cp_Imethod_desc*/, CONSTANT_Class, + CONSTANT_NameandType, cpMap, len); + break; + default: + assert(false); + break; + } + } + + cp.expandSignatures(); + cp.initMemberIndexes(); #define SNAME(n, s) #s "\0" - const char *symNames = (ALL_ATTR_DO(SNAME) "<init>"); + const char *symNames = (ALL_ATTR_DO(SNAME) "<init>"); #undef SNAME - for (int sn = 0; sn < constant_pool::s_LIMIT; sn++) - { - assert(symNames[0] >= '0' && symNames[0] <= 'Z'); // sanity - bytes name; - name.set(symNames); - if (name.len > 0 && name.ptr[0] != '0') - { - cp.sym[sn] = cp.ensureUtf8(name); - } - symNames += name.len + 1; // skip trailing nullptr to next name - } - - band::initIndexes(this); + for (int sn = 0; sn < constant_pool::s_LIMIT; sn++) + { + assert(symNames[0] >= '0' && symNames[0] <= 'Z'); // sanity + bytes name; + name.set(symNames); + if (name.len > 0 && name.ptr[0] != '0') + { + cp.sym[sn] = cp.ensureUtf8(name); + } + symNames += name.len + 1; // skip trailing nullptr to next name + } + + band::initIndexes(this); } static band *no_bands[] = {nullptr}; // shared empty body inline band &unpacker::attr_definitions::fixed_band(int e_class_xxx) { - return u->all_bands[xxx_flags_hi_bn + (e_class_xxx - e_class_flags_hi)]; + return u->all_bands[xxx_flags_hi_bn + (e_class_xxx - e_class_flags_hi)]; } inline band &unpacker::attr_definitions::xxx_flags_hi() { - return fixed_band(e_class_flags_hi); + return fixed_band(e_class_flags_hi); } inline band &unpacker::attr_definitions::xxx_flags_lo() { - return fixed_band(e_class_flags_lo); + return fixed_band(e_class_flags_lo); } inline band &unpacker::attr_definitions::xxx_attr_count() { - return fixed_band(e_class_attr_count); + return fixed_band(e_class_attr_count); } inline band &unpacker::attr_definitions::xxx_attr_indexes() { - return fixed_band(e_class_attr_indexes); + return fixed_band(e_class_attr_indexes); } inline band &unpacker::attr_definitions::xxx_attr_calls() { - return fixed_band(e_class_attr_calls); + return fixed_band(e_class_attr_calls); } inline unpacker::layout_definition * unpacker::attr_definitions::defineLayout(int idx, entry *nameEntry, const char *layout) { - const char *name = nameEntry->value.b.strval(); - layout_definition *lo = defineLayout(idx, name, layout); - lo->nameEntry = nameEntry; - return lo; + const char *name = nameEntry->value.b.strval(); + layout_definition *lo = defineLayout(idx, name, layout); + lo->nameEntry = nameEntry; + return lo; } unpacker::layout_definition *unpacker::attr_definitions::defineLayout(int idx, const char *name, - const char *layout) + const char *layout) { - assert(flag_limit != 0); // must be set up already - if (idx >= 0) - { - // Fixed attr. - if (idx >= (int)flag_limit) - unpack_abort("attribute index too large"); - if (isRedefined(idx)) - unpack_abort("redefined attribute index"); - redef |= ((uint64_t)1 << idx); - } - else - { - idx = flag_limit + overflow_count.length(); - overflow_count.add(0); // make a new counter - } - layout_definition *lo = U_NEW(layout_definition, 1); - lo->idx = idx; - lo->name = name; - lo->layout = layout; - for (int adds = (idx + 1) - layouts.length(); adds > 0; adds--) - { - layouts.add(nullptr); - } - layouts.get(idx) = lo; - return lo; + assert(flag_limit != 0); // must be set up already + if (idx >= 0) + { + // Fixed attr. + if (idx >= (int)flag_limit) + unpack_abort("attribute index too large"); + if (isRedefined(idx)) + unpack_abort("redefined attribute index"); + redef |= ((uint64_t)1 << idx); + } + else + { + idx = flag_limit + overflow_count.length(); + overflow_count.add(0); // make a new counter + } + layout_definition *lo = U_NEW(layout_definition, 1); + lo->idx = idx; + lo->name = name; + lo->layout = layout; + for (int adds = (idx + 1) - layouts.length(); adds > 0; adds--) + { + layouts.add(nullptr); + } + layouts.get(idx) = lo; + return lo; } band **unpacker::attr_definitions::buildBands(unpacker::layout_definition *lo) { - int i; - if (lo->elems != nullptr) - return lo->bands(); - if (lo->layout[0] == '\0') - { - lo->elems = no_bands; - } - else - { - // Create bands for this attribute by parsing the layout. - bool hasCallables = lo->hasCallables(); - bands_made = 0x10000; // base number for bands made - const char *lp = lo->layout; - lp = parseLayout(lp, lo->elems, -1); - if (lp[0] != '\0' || band_stack.length() > 0) - { - unpack_abort("garbage at end of layout"); - } - band_stack.popTo(0); - - // Fix up callables to point at their callees. - band **bands = lo->elems; - assert(bands == lo->bands()); - int num_callables = 0; - if (hasCallables) - { - while (bands[num_callables] != nullptr) - { - if (bands[num_callables]->le_kind != EK_CBLE) - { - unpack_abort("garbage mixed with callables"); - break; - } - num_callables += 1; - } - } - for (i = 0; i < calls_to_link.length(); i++) - { - band &call = *(band *)calls_to_link.get(i); - assert(call.le_kind == EK_CALL); - // Determine the callee. - int call_num = call.le_len; - if (call_num < 0 || call_num >= num_callables) - { - unpack_abort("bad call in layout"); - break; - } - band &cble = *bands[call_num]; - // Link the call to it. - call.le_body[0] = &cble; - // Distinguish backward calls and callables: - assert(cble.le_kind == EK_CBLE); - // FIXME: hit this one - // assert(cble.le_len == call_num); - cble.le_back |= call.le_back; - } - calls_to_link.popTo(0); - } - return lo->elems; + int i; + if (lo->elems != nullptr) + return lo->bands(); + if (lo->layout[0] == '\0') + { + lo->elems = no_bands; + } + else + { + // Create bands for this attribute by parsing the layout. + bool hasCallables = lo->hasCallables(); + bands_made = 0x10000; // base number for bands made + const char *lp = lo->layout; + lp = parseLayout(lp, lo->elems, -1); + if (lp[0] != '\0' || band_stack.length() > 0) + { + unpack_abort("garbage at end of layout"); + } + band_stack.popTo(0); + + // Fix up callables to point at their callees. + band **bands = lo->elems; + assert(bands == lo->bands()); + int num_callables = 0; + if (hasCallables) + { + while (bands[num_callables] != nullptr) + { + if (bands[num_callables]->le_kind != EK_CBLE) + { + unpack_abort("garbage mixed with callables"); + break; + } + num_callables += 1; + } + } + for (i = 0; i < calls_to_link.length(); i++) + { + band &call = *(band *)calls_to_link.get(i); + assert(call.le_kind == EK_CALL); + // Determine the callee. + int call_num = call.le_len; + if (call_num < 0 || call_num >= num_callables) + { + unpack_abort("bad call in layout"); + break; + } + band &cble = *bands[call_num]; + // Link the call to it. + call.le_body[0] = &cble; + // Distinguish backward calls and callables: + assert(cble.le_kind == EK_CBLE); + // FIXME: hit this one + // assert(cble.le_len == call_num); + cble.le_back |= call.le_back; + } + calls_to_link.popTo(0); + } + return lo->elems; } /* attribute layout language parser attribute_layout: - ( layout_element )* | ( callable )+ + ( layout_element )* | ( callable )+ layout_element: - ( integral | replication | union | call | reference ) + ( integral | replication | union | call | reference ) callable: - '[' body ']' + '[' body ']' body: - ( layout_element )+ + ( layout_element )+ integral: - ( unsigned_int | signed_int | bc_index | bc_offset | flag ) + ( unsigned_int | signed_int | bc_index | bc_offset | flag ) unsigned_int: - uint_type + uint_type signed_int: - 'S' uint_type + 'S' uint_type any_int: - ( unsigned_int | signed_int ) + ( unsigned_int | signed_int ) bc_index: - ( 'P' uint_type | 'PO' uint_type ) + ( 'P' uint_type | 'PO' uint_type ) bc_offset: - 'O' any_int + 'O' any_int flag: - 'F' uint_type + 'F' uint_type uint_type: - ( 'B' | 'H' | 'I' | 'V' ) + ( 'B' | 'H' | 'I' | 'V' ) replication: - 'N' uint_type '[' body ']' + 'N' uint_type '[' body ']' union: - 'T' any_int (union_case)* '(' ')' '[' (body)? ']' + 'T' any_int (union_case)* '(' ')' '[' (body)? ']' union_case: - '(' union_case_tag (',' union_case_tag)* ')' '[' (body)? ']' + '(' union_case_tag (',' union_case_tag)* ')' '[' (body)? ']' union_case_tag: - ( numeral | numeral '-' numeral ) + ( numeral | numeral '-' numeral ) call: - '(' numeral ')' + '(' numeral ')' reference: - reference_type ( 'N' )? uint_type + reference_type ( 'N' )? uint_type reference_type: - ( constant_ref | schema_ref | utf8_ref | untyped_ref ) + ( constant_ref | schema_ref | utf8_ref | untyped_ref ) constant_ref: - ( 'KI' | 'KJ' | 'KF' | 'KD' | 'KS' | 'KQ' ) + ( 'KI' | 'KJ' | 'KF' | 'KD' | 'KS' | 'KQ' ) schema_ref: - ( 'RC' | 'RS' | 'RD' | 'RF' | 'RM' | 'RI' ) + ( 'RC' | 'RS' | 'RD' | 'RF' | 'RM' | 'RI' ) utf8_ref: - 'RU' + 'RU' untyped_ref: - 'RQ' + 'RQ' numeral: - '(' ('-')? (digit)+ ')' + '(' ('-')? (digit)+ ')' digit: - ( '0' | '1' | '2' | '3' | '4' | '5' | '6' | '7' | '8' | '9' ) + ( '0' | '1' | '2' | '3' | '4' | '5' | '6' | '7' | '8' | '9' ) */ const char *unpacker::attr_definitions::parseIntLayout(const char *lp, band *&res, byte le_kind, - bool can_be_signed) + bool can_be_signed) { - band *b = U_NEW(band, 1); - char le = *lp++; - int spec = UNSIGNED5_spec; - if (le == 'S' && can_be_signed) - { - // Note: This is the last use of sign. There is no 'EF_SIGN'. - spec = SIGNED5_spec; - le = *lp++; - } - else if (le == 'B') - { - spec = BYTE1_spec; // unsigned byte - } - b->init(u, bands_made++, spec); - b->le_kind = le_kind; - int le_len = 0; - switch (le) - { - case 'B': - le_len = 1; - break; - case 'H': - le_len = 2; - break; - case 'I': - le_len = 4; - break; - case 'V': - le_len = 0; - break; - default: - unpack_abort("bad layout element"); - } - b->le_len = le_len; - band_stack.add(b); - res = b; - return lp; + band *b = U_NEW(band, 1); + char le = *lp++; + int spec = UNSIGNED5_spec; + if (le == 'S' && can_be_signed) + { + // Note: This is the last use of sign. There is no 'EF_SIGN'. + spec = SIGNED5_spec; + le = *lp++; + } + else if (le == 'B') + { + spec = BYTE1_spec; // unsigned byte + } + b->init(u, bands_made++, spec); + b->le_kind = le_kind; + int le_len = 0; + switch (le) + { + case 'B': + le_len = 1; + break; + case 'H': + le_len = 2; + break; + case 'I': + le_len = 4; + break; + case 'V': + le_len = 0; + break; + default: + unpack_abort("bad layout element"); + } + b->le_len = le_len; + band_stack.add(b); + res = b; + return lp; } const char *unpacker::attr_definitions::parseNumeral(const char *lp, int &res) { - bool sgn = false; - if (*lp == '0') - { - res = 0; - return lp + 1; - } // special case '0' - if (*lp == '-') - { - sgn = true; - lp++; - } - const char *dp = lp; - int con = 0; - while (*dp >= '0' && *dp <= '9') - { - int con0 = con; - con *= 10; - con += (*dp++) - '0'; - if (con <= con0) - { - con = -1; - break; - } // numeral overflow - } - if (lp == dp) - { - unpack_abort("missing numeral in layout"); - } - lp = dp; - if (con < 0 && !(sgn && con == -con)) - { - // (Portability note: Misses the error if int is not 32 bits.) - unpack_abort("numeral overflow"); - } - if (sgn) - con = -con; - res = con; - return lp; + bool sgn = false; + if (*lp == '0') + { + res = 0; + return lp + 1; + } // special case '0' + if (*lp == '-') + { + sgn = true; + lp++; + } + const char *dp = lp; + int con = 0; + while (*dp >= '0' && *dp <= '9') + { + int con0 = con; + con *= 10; + con += (*dp++) - '0'; + if (con <= con0) + { + con = -1; + break; + } // numeral overflow + } + if (lp == dp) + { + unpack_abort("missing numeral in layout"); + } + lp = dp; + if (con < 0 && !(sgn && con == -con)) + { + // (Portability note: Misses the error if int is not 32 bits.) + unpack_abort("numeral overflow"); + } + if (sgn) + con = -con; + res = con; + return lp; } band **unpacker::attr_definitions::popBody(int bs_base) { - // Return everything that was pushed, as a nullptr-terminated pointer array. - int bs_limit = band_stack.length(); - if (bs_base == bs_limit) - { - return no_bands; - } - else - { - int nb = bs_limit - bs_base; - band **res = U_NEW(band *, add_size(nb, 1)); - for (int i = 0; i < nb; i++) - { - band *b = (band *)band_stack.get(bs_base + i); - res[i] = b; - } - band_stack.popTo(bs_base); - return res; - } + // Return everything that was pushed, as a nullptr-terminated pointer array. + int bs_limit = band_stack.length(); + if (bs_base == bs_limit) + { + return no_bands; + } + else + { + int nb = bs_limit - bs_base; + band **res = U_NEW(band *, add_size(nb, 1)); + for (int i = 0; i < nb; i++) + { + band *b = (band *)band_stack.get(bs_base + i); + res[i] = b; + } + band_stack.popTo(bs_base); + return res; + } } const char *unpacker::attr_definitions::parseLayout(const char *lp, band **&res, int curCble) { - int bs_base = band_stack.length(); - bool top_level = (bs_base == 0); - band *b; - enum - { - can_be_signed = true - }; // optional arg to parseIntLayout - - for (bool done = false; !done;) - { - switch (*lp++) - { - case 'B': - case 'H': - case 'I': - case 'V': // unsigned_int - case 'S': // signed_int - --lp; // reparse - case 'F': - lp = parseIntLayout(lp, b, EK_INT); - break; - case 'P': - { - int le_bci = EK_BCI; - if (*lp == 'O') - { - ++lp; - le_bci = EK_BCID; - } - assert(*lp != 'S'); // no PSH, etc. - lp = parseIntLayout(lp, b, EK_INT); - b->le_bci = le_bci; - if (le_bci == EK_BCI) - b->defc = coding::findBySpec(BCI5_spec); - else - b->defc = coding::findBySpec(BRANCH5_spec); - } - break; - case 'O': - lp = parseIntLayout(lp, b, EK_INT, can_be_signed); - b->le_bci = EK_BCO; - b->defc = coding::findBySpec(BRANCH5_spec); - break; - case 'N': // replication: 'N' uint32_t '[' elem ... ']' - lp = parseIntLayout(lp, b, EK_REPL); - assert(*lp == '['); - ++lp; - lp = parseLayout(lp, b->le_body, curCble); - break; - case 'T': // union: 'T' any_int union_case* '(' ')' '[' body ']' - lp = parseIntLayout(lp, b, EK_UN, can_be_signed); - { - int union_base = band_stack.length(); - for (;;) - { // for each case - band &k_case = *U_NEW(band, 1); - band_stack.add(&k_case); - k_case.le_kind = EK_CASE; - k_case.bn = bands_made++; - if (*lp++ != '(') - { - unpack_abort("bad union case"); - return ""; - } - if (*lp++ != ')') - { - --lp; // reparse - // Read some case values. (Use band_stack for temp. storage.) - int case_base = band_stack.length(); - for (;;) - { - int caseval = 0; - lp = parseNumeral(lp, caseval); - band_stack.add((void *)(size_t)caseval); - if (*lp == '-') - { - // new in version 160, allow (1-5) for (1,2,3,4,5) - if (u->majver < JAVA6_PACKAGE_MAJOR_VERSION) - { - unpack_abort( - "bad range in union case label (old archive format)"); - return ""; - } - int caselimit = caseval; - lp++; - lp = parseNumeral(lp, caselimit); - if (caseval >= caselimit || - (uint32_t)(caselimit - caseval) > 0x10000) - { - // Note: 0x10000 is arbitrary implementation restriction. - // We can remove it later if it's important to. - unpack_abort("bad range in union case label"); - } - for (;;) - { - ++caseval; - band_stack.add((void *)(size_t)caseval); - if (caseval == caselimit) - break; - } - } - if (*lp != ',') - break; - lp++; - } - if (*lp++ != ')') - { - unpack_abort("bad case label"); - } - // save away the case labels - int ntags = band_stack.length() - case_base; - int *tags = U_NEW(int, add_size(ntags, 1)); - k_case.le_casetags = tags; - *tags++ = ntags; - for (int i = 0; i < ntags; i++) - { - *tags++ = ptrlowbits(band_stack.get(case_base + i)); - } - band_stack.popTo(case_base); - } - // Got le_casetags. Now grab the body. - assert(*lp == '['); - ++lp; - lp = parseLayout(lp, k_case.le_body, curCble); - if (k_case.le_casetags == nullptr) - break; // done - } - b->le_body = popBody(union_base); - } - break; - case '(': // call: '(' -?NN* ')' - { - band &call = *U_NEW(band, 1); - band_stack.add(&call); - call.le_kind = EK_CALL; - call.bn = bands_made++; - call.le_body = U_NEW(band *, 2); // fill in later - int call_num = 0; - lp = parseNumeral(lp, call_num); - call.le_back = (call_num <= 0); - call_num += curCble; // numeral is self-relative offset - call.le_len = call_num; // use le_len as scratch - calls_to_link.add(&call); - if (*lp++ != ')') - { - unpack_abort("bad call label"); - } - } - break; - case 'K': // reference_type: constant_ref - case 'R': // reference_type: schema_ref - { - int ixTag = CONSTANT_None; - if (lp[-1] == 'K') - { - switch (*lp++) - { - case 'I': - ixTag = CONSTANT_Integer; - break; - case 'J': - ixTag = CONSTANT_Long; - break; - case 'F': - ixTag = CONSTANT_Float; - break; - case 'D': - ixTag = CONSTANT_Double; - break; - case 'S': - ixTag = CONSTANT_String; - break; - case 'Q': - ixTag = CONSTANT_Literal; - break; - } - } - else - { - switch (*lp++) - { - case 'C': - ixTag = CONSTANT_Class; - break; - case 'S': - ixTag = CONSTANT_Signature; - break; - case 'D': - ixTag = CONSTANT_NameandType; - break; - case 'F': - ixTag = CONSTANT_Fieldref; - break; - case 'M': - ixTag = CONSTANT_Methodref; - break; - case 'I': - ixTag = CONSTANT_InterfaceMethodref; - break; - case 'U': - ixTag = CONSTANT_Utf8; - break; // utf8_ref - case 'Q': - ixTag = CONSTANT_All; - break; // untyped_ref - } - } - if (ixTag == CONSTANT_None) - { - unpack_abort("bad reference layout"); - break; - } - bool nullOK = false; - if (*lp == 'N') - { - nullOK = true; - lp++; - } - lp = parseIntLayout(lp, b, EK_REF); - b->defc = coding::findBySpec(UNSIGNED5_spec); - b->initRef(ixTag, nullOK); - } - break; - case '[': - { - // [callable1][callable2]... - if (!top_level) - { - unpack_abort("bad nested callable"); - break; - } - curCble += 1; - band &cble = *U_NEW(band, 1); - band_stack.add(&cble); - cble.le_kind = EK_CBLE; - cble.bn = bands_made++; - lp = parseLayout(lp, cble.le_body, curCble); - } - break; - case ']': - // Hit a closing brace. This ends whatever body we were in. - done = true; - break; - case '\0': - // Hit a nullptr. Also ends the (top-level) body. - --lp; // back up, so caller can see the nullptr also - done = true; - break; - default: - unpack_abort("bad layout"); - } - } - - // Return the accumulated bands: - res = popBody(bs_base); - return lp; + int bs_base = band_stack.length(); + bool top_level = (bs_base == 0); + band *b; + enum + { + can_be_signed = true + }; // optional arg to parseIntLayout + + for (bool done = false; !done;) + { + switch (*lp++) + { + case 'B': + case 'H': + case 'I': + case 'V': // unsigned_int + case 'S': // signed_int + --lp; // reparse + case 'F': + lp = parseIntLayout(lp, b, EK_INT); + break; + case 'P': + { + int le_bci = EK_BCI; + if (*lp == 'O') + { + ++lp; + le_bci = EK_BCID; + } + assert(*lp != 'S'); // no PSH, etc. + lp = parseIntLayout(lp, b, EK_INT); + b->le_bci = le_bci; + if (le_bci == EK_BCI) + b->defc = coding::findBySpec(BCI5_spec); + else + b->defc = coding::findBySpec(BRANCH5_spec); + } + break; + case 'O': + lp = parseIntLayout(lp, b, EK_INT, can_be_signed); + b->le_bci = EK_BCO; + b->defc = coding::findBySpec(BRANCH5_spec); + break; + case 'N': // replication: 'N' uint32_t '[' elem ... ']' + lp = parseIntLayout(lp, b, EK_REPL); + assert(*lp == '['); + ++lp; + lp = parseLayout(lp, b->le_body, curCble); + break; + case 'T': // union: 'T' any_int union_case* '(' ')' '[' body ']' + lp = parseIntLayout(lp, b, EK_UN, can_be_signed); + { + int union_base = band_stack.length(); + for (;;) + { // for each case + band &k_case = *U_NEW(band, 1); + band_stack.add(&k_case); + k_case.le_kind = EK_CASE; + k_case.bn = bands_made++; + if (*lp++ != '(') + { + unpack_abort("bad union case"); + return ""; + } + if (*lp++ != ')') + { + --lp; // reparse + // Read some case values. (Use band_stack for temp. storage.) + int case_base = band_stack.length(); + for (;;) + { + int caseval = 0; + lp = parseNumeral(lp, caseval); + band_stack.add((void *)(size_t)caseval); + if (*lp == '-') + { + // new in version 160, allow (1-5) for (1,2,3,4,5) + if (u->majver < JAVA6_PACKAGE_MAJOR_VERSION) + { + unpack_abort( + "bad range in union case label (old archive format)"); + return ""; + } + int caselimit = caseval; + lp++; + lp = parseNumeral(lp, caselimit); + if (caseval >= caselimit || + (uint32_t)(caselimit - caseval) > 0x10000) + { + // Note: 0x10000 is arbitrary implementation restriction. + // We can remove it later if it's important to. + unpack_abort("bad range in union case label"); + } + for (;;) + { + ++caseval; + band_stack.add((void *)(size_t)caseval); + if (caseval == caselimit) + break; + } + } + if (*lp != ',') + break; + lp++; + } + if (*lp++ != ')') + { + unpack_abort("bad case label"); + } + // save away the case labels + int ntags = band_stack.length() - case_base; + int *tags = U_NEW(int, add_size(ntags, 1)); + k_case.le_casetags = tags; + *tags++ = ntags; + for (int i = 0; i < ntags; i++) + { + *tags++ = ptrlowbits(band_stack.get(case_base + i)); + } + band_stack.popTo(case_base); + } + // Got le_casetags. Now grab the body. + assert(*lp == '['); + ++lp; + lp = parseLayout(lp, k_case.le_body, curCble); + if (k_case.le_casetags == nullptr) + break; // done + } + b->le_body = popBody(union_base); + } + break; + case '(': // call: '(' -?NN* ')' + { + band &call = *U_NEW(band, 1); + band_stack.add(&call); + call.le_kind = EK_CALL; + call.bn = bands_made++; + call.le_body = U_NEW(band *, 2); // fill in later + int call_num = 0; + lp = parseNumeral(lp, call_num); + call.le_back = (call_num <= 0); + call_num += curCble; // numeral is self-relative offset + call.le_len = call_num; // use le_len as scratch + calls_to_link.add(&call); + if (*lp++ != ')') + { + unpack_abort("bad call label"); + } + } + break; + case 'K': // reference_type: constant_ref + case 'R': // reference_type: schema_ref + { + int ixTag = CONSTANT_None; + if (lp[-1] == 'K') + { + switch (*lp++) + { + case 'I': + ixTag = CONSTANT_Integer; + break; + case 'J': + ixTag = CONSTANT_Long; + break; + case 'F': + ixTag = CONSTANT_Float; + break; + case 'D': + ixTag = CONSTANT_Double; + break; + case 'S': + ixTag = CONSTANT_String; + break; + case 'Q': + ixTag = CONSTANT_Literal; + break; + } + } + else + { + switch (*lp++) + { + case 'C': + ixTag = CONSTANT_Class; + break; + case 'S': + ixTag = CONSTANT_Signature; + break; + case 'D': + ixTag = CONSTANT_NameandType; + break; + case 'F': + ixTag = CONSTANT_Fieldref; + break; + case 'M': + ixTag = CONSTANT_Methodref; + break; + case 'I': + ixTag = CONSTANT_InterfaceMethodref; + break; + case 'U': + ixTag = CONSTANT_Utf8; + break; // utf8_ref + case 'Q': + ixTag = CONSTANT_All; + break; // untyped_ref + } + } + if (ixTag == CONSTANT_None) + { + unpack_abort("bad reference layout"); + break; + } + bool nullOK = false; + if (*lp == 'N') + { + nullOK = true; + lp++; + } + lp = parseIntLayout(lp, b, EK_REF); + b->defc = coding::findBySpec(UNSIGNED5_spec); + b->initRef(ixTag, nullOK); + } + break; + case '[': + { + // [callable1][callable2]... + if (!top_level) + { + unpack_abort("bad nested callable"); + break; + } + curCble += 1; + band &cble = *U_NEW(band, 1); + band_stack.add(&cble); + cble.le_kind = EK_CBLE; + cble.bn = bands_made++; + lp = parseLayout(lp, cble.le_body, curCble); + } + break; + case ']': + // Hit a closing brace. This ends whatever body we were in. + done = true; + break; + case '\0': + // Hit a nullptr. Also ends the (top-level) body. + --lp; // back up, so caller can see the nullptr also + done = true; + break; + default: + unpack_abort("bad layout"); + } + } + + // Return the accumulated bands: + res = popBody(bs_base); + return lp; } void unpacker::read_attr_defs() { - int i; - - // Tell each AD which attrc it is and where its fixed flags are: - attr_defs[ATTR_CONTEXT_CLASS].attrc = ATTR_CONTEXT_CLASS; - attr_defs[ATTR_CONTEXT_CLASS].xxx_flags_hi_bn = e_class_flags_hi; - attr_defs[ATTR_CONTEXT_FIELD].attrc = ATTR_CONTEXT_FIELD; - attr_defs[ATTR_CONTEXT_FIELD].xxx_flags_hi_bn = e_field_flags_hi; - attr_defs[ATTR_CONTEXT_METHOD].attrc = ATTR_CONTEXT_METHOD; - attr_defs[ATTR_CONTEXT_METHOD].xxx_flags_hi_bn = e_method_flags_hi; - attr_defs[ATTR_CONTEXT_CODE].attrc = ATTR_CONTEXT_CODE; - attr_defs[ATTR_CONTEXT_CODE].xxx_flags_hi_bn = e_code_flags_hi; - - // Decide whether bands for the optional high flag words are present. - attr_defs[ATTR_CONTEXT_CLASS] - .setHaveLongFlags((archive_options & AO_HAVE_CLASS_FLAGS_HI) != 0); - attr_defs[ATTR_CONTEXT_FIELD] - .setHaveLongFlags((archive_options & AO_HAVE_FIELD_FLAGS_HI) != 0); - attr_defs[ATTR_CONTEXT_METHOD] - .setHaveLongFlags((archive_options & AO_HAVE_METHOD_FLAGS_HI) != 0); - attr_defs[ATTR_CONTEXT_CODE] - .setHaveLongFlags((archive_options & AO_HAVE_CODE_FLAGS_HI) != 0); - - // Set up built-in attrs. - // (The simple ones are hard-coded. The metadata layouts are not.) - const char *md_layout = ( + int i; + + // Tell each AD which attrc it is and where its fixed flags are: + attr_defs[ATTR_CONTEXT_CLASS].attrc = ATTR_CONTEXT_CLASS; + attr_defs[ATTR_CONTEXT_CLASS].xxx_flags_hi_bn = e_class_flags_hi; + attr_defs[ATTR_CONTEXT_FIELD].attrc = ATTR_CONTEXT_FIELD; + attr_defs[ATTR_CONTEXT_FIELD].xxx_flags_hi_bn = e_field_flags_hi; + attr_defs[ATTR_CONTEXT_METHOD].attrc = ATTR_CONTEXT_METHOD; + attr_defs[ATTR_CONTEXT_METHOD].xxx_flags_hi_bn = e_method_flags_hi; + attr_defs[ATTR_CONTEXT_CODE].attrc = ATTR_CONTEXT_CODE; + attr_defs[ATTR_CONTEXT_CODE].xxx_flags_hi_bn = e_code_flags_hi; + + // Decide whether bands for the optional high flag words are present. + attr_defs[ATTR_CONTEXT_CLASS] + .setHaveLongFlags((archive_options & AO_HAVE_CLASS_FLAGS_HI) != 0); + attr_defs[ATTR_CONTEXT_FIELD] + .setHaveLongFlags((archive_options & AO_HAVE_FIELD_FLAGS_HI) != 0); + attr_defs[ATTR_CONTEXT_METHOD] + .setHaveLongFlags((archive_options & AO_HAVE_METHOD_FLAGS_HI) != 0); + attr_defs[ATTR_CONTEXT_CODE] + .setHaveLongFlags((archive_options & AO_HAVE_CODE_FLAGS_HI) != 0); + + // Set up built-in attrs. + // (The simple ones are hard-coded. The metadata layouts are not.) + const char *md_layout = ( // parameter annotations: #define MDL0 "[NB[(1)]]" - MDL0 + MDL0 // annotations: #define MDL1 \ - "[NH[(1)]]" \ - "[RSHNH[RUH(1)]]" - MDL1 - // member_value: - "[TB" - "(66,67,73,83,90)[KIH]" - "(68)[KDH]" - "(70)[KFH]" - "(74)[KJH]" - "(99)[RSH]" - "(101)[RSHRUH]" - "(115)[RUH]" - "(91)[NH[(0)]]" - "(64)[" - // nested annotation: - "RSH" - "NH[RUH(0)]" - "]" - "()[]" - "]"); - - const char *md_layout_P = md_layout; - const char *md_layout_A = md_layout + strlen(MDL0); - const char *md_layout_V = md_layout + strlen(MDL0 MDL1); - assert(0 == strncmp(&md_layout_A[-3], ")]][", 4)); - assert(0 == strncmp(&md_layout_V[-3], ")]][", 4)); - - for (i = 0; i < ATTR_CONTEXT_LIMIT; i++) - { - attr_definitions &ad = attr_defs[i]; - ad.defineLayout(X_ATTR_RuntimeVisibleAnnotations, "RuntimeVisibleAnnotations", - md_layout_A); - ad.defineLayout(X_ATTR_RuntimeInvisibleAnnotations, "RuntimeInvisibleAnnotations", - md_layout_A); - if (i != ATTR_CONTEXT_METHOD) - continue; - ad.defineLayout(METHOD_ATTR_RuntimeVisibleParameterAnnotations, - "RuntimeVisibleParameterAnnotations", md_layout_P); - ad.defineLayout(METHOD_ATTR_RuntimeInvisibleParameterAnnotations, - "RuntimeInvisibleParameterAnnotations", md_layout_P); - ad.defineLayout(METHOD_ATTR_AnnotationDefault, "AnnotationDefault", md_layout_V); - } - - attr_definition_headers.readData(attr_definition_count); - attr_definition_name.readData(attr_definition_count); - attr_definition_layout.readData(attr_definition_count); + "[NH[(1)]]" \ + "[RSHNH[RUH(1)]]" + MDL1 + // member_value: + "[TB" + "(66,67,73,83,90)[KIH]" + "(68)[KDH]" + "(70)[KFH]" + "(74)[KJH]" + "(99)[RSH]" + "(101)[RSHRUH]" + "(115)[RUH]" + "(91)[NH[(0)]]" + "(64)[" + // nested annotation: + "RSH" + "NH[RUH(0)]" + "]" + "()[]" + "]"); + + const char *md_layout_P = md_layout; + const char *md_layout_A = md_layout + strlen(MDL0); + const char *md_layout_V = md_layout + strlen(MDL0 MDL1); + assert(0 == strncmp(&md_layout_A[-3], ")]][", 4)); + assert(0 == strncmp(&md_layout_V[-3], ")]][", 4)); + + for (i = 0; i < ATTR_CONTEXT_LIMIT; i++) + { + attr_definitions &ad = attr_defs[i]; + ad.defineLayout(X_ATTR_RuntimeVisibleAnnotations, "RuntimeVisibleAnnotations", + md_layout_A); + ad.defineLayout(X_ATTR_RuntimeInvisibleAnnotations, "RuntimeInvisibleAnnotations", + md_layout_A); + if (i != ATTR_CONTEXT_METHOD) + continue; + ad.defineLayout(METHOD_ATTR_RuntimeVisibleParameterAnnotations, + "RuntimeVisibleParameterAnnotations", md_layout_P); + ad.defineLayout(METHOD_ATTR_RuntimeInvisibleParameterAnnotations, + "RuntimeInvisibleParameterAnnotations", md_layout_P); + ad.defineLayout(METHOD_ATTR_AnnotationDefault, "AnnotationDefault", md_layout_V); + } + + attr_definition_headers.readData(attr_definition_count); + attr_definition_name.readData(attr_definition_count); + attr_definition_layout.readData(attr_definition_count); // Initialize correct predef bits, to distinguish predefs from new defs. #define ORBIT(n, s) | ((uint64_t)1 << n) - attr_defs[ATTR_CONTEXT_CLASS].predef = (0 X_ATTR_DO(ORBIT) CLASS_ATTR_DO(ORBIT)); - attr_defs[ATTR_CONTEXT_FIELD].predef = (0 X_ATTR_DO(ORBIT) FIELD_ATTR_DO(ORBIT)); - attr_defs[ATTR_CONTEXT_METHOD].predef = (0 X_ATTR_DO(ORBIT) METHOD_ATTR_DO(ORBIT)); - attr_defs[ATTR_CONTEXT_CODE].predef = (0 O_ATTR_DO(ORBIT) CODE_ATTR_DO(ORBIT)); + attr_defs[ATTR_CONTEXT_CLASS].predef = (0 X_ATTR_DO(ORBIT) CLASS_ATTR_DO(ORBIT)); + attr_defs[ATTR_CONTEXT_FIELD].predef = (0 X_ATTR_DO(ORBIT) FIELD_ATTR_DO(ORBIT)); + attr_defs[ATTR_CONTEXT_METHOD].predef = (0 X_ATTR_DO(ORBIT) METHOD_ATTR_DO(ORBIT)); + attr_defs[ATTR_CONTEXT_CODE].predef = (0 O_ATTR_DO(ORBIT) CODE_ATTR_DO(ORBIT)); #undef ORBIT - // Clear out the redef bits, folding them back into predef. - for (i = 0; i < ATTR_CONTEXT_LIMIT; i++) - { - attr_defs[i].predef |= attr_defs[i].redef; - attr_defs[i].redef = 0; - } - - // Now read the transmitted locally defined attrs. - // This will set redef bits again. - for (i = 0; i < attr_definition_count; i++) - { - int header = attr_definition_headers.getByte(); - int attrc = ADH_BYTE_CONTEXT(header); - int idx = ADH_BYTE_INDEX(header); - entry *name = attr_definition_name.getRef(); - entry *layout = attr_definition_layout.getRef(); - attr_defs[attrc].defineLayout(idx, name, layout->value.b.strval()); - } + // Clear out the redef bits, folding them back into predef. + for (i = 0; i < ATTR_CONTEXT_LIMIT; i++) + { + attr_defs[i].predef |= attr_defs[i].redef; + attr_defs[i].redef = 0; + } + + // Now read the transmitted locally defined attrs. + // This will set redef bits again. + for (i = 0; i < attr_definition_count; i++) + { + int header = attr_definition_headers.getByte(); + int attrc = ADH_BYTE_CONTEXT(header); + int idx = ADH_BYTE_INDEX(header); + entry *name = attr_definition_name.getRef(); + entry *layout = attr_definition_layout.getRef(); + attr_defs[attrc].defineLayout(idx, name, layout->value.b.strval()); + } } #define NO_ENTRY_YET ((entry *)-1) static bool isDigitString(bytes &x, int beg, int end) { - if (beg == end) - return false; // nullptr string - byte *xptr = x.ptr; - for (int i = beg; i < end; i++) - { - char ch = xptr[i]; - if (!(ch >= '0' && ch <= '9')) - return false; - } - return true; + if (beg == end) + return false; // nullptr string + byte *xptr = x.ptr; + for (int i = beg; i < end; i++) + { + char ch = xptr[i]; + if (!(ch >= '0' && ch <= '9')) + return false; + } + return true; } enum { // constants for parsing class names - SLASH_MIN = '.', - SLASH_MAX = '/', - DOLLAR_MIN = 0, - DOLLAR_MAX = '-'}; + SLASH_MIN = '.', + SLASH_MAX = '/', + DOLLAR_MIN = 0, + DOLLAR_MAX = '-'}; static int lastIndexOf(int chmin, int chmax, bytes &x, int pos) { - byte *ptr = x.ptr; - for (byte *cp = ptr + pos; --cp >= ptr;) - { - assert(x.inBounds(cp)); - if (*cp >= chmin && *cp <= chmax) - return (int)(cp - ptr); - } - return -1; + byte *ptr = x.ptr; + for (byte *cp = ptr + pos; --cp >= ptr;) + { + assert(x.inBounds(cp)); + if (*cp >= chmin && *cp <= chmax) + return (int)(cp - ptr); + } + return -1; } inner_class *constant_pool::getIC(entry *inner) { - if (inner == nullptr) - return nullptr; - assert(inner->tag == CONSTANT_Class); - if (inner->inord == NO_INORD) - return nullptr; - inner_class *ic = ic_index[inner->inord]; - assert(ic == nullptr || ic->inner == inner); - return ic; + if (inner == nullptr) + return nullptr; + assert(inner->tag == CONSTANT_Class); + if (inner->inord == NO_INORD) + return nullptr; + inner_class *ic = ic_index[inner->inord]; + assert(ic == nullptr || ic->inner == inner); + return ic; } inner_class *constant_pool::getFirstChildIC(entry *outer) { - if (outer == nullptr) - return nullptr; - assert(outer->tag == CONSTANT_Class); - if (outer->inord == NO_INORD) - return nullptr; - inner_class *ic = ic_child_index[outer->inord]; - assert(ic == nullptr || ic->outer == outer); - return ic; + if (outer == nullptr) + return nullptr; + assert(outer->tag == CONSTANT_Class); + if (outer->inord == NO_INORD) + return nullptr; + inner_class *ic = ic_child_index[outer->inord]; + assert(ic == nullptr || ic->outer == outer); + return ic; } inner_class *constant_pool::getNextChildIC(inner_class *child) { - inner_class *ic = child->next_sibling; - assert(ic == nullptr || ic->outer == child->outer); - return ic; + inner_class *ic = child->next_sibling; + assert(ic == nullptr || ic->outer == child->outer); + return ic; } void unpacker::read_ics() { - int i; - int index_size = cp.tag_count[CONSTANT_Class]; - inner_class **ic_index = U_NEW(inner_class *, index_size); - inner_class **ic_child_index = U_NEW(inner_class *, index_size); - cp.ic_index = ic_index; - cp.ic_child_index = ic_child_index; - ics = U_NEW(inner_class, ic_count); - ic_this_class.readData(ic_count); - ic_flags.readData(ic_count); - // Scan flags to get count of long-form bands. - int long_forms = 0; - for (i = 0; i < ic_count; i++) - { - int flags = ic_flags.getInt(); // may be long form! - if ((flags & ACC_IC_LONG_FORM) != 0) - { - long_forms += 1; - ics[i].name = NO_ENTRY_YET; - } - flags &= ~ACC_IC_LONG_FORM; - entry *inner = ic_this_class.getRef(); - uint32_t inord = inner->inord; - assert(inord < (uint32_t)cp.tag_count[CONSTANT_Class]); - if (ic_index[inord] != nullptr) - { - unpack_abort("identical inner class"); - break; - } - ic_index[inord] = &ics[i]; - ics[i].inner = inner; - ics[i].flags = flags; - assert(cp.getIC(inner) == &ics[i]); - } - // ic_this_class.done(); - // ic_flags.done(); - ic_outer_class.readData(long_forms); - ic_name.readData(long_forms); - for (i = 0; i < ic_count; i++) - { - if (ics[i].name == NO_ENTRY_YET) - { - // Long form. - ics[i].outer = ic_outer_class.getRefN(); - ics[i].name = ic_name.getRefN(); - } - else - { - // Fill in outer and name based on inner. - bytes &n = ics[i].inner->value.b; - bytes pkgOuter; - bytes number; - bytes name; - // Parse n into pkgOuter and name (and number). - int dollar1, dollar2; // pointers to $ in the pattern - // parse n = (<pkg>/)*<outer>($<number>)?($<name>)? - int nlen = (int)n.len; - int pkglen = lastIndexOf(SLASH_MIN, SLASH_MAX, n, nlen) + 1; - dollar2 = lastIndexOf(DOLLAR_MIN, DOLLAR_MAX, n, nlen); - if (dollar2 < 0) - { - unpack_abort(); - } - assert(dollar2 >= pkglen); - if (isDigitString(n, dollar2 + 1, nlen)) - { - // n = (<pkg>/)*<outer>$<number> - number = n.slice(dollar2 + 1, nlen); - name.set(nullptr, 0); - dollar1 = dollar2; - } - else if (pkglen < (dollar1 = lastIndexOf(DOLLAR_MIN, DOLLAR_MAX, n, dollar2 - 1)) && - isDigitString(n, dollar1 + 1, dollar2)) - { - // n = (<pkg>/)*<outer>$<number>$<name> - number = n.slice(dollar1 + 1, dollar2); - name = n.slice(dollar2 + 1, nlen); - } - else - { - // n = (<pkg>/)*<outer>$<name> - dollar1 = dollar2; - number.set(nullptr, 0); - name = n.slice(dollar2 + 1, nlen); - } - if (number.ptr == nullptr) - pkgOuter = n.slice(0, dollar1); - else - pkgOuter.set(nullptr, 0); - - if (pkgOuter.ptr != nullptr) - ics[i].outer = cp.ensureClass(pkgOuter); - - if (name.ptr != nullptr) - ics[i].name = cp.ensureUtf8(name); - } - - // update child/sibling list - if (ics[i].outer != nullptr) - { - uint32_t outord = ics[i].outer->inord; - if (outord != NO_INORD) - { - assert(outord < (uint32_t)cp.tag_count[CONSTANT_Class]); - ics[i].next_sibling = ic_child_index[outord]; - ic_child_index[outord] = &ics[i]; - } - } - } - // ic_outer_class.done(); - // ic_name.done(); + int i; + int index_size = cp.tag_count[CONSTANT_Class]; + inner_class **ic_index = U_NEW(inner_class *, index_size); + inner_class **ic_child_index = U_NEW(inner_class *, index_size); + cp.ic_index = ic_index; + cp.ic_child_index = ic_child_index; + ics = U_NEW(inner_class, ic_count); + ic_this_class.readData(ic_count); + ic_flags.readData(ic_count); + // Scan flags to get count of long-form bands. + int long_forms = 0; + for (i = 0; i < ic_count; i++) + { + int flags = ic_flags.getInt(); // may be long form! + if ((flags & ACC_IC_LONG_FORM) != 0) + { + long_forms += 1; + ics[i].name = NO_ENTRY_YET; + } + flags &= ~ACC_IC_LONG_FORM; + entry *inner = ic_this_class.getRef(); + uint32_t inord = inner->inord; + assert(inord < (uint32_t)cp.tag_count[CONSTANT_Class]); + if (ic_index[inord] != nullptr) + { + unpack_abort("identical inner class"); + break; + } + ic_index[inord] = &ics[i]; + ics[i].inner = inner; + ics[i].flags = flags; + assert(cp.getIC(inner) == &ics[i]); + } + // ic_this_class.done(); + // ic_flags.done(); + ic_outer_class.readData(long_forms); + ic_name.readData(long_forms); + for (i = 0; i < ic_count; i++) + { + if (ics[i].name == NO_ENTRY_YET) + { + // Long form. + ics[i].outer = ic_outer_class.getRefN(); + ics[i].name = ic_name.getRefN(); + } + else + { + // Fill in outer and name based on inner. + bytes &n = ics[i].inner->value.b; + bytes pkgOuter; + bytes number; + bytes name; + // Parse n into pkgOuter and name (and number). + int dollar1, dollar2; // pointers to $ in the pattern + // parse n = (<pkg>/)*<outer>($<number>)?($<name>)? + int nlen = (int)n.len; + int pkglen = lastIndexOf(SLASH_MIN, SLASH_MAX, n, nlen) + 1; + dollar2 = lastIndexOf(DOLLAR_MIN, DOLLAR_MAX, n, nlen); + if (dollar2 < 0) + { + unpack_abort(); + } + assert(dollar2 >= pkglen); + if (isDigitString(n, dollar2 + 1, nlen)) + { + // n = (<pkg>/)*<outer>$<number> + number = n.slice(dollar2 + 1, nlen); + name.set(nullptr, 0); + dollar1 = dollar2; + } + else if (pkglen < (dollar1 = lastIndexOf(DOLLAR_MIN, DOLLAR_MAX, n, dollar2 - 1)) && + isDigitString(n, dollar1 + 1, dollar2)) + { + // n = (<pkg>/)*<outer>$<number>$<name> + number = n.slice(dollar1 + 1, dollar2); + name = n.slice(dollar2 + 1, nlen); + } + else + { + // n = (<pkg>/)*<outer>$<name> + dollar1 = dollar2; + number.set(nullptr, 0); + name = n.slice(dollar2 + 1, nlen); + } + if (number.ptr == nullptr) + pkgOuter = n.slice(0, dollar1); + else + pkgOuter.set(nullptr, 0); + + if (pkgOuter.ptr != nullptr) + ics[i].outer = cp.ensureClass(pkgOuter); + + if (name.ptr != nullptr) + ics[i].name = cp.ensureUtf8(name); + } + + // update child/sibling list + if (ics[i].outer != nullptr) + { + uint32_t outord = ics[i].outer->inord; + if (outord != NO_INORD) + { + assert(outord < (uint32_t)cp.tag_count[CONSTANT_Class]); + ics[i].next_sibling = ic_child_index[outord]; + ic_child_index[outord] = &ics[i]; + } + } + } + // ic_outer_class.done(); + // ic_name.done(); } void unpacker::read_classes() { - class_this.readData(class_count); - class_super.readData(class_count); - class_interface_count.readData(class_count); - class_interface.readData(class_interface_count.getIntTotal()); + class_this.readData(class_count); + class_super.readData(class_count); + class_interface_count.readData(class_count); + class_interface.readData(class_interface_count.getIntTotal()); #if 0 int i; @@ -2256,1218 +2250,1220 @@ void unpacker::read_classes() class_super.rewind(); #endif - // Members. - class_field_count.readData(class_count); - class_method_count.readData(class_count); + // Members. + class_field_count.readData(class_count); + class_method_count.readData(class_count); - int field_count = class_field_count.getIntTotal(); - int method_count = class_method_count.getIntTotal(); + int field_count = class_field_count.getIntTotal(); + int method_count = class_method_count.getIntTotal(); - field_descr.readData(field_count); - read_attrs(ATTR_CONTEXT_FIELD, field_count); - method_descr.readData(method_count); - read_attrs(ATTR_CONTEXT_METHOD, method_count); - read_attrs(ATTR_CONTEXT_CLASS, class_count); - read_code_headers(); + field_descr.readData(field_count); + read_attrs(ATTR_CONTEXT_FIELD, field_count); + method_descr.readData(method_count); + read_attrs(ATTR_CONTEXT_METHOD, method_count); + read_attrs(ATTR_CONTEXT_CLASS, class_count); + read_code_headers(); } int unpacker::attr_definitions::predefCount(uint32_t idx) { - return isPredefined(idx) ? flag_count[idx] : 0; + return isPredefined(idx) ? flag_count[idx] : 0; } void unpacker::read_attrs(int attrc, int obj_count) { - attr_definitions &ad = attr_defs[attrc]; - assert(ad.attrc == attrc); - - int i, idx, count; - - bool haveLongFlags = ad.haveLongFlags(); - - band &xxx_flags_hi = ad.xxx_flags_hi(); - if (haveLongFlags) - xxx_flags_hi.readData(obj_count); - - band &xxx_flags_lo = ad.xxx_flags_lo(); - xxx_flags_lo.readData(obj_count); - - // pre-scan flags, counting occurrences of each index bit - uint64_t indexMask = ad.flagIndexMask(); // which flag bits are index bits? - for (i = 0; i < obj_count; i++) - { - uint64_t indexBits = xxx_flags_hi.getLong(xxx_flags_lo, haveLongFlags); - if ((indexBits & ~indexMask) > (ushort) - 1) - { - unpack_abort("undefined attribute flag bit"); - return; - } - indexBits &= indexMask; // ignore classfile flag bits - for (idx = 0; indexBits != 0; idx++, indexBits >>= 1) - { - ad.flag_count[idx] += (int)(indexBits & 1); - } - } - // we'll scan these again later for output: - xxx_flags_lo.rewind(); - xxx_flags_hi.rewind(); - - band &xxx_attr_count = ad.xxx_attr_count(); - // There is one count element for each 1<<16 bit set in flags: - xxx_attr_count.readData(ad.predefCount(X_ATTR_OVERFLOW)); - - band &xxx_attr_indexes = ad.xxx_attr_indexes(); - int overflowIndexCount = xxx_attr_count.getIntTotal(); - xxx_attr_indexes.readData(overflowIndexCount); - // pre-scan attr indexes, counting occurrences of each value - for (i = 0; i < overflowIndexCount; i++) - { - idx = xxx_attr_indexes.getInt(); - if (!ad.isIndex(idx)) - { - unpack_abort("attribute index out of bounds"); - return; - } - ad.getCount(idx) += 1; - } - xxx_attr_indexes.rewind(); // we'll scan it again later for output - - // We will need a backward call count for each used backward callable. - int backwardCounts = 0; - for (idx = 0; idx < ad.layouts.length(); idx++) - { - layout_definition *lo = ad.getLayout(idx); - if (lo != nullptr && ad.getCount(idx) != 0) - { - // Build the bands lazily, only when they are used. - band **bands = ad.buildBands(lo); - if (lo->hasCallables()) - { - for (i = 0; bands[i] != nullptr; i++) - { - if (bands[i]->le_back) - { - assert(bands[i]->le_kind == EK_CBLE); - backwardCounts += 1; - } - } - } - } - } - ad.xxx_attr_calls().readData(backwardCounts); - - // Read built-in bands. - // Mostly, these are hand-coded equivalents to readBandData(). - switch (attrc) - { - case ATTR_CONTEXT_CLASS: - - count = ad.predefCount(CLASS_ATTR_SourceFile); - class_SourceFile_RUN.readData(count); - - count = ad.predefCount(CLASS_ATTR_EnclosingMethod); - class_EnclosingMethod_RC.readData(count); - class_EnclosingMethod_RDN.readData(count); - - count = ad.predefCount(X_ATTR_Signature); - class_Signature_RS.readData(count); - - ad.readBandData(X_ATTR_RuntimeVisibleAnnotations); - ad.readBandData(X_ATTR_RuntimeInvisibleAnnotations); - - count = ad.predefCount(CLASS_ATTR_InnerClasses); - class_InnerClasses_N.readData(count); - - count = class_InnerClasses_N.getIntTotal(); - class_InnerClasses_RC.readData(count); - class_InnerClasses_F.readData(count); - - // Drop remaining columns wherever flags are zero: - count -= class_InnerClasses_F.getIntCount(0); - class_InnerClasses_outer_RCN.readData(count); - class_InnerClasses_name_RUN.readData(count); - - count = ad.predefCount(CLASS_ATTR_ClassFile_version); - class_ClassFile_version_minor_H.readData(count); - class_ClassFile_version_major_H.readData(count); - break; - - case ATTR_CONTEXT_FIELD: - - count = ad.predefCount(FIELD_ATTR_ConstantValue); - field_ConstantValue_KQ.readData(count); - - count = ad.predefCount(X_ATTR_Signature); - field_Signature_RS.readData(count); - - ad.readBandData(X_ATTR_RuntimeVisibleAnnotations); - ad.readBandData(X_ATTR_RuntimeInvisibleAnnotations); - break; - - case ATTR_CONTEXT_METHOD: - - code_count = ad.predefCount(METHOD_ATTR_Code); - // Code attrs are handled very specially below... - - count = ad.predefCount(METHOD_ATTR_Exceptions); - method_Exceptions_N.readData(count); - count = method_Exceptions_N.getIntTotal(); - method_Exceptions_RC.readData(count); - - count = ad.predefCount(X_ATTR_Signature); - method_Signature_RS.readData(count); - - ad.readBandData(X_ATTR_RuntimeVisibleAnnotations); - ad.readBandData(X_ATTR_RuntimeInvisibleAnnotations); - ad.readBandData(METHOD_ATTR_RuntimeVisibleParameterAnnotations); - ad.readBandData(METHOD_ATTR_RuntimeInvisibleParameterAnnotations); - ad.readBandData(METHOD_ATTR_AnnotationDefault); - break; - - case ATTR_CONTEXT_CODE: - // (keep this code aligned with its brother in unpacker::write_attrs) - count = ad.predefCount(CODE_ATTR_StackMapTable); - // disable this feature in old archives! - if (count != 0 && majver < JAVA6_PACKAGE_MAJOR_VERSION) - { - unpack_abort("undefined StackMapTable attribute (old archive format)"); - return; - } - code_StackMapTable_N.readData(count); - count = code_StackMapTable_N.getIntTotal(); - code_StackMapTable_frame_T.readData(count); - // the rest of it depends in a complicated way on frame tags - { - int fat_frame_count = 0; - int offset_count = 0; - int type_count = 0; - for (int k = 0; k < count; k++) - { - int tag = code_StackMapTable_frame_T.getByte(); - if (tag <= 127) - { - // (64-127) [(2)] - if (tag >= 64) - type_count++; - } - else if (tag <= 251) - { - // (247) [(1)(2)] - // (248-251) [(1)] - if (tag >= 247) - offset_count++; - if (tag == 247) - type_count++; - } - else if (tag <= 254) - { - // (252) [(1)(2)] - // (253) [(1)(2)(2)] - // (254) [(1)(2)(2)(2)] - offset_count++; - type_count += (tag - 251); - } - else - { - // (255) [(1)NH[(2)]NH[(2)]] - fat_frame_count++; - } - } - - // done pre-scanning frame tags: - code_StackMapTable_frame_T.rewind(); - - // deal completely with fat frames: - offset_count += fat_frame_count; - code_StackMapTable_local_N.readData(fat_frame_count); - type_count += code_StackMapTable_local_N.getIntTotal(); - code_StackMapTable_stack_N.readData(fat_frame_count); - type_count += code_StackMapTable_stack_N.getIntTotal(); - // read the rest: - code_StackMapTable_offset.readData(offset_count); - code_StackMapTable_T.readData(type_count); - // (7) [RCH] - count = code_StackMapTable_T.getIntCount(7); - code_StackMapTable_RC.readData(count); - // (8) [PH] - count = code_StackMapTable_T.getIntCount(8); - code_StackMapTable_P.readData(count); - } - - count = ad.predefCount(CODE_ATTR_LineNumberTable); - code_LineNumberTable_N.readData(count); - count = code_LineNumberTable_N.getIntTotal(); - code_LineNumberTable_bci_P.readData(count); - code_LineNumberTable_line.readData(count); - - count = ad.predefCount(CODE_ATTR_LocalVariableTable); - code_LocalVariableTable_N.readData(count); - count = code_LocalVariableTable_N.getIntTotal(); - code_LocalVariableTable_bci_P.readData(count); - code_LocalVariableTable_span_O.readData(count); - code_LocalVariableTable_name_RU.readData(count); - code_LocalVariableTable_type_RS.readData(count); - code_LocalVariableTable_slot.readData(count); - - count = ad.predefCount(CODE_ATTR_LocalVariableTypeTable); - code_LocalVariableTypeTable_N.readData(count); - count = code_LocalVariableTypeTable_N.getIntTotal(); - code_LocalVariableTypeTable_bci_P.readData(count); - code_LocalVariableTypeTable_span_O.readData(count); - code_LocalVariableTypeTable_name_RU.readData(count); - code_LocalVariableTypeTable_type_RS.readData(count); - code_LocalVariableTypeTable_slot.readData(count); - break; - } - - // Read compressor-defined bands. - for (idx = 0; idx < ad.layouts.length(); idx++) - { - if (ad.getLayout(idx) == nullptr) - continue; // none at this fixed index <32 - if (idx < (int)ad.flag_limit && ad.isPredefined(idx)) - continue; // already handled - if (ad.getCount(idx) == 0) - continue; // no attributes of this type (then why transmit layouts?) - ad.readBandData(idx); - } + attr_definitions &ad = attr_defs[attrc]; + assert(ad.attrc == attrc); + + int i, idx, count; + + bool haveLongFlags = ad.haveLongFlags(); + + band &xxx_flags_hi = ad.xxx_flags_hi(); + if (haveLongFlags) + xxx_flags_hi.readData(obj_count); + + band &xxx_flags_lo = ad.xxx_flags_lo(); + xxx_flags_lo.readData(obj_count); + + // pre-scan flags, counting occurrences of each index bit + uint64_t indexMask = ad.flagIndexMask(); // which flag bits are index bits? + for (i = 0; i < obj_count; i++) + { + uint64_t indexBits = xxx_flags_hi.getLong(xxx_flags_lo, haveLongFlags); + if ((indexBits & ~indexMask) > (ushort) - 1) + { + unpack_abort("undefined attribute flag bit"); + return; + } + indexBits &= indexMask; // ignore classfile flag bits + for (idx = 0; indexBits != 0; idx++, indexBits >>= 1) + { + ad.flag_count[idx] += (int)(indexBits & 1); + } + } + // we'll scan these again later for output: + xxx_flags_lo.rewind(); + xxx_flags_hi.rewind(); + + band &xxx_attr_count = ad.xxx_attr_count(); + // There is one count element for each 1<<16 bit set in flags: + xxx_attr_count.readData(ad.predefCount(X_ATTR_OVERFLOW)); + + band &xxx_attr_indexes = ad.xxx_attr_indexes(); + int overflowIndexCount = xxx_attr_count.getIntTotal(); + xxx_attr_indexes.readData(overflowIndexCount); + // pre-scan attr indexes, counting occurrences of each value + for (i = 0; i < overflowIndexCount; i++) + { + idx = xxx_attr_indexes.getInt(); + if (!ad.isIndex(idx)) + { + unpack_abort("attribute index out of bounds"); + return; + } + ad.getCount(idx) += 1; + } + xxx_attr_indexes.rewind(); // we'll scan it again later for output + + // We will need a backward call count for each used backward callable. + int backwardCounts = 0; + for (idx = 0; idx < ad.layouts.length(); idx++) + { + layout_definition *lo = ad.getLayout(idx); + if (lo != nullptr && ad.getCount(idx) != 0) + { + // Build the bands lazily, only when they are used. + band **bands = ad.buildBands(lo); + if (lo->hasCallables()) + { + for (i = 0; bands[i] != nullptr; i++) + { + if (bands[i]->le_back) + { + assert(bands[i]->le_kind == EK_CBLE); + backwardCounts += 1; + } + } + } + } + } + ad.xxx_attr_calls().readData(backwardCounts); + + // Read built-in bands. + // Mostly, these are hand-coded equivalents to readBandData(). + switch (attrc) + { + case ATTR_CONTEXT_CLASS: + + count = ad.predefCount(CLASS_ATTR_SourceFile); + class_SourceFile_RUN.readData(count); + + count = ad.predefCount(CLASS_ATTR_EnclosingMethod); + class_EnclosingMethod_RC.readData(count); + class_EnclosingMethod_RDN.readData(count); + + count = ad.predefCount(X_ATTR_Signature); + class_Signature_RS.readData(count); + + ad.readBandData(X_ATTR_RuntimeVisibleAnnotations); + ad.readBandData(X_ATTR_RuntimeInvisibleAnnotations); + + count = ad.predefCount(CLASS_ATTR_InnerClasses); + class_InnerClasses_N.readData(count); + + count = class_InnerClasses_N.getIntTotal(); + class_InnerClasses_RC.readData(count); + class_InnerClasses_F.readData(count); + + // Drop remaining columns wherever flags are zero: + count -= class_InnerClasses_F.getIntCount(0); + class_InnerClasses_outer_RCN.readData(count); + class_InnerClasses_name_RUN.readData(count); + + count = ad.predefCount(CLASS_ATTR_ClassFile_version); + class_ClassFile_version_minor_H.readData(count); + class_ClassFile_version_major_H.readData(count); + break; + + case ATTR_CONTEXT_FIELD: + + count = ad.predefCount(FIELD_ATTR_ConstantValue); + field_ConstantValue_KQ.readData(count); + + count = ad.predefCount(X_ATTR_Signature); + field_Signature_RS.readData(count); + + ad.readBandData(X_ATTR_RuntimeVisibleAnnotations); + ad.readBandData(X_ATTR_RuntimeInvisibleAnnotations); + break; + + case ATTR_CONTEXT_METHOD: + + code_count = ad.predefCount(METHOD_ATTR_Code); + // Code attrs are handled very specially below... + + count = ad.predefCount(METHOD_ATTR_Exceptions); + method_Exceptions_N.readData(count); + count = method_Exceptions_N.getIntTotal(); + method_Exceptions_RC.readData(count); + + count = ad.predefCount(X_ATTR_Signature); + method_Signature_RS.readData(count); + + ad.readBandData(X_ATTR_RuntimeVisibleAnnotations); + ad.readBandData(X_ATTR_RuntimeInvisibleAnnotations); + ad.readBandData(METHOD_ATTR_RuntimeVisibleParameterAnnotations); + ad.readBandData(METHOD_ATTR_RuntimeInvisibleParameterAnnotations); + ad.readBandData(METHOD_ATTR_AnnotationDefault); + break; + + case ATTR_CONTEXT_CODE: + // (keep this code aligned with its brother in unpacker::write_attrs) + count = ad.predefCount(CODE_ATTR_StackMapTable); + // disable this feature in old archives! + if (count != 0 && majver < JAVA6_PACKAGE_MAJOR_VERSION) + { + unpack_abort("undefined StackMapTable attribute (old archive format)"); + return; + } + code_StackMapTable_N.readData(count); + count = code_StackMapTable_N.getIntTotal(); + code_StackMapTable_frame_T.readData(count); + // the rest of it depends in a complicated way on frame tags + { + int fat_frame_count = 0; + int offset_count = 0; + int type_count = 0; + for (int k = 0; k < count; k++) + { + int tag = code_StackMapTable_frame_T.getByte(); + if (tag <= 127) + { + // (64-127) [(2)] + if (tag >= 64) + type_count++; + } + else if (tag <= 251) + { + // (247) [(1)(2)] + // (248-251) [(1)] + if (tag >= 247) + offset_count++; + if (tag == 247) + type_count++; + } + else if (tag <= 254) + { + // (252) [(1)(2)] + // (253) [(1)(2)(2)] + // (254) [(1)(2)(2)(2)] + offset_count++; + type_count += (tag - 251); + } + else + { + // (255) [(1)NH[(2)]NH[(2)]] + fat_frame_count++; + } + } + + // done pre-scanning frame tags: + code_StackMapTable_frame_T.rewind(); + + // deal completely with fat frames: + offset_count += fat_frame_count; + code_StackMapTable_local_N.readData(fat_frame_count); + type_count += code_StackMapTable_local_N.getIntTotal(); + code_StackMapTable_stack_N.readData(fat_frame_count); + type_count += code_StackMapTable_stack_N.getIntTotal(); + // read the rest: + code_StackMapTable_offset.readData(offset_count); + code_StackMapTable_T.readData(type_count); + // (7) [RCH] + count = code_StackMapTable_T.getIntCount(7); + code_StackMapTable_RC.readData(count); + // (8) [PH] + count = code_StackMapTable_T.getIntCount(8); + code_StackMapTable_P.readData(count); + } + + count = ad.predefCount(CODE_ATTR_LineNumberTable); + code_LineNumberTable_N.readData(count); + count = code_LineNumberTable_N.getIntTotal(); + code_LineNumberTable_bci_P.readData(count); + code_LineNumberTable_line.readData(count); + + count = ad.predefCount(CODE_ATTR_LocalVariableTable); + code_LocalVariableTable_N.readData(count); + count = code_LocalVariableTable_N.getIntTotal(); + code_LocalVariableTable_bci_P.readData(count); + code_LocalVariableTable_span_O.readData(count); + code_LocalVariableTable_name_RU.readData(count); + code_LocalVariableTable_type_RS.readData(count); + code_LocalVariableTable_slot.readData(count); + + count = ad.predefCount(CODE_ATTR_LocalVariableTypeTable); + code_LocalVariableTypeTable_N.readData(count); + count = code_LocalVariableTypeTable_N.getIntTotal(); + code_LocalVariableTypeTable_bci_P.readData(count); + code_LocalVariableTypeTable_span_O.readData(count); + code_LocalVariableTypeTable_name_RU.readData(count); + code_LocalVariableTypeTable_type_RS.readData(count); + code_LocalVariableTypeTable_slot.readData(count); + break; + } + + // Read compressor-defined bands. + for (idx = 0; idx < ad.layouts.length(); idx++) + { + if (ad.getLayout(idx) == nullptr) + continue; // none at this fixed index <32 + if (idx < (int)ad.flag_limit && ad.isPredefined(idx)) + continue; // already handled + if (ad.getCount(idx) == 0) + continue; // no attributes of this type (then why transmit layouts?) + ad.readBandData(idx); + } } void unpacker::attr_definitions::readBandData(int idx) { - int j; - uint32_t count = getCount(idx); - if (count == 0) - return; - layout_definition *lo = getLayout(idx); - bool hasCallables = lo->hasCallables(); - band **bands = lo->bands(); - if (!hasCallables) - { - // Read through the rest of the bands in a regular way. - readBandData(bands, count); - } - else - { - // Deal with the callables. - // First set up the forward entry count for each callable. - // This is stored on band::length of the callable. - bands[0]->expectMoreLength(count); - for (j = 0; bands[j] != nullptr; j++) - { - band &j_cble = *bands[j]; - assert(j_cble.le_kind == EK_CBLE); - if (j_cble.le_back) - { - // Add in the predicted effects of backward calls, too. - int back_calls = xxx_attr_calls().getInt(); - j_cble.expectMoreLength(back_calls); - // In a moment, more forward calls may increment j_cble.length. - } - } - // Now consult whichever callables have non-zero entry counts. - readBandData(bands, (uint32_t) - 1); - } + int j; + uint32_t count = getCount(idx); + if (count == 0) + return; + layout_definition *lo = getLayout(idx); + bool hasCallables = lo->hasCallables(); + band **bands = lo->bands(); + if (!hasCallables) + { + // Read through the rest of the bands in a regular way. + readBandData(bands, count); + } + else + { + // Deal with the callables. + // First set up the forward entry count for each callable. + // This is stored on band::length of the callable. + bands[0]->expectMoreLength(count); + for (j = 0; bands[j] != nullptr; j++) + { + band &j_cble = *bands[j]; + assert(j_cble.le_kind == EK_CBLE); + if (j_cble.le_back) + { + // Add in the predicted effects of backward calls, too. + int back_calls = xxx_attr_calls().getInt(); + j_cble.expectMoreLength(back_calls); + // In a moment, more forward calls may increment j_cble.length. + } + } + // Now consult whichever callables have non-zero entry counts. + readBandData(bands, (uint32_t) - 1); + } } // Recursive helper to the previous function: void unpacker::attr_definitions::readBandData(band **body, uint32_t count) { - int j, k; - for (j = 0; body[j] != nullptr; j++) - { - band &b = *body[j]; - if (b.defc != nullptr) - { - // It has data, so read it. - b.readData(count); - } - switch (b.le_kind) - { - case EK_REPL: - { - int reps = b.getIntTotal(); - readBandData(b.le_body, reps); - } - break; - case EK_UN: - { - int remaining = count; - for (k = 0; b.le_body[k] != nullptr; k++) - { - band &k_case = *b.le_body[k]; - int k_count = 0; - if (k_case.le_casetags == nullptr) - { - k_count = remaining; // last (empty) case - } - else - { - int *tags = k_case.le_casetags; - int ntags = *tags++; // 1st element is length (why not?) - while (ntags-- > 0) - { - int tag = *tags++; - k_count += b.getIntCount(tag); - } - } - readBandData(k_case.le_body, k_count); - remaining -= k_count; - } - assert(remaining == 0); - } - break; - case EK_CALL: - // Push the count forward, if it is not a backward call. - if (!b.le_back) - { - band &cble = *b.le_body[0]; - assert(cble.le_kind == EK_CBLE); - cble.expectMoreLength(count); - } - break; - case EK_CBLE: - assert((int)count == -1); // incoming count is meaningless - k = b.length; - assert(k >= 0); - // This is intended and required for non production mode. - assert((b.length = -1)); // make it unable to accept more calls now. - readBandData(b.le_body, k); - break; - } - } + int j, k; + for (j = 0; body[j] != nullptr; j++) + { + band &b = *body[j]; + if (b.defc != nullptr) + { + // It has data, so read it. + b.readData(count); + } + switch (b.le_kind) + { + case EK_REPL: + { + int reps = b.getIntTotal(); + readBandData(b.le_body, reps); + } + break; + case EK_UN: + { + int remaining = count; + for (k = 0; b.le_body[k] != nullptr; k++) + { + band &k_case = *b.le_body[k]; + int k_count = 0; + if (k_case.le_casetags == nullptr) + { + k_count = remaining; // last (empty) case + } + else + { + int *tags = k_case.le_casetags; + int ntags = *tags++; // 1st element is length (why not?) + while (ntags-- > 0) + { + int tag = *tags++; + k_count += b.getIntCount(tag); + } + } + readBandData(k_case.le_body, k_count); + remaining -= k_count; + } + assert(remaining == 0); + } + break; + case EK_CALL: + // Push the count forward, if it is not a backward call. + if (!b.le_back) + { + band &cble = *b.le_body[0]; + assert(cble.le_kind == EK_CBLE); + cble.expectMoreLength(count); + } + break; + case EK_CBLE: + assert((int)count == -1); // incoming count is meaningless + k = b.length; + assert(k >= 0); + // This is intended and required for non production mode. + assert((b.length = -1)); // make it unable to accept more calls now. + readBandData(b.le_body, k); + break; + } + } } static inline band **findMatchingCase(int matchTag, band **cases) { - for (int k = 0; cases[k] != nullptr; k++) - { - band &k_case = *cases[k]; - if (k_case.le_casetags != nullptr) - { - // If it has tags, it must match a tag. - int *tags = k_case.le_casetags; - int ntags = *tags++; // 1st element is length - for (; ntags > 0; ntags--) - { - int tag = *tags++; - if (tag == matchTag) - break; - } - if (ntags == 0) - continue; // does not match - } - return k_case.le_body; - } - return nullptr; + for (int k = 0; cases[k] != nullptr; k++) + { + band &k_case = *cases[k]; + if (k_case.le_casetags != nullptr) + { + // If it has tags, it must match a tag. + int *tags = k_case.le_casetags; + int ntags = *tags++; // 1st element is length + for (; ntags > 0; ntags--) + { + int tag = *tags++; + if (tag == matchTag) + break; + } + if (ntags == 0) + continue; // does not match + } + return k_case.le_body; + } + return nullptr; } // write attribute band data: void unpacker::putlayout(band **body) { - int i; - int prevBII = -1; - int prevBCI = -1; - if (body == NULL) - { - unpack_abort("putlayout: unexpected NULL for body"); - return; - } - for (i = 0; body[i] != nullptr; i++) - { - band &b = *body[i]; - byte le_kind = b.le_kind; - - // Handle scalar part, if any. - int x = 0; - entry *e = nullptr; - if (b.defc != nullptr) - { - // It has data, so unparse an element. - if (b.ixTag != CONSTANT_None) - { - assert(le_kind == EK_REF); - if (b.ixTag == CONSTANT_Literal) - e = b.getRefUsing(cp.getKQIndex()); - else - e = b.getRefN(); - switch (b.le_len) - { - case 0: - break; - case 1: - putu1ref(e); - break; - case 2: - putref(e); - break; - case 4: - putu2(0); - putref(e); - break; - default: - assert(false); - } - } - else - { - assert(le_kind == EK_INT || le_kind == EK_REPL || le_kind == EK_UN); - x = b.getInt(); - - assert(!b.le_bci || prevBCI == (int)to_bci(prevBII)); - switch (b.le_bci) - { - case EK_BCI: // PH: transmit R(bci), store bci - x = to_bci(prevBII = x); - prevBCI = x; - break; - case EK_BCID: // POH: transmit D(R(bci)), store bci - x = to_bci(prevBII += x); - prevBCI = x; - break; - case EK_BCO: // OH: transmit D(R(bci)), store D(bci) - x = to_bci(prevBII += x) - prevBCI; - prevBCI += x; - break; - } - assert(!b.le_bci || prevBCI == (int)to_bci(prevBII)); - - switch (b.le_len) - { - case 0: - break; - case 1: - putu1(x); - break; - case 2: - putu2(x); - break; - case 4: - putu4(x); - break; - default: - assert(false); - } - } - } - - // Handle subparts, if any. - switch (le_kind) - { - case EK_REPL: - // x is the repeat count - while (x-- > 0) - { - putlayout(b.le_body); - } - break; - case EK_UN: - // x is the tag - putlayout(findMatchingCase(x, b.le_body)); - break; - case EK_CALL: - { - band &cble = *b.le_body[0]; - assert(cble.le_kind == EK_CBLE); - // FIXME: hit this one - // assert(cble.le_len == b.le_len); - putlayout(cble.le_body); - } - break; - - case EK_CBLE: - case EK_CASE: - assert(false); // should not reach here - } - } + int i; + int prevBII = -1; + int prevBCI = -1; + if (body == NULL) + { + unpack_abort("putlayout: unexpected NULL for body"); + return; + } + for (i = 0; body[i] != nullptr; i++) + { + band &b = *body[i]; + byte le_kind = b.le_kind; + + // Handle scalar part, if any. + int x = 0; + entry *e = nullptr; + if (b.defc != nullptr) + { + // It has data, so unparse an element. + if (b.ixTag != CONSTANT_None) + { + assert(le_kind == EK_REF); + if (b.ixTag == CONSTANT_Literal) + e = b.getRefUsing(cp.getKQIndex()); + else + e = b.getRefN(); + switch (b.le_len) + { + case 0: + break; + case 1: + putu1ref(e); + break; + case 2: + putref(e); + break; + case 4: + putu2(0); + putref(e); + break; + default: + assert(false); + } + } + else + { + assert(le_kind == EK_INT || le_kind == EK_REPL || le_kind == EK_UN); + x = b.getInt(); + + assert(!b.le_bci || prevBCI == (int)to_bci(prevBII)); + switch (b.le_bci) + { + case EK_BCI: // PH: transmit R(bci), store bci + x = to_bci(prevBII = x); + prevBCI = x; + break; + case EK_BCID: // POH: transmit D(R(bci)), store bci + x = to_bci(prevBII += x); + prevBCI = x; + break; + case EK_BCO: // OH: transmit D(R(bci)), store D(bci) + x = to_bci(prevBII += x) - prevBCI; + prevBCI += x; + break; + } + assert(!b.le_bci || prevBCI == (int)to_bci(prevBII)); + + switch (b.le_len) + { + case 0: + break; + case 1: + putu1(x); + break; + case 2: + putu2(x); + break; + case 4: + putu4(x); + break; + default: + assert(false); + } + } + } + + // Handle subparts, if any. + switch (le_kind) + { + case EK_REPL: + // x is the repeat count + while (x-- > 0) + { + putlayout(b.le_body); + } + break; + case EK_UN: + // x is the tag + putlayout(findMatchingCase(x, b.le_body)); + break; + case EK_CALL: + { + band &cble = *b.le_body[0]; + assert(cble.le_kind == EK_CBLE); + // FIXME: hit this one + // assert(cble.le_len == b.le_len); + putlayout(cble.le_body); + } + break; + + case EK_CBLE: + case EK_CASE: + assert(false); // should not reach here + } + } } void unpacker::read_files() { - file_name.readData(file_count); - if ((archive_options & AO_HAVE_FILE_SIZE_HI) != 0) - file_size_hi.readData(file_count); - file_size_lo.readData(file_count); - if ((archive_options & AO_HAVE_FILE_MODTIME) != 0) - file_modtime.readData(file_count); - int allFiles = file_count + class_count; - if ((archive_options & AO_HAVE_FILE_OPTIONS) != 0) - { - file_options.readData(file_count); - // FO_IS_CLASS_STUB might be set, causing overlap between classes and files - for (int i = 0; i < file_count; i++) - { - if ((file_options.getInt() & FO_IS_CLASS_STUB) != 0) - { - allFiles -= 1; // this one counts as both class and file - } - } - file_options.rewind(); - } - assert((default_file_options & FO_IS_CLASS_STUB) == 0); - files_remaining = allFiles; + file_name.readData(file_count); + if ((archive_options & AO_HAVE_FILE_SIZE_HI) != 0) + file_size_hi.readData(file_count); + file_size_lo.readData(file_count); + if ((archive_options & AO_HAVE_FILE_MODTIME) != 0) + file_modtime.readData(file_count); + int allFiles = file_count + class_count; + if ((archive_options & AO_HAVE_FILE_OPTIONS) != 0) + { + file_options.readData(file_count); + // FO_IS_CLASS_STUB might be set, causing overlap between classes and files + for (int i = 0; i < file_count; i++) + { + if ((file_options.getInt() & FO_IS_CLASS_STUB) != 0) + { + allFiles -= 1; // this one counts as both class and file + } + } + file_options.rewind(); + } + assert((default_file_options & FO_IS_CLASS_STUB) == 0); + files_remaining = allFiles; } void unpacker::get_code_header(int &max_stack, int &max_na_locals, int &handler_count, - int &cflags) + int &cflags) { - int sc = code_headers.getByte(); - if (sc == 0) - { - max_stack = max_na_locals = handler_count = cflags = -1; - return; - } - // Short code header is the usual case: - int nh; - int mod; - if (sc < 1 + 12 * 12) - { - sc -= 1; - nh = 0; - mod = 12; - } - else if (sc < 1 + 12 * 12 + 8 * 8) - { - sc -= 1 + 12 * 12; - nh = 1; - mod = 8; - } - else - { - assert(sc < 1 + 12 * 12 + 8 * 8 + 7 * 7); - sc -= 1 + 12 * 12 + 8 * 8; - nh = 2; - mod = 7; - } - max_stack = sc % mod; - max_na_locals = sc / mod; // caller must add static, siglen - handler_count = nh; - if ((archive_options & AO_HAVE_ALL_CODE_FLAGS) != 0) - cflags = -1; - else - cflags = 0; // this one has no attributes + int sc = code_headers.getByte(); + if (sc == 0) + { + max_stack = max_na_locals = handler_count = cflags = -1; + return; + } + // Short code header is the usual case: + int nh; + int mod; + if (sc < 1 + 12 * 12) + { + sc -= 1; + nh = 0; + mod = 12; + } + else if (sc < 1 + 12 * 12 + 8 * 8) + { + sc -= 1 + 12 * 12; + nh = 1; + mod = 8; + } + else + { + assert(sc < 1 + 12 * 12 + 8 * 8 + 7 * 7); + sc -= 1 + 12 * 12 + 8 * 8; + nh = 2; + mod = 7; + } + max_stack = sc % mod; + max_na_locals = sc / mod; // caller must add static, siglen + handler_count = nh; + if ((archive_options & AO_HAVE_ALL_CODE_FLAGS) != 0) + cflags = -1; + else + cflags = 0; // this one has no attributes } // Cf. PackageReader.readCodeHeaders void unpacker::read_code_headers() { - code_headers.readData(code_count); - int totalHandlerCount = 0; - int totalFlagsCount = 0; - for (int i = 0; i < code_count; i++) - { - int max_stack, max_locals, handler_count, cflags; - get_code_header(max_stack, max_locals, handler_count, cflags); - if (max_stack < 0) - code_max_stack.expectMoreLength(1); - if (max_locals < 0) - code_max_na_locals.expectMoreLength(1); - if (handler_count < 0) - code_handler_count.expectMoreLength(1); - else - totalHandlerCount += handler_count; - if (cflags < 0) - totalFlagsCount += 1; - } - code_headers.rewind(); // replay later during writing - - code_max_stack.readData(); - code_max_na_locals.readData(); - code_handler_count.readData(); - totalHandlerCount += code_handler_count.getIntTotal(); - - // Read handler specifications. - // Cf. PackageReader.readCodeHandlers. - code_handler_start_P.readData(totalHandlerCount); - code_handler_end_PO.readData(totalHandlerCount); - code_handler_catch_PO.readData(totalHandlerCount); - code_handler_class_RCN.readData(totalHandlerCount); - - read_attrs(ATTR_CONTEXT_CODE, totalFlagsCount); + code_headers.readData(code_count); + int totalHandlerCount = 0; + int totalFlagsCount = 0; + for (int i = 0; i < code_count; i++) + { + int max_stack, max_locals, handler_count, cflags; + get_code_header(max_stack, max_locals, handler_count, cflags); + if (max_stack < 0) + code_max_stack.expectMoreLength(1); + if (max_locals < 0) + code_max_na_locals.expectMoreLength(1); + if (handler_count < 0) + code_handler_count.expectMoreLength(1); + else + totalHandlerCount += handler_count; + if (cflags < 0) + totalFlagsCount += 1; + } + code_headers.rewind(); // replay later during writing + + code_max_stack.readData(); + code_max_na_locals.readData(); + code_handler_count.readData(); + totalHandlerCount += code_handler_count.getIntTotal(); + + // Read handler specifications. + // Cf. PackageReader.readCodeHandlers. + code_handler_start_P.readData(totalHandlerCount); + code_handler_end_PO.readData(totalHandlerCount); + code_handler_catch_PO.readData(totalHandlerCount); + code_handler_class_RCN.readData(totalHandlerCount); + + read_attrs(ATTR_CONTEXT_CODE, totalFlagsCount); } static inline bool is_in_range(uint32_t n, uint32_t min, uint32_t max) { - return n - min <= max - min; // unsigned arithmetic! + return n - min <= max - min; // unsigned arithmetic! } static inline bool is_field_op(int bc) { - return is_in_range(bc, bc_getstatic, bc_putfield); + return is_in_range(bc, bc_getstatic, bc_putfield); } static inline bool is_invoke_init_op(int bc) { - return is_in_range(bc, _invokeinit_op, _invokeinit_limit - 1); + return is_in_range(bc, _invokeinit_op, _invokeinit_limit - 1); } static inline bool is_self_linker_op(int bc) { - return is_in_range(bc, _self_linker_op, _self_linker_limit - 1); + return is_in_range(bc, _self_linker_op, _self_linker_limit - 1); } static bool is_branch_op(int bc) { - return is_in_range(bc, bc_ifeq, bc_jsr) || is_in_range(bc, bc_ifnull, bc_jsr_w); + return is_in_range(bc, bc_ifeq, bc_jsr) || is_in_range(bc, bc_ifnull, bc_jsr_w); } static bool is_local_slot_op(int bc) { - return is_in_range(bc, bc_iload, bc_aload) || is_in_range(bc, bc_istore, bc_astore) || - bc == bc_iinc || bc == bc_ret; + return is_in_range(bc, bc_iload, bc_aload) || is_in_range(bc, bc_istore, bc_astore) || + bc == bc_iinc || bc == bc_ret; } band *unpacker::ref_band_for_op(int bc) { - switch (bc) - { - case bc_ildc: - case bc_ildc_w: - return &bc_intref; - case bc_fldc: - case bc_fldc_w: - return &bc_floatref; - case bc_lldc2_w: - return &bc_longref; - case bc_dldc2_w: - return &bc_doubleref; - case bc_aldc: - case bc_aldc_w: - return &bc_stringref; - case bc_cldc: - case bc_cldc_w: - return &bc_classref; - - case bc_getstatic: - case bc_putstatic: - case bc_getfield: - case bc_putfield: - return &bc_fieldref; - - case bc_invokevirtual: - case bc_invokespecial: - case bc_invokestatic: - return &bc_methodref; - case bc_invokeinterface: - return &bc_imethodref; - - case bc_new: - case bc_anewarray: - case bc_checkcast: - case bc_instanceof: - case bc_multianewarray: - return &bc_classref; - } - return nullptr; + switch (bc) + { + case bc_ildc: + case bc_ildc_w: + return &bc_intref; + case bc_fldc: + case bc_fldc_w: + return &bc_floatref; + case bc_lldc2_w: + return &bc_longref; + case bc_dldc2_w: + return &bc_doubleref; + case bc_aldc: + case bc_aldc_w: + return &bc_stringref; + case bc_cldc: + case bc_cldc_w: + return &bc_classref; + + case bc_getstatic: + case bc_putstatic: + case bc_getfield: + case bc_putfield: + return &bc_fieldref; + + case bc_invokevirtual: + case bc_invokespecial: + case bc_invokestatic: + return &bc_methodref; + case bc_invokeinterface: + return &bc_imethodref; + + case bc_new: + case bc_anewarray: + case bc_checkcast: + case bc_instanceof: + case bc_multianewarray: + return &bc_classref; + } + return nullptr; } band *unpacker::ref_band_for_self_op(int bc, bool &isAloadVar, int &origBCVar) { - if (!is_self_linker_op(bc)) - return nullptr; - int idx = (bc - _self_linker_op); - bool isSuper = (idx >= _self_linker_super_flag); - if (isSuper) - idx -= _self_linker_super_flag; - bool isAload = (idx >= _self_linker_aload_flag); - if (isAload) - idx -= _self_linker_aload_flag; - int origBC = _first_linker_op + idx; - bool isField = is_field_op(origBC); - isAloadVar = isAload; - origBCVar = _first_linker_op + idx; - if (!isSuper) - return isField ? &bc_thisfield : &bc_thismethod; - else - return isField ? &bc_superfield : &bc_supermethod; + if (!is_self_linker_op(bc)) + return nullptr; + int idx = (bc - _self_linker_op); + bool isSuper = (idx >= _self_linker_super_flag); + if (isSuper) + idx -= _self_linker_super_flag; + bool isAload = (idx >= _self_linker_aload_flag); + if (isAload) + idx -= _self_linker_aload_flag; + int origBC = _first_linker_op + idx; + bool isField = is_field_op(origBC); + isAloadVar = isAload; + origBCVar = _first_linker_op + idx; + if (!isSuper) + return isField ? &bc_thisfield : &bc_thismethod; + else + return isField ? &bc_superfield : &bc_supermethod; } // Cf. PackageReader.readByteCodes inline // called exactly once => inline - void + void unpacker::read_bcs() { - // read from bc_codes and bc_case_count - fillbytes all_switch_ops; - all_switch_ops.init(); - - // Read directly from rp/rplimit. - // Do this later: bc_codes.readData(...) - byte *rp0 = rp; - - band *bc_which; - byte *opptr = rp; - byte *oplimit = rplimit; - - bool isAload; // passed by ref and then ignored - int junkBC; // passed by ref and then ignored - for (int k = 0; k < code_count; k++) - { - // Scan one method: - for (;;) - { - if (opptr + 2 > oplimit) - { - rp = opptr; - ensure_input(2); - oplimit = rplimit; - rp = rp0; // back up - } - if (opptr == oplimit) - { - unpack_abort(); - } - int bc = *opptr++ & 0xFF; - bool isWide = false; - if (bc == bc_wide) - { - if (opptr == oplimit) - { - unpack_abort(); - } - bc = *opptr++ & 0xFF; - isWide = true; - } - // Adjust expectations of various band sizes. - switch (bc) - { - case bc_tableswitch: - case bc_lookupswitch: - all_switch_ops.addByte(bc); - break; - case bc_iinc: - bc_local.expectMoreLength(1); - bc_which = isWide ? &bc_short : &bc_byte; - bc_which->expectMoreLength(1); - break; - case bc_sipush: - bc_short.expectMoreLength(1); - break; - case bc_bipush: - bc_byte.expectMoreLength(1); - break; - case bc_newarray: - bc_byte.expectMoreLength(1); - break; - case bc_multianewarray: - assert(ref_band_for_op(bc) == &bc_classref); - bc_classref.expectMoreLength(1); - bc_byte.expectMoreLength(1); - break; - case bc_ref_escape: - bc_escrefsize.expectMoreLength(1); - bc_escref.expectMoreLength(1); - break; - case bc_byte_escape: - bc_escsize.expectMoreLength(1); - // bc_escbyte will have to be counted too - break; - default: - if (is_invoke_init_op(bc)) - { - bc_initref.expectMoreLength(1); - break; - } - bc_which = ref_band_for_self_op(bc, isAload, junkBC); - if (bc_which != nullptr) - { - bc_which->expectMoreLength(1); - break; - } - if (is_branch_op(bc)) - { - bc_label.expectMoreLength(1); - break; - } - bc_which = ref_band_for_op(bc); - if (bc_which != nullptr) - { - bc_which->expectMoreLength(1); - assert(bc != bc_multianewarray); // handled elsewhere - break; - } - if (is_local_slot_op(bc)) - { - bc_local.expectMoreLength(1); - break; - } - break; - case bc_end_marker: - // Increment k and test against code_count. - goto doneScanningMethod; - } - } - doneScanningMethod: - { - } - } - - // Go through the formality, so we can use it in a regular fashion later: - assert(rp == rp0); - bc_codes.readData((int)(opptr - rp)); - - int i = 0; - - // To size instruction bands correctly, we need info on switches: - bc_case_count.readData((int)all_switch_ops.size()); - for (i = 0; i < (int)all_switch_ops.size(); i++) - { - int caseCount = bc_case_count.getInt(); - int bc = all_switch_ops.getByte(i); - bc_label.expectMoreLength(1 + caseCount); // default label + cases - bc_case_value.expectMoreLength(bc == bc_tableswitch ? 1 : caseCount); - } - bc_case_count.rewind(); // uses again for output - - all_switch_ops.free(); - - for (i = e_bc_case_value; i <= e_bc_escsize; i++) - { - all_bands[i].readData(); - } - - // The bc_escbyte band is counted by the immediately previous band. - bc_escbyte.readData(bc_escsize.getIntTotal()); + // read from bc_codes and bc_case_count + fillbytes all_switch_ops; + all_switch_ops.init(); + + // Read directly from rp/rplimit. + // Do this later: bc_codes.readData(...) + byte *rp0 = rp; + + band *bc_which; + byte *opptr = rp; + byte *oplimit = rplimit; + + bool isAload; // passed by ref and then ignored + int junkBC; // passed by ref and then ignored + for (int k = 0; k < code_count; k++) + { + // Scan one method: + for (;;) + { + if (opptr + 2 > oplimit) + { + rp = opptr; + ensure_input(2); + oplimit = rplimit; + rp = rp0; // back up + } + if (opptr == oplimit) + { + unpack_abort(); + } + int bc = *opptr++ & 0xFF; + bool isWide = false; + if (bc == bc_wide) + { + if (opptr == oplimit) + { + unpack_abort(); + } + bc = *opptr++ & 0xFF; + isWide = true; + } + // Adjust expectations of various band sizes. + switch (bc) + { + case bc_tableswitch: + case bc_lookupswitch: + all_switch_ops.addByte(bc); + break; + case bc_iinc: + bc_local.expectMoreLength(1); + bc_which = isWide ? &bc_short : &bc_byte; + bc_which->expectMoreLength(1); + break; + case bc_sipush: + bc_short.expectMoreLength(1); + break; + case bc_bipush: + bc_byte.expectMoreLength(1); + break; + case bc_newarray: + bc_byte.expectMoreLength(1); + break; + case bc_multianewarray: + assert(ref_band_for_op(bc) == &bc_classref); + bc_classref.expectMoreLength(1); + bc_byte.expectMoreLength(1); + break; + case bc_ref_escape: + bc_escrefsize.expectMoreLength(1); + bc_escref.expectMoreLength(1); + break; + case bc_byte_escape: + bc_escsize.expectMoreLength(1); + // bc_escbyte will have to be counted too + break; + default: + if (is_invoke_init_op(bc)) + { + bc_initref.expectMoreLength(1); + break; + } + bc_which = ref_band_for_self_op(bc, isAload, junkBC); + if (bc_which != nullptr) + { + bc_which->expectMoreLength(1); + break; + } + if (is_branch_op(bc)) + { + bc_label.expectMoreLength(1); + break; + } + bc_which = ref_band_for_op(bc); + if (bc_which != nullptr) + { + bc_which->expectMoreLength(1); + assert(bc != bc_multianewarray); // handled elsewhere + break; + } + if (is_local_slot_op(bc)) + { + bc_local.expectMoreLength(1); + break; + } + break; + case bc_end_marker: + // Increment k and test against code_count. + goto doneScanningMethod; + } + } + doneScanningMethod: + { + } + } + + // Go through the formality, so we can use it in a regular fashion later: + assert(rp == rp0); + bc_codes.readData((int)(opptr - rp)); + + int i = 0; + + // To size instruction bands correctly, we need info on switches: + bc_case_count.readData((int)all_switch_ops.size()); + for (i = 0; i < (int)all_switch_ops.size(); i++) + { + int caseCount = bc_case_count.getInt(); + int bc = all_switch_ops.getByte(i); + bc_label.expectMoreLength(1 + caseCount); // default label + cases + bc_case_value.expectMoreLength(bc == bc_tableswitch ? 1 : caseCount); + } + bc_case_count.rewind(); // uses again for output + + all_switch_ops.free(); + + for (i = e_bc_case_value; i <= e_bc_escsize; i++) + { + all_bands[i].readData(); + } + + // The bc_escbyte band is counted by the immediately previous band. + bc_escbyte.readData(bc_escsize.getIntTotal()); } void unpacker::read_bands() { - read_file_header(); - - if (cp.nentries == 0) - { - // read_file_header failed to read a CP, because it copied a JAR. - return; - } - - // Do this after the file header has been read: - check_options(); - - read_cp(); - read_attr_defs(); - read_ics(); - read_classes(); - read_bcs(); - read_files(); + read_file_header(); + + if (cp.nentries == 0) + { + // read_file_header failed to read a CP, because it copied a JAR. + return; + } + + // Do this after the file header has been read: + check_options(); + + read_cp(); + read_attr_defs(); + read_ics(); + read_classes(); + read_bcs(); + read_files(); } /// CP routines entry *&constant_pool::hashTabRef(byte tag, bytes &b) { - uint32_t hash = tag + (int)b.len; - for (int i = 0; i < (int)b.len; i++) - { - hash = hash * 31 + (0xFF & b.ptr[i]); - } - entry **ht = hashTab; - int hlen = hashTabLength; - assert((hlen & (hlen - 1)) == 0); // must be power of 2 - uint32_t hash1 = hash & (hlen - 1); // == hash % hlen - uint32_t hash2 = 0; // lazily computed (requires mod op.) - int probes = 0; - while (ht[hash1] != nullptr) - { - entry &e = *ht[hash1]; - if (e.value.b.equals(b) && e.tag == tag) - break; - if (hash2 == 0) - // Note: hash2 must be relatively prime to hlen, hence the "|1". - hash2 = (((hash % 499) & (hlen - 1)) | 1); - hash1 += hash2; - if (hash1 >= (uint32_t)hlen) - hash1 -= hlen; - assert(hash1 < (uint32_t)hlen); - assert(++probes < hlen); - } - return ht[hash1]; + uint32_t hash = tag + (int)b.len; + for (int i = 0; i < (int)b.len; i++) + { + hash = hash * 31 + (0xFF & b.ptr[i]); + } + entry **ht = hashTab; + int hlen = hashTabLength; + assert((hlen & (hlen - 1)) == 0); // must be power of 2 + uint32_t hash1 = hash & (hlen - 1); // == hash % hlen + uint32_t hash2 = 0; // lazily computed (requires mod op.) +#ifndef NDEBUG + int probes = 0; +#endif + while (ht[hash1] != nullptr) + { + entry &e = *ht[hash1]; + if (e.value.b.equals(b) && e.tag == tag) + break; + if (hash2 == 0) + // Note: hash2 must be relatively prime to hlen, hence the "|1". + hash2 = (((hash % 499) & (hlen - 1)) | 1); + hash1 += hash2; + if (hash1 >= (uint32_t)hlen) + hash1 -= hlen; + assert(hash1 < (uint32_t)hlen); + assert(++probes < hlen); + } + return ht[hash1]; } static void insert_extra(entry *e, ptrlist &extras) { - // This ordering helps implement the Pack200 requirement - // of a predictable CP order in the class files produced. - e->inord = NO_INORD; // mark as an "extra" - extras.add(e); - // Note: We will sort the list (by string-name) later. + // This ordering helps implement the Pack200 requirement + // of a predictable CP order in the class files produced. + e->inord = NO_INORD; // mark as an "extra" + extras.add(e); + // Note: We will sort the list (by string-name) later. } entry *constant_pool::ensureUtf8(bytes &b) { - entry *&ix = hashTabRef(CONSTANT_Utf8, b); - if (ix != nullptr) - return ix; - // Make one. - if (nentries == maxentries) - { - unpack_abort("cp utf8 overflow"); - return &entries[tag_base[CONSTANT_Utf8]]; // return something - } - entry &e = entries[nentries++]; - e.tag = CONSTANT_Utf8; - u->saveTo(e.value.b, b); - assert(&e >= first_extra_entry); - insert_extra(&e, tag_extras[CONSTANT_Utf8]); - return ix = &e; + entry *&ix = hashTabRef(CONSTANT_Utf8, b); + if (ix != nullptr) + return ix; + // Make one. + if (nentries == maxentries) + { + unpack_abort("cp utf8 overflow"); + return &entries[tag_base[CONSTANT_Utf8]]; // return something + } + entry &e = entries[nentries++]; + e.tag = CONSTANT_Utf8; + u->saveTo(e.value.b, b); + assert(&e >= first_extra_entry); + insert_extra(&e, tag_extras[CONSTANT_Utf8]); + return ix = &e; } entry *constant_pool::ensureClass(bytes &b) { - entry *&ix = hashTabRef(CONSTANT_Class, b); - if (ix != nullptr) - return ix; - // Make one. - if (nentries == maxentries) - { - unpack_abort("cp class overflow"); - return &entries[tag_base[CONSTANT_Class]]; // return something - } - entry &e = entries[nentries++]; - e.tag = CONSTANT_Class; - e.nrefs = 1; - e.refs = U_NEW(entry *, 1); - ix = &e; // hold my spot in the index - entry *utf = ensureUtf8(b); - e.refs[0] = utf; - e.value.b = utf->value.b; - assert(&e >= first_extra_entry); - insert_extra(&e, tag_extras[CONSTANT_Class]); - return &e; + entry *&ix = hashTabRef(CONSTANT_Class, b); + if (ix != nullptr) + return ix; + // Make one. + if (nentries == maxentries) + { + unpack_abort("cp class overflow"); + return &entries[tag_base[CONSTANT_Class]]; // return something + } + entry &e = entries[nentries++]; + e.tag = CONSTANT_Class; + e.nrefs = 1; + e.refs = U_NEW(entry *, 1); + ix = &e; // hold my spot in the index + entry *utf = ensureUtf8(b); + e.refs[0] = utf; + e.value.b = utf->value.b; + assert(&e >= first_extra_entry); + insert_extra(&e, tag_extras[CONSTANT_Class]); + return &e; } void constant_pool::expandSignatures() { - int i; - int nsigs = 0; - int nreused = 0; - int first_sig = tag_base[CONSTANT_Signature]; - int sig_limit = tag_count[CONSTANT_Signature] + first_sig; - fillbytes buf; - buf.init(1 << 10); - for (i = first_sig; i < sig_limit; i++) - { - entry &e = entries[i]; - assert(e.tag == CONSTANT_Signature); - int refnum = 0; - bytes form = e.refs[refnum++]->asUtf8(); - buf.empty(); - for (int j = 0; j < (int)form.len; j++) - { - int c = form.ptr[j]; - buf.addByte(c); - if (c == 'L') - { - entry *cls = e.refs[refnum++]; - buf.append(cls->className()->asUtf8()); - } - } - assert(refnum == e.nrefs); - bytes &sig = buf.b; - - // try to find a pre-existing Utf8: - entry *&e2 = hashTabRef(CONSTANT_Utf8, sig); - if (e2 != nullptr) - { - assert(e2->isUtf8(sig)); - e.value.b = e2->value.b; - e.refs[0] = e2; - e.nrefs = 1; - nreused++; - } - else - { - // there is no other replacement; reuse this CP entry as a Utf8 - u->saveTo(e.value.b, sig); - e.tag = CONSTANT_Utf8; - e.nrefs = 0; - e2 = &e; - } - nsigs++; - } - buf.free(); - - // go expunge all references to remaining signatures: - for (i = 0; i < (int)nentries; i++) - { - entry &e = entries[i]; - for (int j = 0; j < e.nrefs; j++) - { - entry *&e2 = e.refs[j]; - if (e2 != nullptr && e2->tag == CONSTANT_Signature) - e2 = e2->refs[0]; - } - } + int i; + int nsigs = 0; + int nreused = 0; + int first_sig = tag_base[CONSTANT_Signature]; + int sig_limit = tag_count[CONSTANT_Signature] + first_sig; + fillbytes buf; + buf.init(1 << 10); + for (i = first_sig; i < sig_limit; i++) + { + entry &e = entries[i]; + assert(e.tag == CONSTANT_Signature); + int refnum = 0; + bytes form = e.refs[refnum++]->asUtf8(); + buf.empty(); + for (int j = 0; j < (int)form.len; j++) + { + int c = form.ptr[j]; + buf.addByte(c); + if (c == 'L') + { + entry *cls = e.refs[refnum++]; + buf.append(cls->className()->asUtf8()); + } + } + assert(refnum == e.nrefs); + bytes &sig = buf.b; + + // try to find a pre-existing Utf8: + entry *&e2 = hashTabRef(CONSTANT_Utf8, sig); + if (e2 != nullptr) + { + assert(e2->isUtf8(sig)); + e.value.b = e2->value.b; + e.refs[0] = e2; + e.nrefs = 1; + nreused++; + } + else + { + // there is no other replacement; reuse this CP entry as a Utf8 + u->saveTo(e.value.b, sig); + e.tag = CONSTANT_Utf8; + e.nrefs = 0; + e2 = &e; + } + nsigs++; + } + buf.free(); + + // go expunge all references to remaining signatures: + for (i = 0; i < (int)nentries; i++) + { + entry &e = entries[i]; + for (int j = 0; j < e.nrefs; j++) + { + entry *&e2 = e.refs[j]; + if (e2 != nullptr && e2->tag == CONSTANT_Signature) + e2 = e2->refs[0]; + } + } } void constant_pool::initMemberIndexes() { - // This function does NOT refer to any class schema. - // It is totally internal to the cpool. - int i, j; - - // Get the pre-existing indexes: - int nclasses = tag_count[CONSTANT_Class]; - // entry *classes = tag_base[CONSTANT_Class] + entries; // UNUSED - int nfields = tag_count[CONSTANT_Fieldref]; - entry *fields = tag_base[CONSTANT_Fieldref] + entries; - int nmethods = tag_count[CONSTANT_Methodref]; - entry *methods = tag_base[CONSTANT_Methodref] + entries; - - int *field_counts = T_NEW(int, nclasses); - int *method_counts = T_NEW(int, nclasses); - cpindex *all_indexes = U_NEW(cpindex, nclasses * 2); - entry **field_ix = U_NEW(entry *, add_size(nfields, nclasses)); - entry **method_ix = U_NEW(entry *, add_size(nmethods, nclasses)); - - for (j = 0; j < nfields; j++) - { - entry &f = fields[j]; - i = f.memberClass()->inord; - assert(i < nclasses); - field_counts[i]++; - } - for (j = 0; j < nmethods; j++) - { - entry &m = methods[j]; - i = m.memberClass()->inord; - assert(i < nclasses); - method_counts[i]++; - } - - int fbase = 0, mbase = 0; - for (i = 0; i < nclasses; i++) - { - int fc = field_counts[i]; - int mc = method_counts[i]; - all_indexes[i * 2 + 0].init(fc, field_ix + fbase, CONSTANT_Fieldref + SUBINDEX_BIT); - all_indexes[i * 2 + 1].init(mc, method_ix + mbase, CONSTANT_Methodref + SUBINDEX_BIT); - // reuse field_counts and member_counts as fill pointers: - field_counts[i] = fbase; - method_counts[i] = mbase; - fbase += fc + 1; - mbase += mc + 1; - // (the +1 leaves a space between every subarray) - } - assert(fbase == nfields + nclasses); - assert(mbase == nmethods + nclasses); - - for (j = 0; j < nfields; j++) - { - entry &f = fields[j]; - i = f.memberClass()->inord; - field_ix[field_counts[i]++] = &f; - } - for (j = 0; j < nmethods; j++) - { - entry &m = methods[j]; - i = m.memberClass()->inord; - method_ix[method_counts[i]++] = &m; - } - - member_indexes = all_indexes; - - // Free intermediate buffers. - u->free_temps(); + // This function does NOT refer to any class schema. + // It is totally internal to the cpool. + int i, j; + + // Get the pre-existing indexes: + int nclasses = tag_count[CONSTANT_Class]; + // entry *classes = tag_base[CONSTANT_Class] + entries; // UNUSED + int nfields = tag_count[CONSTANT_Fieldref]; + entry *fields = tag_base[CONSTANT_Fieldref] + entries; + int nmethods = tag_count[CONSTANT_Methodref]; + entry *methods = tag_base[CONSTANT_Methodref] + entries; + + int *field_counts = T_NEW(int, nclasses); + int *method_counts = T_NEW(int, nclasses); + cpindex *all_indexes = U_NEW(cpindex, nclasses * 2); + entry **field_ix = U_NEW(entry *, add_size(nfields, nclasses)); + entry **method_ix = U_NEW(entry *, add_size(nmethods, nclasses)); + + for (j = 0; j < nfields; j++) + { + entry &f = fields[j]; + i = f.memberClass()->inord; + assert(i < nclasses); + field_counts[i]++; + } + for (j = 0; j < nmethods; j++) + { + entry &m = methods[j]; + i = m.memberClass()->inord; + assert(i < nclasses); + method_counts[i]++; + } + + int fbase = 0, mbase = 0; + for (i = 0; i < nclasses; i++) + { + int fc = field_counts[i]; + int mc = method_counts[i]; + all_indexes[i * 2 + 0].init(fc, field_ix + fbase, CONSTANT_Fieldref + SUBINDEX_BIT); + all_indexes[i * 2 + 1].init(mc, method_ix + mbase, CONSTANT_Methodref + SUBINDEX_BIT); + // reuse field_counts and member_counts as fill pointers: + field_counts[i] = fbase; + method_counts[i] = mbase; + fbase += fc + 1; + mbase += mc + 1; + // (the +1 leaves a space between every subarray) + } + assert(fbase == nfields + nclasses); + assert(mbase == nmethods + nclasses); + + for (j = 0; j < nfields; j++) + { + entry &f = fields[j]; + i = f.memberClass()->inord; + field_ix[field_counts[i]++] = &f; + } + for (j = 0; j < nmethods; j++) + { + entry &m = methods[j]; + i = m.memberClass()->inord; + method_ix[method_counts[i]++] = &m; + } + + member_indexes = all_indexes; + + // Free intermediate buffers. + u->free_temps(); } void entry::requestOutputIndex(constant_pool &cp, int req) { - assert(outputIndex <= NOT_REQUESTED); // must not have assigned indexes yet - if (tag == CONSTANT_Signature) - { - ref(0)->requestOutputIndex(cp, req); - return; - } - assert(req == REQUESTED || req == REQUESTED_LDC); - if (outputIndex != NOT_REQUESTED) - { - if (req == REQUESTED_LDC) - outputIndex = req; // this kind has precedence - return; - } - outputIndex = req; - // assert(!cp.outputEntries.contains(this)); - assert(tag != CONSTANT_Signature); - cp.outputEntries.add(this); - for (int j = 0; j < nrefs; j++) - { - ref(j)->requestOutputIndex(cp); - } + assert(outputIndex <= NOT_REQUESTED); // must not have assigned indexes yet + if (tag == CONSTANT_Signature) + { + ref(0)->requestOutputIndex(cp, req); + return; + } + assert(req == REQUESTED || req == REQUESTED_LDC); + if (outputIndex != NOT_REQUESTED) + { + if (req == REQUESTED_LDC) + outputIndex = req; // this kind has precedence + return; + } + outputIndex = req; + // assert(!cp.outputEntries.contains(this)); + assert(tag != CONSTANT_Signature); + cp.outputEntries.add(this); + for (int j = 0; j < nrefs; j++) + { + ref(j)->requestOutputIndex(cp); + } } void constant_pool::resetOutputIndexes() { - int i; - int noes = outputEntries.length(); - entry **oes = (entry **)outputEntries.base(); - for (i = 0; i < noes; i++) - { - entry &e = *oes[i]; - e.outputIndex = NOT_REQUESTED; - } - outputIndexLimit = 0; - outputEntries.empty(); + int i; + int noes = outputEntries.length(); + entry **oes = (entry **)outputEntries.base(); + for (i = 0; i < noes; i++) + { + entry &e = *oes[i]; + e.outputIndex = NOT_REQUESTED; + } + outputIndexLimit = 0; + outputEntries.empty(); } static const byte TAG_ORDER[CONSTANT_Limit] = {0, 1, 0, 2, 3, 4, 5, 7, 6, 10, 11, 12, 9, 8}; extern "C" int outputEntry_cmp(const void *e1p, const void *e2p) { - // Sort entries according to the Pack200 rules for deterministic - // constant pool ordering. - // - // The four sort keys as follows, in order of decreasing importance: - // 1. ldc first, then non-ldc guys - // 2. normal cp_All entries by input order (i.e., address order) - // 3. after that, extra entries by lexical order (as in tag_extras[*]) - entry &e1 = *(entry *)*(void **)e1p; - entry &e2 = *(entry *)*(void **)e2p; - int oi1 = e1.outputIndex; - int oi2 = e2.outputIndex; - assert(oi1 == REQUESTED || oi1 == REQUESTED_LDC); - assert(oi2 == REQUESTED || oi2 == REQUESTED_LDC); - if (oi1 != oi2) - { - if (oi1 == REQUESTED_LDC) - return 0 - 1; - if (oi2 == REQUESTED_LDC) - return 1 - 0; - // Else fall through; neither is an ldc request. - } - if (e1.inord != NO_INORD || e2.inord != NO_INORD) - { - // One or both is normal. Use input order. - if (&e1 > &e2) - return 1 - 0; - if (&e1 < &e2) - return 0 - 1; - return 0; // equal pointers - } - // Both are extras. Sort by tag and then by value. - if (e1.tag != e2.tag) - { - return TAG_ORDER[e1.tag] - TAG_ORDER[e2.tag]; - } - // If the tags are the same, use string comparison. - return compare_Utf8_chars(e1.value.b, e2.value.b); + // Sort entries according to the Pack200 rules for deterministic + // constant pool ordering. + // + // The four sort keys as follows, in order of decreasing importance: + // 1. ldc first, then non-ldc guys + // 2. normal cp_All entries by input order (i.e., address order) + // 3. after that, extra entries by lexical order (as in tag_extras[*]) + entry &e1 = *(entry *)*(void **)e1p; + entry &e2 = *(entry *)*(void **)e2p; + int oi1 = e1.outputIndex; + int oi2 = e2.outputIndex; + assert(oi1 == REQUESTED || oi1 == REQUESTED_LDC); + assert(oi2 == REQUESTED || oi2 == REQUESTED_LDC); + if (oi1 != oi2) + { + if (oi1 == REQUESTED_LDC) + return 0 - 1; + if (oi2 == REQUESTED_LDC) + return 1 - 0; + // Else fall through; neither is an ldc request. + } + if (e1.inord != NO_INORD || e2.inord != NO_INORD) + { + // One or both is normal. Use input order. + if (&e1 > &e2) + return 1 - 0; + if (&e1 < &e2) + return 0 - 1; + return 0; // equal pointers + } + // Both are extras. Sort by tag and then by value. + if (e1.tag != e2.tag) + { + return TAG_ORDER[e1.tag] - TAG_ORDER[e2.tag]; + } + // If the tags are the same, use string comparison. + return compare_Utf8_chars(e1.value.b, e2.value.b); } void constant_pool::computeOutputIndexes() { - int i; - - int noes = outputEntries.length(); - entry **oes = (entry **)outputEntries.base(); - - // Sort the output constant pool into the order required by Pack200. - PTRLIST_QSORT(outputEntries, outputEntry_cmp); - - // Allocate a new index for each entry that needs one. - // We do this in two passes, one for LDC entries and one for the rest. - int nextIndex = 1; // always skip index #0 in output cpool - for (i = 0; i < noes; i++) - { - entry &e = *oes[i]; - assert(e.outputIndex == REQUESTED || e.outputIndex == REQUESTED_LDC); - e.outputIndex = nextIndex++; - if (e.isDoubleWord()) - nextIndex++; // do not use the next index - } - outputIndexLimit = nextIndex; + int i; + + int noes = outputEntries.length(); + entry **oes = (entry **)outputEntries.base(); + + // Sort the output constant pool into the order required by Pack200. + PTRLIST_QSORT(outputEntries, outputEntry_cmp); + + // Allocate a new index for each entry that needs one. + // We do this in two passes, one for LDC entries and one for the rest. + int nextIndex = 1; // always skip index #0 in output cpool + for (i = 0; i < noes; i++) + { + entry &e = *oes[i]; + assert(e.outputIndex == REQUESTED || e.outputIndex == REQUESTED_LDC); + e.outputIndex = nextIndex++; + if (e.isDoubleWord()) + nextIndex++; // do not use the next index + } + outputIndexLimit = nextIndex; } // Unpacker Start @@ -3477,64 +3473,63 @@ void constant_pool::computeOutputIndexes() // Do not reset any unpack options. void unpacker::reset() { - bytes_read_before_reset += bytes_read; - bytes_written_before_reset += bytes_written; - files_written_before_reset += files_written; - classes_written_before_reset += classes_written; - segments_read_before_reset += 1; - if (verbose >= 2) - { - fprintf(stderr, "After segment %d, " LONG_LONG_FORMAT - " bytes read and " LONG_LONG_FORMAT " bytes written.\n", - segments_read_before_reset - 1, bytes_read_before_reset, - bytes_written_before_reset); - fprintf(stderr, - "After segment %d, %d files (of which %d are classes) written to output.\n", - segments_read_before_reset - 1, files_written_before_reset, - classes_written_before_reset); - if (archive_next_count != 0) - { - fprintf(stderr, "After segment %d, %d segment%s remaining (estimated).\n", - segments_read_before_reset - 1, archive_next_count, - archive_next_count == 1 ? "" : "s"); - } - } - - unpacker save_u = (*this); // save bytewise image - infileptr = nullptr; // make asserts happy - jarout = nullptr; // do not close the output jar - gzin = nullptr; // do not close the input gzip stream - this->free(); - this->init(read_input_fn); - - // restore selected interface state: - infileptr = save_u.infileptr; - inbytes = save_u.inbytes; - jarout = save_u.jarout; - gzin = save_u.gzin; - verbose = save_u.verbose; - deflate_hint_or_zero = save_u.deflate_hint_or_zero; - modification_time_or_zero = save_u.modification_time_or_zero; - bytes_read_before_reset = save_u.bytes_read_before_reset; - bytes_written_before_reset = save_u.bytes_written_before_reset; - files_written_before_reset = save_u.files_written_before_reset; - classes_written_before_reset = save_u.classes_written_before_reset; - segments_read_before_reset = save_u.segments_read_before_reset; - // Note: If we use strip_names, watch out: They get nuked here. + bytes_read_before_reset += bytes_read; + bytes_written_before_reset += bytes_written; + files_written_before_reset += files_written; + classes_written_before_reset += classes_written; + segments_read_before_reset += 1; + if (verbose >= 2) + { + fprintf(stderr, "After segment %d, %" PRIu64 " bytes read and %" PRIu64 " bytes written.\n", + segments_read_before_reset - 1, bytes_read_before_reset, + bytes_written_before_reset); + fprintf(stderr, + "After segment %d, %d files (of which %d are classes) written to output.\n", + segments_read_before_reset - 1, files_written_before_reset, + classes_written_before_reset); + if (archive_next_count != 0) + { + fprintf(stderr, "After segment %d, %d segment%s remaining (estimated).\n", + segments_read_before_reset - 1, archive_next_count, + archive_next_count == 1 ? "" : "s"); + } + } + + unpacker save_u = (*this); // save bytewise image + infileptr = nullptr; // make asserts happy + jarout = nullptr; // do not close the output jar + gzin = nullptr; // do not close the input gzip stream + this->free(); + this->init(read_input_fn); + + // restore selected interface state: + infileptr = save_u.infileptr; + inbytes = save_u.inbytes; + jarout = save_u.jarout; + gzin = save_u.gzin; + verbose = save_u.verbose; + deflate_hint_or_zero = save_u.deflate_hint_or_zero; + modification_time_or_zero = save_u.modification_time_or_zero; + bytes_read_before_reset = save_u.bytes_read_before_reset; + bytes_written_before_reset = save_u.bytes_written_before_reset; + files_written_before_reset = save_u.files_written_before_reset; + classes_written_before_reset = save_u.classes_written_before_reset; + segments_read_before_reset = save_u.segments_read_before_reset; + // Note: If we use strip_names, watch out: They get nuked here. } void unpacker::init(read_input_fn_t input_fn) { - int i; - BYTES_OF(*this).clear(); - this->u = this; // self-reference for U_NEW macro - read_input_fn = input_fn; - all_bands = band::makeBands(this); - // Make a default jar buffer; caller may safely overwrite it. - jarout = U_NEW(jar, 1); - jarout->init(this); - for (i = 0; i < ATTR_CONTEXT_LIMIT; i++) - attr_defs[i].u = u; // set up outer ptr + int i; + BYTES_OF(*this).clear(); + this->u = this; // self-reference for U_NEW macro + read_input_fn = input_fn; + all_bands = band::makeBands(this); + // Make a default jar buffer; caller may safely overwrite it. + jarout = U_NEW(jar, 1); + jarout->init(this); + for (i = 0; i < ATTR_CONTEXT_LIMIT; i++) + attr_defs[i].u = u; // set up outer ptr } // Usage: unpack a byte buffer @@ -3543,1251 +3538,1253 @@ void unpacker::init(read_input_fn_t input_fn) // If nullptr, the callback is used to fill an internal buffer. void unpacker::start(void *packptr, size_t len) { - if (packptr != nullptr && len != 0) - { - inbytes.set((byte *)packptr, len); - } - read_bands(); + if (packptr != nullptr && len != 0) + { + inbytes.set((byte *)packptr, len); + } + read_bands(); } void unpacker::check_options() { - if (deflate_hint_or_zero != 0) - { - bool force_deflate_hint = (deflate_hint_or_zero > 0); - if (force_deflate_hint) - default_file_options |= FO_DEFLATE_HINT; - else - default_file_options &= ~FO_DEFLATE_HINT; - // Turn off per-file deflate hint by force. - suppress_file_options |= FO_DEFLATE_HINT; - } - if (modification_time_or_zero != 0) - { - default_file_modtime = modification_time_or_zero; - // Turn off per-file modtime by force. - archive_options &= ~AO_HAVE_FILE_MODTIME; - } + if (deflate_hint_or_zero != 0) + { + bool force_deflate_hint = (deflate_hint_or_zero > 0); + if (force_deflate_hint) + default_file_options |= FO_DEFLATE_HINT; + else + default_file_options &= ~FO_DEFLATE_HINT; + // Turn off per-file deflate hint by force. + suppress_file_options |= FO_DEFLATE_HINT; + } + if (modification_time_or_zero != 0) + { + default_file_modtime = modification_time_or_zero; + // Turn off per-file modtime by force. + archive_options &= ~AO_HAVE_FILE_MODTIME; + } } // classfile writing void unpacker::reset_cur_classfile() { - // set defaults - cur_class_minver = default_class_minver; - cur_class_majver = default_class_majver; - - // reset constant pool state - cp.resetOutputIndexes(); - - // reset fixups - class_fixup_type.empty(); - class_fixup_offset.empty(); - class_fixup_ref.empty(); - requested_ics.empty(); + // set defaults + cur_class_minver = default_class_minver; + cur_class_majver = default_class_majver; + + // reset constant pool state + cp.resetOutputIndexes(); + + // reset fixups + class_fixup_type.empty(); + class_fixup_offset.empty(); + class_fixup_ref.empty(); + requested_ics.empty(); } cpindex *constant_pool::getKQIndex() { - char ch = '?'; - if (u->cur_descr != nullptr) - { - entry *type = u->cur_descr->descrType(); - ch = type->value.b.ptr[0]; - } - byte tag = CONSTANT_Integer; - switch (ch) - { - case 'L': - tag = CONSTANT_String; - break; - case 'I': - tag = CONSTANT_Integer; - break; - case 'J': - tag = CONSTANT_Long; - break; - case 'F': - tag = CONSTANT_Float; - break; - case 'D': - tag = CONSTANT_Double; - break; - case 'B': - case 'S': - case 'C': - case 'Z': - tag = CONSTANT_Integer; - break; - default: - unpack_abort("bad KQ reference"); - break; - } - return getIndex(tag); + char ch = '?'; + if (u->cur_descr != nullptr) + { + entry *type = u->cur_descr->descrType(); + ch = type->value.b.ptr[0]; + } + byte tag = CONSTANT_Integer; + switch (ch) + { + case 'L': + tag = CONSTANT_String; + break; + case 'I': + tag = CONSTANT_Integer; + break; + case 'J': + tag = CONSTANT_Long; + break; + case 'F': + tag = CONSTANT_Float; + break; + case 'D': + tag = CONSTANT_Double; + break; + case 'B': + case 'S': + case 'C': + case 'Z': + tag = CONSTANT_Integer; + break; + default: + unpack_abort("bad KQ reference"); + break; + } + return getIndex(tag); } uint32_t unpacker::to_bci(uint32_t bii) { - uint32_t len = bcimap.length(); - uint32_t *map = (uint32_t *)bcimap.base(); - assert(len > 0); // must be initialized before using to_bci - if (bii < len) - return map[bii]; - // Else it's a fractional or out-of-range BCI. - uint32_t key = bii - len; - for (int i = len;; i--) - { - if (map[i - 1] - (i - 1) <= key) - break; - else - --bii; - } - return bii; + uint32_t len = bcimap.length(); + uint32_t *map = (uint32_t *)bcimap.base(); + assert(len > 0); // must be initialized before using to_bci + if (bii < len) + return map[bii]; + // Else it's a fractional or out-of-range BCI. + uint32_t key = bii - len; + for (int i = len;; i--) + { + if (map[i - 1] - (i - 1) <= key) + break; + else + --bii; + } + return bii; } void unpacker::put_stackmap_type() { - int tag = code_StackMapTable_T.getByte(); - putu1(tag); - switch (tag) - { - case 7: // (7) [RCH] - putref(code_StackMapTable_RC.getRef()); - break; - case 8: // (8) [PH] - putu2(to_bci(code_StackMapTable_P.getInt())); - break; - } + int tag = code_StackMapTable_T.getByte(); + putu1(tag); + switch (tag) + { + case 7: // (7) [RCH] + putref(code_StackMapTable_RC.getRef()); + break; + case 8: // (8) [PH] + putu2(to_bci(code_StackMapTable_P.getInt())); + break; + } } // Functions for writing code. void unpacker::put_label(int curIP, int size) { - code_fixup_type.addByte(size); - code_fixup_offset.add((int)put_empty(size)); - code_fixup_source.add(curIP); + code_fixup_type.addByte(size); + code_fixup_offset.add((int)put_empty(size)); + code_fixup_source.add(curIP); } inline // called exactly once => inline - void + void unpacker::write_bc_ops() { - bcimap.empty(); - code_fixup_type.empty(); - code_fixup_offset.empty(); - code_fixup_source.empty(); - - band *bc_which; - - byte *opptr = bc_codes.curRP(); - // No need for oplimit, since the codes are pre-counted. - - size_t codeBase = wpoffset(); - - bool isAload; // copy-out result - int origBC; - - entry *thisClass = cur_class; - entry *superClass = cur_super; - entry *newClass = nullptr; // class of last _new opcode - - // overwrite any prior index on these bands; it changes w/ current class: - bc_thisfield.setIndex(cp.getFieldIndex(thisClass)); - bc_thismethod.setIndex(cp.getMethodIndex(thisClass)); - if (superClass != nullptr) - { - bc_superfield.setIndex(cp.getFieldIndex(superClass)); - bc_supermethod.setIndex(cp.getMethodIndex(superClass)); - } - - for (int curIP = 0;; curIP++) - { - int curPC = (int)(wpoffset() - codeBase); - bcimap.add(curPC); - ensure_put_space(10); // covers most instrs w/o further bounds check - int bc = *opptr++ & 0xFF; - - putu1_fast(bc); - // Note: See '--wp' below for pseudo-bytecodes like bc_end_marker. - - bool isWide = false; - if (bc == bc_wide) - { - bc = *opptr++ & 0xFF; - putu1_fast(bc); - isWide = true; - } - switch (bc) - { - case bc_end_marker: - --wp; // not really part of the code - assert(opptr <= bc_codes.maxRP()); - bc_codes.curRP() = opptr; // advance over this in bc_codes - goto doneScanningMethod; - case bc_tableswitch: // apc: (df, lo, hi, (hi-lo+1)*(label)) - case bc_lookupswitch: // apc: (df, nc, nc*(case, label)) - { - int caseCount = bc_case_count.getInt(); - while (((wpoffset() - codeBase) % 4) != 0) - putu1_fast(0); - ensure_put_space(30 + caseCount * 8); - put_label(curIP, 4); // int df = bc_label.getInt(); - if (bc == bc_tableswitch) - { - int lo = bc_case_value.getInt(); - int hi = lo + caseCount - 1; - putu4(lo); - putu4(hi); - for (int j = 0; j < caseCount; j++) - { - put_label(curIP, 4); // int lVal = bc_label.getInt(); - // int cVal = lo + j; - } - } - else - { - putu4(caseCount); - for (int j = 0; j < caseCount; j++) - { - int cVal = bc_case_value.getInt(); - putu4(cVal); - put_label(curIP, 4); // int lVal = bc_label.getInt(); - } - } - assert((int)to_bci(curIP) == curPC); - continue; - } - case bc_iinc: - { - int local = bc_local.getInt(); - int delta = (isWide ? bc_short : bc_byte).getInt(); - if (isWide) - { - putu2(local); - putu2(delta); - } - else - { - putu1_fast(local); - putu1_fast(delta); - } - continue; - } - case bc_sipush: - { - int val = bc_short.getInt(); - putu2(val); - continue; - } - case bc_bipush: - case bc_newarray: - { - int val = bc_byte.getByte(); - putu1_fast(val); - continue; - } - case bc_ref_escape: - { - // Note that insnMap has one entry for this. - --wp; // not really part of the code - int size = bc_escrefsize.getInt(); - entry *ref = bc_escref.getRefN(); - switch (size) - { - case 1: - putu1ref(ref); - break; - case 2: - putref(ref); - break; - default: - assert(false); - } - continue; - } - case bc_byte_escape: - { - // Note that insnMap has one entry for all these bytes. - --wp; // not really part of the code - int size = bc_escsize.getInt(); - ensure_put_space(size); - for (int j = 0; j < size; j++) - putu1_fast(bc_escbyte.getByte()); - continue; - } - default: - if (is_invoke_init_op(bc)) - { - origBC = bc_invokespecial; - entry *classRef; - switch (bc - _invokeinit_op) - { - case _invokeinit_self_option: - classRef = thisClass; - break; - case _invokeinit_super_option: - classRef = superClass; - break; - default: - assert(bc == _invokeinit_op + _invokeinit_new_option); - case _invokeinit_new_option: - classRef = newClass; - break; - } - wp[-1] = origBC; // overwrite with origBC - int coding = bc_initref.getInt(); - // Find the nth overloading of <init> in classRef. - entry *ref = nullptr; - cpindex *ix = (classRef == nullptr) ? nullptr : cp.getMethodIndex(classRef); - for (int j = 0, which_init = 0;; j++) - { - ref = (ix == nullptr) ? nullptr : ix->get(j); - if (ref == nullptr) - break; // oops, bad input - assert(ref->tag == CONSTANT_Methodref); - if (ref->memberDescr()->descrName() == cp.sym[constant_pool::s_lt_init_gt]) - { - if (which_init++ == coding) - break; - } - } - putref(ref); - continue; - } - bc_which = ref_band_for_self_op(bc, isAload, origBC); - if (bc_which != nullptr) - { - if (!isAload) - { - wp[-1] = origBC; // overwrite with origBC - } - else - { - wp[-1] = bc_aload_0; // overwrite with _aload_0 - // Note: insnMap keeps the _aload_0 separate. - bcimap.add(++curPC); - ++curIP; - putu1_fast(origBC); - } - entry *ref = bc_which->getRef(); - putref(ref); - continue; - } - if (is_branch_op(bc)) - { - // int lVal = bc_label.getInt(); - if (bc < bc_goto_w) - { - put_label(curIP, 2); // putu2(lVal & 0xFFFF); - } - else - { - assert(bc <= bc_jsr_w); - put_label(curIP, 4); // putu4(lVal); - } - assert((int)to_bci(curIP) == curPC); - continue; - } - bc_which = ref_band_for_op(bc); - if (bc_which != nullptr) - { - entry *ref = bc_which->getRefCommon(bc_which->ix, bc_which->nullOK); - if (ref == nullptr && bc_which == &bc_classref) - { - // Shorthand for class self-references. - ref = thisClass; - } - origBC = bc; - switch (bc) - { - case bc_ildc: - case bc_cldc: - case bc_fldc: - case bc_aldc: - origBC = bc_ldc; - break; - case bc_ildc_w: - case bc_cldc_w: - case bc_fldc_w: - case bc_aldc_w: - origBC = bc_ldc_w; - break; - case bc_lldc2_w: - case bc_dldc2_w: - origBC = bc_ldc2_w; - break; - case bc_new: - newClass = ref; - break; - } - wp[-1] = origBC; // overwrite with origBC - if (origBC == bc_ldc) - { - putu1ref(ref); - } - else - { - putref(ref); - } - if (origBC == bc_multianewarray) - { - // Copy the trailing byte also. - int val = bc_byte.getByte(); - putu1_fast(val); - } - else if (origBC == bc_invokeinterface) - { - int argSize = ref->memberDescr()->descrType()->typeSize(); - putu1_fast(1 + argSize); - putu1_fast(0); - } - continue; - } - if (is_local_slot_op(bc)) - { - int local = bc_local.getInt(); - if (isWide) - { - putu2(local); - if (bc == bc_iinc) - { - int iVal = bc_short.getInt(); - putu2(iVal); - } - } - else - { - putu1_fast(local); - if (bc == bc_iinc) - { - int iVal = bc_byte.getByte(); - putu1_fast(iVal); - } - } - continue; - } - // Random bytecode. Just copy it. - assert(bc < bc_bytecode_limit); - } - } + bcimap.empty(); + code_fixup_type.empty(); + code_fixup_offset.empty(); + code_fixup_source.empty(); + + band *bc_which; + + byte *opptr = bc_codes.curRP(); + // No need for oplimit, since the codes are pre-counted. + + size_t codeBase = wpoffset(); + + bool isAload; // copy-out result + int origBC; + + entry *thisClass = cur_class; + entry *superClass = cur_super; + entry *newClass = nullptr; // class of last _new opcode + + // overwrite any prior index on these bands; it changes w/ current class: + bc_thisfield.setIndex(cp.getFieldIndex(thisClass)); + bc_thismethod.setIndex(cp.getMethodIndex(thisClass)); + if (superClass != nullptr) + { + bc_superfield.setIndex(cp.getFieldIndex(superClass)); + bc_supermethod.setIndex(cp.getMethodIndex(superClass)); + } + + for (int curIP = 0;; curIP++) + { + int curPC = (int)(wpoffset() - codeBase); + bcimap.add(curPC); + ensure_put_space(10); // covers most instrs w/o further bounds check + int bc = *opptr++ & 0xFF; + + putu1_fast(bc); + // Note: See '--wp' below for pseudo-bytecodes like bc_end_marker. + + bool isWide = false; + if (bc == bc_wide) + { + bc = *opptr++ & 0xFF; + putu1_fast(bc); + isWide = true; + } + switch (bc) + { + case bc_end_marker: + --wp; // not really part of the code + assert(opptr <= bc_codes.maxRP()); + bc_codes.curRP() = opptr; // advance over this in bc_codes + goto doneScanningMethod; + case bc_tableswitch: // apc: (df, lo, hi, (hi-lo+1)*(label)) + case bc_lookupswitch: // apc: (df, nc, nc*(case, label)) + { + int caseCount = bc_case_count.getInt(); + while (((wpoffset() - codeBase) % 4) != 0) + putu1_fast(0); + ensure_put_space(30 + caseCount * 8); + put_label(curIP, 4); // int df = bc_label.getInt(); + if (bc == bc_tableswitch) + { + int lo = bc_case_value.getInt(); + int hi = lo + caseCount - 1; + putu4(lo); + putu4(hi); + for (int j = 0; j < caseCount; j++) + { + put_label(curIP, 4); // int lVal = bc_label.getInt(); + // int cVal = lo + j; + } + } + else + { + putu4(caseCount); + for (int j = 0; j < caseCount; j++) + { + int cVal = bc_case_value.getInt(); + putu4(cVal); + put_label(curIP, 4); // int lVal = bc_label.getInt(); + } + } + assert((int)to_bci(curIP) == curPC); + continue; + } + case bc_iinc: + { + int local = bc_local.getInt(); + int delta = (isWide ? bc_short : bc_byte).getInt(); + if (isWide) + { + putu2(local); + putu2(delta); + } + else + { + putu1_fast(local); + putu1_fast(delta); + } + continue; + } + case bc_sipush: + { + int val = bc_short.getInt(); + putu2(val); + continue; + } + case bc_bipush: + case bc_newarray: + { + int val = bc_byte.getByte(); + putu1_fast(val); + continue; + } + case bc_ref_escape: + { + // Note that insnMap has one entry for this. + --wp; // not really part of the code + int size = bc_escrefsize.getInt(); + entry *ref = bc_escref.getRefN(); + switch (size) + { + case 1: + putu1ref(ref); + break; + case 2: + putref(ref); + break; + default: + assert(false); + } + continue; + } + case bc_byte_escape: + { + // Note that insnMap has one entry for all these bytes. + --wp; // not really part of the code + int size = bc_escsize.getInt(); + ensure_put_space(size); + for (int j = 0; j < size; j++) + putu1_fast(bc_escbyte.getByte()); + continue; + } + default: + if (is_invoke_init_op(bc)) + { + origBC = bc_invokespecial; + entry *classRef; + switch (bc - _invokeinit_op) + { + case _invokeinit_self_option: + classRef = thisClass; + break; + case _invokeinit_super_option: + classRef = superClass; + break; + default: + assert(bc == _invokeinit_op + _invokeinit_new_option); + case _invokeinit_new_option: + classRef = newClass; + break; + } + wp[-1] = origBC; // overwrite with origBC + int coding = bc_initref.getInt(); + // Find the nth overloading of <init> in classRef. + entry *ref = nullptr; + cpindex *ix = (classRef == nullptr) ? nullptr : cp.getMethodIndex(classRef); + for (int j = 0, which_init = 0;; j++) + { + ref = (ix == nullptr) ? nullptr : ix->get(j); + if (ref == nullptr) + break; // oops, bad input + assert(ref->tag == CONSTANT_Methodref); + if (ref->memberDescr()->descrName() == cp.sym[constant_pool::s_lt_init_gt]) + { + if (which_init++ == coding) + break; + } + } + putref(ref); + continue; + } + bc_which = ref_band_for_self_op(bc, isAload, origBC); + if (bc_which != nullptr) + { + if (!isAload) + { + wp[-1] = origBC; // overwrite with origBC + } + else + { + wp[-1] = bc_aload_0; // overwrite with _aload_0 + // Note: insnMap keeps the _aload_0 separate. + bcimap.add(++curPC); + ++curIP; + putu1_fast(origBC); + } + entry *ref = bc_which->getRef(); + putref(ref); + continue; + } + if (is_branch_op(bc)) + { + // int lVal = bc_label.getInt(); + if (bc < bc_goto_w) + { + put_label(curIP, 2); // putu2(lVal & 0xFFFF); + } + else + { + assert(bc <= bc_jsr_w); + put_label(curIP, 4); // putu4(lVal); + } + assert((int)to_bci(curIP) == curPC); + continue; + } + bc_which = ref_band_for_op(bc); + if (bc_which != nullptr) + { + entry *ref = bc_which->getRefCommon(bc_which->ix, bc_which->nullOK); + if (ref == nullptr && bc_which == &bc_classref) + { + // Shorthand for class self-references. + ref = thisClass; + } + origBC = bc; + switch (bc) + { + case bc_ildc: + case bc_cldc: + case bc_fldc: + case bc_aldc: + origBC = bc_ldc; + break; + case bc_ildc_w: + case bc_cldc_w: + case bc_fldc_w: + case bc_aldc_w: + origBC = bc_ldc_w; + break; + case bc_lldc2_w: + case bc_dldc2_w: + origBC = bc_ldc2_w; + break; + case bc_new: + newClass = ref; + break; + } + wp[-1] = origBC; // overwrite with origBC + if (origBC == bc_ldc) + { + putu1ref(ref); + } + else + { + putref(ref); + } + if (origBC == bc_multianewarray) + { + // Copy the trailing byte also. + int val = bc_byte.getByte(); + putu1_fast(val); + } + else if (origBC == bc_invokeinterface) + { + int argSize = ref->memberDescr()->descrType()->typeSize(); + putu1_fast(1 + argSize); + putu1_fast(0); + } + continue; + } + if (is_local_slot_op(bc)) + { + int local = bc_local.getInt(); + if (isWide) + { + putu2(local); + if (bc == bc_iinc) + { + int iVal = bc_short.getInt(); + putu2(iVal); + } + } + else + { + putu1_fast(local); + if (bc == bc_iinc) + { + int iVal = bc_byte.getByte(); + putu1_fast(iVal); + } + } + continue; + } + // Random bytecode. Just copy it. + assert(bc < bc_bytecode_limit); + } + } doneScanningMethod: { } - // bcimap.add(curPC); // PC limit is already also in map, from bc_end_marker - - // Armed with a bcimap, we can now fix up all the labels. - for (int i = 0; i < (int)code_fixup_type.size(); i++) - { - int type = code_fixup_type.getByte(i); - byte *bp = wp_at(code_fixup_offset.get(i)); - int curIP = code_fixup_source.get(i); - int destIP = curIP + bc_label.getInt(); - int span = to_bci(destIP) - to_bci(curIP); - switch (type) - { - case 2: - putu2_at(bp, (ushort)span); - break; - case 4: - putu4_at(bp, span); - break; - default: - assert(false); - } - } + // bcimap.add(curPC); // PC limit is already also in map, from bc_end_marker + + // Armed with a bcimap, we can now fix up all the labels. + for (int i = 0; i < (int)code_fixup_type.size(); i++) + { + int type = code_fixup_type.getByte(i); + byte *bp = wp_at(code_fixup_offset.get(i)); + int curIP = code_fixup_source.get(i); + int destIP = curIP + bc_label.getInt(); + int span = to_bci(destIP) - to_bci(curIP); + switch (type) + { + case 2: + putu2_at(bp, (ushort)span); + break; + case 4: + putu4_at(bp, span); + break; + default: + assert(false); + } + } } inline // called exactly once => inline - void + void unpacker::write_code() { - int j; - - int max_stack, max_locals, handler_count, cflags; - get_code_header(max_stack, max_locals, handler_count, cflags); - - if (max_stack < 0) - max_stack = code_max_stack.getInt(); - if (max_locals < 0) - max_locals = code_max_na_locals.getInt(); - if (handler_count < 0) - handler_count = code_handler_count.getInt(); - - int siglen = cur_descr->descrType()->typeSize(); - if ((cur_descr_flags & ACC_STATIC) == 0) - siglen++; - max_locals += siglen; - - putu2(max_stack); - putu2(max_locals); - size_t bcbase = put_empty(4); - - // Write the bytecodes themselves. - write_bc_ops(); - - byte *bcbasewp = wp_at(bcbase); - putu4_at(bcbasewp, (int)(wp - (bcbasewp + 4))); // size of code attr - - putu2(handler_count); - for (j = 0; j < handler_count; j++) - { - int bii = code_handler_start_P.getInt(); - putu2(to_bci(bii)); - bii += code_handler_end_PO.getInt(); - putu2(to_bci(bii)); - bii += code_handler_catch_PO.getInt(); - putu2(to_bci(bii)); - putref(code_handler_class_RCN.getRefN()); - } - - uint64_t indexBits = cflags; - if (cflags < 0) - { - bool haveLongFlags = attr_defs[ATTR_CONTEXT_CODE].haveLongFlags(); - indexBits = code_flags_hi.getLong(code_flags_lo, haveLongFlags); - } - write_attrs(ATTR_CONTEXT_CODE, indexBits); + int j; + + int max_stack, max_locals, handler_count, cflags; + get_code_header(max_stack, max_locals, handler_count, cflags); + + if (max_stack < 0) + max_stack = code_max_stack.getInt(); + if (max_locals < 0) + max_locals = code_max_na_locals.getInt(); + if (handler_count < 0) + handler_count = code_handler_count.getInt(); + + int siglen = cur_descr->descrType()->typeSize(); + if ((cur_descr_flags & ACC_STATIC) == 0) + siglen++; + max_locals += siglen; + + putu2(max_stack); + putu2(max_locals); + size_t bcbase = put_empty(4); + + // Write the bytecodes themselves. + write_bc_ops(); + + byte *bcbasewp = wp_at(bcbase); + putu4_at(bcbasewp, (int)(wp - (bcbasewp + 4))); // size of code attr + + putu2(handler_count); + for (j = 0; j < handler_count; j++) + { + int bii = code_handler_start_P.getInt(); + putu2(to_bci(bii)); + bii += code_handler_end_PO.getInt(); + putu2(to_bci(bii)); + bii += code_handler_catch_PO.getInt(); + putu2(to_bci(bii)); + putref(code_handler_class_RCN.getRefN()); + } + + uint64_t indexBits = cflags; + if (cflags < 0) + { + bool haveLongFlags = attr_defs[ATTR_CONTEXT_CODE].haveLongFlags(); + indexBits = code_flags_hi.getLong(code_flags_lo, haveLongFlags); + } + write_attrs(ATTR_CONTEXT_CODE, indexBits); } int unpacker::write_attrs(int attrc, uint64_t indexBits) { - if (indexBits == 0) - { - // Quick short-circuit. - putu2(0); - return 0; - } - - attr_definitions &ad = attr_defs[attrc]; - - int i, j, j2, idx, count; - - int oiCount = 0; - if (ad.isPredefined(X_ATTR_OVERFLOW) && (indexBits & ((uint64_t)1 << X_ATTR_OVERFLOW)) != 0) - { - indexBits -= ((uint64_t)1 << X_ATTR_OVERFLOW); - oiCount = ad.xxx_attr_count().getInt(); - } - - int bitIndexes[X_ATTR_LIMIT_FLAGS_HI]; - int biCount = 0; - - // Fill bitIndexes with index bits, in order. - for (idx = 0; indexBits != 0; idx++, indexBits >>= 1) - { - if ((indexBits & 1) != 0) - bitIndexes[biCount++] = idx; - } - assert(biCount <= (int)lengthof(bitIndexes)); - - // Write a provisional attribute count, perhaps to be corrected later. - int naOffset = (int)wpoffset(); - int na0 = biCount + oiCount; - putu2(na0); - - int na = 0; - for (i = 0; i < na0; i++) - { - if (i < biCount) - idx = bitIndexes[i]; - else - idx = ad.xxx_attr_indexes().getInt(); - assert(ad.isIndex(idx)); - entry *aname = nullptr; - entry *ref; // scratch - size_t abase = put_empty(2 + 4); - if (idx < (int)ad.flag_limit && ad.isPredefined(idx)) - { - // Switch on the attrc and idx simultaneously. - switch (ADH_BYTE(attrc, idx)) - { - - case ADH_BYTE(ATTR_CONTEXT_CLASS, X_ATTR_OVERFLOW) : - case ADH_BYTE(ATTR_CONTEXT_FIELD, X_ATTR_OVERFLOW) : - case ADH_BYTE(ATTR_CONTEXT_METHOD, X_ATTR_OVERFLOW) : - case ADH_BYTE(ATTR_CONTEXT_CODE, X_ATTR_OVERFLOW) : - // no attribute at all, so back up on this one - wp = wp_at(abase); - continue; - - case ADH_BYTE(ATTR_CONTEXT_CLASS, CLASS_ATTR_ClassFile_version) : - cur_class_minver = class_ClassFile_version_minor_H.getInt(); - cur_class_majver = class_ClassFile_version_major_H.getInt(); - // back up; not a real attribute - wp = wp_at(abase); - continue; - - case ADH_BYTE(ATTR_CONTEXT_CLASS, CLASS_ATTR_InnerClasses) : - // note the existence of this attr, but save for later - if (cur_class_has_local_ics) - unpack_abort("too many InnerClasses attrs"); - cur_class_has_local_ics = true; - wp = wp_at(abase); - continue; - - case ADH_BYTE(ATTR_CONTEXT_CLASS, CLASS_ATTR_SourceFile) : - aname = cp.sym[constant_pool::s_SourceFile]; - ref = class_SourceFile_RUN.getRefN(); - if (ref == nullptr) - { - bytes &n = cur_class->ref(0)->value.b; - // parse n = (<pkg>/)*<outer>?($<id>)* - int pkglen = lastIndexOf(SLASH_MIN, SLASH_MAX, n, (int)n.len) + 1; - bytes prefix = n.slice(pkglen, n.len); - for (;;) - { - // Work backwards, finding all '$', '#', etc. - int dollar = - lastIndexOf(DOLLAR_MIN, DOLLAR_MAX, prefix, (int)prefix.len); - if (dollar < 0) - break; - prefix = prefix.slice(0, dollar); - } - const char *suffix = ".java"; - int len = (int)(prefix.len + strlen(suffix)); - bytes name; - name.set(T_NEW(byte, add_size(len, 1)), len); - name.strcat(prefix).strcat(suffix); - ref = cp.ensureUtf8(name); - } - putref(ref); - break; - - case ADH_BYTE(ATTR_CONTEXT_CLASS, CLASS_ATTR_EnclosingMethod) : - aname = cp.sym[constant_pool::s_EnclosingMethod]; - putref(class_EnclosingMethod_RC.getRefN()); - putref(class_EnclosingMethod_RDN.getRefN()); - break; - - case ADH_BYTE(ATTR_CONTEXT_FIELD, FIELD_ATTR_ConstantValue) : - aname = cp.sym[constant_pool::s_ConstantValue]; - putref(field_ConstantValue_KQ.getRefUsing(cp.getKQIndex())); - break; - - case ADH_BYTE(ATTR_CONTEXT_METHOD, METHOD_ATTR_Code) : - aname = cp.sym[constant_pool::s_Code]; - write_code(); - break; - - case ADH_BYTE(ATTR_CONTEXT_METHOD, METHOD_ATTR_Exceptions) : - aname = cp.sym[constant_pool::s_Exceptions]; - putu2(count = method_Exceptions_N.getInt()); - for (j = 0; j < count; j++) - { - putref(method_Exceptions_RC.getRefN()); - } - break; - - case ADH_BYTE(ATTR_CONTEXT_CODE, CODE_ATTR_StackMapTable) : - aname = cp.sym[constant_pool::s_StackMapTable]; - // (keep this code aligned with its brother in unpacker::read_attrs) - putu2(count = code_StackMapTable_N.getInt()); - for (j = 0; j < count; j++) - { - int tag = code_StackMapTable_frame_T.getByte(); - putu1(tag); - if (tag <= 127) - { - // (64-127) [(2)] - if (tag >= 64) - put_stackmap_type(); - } - else if (tag <= 251) - { - // (247) [(1)(2)] - // (248-251) [(1)] - if (tag >= 247) - putu2(code_StackMapTable_offset.getInt()); - if (tag == 247) - put_stackmap_type(); - } - else if (tag <= 254) - { - // (252) [(1)(2)] - // (253) [(1)(2)(2)] - // (254) [(1)(2)(2)(2)] - putu2(code_StackMapTable_offset.getInt()); - for (int k = (tag - 251); k > 0; k--) - { - put_stackmap_type(); - } - } - else - { - // (255) [(1)NH[(2)]NH[(2)]] - putu2(code_StackMapTable_offset.getInt()); - putu2(j2 = code_StackMapTable_local_N.getInt()); - while (j2-- > 0) - put_stackmap_type(); - putu2(j2 = code_StackMapTable_stack_N.getInt()); - while (j2-- > 0) - put_stackmap_type(); - } - } - break; - - case ADH_BYTE(ATTR_CONTEXT_CODE, CODE_ATTR_LineNumberTable) : - aname = cp.sym[constant_pool::s_LineNumberTable]; - putu2(count = code_LineNumberTable_N.getInt()); - for (j = 0; j < count; j++) - { - putu2(to_bci(code_LineNumberTable_bci_P.getInt())); - putu2(code_LineNumberTable_line.getInt()); - } - break; - - case ADH_BYTE(ATTR_CONTEXT_CODE, CODE_ATTR_LocalVariableTable) : - aname = cp.sym[constant_pool::s_LocalVariableTable]; - putu2(count = code_LocalVariableTable_N.getInt()); - for (j = 0; j < count; j++) - { - int bii = code_LocalVariableTable_bci_P.getInt(); - int bci = to_bci(bii); - putu2(bci); - bii += code_LocalVariableTable_span_O.getInt(); - putu2(to_bci(bii) - bci); - putref(code_LocalVariableTable_name_RU.getRefN()); - putref(code_LocalVariableTable_type_RS.getRefN()); - putu2(code_LocalVariableTable_slot.getInt()); - } - break; - - case ADH_BYTE(ATTR_CONTEXT_CODE, CODE_ATTR_LocalVariableTypeTable) : - aname = cp.sym[constant_pool::s_LocalVariableTypeTable]; - putu2(count = code_LocalVariableTypeTable_N.getInt()); - for (j = 0; j < count; j++) - { - int bii = code_LocalVariableTypeTable_bci_P.getInt(); - int bci = to_bci(bii); - putu2(bci); - bii += code_LocalVariableTypeTable_span_O.getInt(); - putu2(to_bci(bii) - bci); - putref(code_LocalVariableTypeTable_name_RU.getRefN()); - putref(code_LocalVariableTypeTable_type_RS.getRefN()); - putu2(code_LocalVariableTypeTable_slot.getInt()); - } - break; - - case ADH_BYTE(ATTR_CONTEXT_CLASS, X_ATTR_Signature) : - aname = cp.sym[constant_pool::s_Signature]; - putref(class_Signature_RS.getRefN()); - break; - - case ADH_BYTE(ATTR_CONTEXT_FIELD, X_ATTR_Signature) : - aname = cp.sym[constant_pool::s_Signature]; - putref(field_Signature_RS.getRefN()); - break; - - case ADH_BYTE(ATTR_CONTEXT_METHOD, X_ATTR_Signature) : - aname = cp.sym[constant_pool::s_Signature]; - putref(method_Signature_RS.getRefN()); - break; - - case ADH_BYTE(ATTR_CONTEXT_CLASS, X_ATTR_Deprecated) : - case ADH_BYTE(ATTR_CONTEXT_FIELD, X_ATTR_Deprecated) : - case ADH_BYTE(ATTR_CONTEXT_METHOD, X_ATTR_Deprecated) : - aname = cp.sym[constant_pool::s_Deprecated]; - // no data - break; - } - } - - if (aname == nullptr) - { - // Unparse a compressor-defined attribute. - layout_definition *lo = ad.getLayout(idx); - if (lo == nullptr) - { - unpack_abort("bad layout index"); - break; - } - assert((int)lo->idx == idx); - aname = lo->nameEntry; - if (aname == nullptr) - { - bytes nameb; - nameb.set(lo->name); - aname = cp.ensureUtf8(nameb); - // Cache the name entry for next time. - lo->nameEntry = aname; - } - // Execute all the layout elements. - band **bands = lo->bands(); - if (lo->hasCallables()) - { - band &cble = *bands[0]; - assert(cble.le_kind == EK_CBLE); - bands = cble.le_body; - } - putlayout(bands); - } - - if (aname == nullptr) - unpack_abort("bad attribute index"); - - byte *wp1 = wp; - wp = wp_at(abase); - - // DTRT if this attr is on the strip-list. - // (Note that we emptied the data out of the band first.) - if (ad.strip_names.contains(aname)) - { - continue; - } - - // patch the name and length - putref(aname); - putu4((int)(wp1 - (wp + 4))); // put the attr size - wp = wp1; - na++; // count the attrs actually written - } - - if (na != na0) - // Refresh changed count. - putu2_at(wp_at(naOffset), na); - return na; + if (indexBits == 0) + { + // Quick short-circuit. + putu2(0); + return 0; + } + + attr_definitions &ad = attr_defs[attrc]; + + int i, j, j2, idx, count; + + int oiCount = 0; + if (ad.isPredefined(X_ATTR_OVERFLOW) && (indexBits & ((uint64_t)1 << X_ATTR_OVERFLOW)) != 0) + { + indexBits -= ((uint64_t)1 << X_ATTR_OVERFLOW); + oiCount = ad.xxx_attr_count().getInt(); + } + + int bitIndexes[X_ATTR_LIMIT_FLAGS_HI]; + int biCount = 0; + + // Fill bitIndexes with index bits, in order. + for (idx = 0; indexBits != 0; idx++, indexBits >>= 1) + { + if ((indexBits & 1) != 0) + bitIndexes[biCount++] = idx; + } + assert(biCount <= (int)lengthof(bitIndexes)); + + // Write a provisional attribute count, perhaps to be corrected later. + int naOffset = (int)wpoffset(); + int na0 = biCount + oiCount; + putu2(na0); + + int na = 0; + for (i = 0; i < na0; i++) + { + if (i < biCount) + idx = bitIndexes[i]; + else + idx = ad.xxx_attr_indexes().getInt(); + assert(ad.isIndex(idx)); + entry *aname = nullptr; + entry *ref; // scratch + size_t abase = put_empty(2 + 4); + if (idx < (int)ad.flag_limit && ad.isPredefined(idx)) + { + // Switch on the attrc and idx simultaneously. + switch (ADH_BYTE(attrc, idx)) + { + + case ADH_BYTE(ATTR_CONTEXT_CLASS, X_ATTR_OVERFLOW) : + case ADH_BYTE(ATTR_CONTEXT_FIELD, X_ATTR_OVERFLOW) : + case ADH_BYTE(ATTR_CONTEXT_METHOD, X_ATTR_OVERFLOW) : + case ADH_BYTE(ATTR_CONTEXT_CODE, X_ATTR_OVERFLOW) : + // no attribute at all, so back up on this one + wp = wp_at(abase); + continue; + + case ADH_BYTE(ATTR_CONTEXT_CLASS, CLASS_ATTR_ClassFile_version) : + cur_class_minver = class_ClassFile_version_minor_H.getInt(); + cur_class_majver = class_ClassFile_version_major_H.getInt(); + // back up; not a real attribute + wp = wp_at(abase); + continue; + + case ADH_BYTE(ATTR_CONTEXT_CLASS, CLASS_ATTR_InnerClasses) : + // note the existence of this attr, but save for later + if (cur_class_has_local_ics) + unpack_abort("too many InnerClasses attrs"); + cur_class_has_local_ics = true; + wp = wp_at(abase); + continue; + + case ADH_BYTE(ATTR_CONTEXT_CLASS, CLASS_ATTR_SourceFile) : + aname = cp.sym[constant_pool::s_SourceFile]; + ref = class_SourceFile_RUN.getRefN(); + if (ref == nullptr) + { + bytes &n = cur_class->ref(0)->value.b; + // parse n = (<pkg>/)*<outer>?($<id>)* + int pkglen = lastIndexOf(SLASH_MIN, SLASH_MAX, n, (int)n.len) + 1; + bytes prefix = n.slice(pkglen, n.len); + for (;;) + { + // Work backwards, finding all '$', '#', etc. + int dollar = + lastIndexOf(DOLLAR_MIN, DOLLAR_MAX, prefix, (int)prefix.len); + if (dollar < 0) + break; + prefix = prefix.slice(0, dollar); + } + const char *suffix = ".java"; + int len = (int)(prefix.len + strlen(suffix)); + bytes name; + name.set(T_NEW(byte, add_size(len, 1)), len); + name.strcat(prefix).strcat(suffix); + ref = cp.ensureUtf8(name); + } + putref(ref); + break; + + case ADH_BYTE(ATTR_CONTEXT_CLASS, CLASS_ATTR_EnclosingMethod) : + aname = cp.sym[constant_pool::s_EnclosingMethod]; + putref(class_EnclosingMethod_RC.getRefN()); + putref(class_EnclosingMethod_RDN.getRefN()); + break; + + case ADH_BYTE(ATTR_CONTEXT_FIELD, FIELD_ATTR_ConstantValue) : + aname = cp.sym[constant_pool::s_ConstantValue]; + putref(field_ConstantValue_KQ.getRefUsing(cp.getKQIndex())); + break; + + case ADH_BYTE(ATTR_CONTEXT_METHOD, METHOD_ATTR_Code) : + aname = cp.sym[constant_pool::s_Code]; + write_code(); + break; + + case ADH_BYTE(ATTR_CONTEXT_METHOD, METHOD_ATTR_Exceptions) : + aname = cp.sym[constant_pool::s_Exceptions]; + putu2(count = method_Exceptions_N.getInt()); + for (j = 0; j < count; j++) + { + putref(method_Exceptions_RC.getRefN()); + } + break; + + case ADH_BYTE(ATTR_CONTEXT_CODE, CODE_ATTR_StackMapTable) : + aname = cp.sym[constant_pool::s_StackMapTable]; + // (keep this code aligned with its brother in unpacker::read_attrs) + putu2(count = code_StackMapTable_N.getInt()); + for (j = 0; j < count; j++) + { + int tag = code_StackMapTable_frame_T.getByte(); + putu1(tag); + if (tag <= 127) + { + // (64-127) [(2)] + if (tag >= 64) + put_stackmap_type(); + } + else if (tag <= 251) + { + // (247) [(1)(2)] + // (248-251) [(1)] + if (tag >= 247) + putu2(code_StackMapTable_offset.getInt()); + if (tag == 247) + put_stackmap_type(); + } + else if (tag <= 254) + { + // (252) [(1)(2)] + // (253) [(1)(2)(2)] + // (254) [(1)(2)(2)(2)] + putu2(code_StackMapTable_offset.getInt()); + for (int k = (tag - 251); k > 0; k--) + { + put_stackmap_type(); + } + } + else + { + // (255) [(1)NH[(2)]NH[(2)]] + putu2(code_StackMapTable_offset.getInt()); + putu2(j2 = code_StackMapTable_local_N.getInt()); + while (j2-- > 0) + put_stackmap_type(); + putu2(j2 = code_StackMapTable_stack_N.getInt()); + while (j2-- > 0) + put_stackmap_type(); + } + } + break; + + case ADH_BYTE(ATTR_CONTEXT_CODE, CODE_ATTR_LineNumberTable) : + aname = cp.sym[constant_pool::s_LineNumberTable]; + putu2(count = code_LineNumberTable_N.getInt()); + for (j = 0; j < count; j++) + { + putu2(to_bci(code_LineNumberTable_bci_P.getInt())); + putu2(code_LineNumberTable_line.getInt()); + } + break; + + case ADH_BYTE(ATTR_CONTEXT_CODE, CODE_ATTR_LocalVariableTable) : + aname = cp.sym[constant_pool::s_LocalVariableTable]; + putu2(count = code_LocalVariableTable_N.getInt()); + for (j = 0; j < count; j++) + { + int bii = code_LocalVariableTable_bci_P.getInt(); + int bci = to_bci(bii); + putu2(bci); + bii += code_LocalVariableTable_span_O.getInt(); + putu2(to_bci(bii) - bci); + putref(code_LocalVariableTable_name_RU.getRefN()); + putref(code_LocalVariableTable_type_RS.getRefN()); + putu2(code_LocalVariableTable_slot.getInt()); + } + break; + + case ADH_BYTE(ATTR_CONTEXT_CODE, CODE_ATTR_LocalVariableTypeTable) : + aname = cp.sym[constant_pool::s_LocalVariableTypeTable]; + putu2(count = code_LocalVariableTypeTable_N.getInt()); + for (j = 0; j < count; j++) + { + int bii = code_LocalVariableTypeTable_bci_P.getInt(); + int bci = to_bci(bii); + putu2(bci); + bii += code_LocalVariableTypeTable_span_O.getInt(); + putu2(to_bci(bii) - bci); + putref(code_LocalVariableTypeTable_name_RU.getRefN()); + putref(code_LocalVariableTypeTable_type_RS.getRefN()); + putu2(code_LocalVariableTypeTable_slot.getInt()); + } + break; + + case ADH_BYTE(ATTR_CONTEXT_CLASS, X_ATTR_Signature) : + aname = cp.sym[constant_pool::s_Signature]; + putref(class_Signature_RS.getRefN()); + break; + + case ADH_BYTE(ATTR_CONTEXT_FIELD, X_ATTR_Signature) : + aname = cp.sym[constant_pool::s_Signature]; + putref(field_Signature_RS.getRefN()); + break; + + case ADH_BYTE(ATTR_CONTEXT_METHOD, X_ATTR_Signature) : + aname = cp.sym[constant_pool::s_Signature]; + putref(method_Signature_RS.getRefN()); + break; + + case ADH_BYTE(ATTR_CONTEXT_CLASS, X_ATTR_Deprecated) : + case ADH_BYTE(ATTR_CONTEXT_FIELD, X_ATTR_Deprecated) : + case ADH_BYTE(ATTR_CONTEXT_METHOD, X_ATTR_Deprecated) : + aname = cp.sym[constant_pool::s_Deprecated]; + // no data + break; + } + } + + if (aname == nullptr) + { + // Unparse a compressor-defined attribute. + layout_definition *lo = ad.getLayout(idx); + if (lo == nullptr) + { + unpack_abort("bad layout index"); + break; + } + assert((int)lo->idx == idx); + aname = lo->nameEntry; + if (aname == nullptr) + { + bytes nameb; + nameb.set(lo->name); + aname = cp.ensureUtf8(nameb); + // Cache the name entry for next time. + lo->nameEntry = aname; + } + // Execute all the layout elements. + band **bands = lo->bands(); + if (lo->hasCallables()) + { + band &cble = *bands[0]; + assert(cble.le_kind == EK_CBLE); + bands = cble.le_body; + } + putlayout(bands); + } + + if (aname == nullptr) + unpack_abort("bad attribute index"); + + byte *wp1 = wp; + wp = wp_at(abase); + + // DTRT if this attr is on the strip-list. + // (Note that we emptied the data out of the band first.) + if (ad.strip_names.contains(aname)) + { + continue; + } + + // patch the name and length + putref(aname); + putu4((int)(wp1 - (wp + 4))); // put the attr size + wp = wp1; + na++; // count the attrs actually written + } + + if (na != na0) + // Refresh changed count. + putu2_at(wp_at(naOffset), na); + return na; } void unpacker::write_members(int num, int attrc) { - attr_definitions &ad = attr_defs[attrc]; - band &member_flags_hi = ad.xxx_flags_hi(); - band &member_flags_lo = ad.xxx_flags_lo(); - band &member_descr = (&member_flags_hi)[e_field_descr - e_field_flags_hi]; - bool haveLongFlags = ad.haveLongFlags(); - - putu2(num); - uint64_t indexMask = attr_defs[attrc].flagIndexMask(); - for (int i = 0; i < num; i++) - { - uint64_t mflags = member_flags_hi.getLong(member_flags_lo, haveLongFlags); - entry *mdescr = member_descr.getRef(); - cur_descr = mdescr; - putu2(cur_descr_flags = (ushort)(mflags & ~indexMask)); - putref(mdescr->descrName()); - putref(mdescr->descrType()); - write_attrs(attrc, (mflags & indexMask)); - } - cur_descr = nullptr; + attr_definitions &ad = attr_defs[attrc]; + band &member_flags_hi = ad.xxx_flags_hi(); + band &member_flags_lo = ad.xxx_flags_lo(); + band &member_descr = (&member_flags_hi)[e_field_descr - e_field_flags_hi]; + bool haveLongFlags = ad.haveLongFlags(); + + putu2(num); + uint64_t indexMask = attr_defs[attrc].flagIndexMask(); + for (int i = 0; i < num; i++) + { + uint64_t mflags = member_flags_hi.getLong(member_flags_lo, haveLongFlags); + entry *mdescr = member_descr.getRef(); + cur_descr = mdescr; + putu2(cur_descr_flags = (ushort)(mflags & ~indexMask)); + putref(mdescr->descrName()); + putref(mdescr->descrType()); + write_attrs(attrc, (mflags & indexMask)); + } + cur_descr = nullptr; } extern "C" int raw_address_cmp(const void *p1p, const void *p2p) { - void *p1 = *(void **)p1p; - void *p2 = *(void **)p2p; - return (p1 > p2) ? 1 : (p1 < p2) ? -1 : 0; + void *p1 = *(void **)p1p; + void *p2 = *(void **)p2p; + return (p1 > p2) ? 1 : (p1 < p2) ? -1 : 0; } void unpacker::write_classfile_tail() { - cur_classfile_tail.empty(); - set_output(&cur_classfile_tail); + cur_classfile_tail.empty(); + set_output(&cur_classfile_tail); - int i, num; + int i, num; - attr_definitions &ad = attr_defs[ATTR_CONTEXT_CLASS]; + attr_definitions &ad = attr_defs[ATTR_CONTEXT_CLASS]; - bool haveLongFlags = ad.haveLongFlags(); - uint64_t kflags = class_flags_hi.getLong(class_flags_lo, haveLongFlags); - uint64_t indexMask = ad.flagIndexMask(); + bool haveLongFlags = ad.haveLongFlags(); + uint64_t kflags = class_flags_hi.getLong(class_flags_lo, haveLongFlags); + uint64_t indexMask = ad.flagIndexMask(); - cur_class = class_this.getRef(); - cur_super = class_super.getRef(); + cur_class = class_this.getRef(); + cur_super = class_super.getRef(); - if (cur_super == cur_class) - cur_super = nullptr; - // special representation for java/lang/Object + if (cur_super == cur_class) + cur_super = nullptr; + // special representation for java/lang/Object - putu2((ushort)(kflags & ~indexMask)); - putref(cur_class); - putref(cur_super); + putu2((ushort)(kflags & ~indexMask)); + putref(cur_class); + putref(cur_super); - putu2(num = class_interface_count.getInt()); - for (i = 0; i < num; i++) - { - putref(class_interface.getRef()); - } + putu2(num = class_interface_count.getInt()); + for (i = 0; i < num; i++) + { + putref(class_interface.getRef()); + } - write_members(class_field_count.getInt(), ATTR_CONTEXT_FIELD); - write_members(class_method_count.getInt(), ATTR_CONTEXT_METHOD); + write_members(class_field_count.getInt(), ATTR_CONTEXT_FIELD); + write_members(class_method_count.getInt(), ATTR_CONTEXT_METHOD); - cur_class_has_local_ics = false; // may be set true by write_attrs + cur_class_has_local_ics = false; // may be set true by write_attrs - int naOffset = (int)wpoffset(); - int na = write_attrs(ATTR_CONTEXT_CLASS, (kflags & indexMask)); + int naOffset = (int)wpoffset(); + int na = write_attrs(ATTR_CONTEXT_CLASS, (kflags & indexMask)); // at the very last, choose which inner classes (if any) pertain to k: #ifdef ASSERT - for (i = 0; i < ic_count; i++) - { - assert(!ics[i].requested); - } + for (i = 0; i < ic_count; i++) + { + assert(!ics[i].requested); + } #endif - // First, consult the global table and the local constant pool, - // and decide on the globally implied inner classes. - // (Note that we read the cpool's outputIndex fields, but we - // do not yet write them, since the local IC attribute might - // reverse a global decision to declare an IC.) - assert(requested_ics.length() == 0); // must start out empty - // Always include all members of the current class. - for (inner_class *child = cp.getFirstChildIC(cur_class); child != nullptr; - child = cp.getNextChildIC(child)) - { - child->requested = true; - requested_ics.add(child); - } - // And, for each inner class mentioned in the constant pool, - // include it and all its outers. - int noes = cp.outputEntries.length(); - entry **oes = (entry **)cp.outputEntries.base(); - for (i = 0; i < noes; i++) - { - entry &e = *oes[i]; - if (e.tag != CONSTANT_Class) - continue; // wrong sort - for (inner_class *ic = cp.getIC(&e); ic != nullptr; ic = cp.getIC(ic->outer)) - { - if (ic->requested) - break; // already processed - ic->requested = true; - requested_ics.add(ic); - } - } - int local_ics = requested_ics.length(); - // Second, consult a local attribute (if any) and adjust the global set. - inner_class *extra_ics = nullptr; - int num_extra_ics = 0; - if (cur_class_has_local_ics) - { - // adjust the set of ICs by symmetric set difference w/ the locals - num_extra_ics = class_InnerClasses_N.getInt(); - if (num_extra_ics == 0) - { - // Explicit zero count has an irregular meaning: It deletes the attr. - local_ics = 0; // (short-circuit all tests of requested bits) - } - else - { - extra_ics = T_NEW(inner_class, num_extra_ics); - // Note: extra_ics will be freed up by next call to get_next_file(). - } - } - for (i = 0; i < num_extra_ics; i++) - { - inner_class &extra_ic = extra_ics[i]; - extra_ic.inner = class_InnerClasses_RC.getRef(); - // Find the corresponding equivalent global IC: - inner_class *global_ic = cp.getIC(extra_ic.inner); - int flags = class_InnerClasses_F.getInt(); - if (flags == 0) - { - // The extra IC is simply a copy of a global IC. - if (global_ic == nullptr) - { - unpack_abort("bad reference to inner class"); - break; - } - extra_ic = (*global_ic); // fill in rest of fields - } - else - { - flags &= ~ACC_IC_LONG_FORM; // clear high bit if set to get clean zero - extra_ic.flags = flags; - extra_ic.outer = class_InnerClasses_outer_RCN.getRefN(); - extra_ic.name = class_InnerClasses_name_RUN.getRefN(); - // Detect if this is an exact copy of the global tuple. - if (global_ic != nullptr) - { - if (global_ic->flags != extra_ic.flags || global_ic->outer != extra_ic.outer || - global_ic->name != extra_ic.name) - { - global_ic = nullptr; // not really the same, so break the link - } - } - } - if (global_ic != nullptr && global_ic->requested) - { - // This local repetition reverses the globally implied request. - global_ic->requested = false; - extra_ic.requested = false; - local_ics -= 1; - } - else - { - // The global either does not exist, or is not yet requested. - extra_ic.requested = true; - local_ics += 1; - } - } - // Finally, if there are any that survived, put them into an attribute. - // (Note that a zero-count attribute is always deleted.) - // The putref calls below will tell the constant pool to add any - // necessary local CP references to support the InnerClasses attribute. - // This step must be the last round of additions to the local CP. - if (local_ics > 0) - { - // append the new attribute: - putref(cp.sym[constant_pool::s_InnerClasses]); - putu4(2 + 2 * 4 * local_ics); - putu2(local_ics); - PTRLIST_QSORT(requested_ics, raw_address_cmp); - int num_global_ics = requested_ics.length(); - for (i = -num_global_ics; i < num_extra_ics; i++) - { - inner_class *ic; - if (i < 0) - ic = (inner_class *)requested_ics.get(num_global_ics + i); - else - ic = &extra_ics[i]; - if (ic->requested) - { - putref(ic->inner); - putref(ic->outer); - putref(ic->name); - putu2(ic->flags); - } - } - putu2_at(wp_at(naOffset), ++na); // increment class attr count - } - - // Tidy up global 'requested' bits: - for (i = requested_ics.length(); --i >= 0;) - { - inner_class *ic = (inner_class *)requested_ics.get(i); - ic->requested = false; - } - requested_ics.empty(); - - close_output(); - - // rewrite CP references in the tail - cp.computeOutputIndexes(); - int nextref = 0; - for (i = 0; i < (int)class_fixup_type.size(); i++) - { - int type = class_fixup_type.getByte(i); - byte *fixp = wp_at(class_fixup_offset.get(i)); - entry *e = (entry *)class_fixup_ref.get(nextref++); - int idx = e->getOutputIndex(); - switch (type) - { - case 1: - putu1_at(fixp, idx); - break; - case 2: - putu2_at(fixp, idx); - break; - default: - assert(false); // should not reach here - } - } + // First, consult the global table and the local constant pool, + // and decide on the globally implied inner classes. + // (Note that we read the cpool's outputIndex fields, but we + // do not yet write them, since the local IC attribute might + // reverse a global decision to declare an IC.) + assert(requested_ics.length() == 0); // must start out empty + // Always include all members of the current class. + for (inner_class *child = cp.getFirstChildIC(cur_class); child != nullptr; + child = cp.getNextChildIC(child)) + { + child->requested = true; + requested_ics.add(child); + } + // And, for each inner class mentioned in the constant pool, + // include it and all its outers. + int noes = cp.outputEntries.length(); + entry **oes = (entry **)cp.outputEntries.base(); + for (i = 0; i < noes; i++) + { + entry &e = *oes[i]; + if (e.tag != CONSTANT_Class) + continue; // wrong sort + for (inner_class *ic = cp.getIC(&e); ic != nullptr; ic = cp.getIC(ic->outer)) + { + if (ic->requested) + break; // already processed + ic->requested = true; + requested_ics.add(ic); + } + } + int local_ics = requested_ics.length(); + // Second, consult a local attribute (if any) and adjust the global set. + inner_class *extra_ics = nullptr; + int num_extra_ics = 0; + if (cur_class_has_local_ics) + { + // adjust the set of ICs by symmetric set difference w/ the locals + num_extra_ics = class_InnerClasses_N.getInt(); + if (num_extra_ics == 0) + { + // Explicit zero count has an irregular meaning: It deletes the attr. + local_ics = 0; // (short-circuit all tests of requested bits) + } + else + { + extra_ics = T_NEW(inner_class, num_extra_ics); + // Note: extra_ics will be freed up by next call to get_next_file(). + } + } + for (i = 0; i < num_extra_ics; i++) + { + inner_class &extra_ic = extra_ics[i]; + extra_ic.inner = class_InnerClasses_RC.getRef(); + // Find the corresponding equivalent global IC: + inner_class *global_ic = cp.getIC(extra_ic.inner); + int flags = class_InnerClasses_F.getInt(); + if (flags == 0) + { + // The extra IC is simply a copy of a global IC. + if (global_ic == nullptr) + { + unpack_abort("bad reference to inner class"); + break; + } + extra_ic = (*global_ic); // fill in rest of fields + } + else + { + flags &= ~ACC_IC_LONG_FORM; // clear high bit if set to get clean zero + extra_ic.flags = flags; + extra_ic.outer = class_InnerClasses_outer_RCN.getRefN(); + extra_ic.name = class_InnerClasses_name_RUN.getRefN(); + // Detect if this is an exact copy of the global tuple. + if (global_ic != nullptr) + { + if (global_ic->flags != extra_ic.flags || global_ic->outer != extra_ic.outer || + global_ic->name != extra_ic.name) + { + global_ic = nullptr; // not really the same, so break the link + } + } + } + if (global_ic != nullptr && global_ic->requested) + { + // This local repetition reverses the globally implied request. + global_ic->requested = false; + extra_ic.requested = false; + local_ics -= 1; + } + else + { + // The global either does not exist, or is not yet requested. + extra_ic.requested = true; + local_ics += 1; + } + } + // Finally, if there are any that survived, put them into an attribute. + // (Note that a zero-count attribute is always deleted.) + // The putref calls below will tell the constant pool to add any + // necessary local CP references to support the InnerClasses attribute. + // This step must be the last round of additions to the local CP. + if (local_ics > 0) + { + // append the new attribute: + putref(cp.sym[constant_pool::s_InnerClasses]); + putu4(2 + 2 * 4 * local_ics); + putu2(local_ics); + PTRLIST_QSORT(requested_ics, raw_address_cmp); + int num_global_ics = requested_ics.length(); + for (i = -num_global_ics; i < num_extra_ics; i++) + { + inner_class *ic; + if (i < 0) + ic = (inner_class *)requested_ics.get(num_global_ics + i); + else + ic = &extra_ics[i]; + if (ic->requested) + { + putref(ic->inner); + putref(ic->outer); + putref(ic->name); + putu2(ic->flags); + } + } + putu2_at(wp_at(naOffset), ++na); // increment class attr count + } + + // Tidy up global 'requested' bits: + for (i = requested_ics.length(); --i >= 0;) + { + inner_class *ic = (inner_class *)requested_ics.get(i); + ic->requested = false; + } + requested_ics.empty(); + + close_output(); + + // rewrite CP references in the tail + cp.computeOutputIndexes(); + int nextref = 0; + for (i = 0; i < (int)class_fixup_type.size(); i++) + { + int type = class_fixup_type.getByte(i); + byte *fixp = wp_at(class_fixup_offset.get(i)); + entry *e = (entry *)class_fixup_ref.get(nextref++); + int idx = e->getOutputIndex(); + switch (type) + { + case 1: + putu1_at(fixp, idx); + break; + case 2: + putu2_at(fixp, idx); + break; + default: + assert(false); // should not reach here + } + } } void unpacker::write_classfile_head() { - cur_classfile_head.empty(); - set_output(&cur_classfile_head); - - putu4(JAVA_MAGIC); - putu2(cur_class_minver); - putu2(cur_class_majver); - putu2(cp.outputIndexLimit); - - int checkIndex = 1; - int noes = cp.outputEntries.length(); - entry **oes = (entry **)cp.outputEntries.base(); - for (int i = 0; i < noes; i++) - { - entry &e = *oes[i]; - assert(e.getOutputIndex() == checkIndex++); - byte tag = e.tag; - assert(tag != CONSTANT_Signature); - putu1(tag); - switch (tag) - { - case CONSTANT_Utf8: - putu2((int)e.value.b.len); - put_bytes(e.value.b); - break; - case CONSTANT_Integer: - case CONSTANT_Float: - putu4(e.value.i); - break; - case CONSTANT_Long: - case CONSTANT_Double: - putu8(e.value.l); - assert(checkIndex++); - break; - case CONSTANT_Class: - case CONSTANT_String: - // just write the ref - putu2(e.refs[0]->getOutputIndex()); - break; - case CONSTANT_Fieldref: - case CONSTANT_Methodref: - case CONSTANT_InterfaceMethodref: - case CONSTANT_NameandType: - putu2(e.refs[0]->getOutputIndex()); - putu2(e.refs[1]->getOutputIndex()); - break; - default: - unpack_abort(ERROR_INTERNAL); - } - } - close_output(); + cur_classfile_head.empty(); + set_output(&cur_classfile_head); + + putu4(JAVA_MAGIC); + putu2(cur_class_minver); + putu2(cur_class_majver); + putu2(cp.outputIndexLimit); + +#ifndef NDEBUG + int checkIndex = 1; +#endif + int noes = cp.outputEntries.length(); + entry **oes = (entry **)cp.outputEntries.base(); + for (int i = 0; i < noes; i++) + { + entry &e = *oes[i]; + assert(e.getOutputIndex() == checkIndex++); + byte tag = e.tag; + assert(tag != CONSTANT_Signature); + putu1(tag); + switch (tag) + { + case CONSTANT_Utf8: + putu2((int)e.value.b.len); + put_bytes(e.value.b); + break; + case CONSTANT_Integer: + case CONSTANT_Float: + putu4(e.value.i); + break; + case CONSTANT_Long: + case CONSTANT_Double: + putu8(e.value.l); + assert(checkIndex++); + break; + case CONSTANT_Class: + case CONSTANT_String: + // just write the ref + putu2(e.refs[0]->getOutputIndex()); + break; + case CONSTANT_Fieldref: + case CONSTANT_Methodref: + case CONSTANT_InterfaceMethodref: + case CONSTANT_NameandType: + putu2(e.refs[0]->getOutputIndex()); + putu2(e.refs[1]->getOutputIndex()); + break; + default: + unpack_abort(ERROR_INTERNAL); + } + } + close_output(); } unpacker::file *unpacker::get_next_file() { - free_temps(); - if (files_remaining == 0) - { - // Leave a clue that we're exhausted. - cur_file.name = nullptr; - cur_file.size = 0; - if (archive_size != 0) - { - uint64_t predicted_size = unsized_bytes_read + archive_size; - if (predicted_size != bytes_read) - unpack_abort("archive header had incorrect size"); - } - return nullptr; - } - files_remaining -= 1; - assert(files_written < file_count || classes_written < class_count); - cur_file.name = ""; - cur_file.size = 0; - cur_file.modtime = default_file_modtime; - cur_file.options = default_file_options; - cur_file.data[0].set(nullptr, 0); - cur_file.data[1].set(nullptr, 0); - if (files_written < file_count) - { - entry *e = file_name.getRef(); - cur_file.name = e->utf8String(); - bool haveLongSize = ((archive_options & AO_HAVE_FILE_SIZE_HI) != 0); - cur_file.size = file_size_hi.getLong(file_size_lo, haveLongSize); - if ((archive_options & AO_HAVE_FILE_MODTIME) != 0) - cur_file.modtime += file_modtime.getInt(); // relative to archive modtime - if ((archive_options & AO_HAVE_FILE_OPTIONS) != 0) - cur_file.options |= file_options.getInt() & ~suppress_file_options; - } - else if (classes_written < class_count) - { - // there is a class for a missing file record - cur_file.options |= FO_IS_CLASS_STUB; - } - if ((cur_file.options & FO_IS_CLASS_STUB) != 0) - { - assert(classes_written < class_count); - classes_written += 1; - if (cur_file.size != 0) - { - unpack_abort("class file size transmitted"); - } - reset_cur_classfile(); - - // write the meat of the classfile: - write_classfile_tail(); - cur_file.data[1] = cur_classfile_tail.b; - - // write the CP of the classfile, second: - write_classfile_head(); - cur_file.data[0] = cur_classfile_head.b; - - cur_file.size += cur_file.data[0].len; - cur_file.size += cur_file.data[1].len; - if (cur_file.name[0] == '\0') - { - bytes &prefix = cur_class->ref(0)->value.b; - const char *suffix = ".class"; - int len = (int)(prefix.len + strlen(suffix)); - bytes name; - name.set(T_NEW(byte, add_size(len, 1)), len); - cur_file.name = name.strcat(prefix).strcat(suffix).strval(); - } - } - else - { - // If there is buffered file data, produce a pointer to it. - if (cur_file.size != (size_t)cur_file.size) - { - // Silly size specified. - unpack_abort("resource file too large"); - } - size_t rpleft = input_remaining(); - if (rpleft > 0) - { - if (rpleft > cur_file.size) - rpleft = (size_t)cur_file.size; - cur_file.data[0].set(rp, rpleft); - rp += rpleft; - } - if (rpleft < cur_file.size) - { - // Caller must read the rest. - size_t fleft = (size_t)cur_file.size - rpleft; - bytes_read += fleft; // Credit it to the overall archive size. - } - } - bytes_written += cur_file.size; - files_written += 1; - return &cur_file; + free_temps(); + if (files_remaining == 0) + { + // Leave a clue that we're exhausted. + cur_file.name = nullptr; + cur_file.size = 0; + if (archive_size != 0) + { + uint64_t predicted_size = unsized_bytes_read + archive_size; + if (predicted_size != bytes_read) + unpack_abort("archive header had incorrect size"); + } + return nullptr; + } + files_remaining -= 1; + assert(files_written < file_count || classes_written < class_count); + cur_file.name = ""; + cur_file.size = 0; + cur_file.modtime = default_file_modtime; + cur_file.options = default_file_options; + cur_file.data[0].set(nullptr, 0); + cur_file.data[1].set(nullptr, 0); + if (files_written < file_count) + { + entry *e = file_name.getRef(); + cur_file.name = e->utf8String(); + bool haveLongSize = ((archive_options & AO_HAVE_FILE_SIZE_HI) != 0); + cur_file.size = file_size_hi.getLong(file_size_lo, haveLongSize); + if ((archive_options & AO_HAVE_FILE_MODTIME) != 0) + cur_file.modtime += file_modtime.getInt(); // relative to archive modtime + if ((archive_options & AO_HAVE_FILE_OPTIONS) != 0) + cur_file.options |= file_options.getInt() & ~suppress_file_options; + } + else if (classes_written < class_count) + { + // there is a class for a missing file record + cur_file.options |= FO_IS_CLASS_STUB; + } + if ((cur_file.options & FO_IS_CLASS_STUB) != 0) + { + assert(classes_written < class_count); + classes_written += 1; + if (cur_file.size != 0) + { + unpack_abort("class file size transmitted"); + } + reset_cur_classfile(); + + // write the meat of the classfile: + write_classfile_tail(); + cur_file.data[1] = cur_classfile_tail.b; + + // write the CP of the classfile, second: + write_classfile_head(); + cur_file.data[0] = cur_classfile_head.b; + + cur_file.size += cur_file.data[0].len; + cur_file.size += cur_file.data[1].len; + if (cur_file.name[0] == '\0') + { + bytes &prefix = cur_class->ref(0)->value.b; + const char *suffix = ".class"; + int len = (int)(prefix.len + strlen(suffix)); + bytes name; + name.set(T_NEW(byte, add_size(len, 1)), len); + cur_file.name = name.strcat(prefix).strcat(suffix).strval(); + } + } + else + { + // If there is buffered file data, produce a pointer to it. + if (cur_file.size != (size_t)cur_file.size) + { + // Silly size specified. + unpack_abort("resource file too large"); + } + size_t rpleft = input_remaining(); + if (rpleft > 0) + { + if (rpleft > cur_file.size) + rpleft = (size_t)cur_file.size; + cur_file.data[0].set(rp, rpleft); + rp += rpleft; + } + if (rpleft < cur_file.size) + { + // Caller must read the rest. + size_t fleft = (size_t)cur_file.size - rpleft; + bytes_read += fleft; // Credit it to the overall archive size. + } + } + bytes_written += cur_file.size; + files_written += 1; + return &cur_file; } // Write a file to jarout. void unpacker::write_file_to_jar(unpacker::file *f) { - size_t htsize = f->data[0].len + f->data[1].len; - uint64_t fsize = f->size; - if (htsize == fsize) - { - jarout->addJarEntry(f->name, f->deflate_hint(), f->modtime, f->data[0], f->data[1]); - } - else - { - assert(input_remaining() == 0); - bytes part1, part2; - part1.len = f->data[0].len; - part1.set(T_NEW(byte, part1.len), part1.len); - part1.copyFrom(f->data[0]); - assert(f->data[1].len == 0); - part2.set(nullptr, 0); - size_t fleft = (size_t)fsize - part1.len; - assert(bytes_read > fleft); // part2 already credited by get_next_file - bytes_read -= fleft; - if (fleft > 0) - { - // Must read some more. - if (live_input) - { - // Stop using the input buffer. Make a new one: - if (free_input) - input.free(); - input.init(fleft > (1 << 12) ? fleft : (1 << 12)); - free_input = true; - live_input = false; - } - else - { - // Make it large enough. - assert(free_input); // must be reallocable - input.ensureSize(fleft); - } - rplimit = rp = input.base(); - input.setLimit(rp + fleft); - if (!ensure_input(fleft)) - unpack_abort("EOF reading resource file"); - part2.ptr = input_scan(); - part2.len = input_remaining(); - rplimit = rp = input.base(); - } - jarout->addJarEntry(f->name, f->deflate_hint(), f->modtime, part1, part2); - } - if (verbose >= 3) - { - fprintf(stderr, "Wrote " LONG_LONG_FORMAT " bytes to: %s\n", fsize, f->name); - } + size_t htsize = f->data[0].len + f->data[1].len; + uint64_t fsize = f->size; + if (htsize == fsize) + { + jarout->addJarEntry(f->name, f->deflate_hint(), f->modtime, f->data[0], f->data[1]); + } + else + { + assert(input_remaining() == 0); + bytes part1, part2; + part1.len = f->data[0].len; + part1.set(T_NEW(byte, part1.len), part1.len); + part1.copyFrom(f->data[0]); + assert(f->data[1].len == 0); + part2.set(nullptr, 0); + size_t fleft = (size_t)fsize - part1.len; + assert(bytes_read > fleft); // part2 already credited by get_next_file + bytes_read -= fleft; + if (fleft > 0) + { + // Must read some more. + if (live_input) + { + // Stop using the input buffer. Make a new one: + if (free_input) + input.free(); + input.init(fleft > (1 << 12) ? fleft : (1 << 12)); + free_input = true; + live_input = false; + } + else + { + // Make it large enough. + assert(free_input); // must be reallocable + input.ensureSize(fleft); + } + rplimit = rp = input.base(); + input.setLimit(rp + fleft); + if (!ensure_input(fleft)) + unpack_abort("EOF reading resource file"); + part2.ptr = input_scan(); + part2.len = input_remaining(); + rplimit = rp = input.base(); + } + jarout->addJarEntry(f->name, f->deflate_hint(), f->modtime, part1, part2); + } + if (verbose >= 3) + { + fprintf(stderr, "Wrote %" PRIu64 " bytes to: %s\n", fsize, f->name); + } } diff --git a/libraries/pack200/src/unpack.h b/libraries/pack200/src/unpack.h index 0100700d..cc5dd60a 100644 --- a/libraries/pack200/src/unpack.h +++ b/libraries/pack200/src/unpack.h @@ -23,6 +23,8 @@ * questions. */ +#pragma once + // Global Structures struct jar; struct gunzip; @@ -33,106 +35,108 @@ struct cpindex; struct inner_class; struct value_stream; +typedef int64_t (*read_input_fn_t)(unpacker *self, void *buf, int64_t minlen, int64_t maxlen); + struct cpindex { - uint32_t len; - entry *base1; // base of primary index - entry **base2; // base of secondary index - byte ixTag; // type of entries (!= CONSTANT_None), plus 64 if sub-index - enum - { - SUB_TAG = 64 - }; - - entry *get(uint32_t i); - - void init(int len_, entry *base1_, int ixTag_) - { - len = len_; - base1 = base1_; - base2 = nullptr; - ixTag = ixTag_; - } - void init(int len_, entry **base2_, int ixTag_) - { - len = len_; - base1 = nullptr; - base2 = base2_; - ixTag = ixTag_; - } + uint32_t len; + entry *base1; // base of primary index + entry **base2; // base of secondary index + byte ixTag; // type of entries (!= CONSTANT_None), plus 64 if sub-index + enum + { + SUB_TAG = 64 + }; + + entry *get(uint32_t i); + + void init(int len_, entry *base1_, int ixTag_) + { + len = len_; + base1 = base1_; + base2 = nullptr; + ixTag = ixTag_; + } + void init(int len_, entry **base2_, int ixTag_) + { + len = len_; + base1 = nullptr; + base2 = base2_; + ixTag = ixTag_; + } }; struct constant_pool { - uint32_t nentries; - entry *entries; - entry *first_extra_entry; - uint32_t maxentries; // total allocated size of entries - - // Position and size of each homogeneous subrange: - int tag_count[CONSTANT_Limit]; - int tag_base[CONSTANT_Limit]; - cpindex tag_index[CONSTANT_Limit]; - ptrlist tag_extras[CONSTANT_Limit]; - - cpindex *member_indexes; // indexed by 2*CONSTANT_Class.inord - cpindex *getFieldIndex(entry *classRef); - cpindex *getMethodIndex(entry *classRef); - - inner_class **ic_index; - inner_class **ic_child_index; - inner_class *getIC(entry *inner); - inner_class *getFirstChildIC(entry *outer); - inner_class *getNextChildIC(inner_class *child); - - int outputIndexLimit; // index limit after renumbering - ptrlist outputEntries; // list of entry* needing output idx assigned - - entry **hashTab; - uint32_t hashTabLength; - entry *&hashTabRef(byte tag, bytes &b); - entry *ensureUtf8(bytes &b); - entry *ensureClass(bytes &b); - - // Well-known Utf8 symbols. - enum - { + uint32_t nentries; + entry *entries; + entry *first_extra_entry; + uint32_t maxentries; // total allocated size of entries + + // Position and size of each homogeneous subrange: + int tag_count[CONSTANT_Limit]; + int tag_base[CONSTANT_Limit]; + cpindex tag_index[CONSTANT_Limit]; + ptrlist tag_extras[CONSTANT_Limit]; + + cpindex *member_indexes; // indexed by 2*CONSTANT_Class.inord + cpindex *getFieldIndex(entry *classRef); + cpindex *getMethodIndex(entry *classRef); + + inner_class **ic_index; + inner_class **ic_child_index; + inner_class *getIC(entry *inner); + inner_class *getFirstChildIC(entry *outer); + inner_class *getNextChildIC(inner_class *child); + + int outputIndexLimit; // index limit after renumbering + ptrlist outputEntries; // list of entry* needing output idx assigned + + entry **hashTab; + uint32_t hashTabLength; + entry *&hashTabRef(byte tag, bytes &b); + entry *ensureUtf8(bytes &b); + entry *ensureClass(bytes &b); + + // Well-known Utf8 symbols. + enum + { #define SNAME(n, s) s_##s, - ALL_ATTR_DO(SNAME) + ALL_ATTR_DO(SNAME) #undef SNAME - s_lt_init_gt, // <init> - s_LIMIT - }; - entry *sym[s_LIMIT]; - - // read counts from hdr, allocate main arrays - enum - { - NUM_COUNTS = 12 - }; - void init(unpacker *u, int counts[NUM_COUNTS]); - - // pointer to outer unpacker, for error checks etc. - unpacker *u; - - int getCount(byte tag) - { - assert((uint32_t)tag < CONSTANT_Limit); - return tag_count[tag]; - } - cpindex *getIndex(byte tag) - { - assert((uint32_t)tag < CONSTANT_Limit); - return &tag_index[tag]; - } - cpindex *getKQIndex(); // uses cur_descr - - void expandSignatures(); - void initMemberIndexes(); - - void computeOutputOrder(); - void computeOutputIndexes(); - void resetOutputIndexes(); + s_lt_init_gt, // <init> + s_LIMIT + }; + entry *sym[s_LIMIT]; + + // read counts from hdr, allocate main arrays + enum + { + NUM_COUNTS = 12 + }; + void init(unpacker *u, int counts[NUM_COUNTS]); + + // pointer to outer unpacker, for error checks etc. + unpacker *u; + + int getCount(byte tag) + { + assert((uint32_t)tag < CONSTANT_Limit); + return tag_count[tag]; + } + cpindex *getIndex(byte tag) + { + assert((uint32_t)tag < CONSTANT_Limit); + return &tag_index[tag]; + } + cpindex *getKQIndex(); // uses cur_descr + + void expandSignatures(); + void initMemberIndexes(); + + void computeOutputOrder(); + void computeOutputIndexes(); + void resetOutputIndexes(); }; /* @@ -141,407 +145,405 @@ struct constant_pool */ struct unpacker { - // One element of the resulting JAR. - struct file - { - const char *name; - uint64_t size; - int modtime; - int options; - bytes data[2]; - // Note: If Sum(data[*].len) < size, - // remaining bytes must be read directly from the input stream. - bool deflate_hint() - { - return ((options & FO_DEFLATE_HINT) != 0); - } - }; - - // if running Unix-style, here are the inputs and outputs - FILE *infileptr; // buffered - bytes inbytes; // direct - gunzip *gzin; // gunzip filter, if any - jar *jarout; // output JAR file - - // pointer to self, for U_NEW macro - unpacker *u; - - ptrlist mallocs; // list of guys to free when we are all done - ptrlist tmallocs; // list of guys to free on next client request - fillbytes smallbuf; // supplies small alloc requests - fillbytes tsmallbuf; // supplies temporary small alloc requests - - // option management members - int verbose; // verbose level, 0 means no output - int deflate_hint_or_zero; // ==0 means not set, otherwise -1 or 1 - int modification_time_or_zero; - - // input stream - fillbytes input; // the whole block (size is predicted, has slop too) - bool live_input; // is the data in this block live? - bool free_input; // must the input buffer be freed? - byte *rp; // read pointer (< rplimit <= input.limit()) - byte *rplimit; // how much of the input block has been read? - uint64_t bytes_read; - int unsized_bytes_read; - - // callback to read at least one byte, up to available input - typedef int64_t (*read_input_fn_t)(unpacker *self, void *buf, int64_t minlen, - int64_t maxlen); - read_input_fn_t read_input_fn; - - // archive header fields - int magic, minver, majver; - size_t archive_size; - int archive_next_count, archive_options, archive_modtime; - int band_headers_size; - int file_count, attr_definition_count, ic_count, class_count; - int default_class_minver, default_class_majver; - int default_file_options, suppress_file_options; // not header fields - int default_archive_modtime, default_file_modtime; // not header fields - int code_count; // not a header field - int files_remaining; // not a header field - - // engine state - band *all_bands; // indexed by band_number - byte *meta_rp; // read-pointer into (copy of) band_headers - constant_pool cp; // all constant pool information - inner_class *ics; // InnerClasses - - // output stream - bytes output; // output block (either classfile head or tail) - byte *wp; // write pointer (< wplimit == output.limit()) - byte *wpbase; // write pointer starting address (<= wp) - byte *wplimit; // how much of the output block has been written? - - // output state - file cur_file; - entry *cur_class; // CONSTANT_Class entry - entry *cur_super; // CONSTANT_Class entry or nullptr - entry *cur_descr; // CONSTANT_NameandType entry - int cur_descr_flags; // flags corresponding to cur_descr - int cur_class_minver, cur_class_majver; - bool cur_class_has_local_ics; - fillbytes cur_classfile_head; - fillbytes cur_classfile_tail; - int files_written; // also tells which file we're working on - int classes_written; // also tells which class we're working on - uint64_t bytes_written; - intlist bcimap; - fillbytes class_fixup_type; - intlist class_fixup_offset; - ptrlist class_fixup_ref; - fillbytes code_fixup_type; // which format of branch operand? - intlist code_fixup_offset; // location of operand needing fixup - intlist code_fixup_source; // encoded ID of branch insn - ptrlist requested_ics; // which ics need output? - - // stats pertaining to multiple segments (updated on reset) - uint64_t bytes_read_before_reset; - uint64_t bytes_written_before_reset; - int files_written_before_reset; - int classes_written_before_reset; - int segments_read_before_reset; - - // attribute state - struct layout_definition - { - uint32_t idx; // index (0..31...) which identifies this layout - const char *name; // name of layout - entry *nameEntry; - const char *layout; // string of layout (not yet parsed) - band **elems; // array of top-level layout elems (or callables) - - bool hasCallables() - { - return layout[0] == '['; - } - band **bands() - { - assert(elems != nullptr); - return elems; - } - }; - struct attr_definitions - { - unpacker *u; // pointer to self, for U_NEW macro - int xxx_flags_hi_bn; // locator for flags, count, indexes, calls bands - int attrc; // ATTR_CONTEXT_CLASS, etc. - uint32_t flag_limit; // 32 or 63, depending on archive_options bit - uint64_t predef; // mask of built-in definitions - uint64_t redef; // mask of local flag definitions or redefinitions - ptrlist layouts; // local (compressor-defined) defs, in index order - int flag_count[X_ATTR_LIMIT_FLAGS_HI]; - intlist overflow_count; - ptrlist strip_names; // what attribute names are being stripped? - ptrlist band_stack; // Temp., used during layout parsing. - ptrlist calls_to_link; // (ditto) - int bands_made; // (ditto) - - void free() - { - layouts.free(); - overflow_count.free(); - strip_names.free(); - band_stack.free(); - calls_to_link.free(); - } - - // Locate the five fixed bands. - band &xxx_flags_hi(); - band &xxx_flags_lo(); - band &xxx_attr_count(); - band &xxx_attr_indexes(); - band &xxx_attr_calls(); - band &fixed_band(int e_class_xxx); - - // Register a new layout, and make bands for it. - layout_definition *defineLayout(int idx, const char *name, const char *layout); - layout_definition *defineLayout(int idx, entry *nameEntry, const char *layout); - band **buildBands(layout_definition *lo); - - // Parse a layout string or part of one, recursively if necessary. - const char *parseLayout(const char *lp, band **&res, int curCble); - const char *parseNumeral(const char *lp, int &res); - const char *parseIntLayout(const char *lp, band *&res, byte le_kind, - bool can_be_signed = false); - band **popBody(int band_stack_base); // pops a body off band_stack - - // Read data into the bands of the idx-th layout. - void readBandData(int idx); // parse layout, make bands, read data - void readBandData(band **body, uint32_t count); // recursive helper - - layout_definition *getLayout(uint32_t idx) - { - if (idx >= (uint32_t)layouts.length()) - return nullptr; - return (layout_definition *)layouts.get(idx); - } - - void setHaveLongFlags(bool z) - { - assert(flag_limit == 0); // not set up yet - flag_limit = (z ? X_ATTR_LIMIT_FLAGS_HI : X_ATTR_LIMIT_NO_FLAGS_HI); - } - bool haveLongFlags() - { - assert(flag_limit == X_ATTR_LIMIT_NO_FLAGS_HI || - flag_limit == X_ATTR_LIMIT_FLAGS_HI); - return flag_limit == X_ATTR_LIMIT_FLAGS_HI; - } - - // Return flag_count if idx is predef and not redef, else zero. - int predefCount(uint32_t idx); - - bool isRedefined(uint32_t idx) - { - if (idx >= flag_limit) - return false; - return (bool)((redef >> idx) & 1); - } - bool isPredefined(uint32_t idx) - { - if (idx >= flag_limit) - return false; - return (bool)(((predef & ~redef) >> idx) & 1); - } - uint64_t flagIndexMask() - { - return (predef | redef); - } - bool isIndex(uint32_t idx) - { - assert(flag_limit != 0); // must be set up already - if (idx < flag_limit) - return (bool)(((predef | redef) >> idx) & 1); - else - return (idx - flag_limit < (uint32_t)overflow_count.length()); - } - int &getCount(uint32_t idx) - { - assert(isIndex(idx)); - if (idx < flag_limit) - return flag_count[idx]; - else - return overflow_count.get(idx - flag_limit); - } - }; - - attr_definitions attr_defs[ATTR_CONTEXT_LIMIT]; - - // Initialization - void init(read_input_fn_t input_fn = nullptr); - // Resets to a known sane state - void reset(); - // Deallocates all storage. - void free(); - // Deallocates temporary storage (volatile after next client call). - void free_temps() - { - tsmallbuf.init(); - tmallocs.freeAll(); - } - - // Option management methods - bool set_option(const char *option, const char *value); - const char *get_option(const char *option); - - // Fetching input. - bool ensure_input(int64_t more); - byte *input_scan() - { - return rp; - } - size_t input_remaining() - { - return rplimit - rp; - } - size_t input_consumed() - { - return rp - input.base(); - } - - // Entry points to the unpack engine - static int run(int argc, char **argv); // Unix-style entry point. - void check_options(); - void start(void *packptr = nullptr, size_t len = 0); - void write_file_to_jar(file *f); - void finish(); - - // Public post unpack methods - int get_files_remaining() - { - return files_remaining; - } - int get_segments_remaining() - { - return archive_next_count; - } - file *get_next_file(); // returns nullptr on last file - - // General purpose methods - void *alloc(size_t size) - { - return alloc_heap(size, true); - } - void *temp_alloc(size_t size) - { - return alloc_heap(size, true, true); - } - void *alloc_heap(size_t size, bool smallOK = false, bool temp = false); - void saveTo(bytes &b, const char *str) - { - saveTo(b, (byte *)str, strlen(str)); - } - void saveTo(bytes &b, bytes &data) - { - saveTo(b, data.ptr, data.len); - } - void saveTo(bytes &b, byte *ptr, size_t len); //{ b.ptr = U_NEW...} - const char *saveStr(const char *str) - { - bytes buf; - saveTo(buf, str); - return buf.strval(); - } - const char *saveIntStr(int num) - { - char buf[30]; - sprintf(buf, "%d", num); - return saveStr(buf); - } - static unpacker *current(); // find current instance - - // Output management - void set_output(fillbytes *which) - { - assert(wp == nullptr); - which->ensureSize(1 << 12); // covers the average classfile - wpbase = which->base(); - wp = which->limit(); - wplimit = which->end(); - } - fillbytes *close_output(fillbytes *which = nullptr); // inverse of set_output - - // These take an implicit parameter of wp/wplimit, and resize as necessary: - byte *put_space(size_t len); // allocates space at wp, returns pointer - size_t put_empty(size_t s) - { - byte *p = put_space(s); - return p - wpbase; - } - void ensure_put_space(size_t len); - void put_bytes(bytes &b) - { - b.writeTo(put_space(b.len)); - } - void putu1(int n) - { - putu1_at(put_space(1), n); - } - void putu1_fast(int n) - { - putu1_at(wp++, n); - } - void putu2(int n); // { putu2_at(put_space(2), n); } - void putu4(int n); // { putu4_at(put_space(4), n); } - void putu8(int64_t n); // { putu8_at(put_space(8), n); } - void putref(entry *e); // { putu2_at(put_space(2), putref_index(e, 2)); } - void putu1ref(entry *e); // { putu1_at(put_space(1), putref_index(e, 1)); } - int putref_index(entry *e, int size); // size in [1..2] - void put_label(int curIP, int size); // size in {2,4} - void putlayout(band **body); - void put_stackmap_type(); - - size_t wpoffset() - { - return (size_t)(wp - wpbase); - } // (unvariant across overflow) - byte *wp_at(size_t offset) - { - return wpbase + offset; - } - uint32_t to_bci(uint32_t bii); - void get_code_header(int &max_stack, int &max_na_locals, int &handler_count, int &cflags); - band *ref_band_for_self_op(int bc, bool &isAloadVar, int &origBCVar); - band *ref_band_for_op(int bc); - - // Definitions of standard classfile int formats: - static void putu1_at(byte *wp, int n) - { - assert(n == (n & 0xFF)); - wp[0] = n; - } - static void putu2_at(byte *wp, int n); - static void putu4_at(byte *wp, int n); - static void putu8_at(byte *wp, int64_t n); - - // Private stuff - void reset_cur_classfile(); - void write_classfile_tail(); - void write_classfile_head(); - void write_code(); - void write_bc_ops(); - void write_members(int num, int attrc); // attrc=ATTR_CONTEXT_FIELD/METHOD - int write_attrs(int attrc, uint64_t indexBits); - - // The readers - void read_bands(); - void read_file_header(); - void read_cp(); - void read_cp_counts(value_stream &hdr); - void read_attr_defs(); - void read_ics(); - void read_attrs(int attrc, int obj_count); - void read_classes(); - void read_code_headers(); - void read_bcs(); - void read_bc_ops(); - void read_files(); - void read_Utf8_values(entry *cpMap, int len); - void read_single_words(band &cp_band, entry *cpMap, int len); - void read_double_words(band &cp_bands, entry *cpMap, int len); - void read_single_refs(band &cp_band, byte refTag, entry *cpMap, int len); - void read_double_refs(band &cp_band, byte ref1Tag, byte ref2Tag, entry *cpMap, int len); - void read_signature_values(entry *cpMap, int len); + // One element of the resulting JAR. + struct file + { + const char *name; + uint64_t size; + int modtime; + int options; + bytes data[2]; + // Note: If Sum(data[*].len) < size, + // remaining bytes must be read directly from the input stream. + bool deflate_hint() + { + return ((options & FO_DEFLATE_HINT) != 0); + } + }; + + // if running Unix-style, here are the inputs and outputs + FILE *infileptr; // buffered + bytes inbytes; // direct + gunzip *gzin; // gunzip filter, if any + jar *jarout; // output JAR file + + // pointer to self, for U_NEW macro + unpacker *u; + + ptrlist mallocs; // list of guys to free when we are all done + ptrlist tmallocs; // list of guys to free on next client request + fillbytes smallbuf; // supplies small alloc requests + fillbytes tsmallbuf; // supplies temporary small alloc requests + + // option management members + int verbose; // verbose level, 0 means no output + int deflate_hint_or_zero; // ==0 means not set, otherwise -1 or 1 + int modification_time_or_zero; + + // input stream + fillbytes input; // the whole block (size is predicted, has slop too) + bool live_input; // is the data in this block live? + bool free_input; // must the input buffer be freed? + byte *rp; // read pointer (< rplimit <= input.limit()) + byte *rplimit; // how much of the input block has been read? + uint64_t bytes_read; + int unsized_bytes_read; + + // callback to read at least one byte, up to available input + read_input_fn_t read_input_fn; + + // archive header fields + int magic, minver, majver; + size_t archive_size; + int archive_next_count, archive_options, archive_modtime; + int band_headers_size; + int file_count, attr_definition_count, ic_count, class_count; + int default_class_minver, default_class_majver; + int default_file_options, suppress_file_options; // not header fields + int default_archive_modtime, default_file_modtime; // not header fields + int code_count; // not a header field + int files_remaining; // not a header field + + // engine state + band *all_bands; // indexed by band_number + byte *meta_rp; // read-pointer into (copy of) band_headers + constant_pool cp; // all constant pool information + inner_class *ics; // InnerClasses + + // output stream + bytes output; // output block (either classfile head or tail) + byte *wp; // write pointer (< wplimit == output.limit()) + byte *wpbase; // write pointer starting address (<= wp) + byte *wplimit; // how much of the output block has been written? + + // output state + file cur_file; + entry *cur_class; // CONSTANT_Class entry + entry *cur_super; // CONSTANT_Class entry or nullptr + entry *cur_descr; // CONSTANT_NameandType entry + int cur_descr_flags; // flags corresponding to cur_descr + int cur_class_minver, cur_class_majver; + bool cur_class_has_local_ics; + fillbytes cur_classfile_head; + fillbytes cur_classfile_tail; + int files_written; // also tells which file we're working on + int classes_written; // also tells which class we're working on + uint64_t bytes_written; + intlist bcimap; + fillbytes class_fixup_type; + intlist class_fixup_offset; + ptrlist class_fixup_ref; + fillbytes code_fixup_type; // which format of branch operand? + intlist code_fixup_offset; // location of operand needing fixup + intlist code_fixup_source; // encoded ID of branch insn + ptrlist requested_ics; // which ics need output? + + // stats pertaining to multiple segments (updated on reset) + uint64_t bytes_read_before_reset; + uint64_t bytes_written_before_reset; + int files_written_before_reset; + int classes_written_before_reset; + int segments_read_before_reset; + + // attribute state + struct layout_definition + { + uint32_t idx; // index (0..31...) which identifies this layout + const char *name; // name of layout + entry *nameEntry; + const char *layout; // string of layout (not yet parsed) + band **elems; // array of top-level layout elems (or callables) + + bool hasCallables() + { + return layout[0] == '['; + } + band **bands() + { + assert(elems != nullptr); + return elems; + } + }; + struct attr_definitions + { + unpacker *u; // pointer to self, for U_NEW macro + int xxx_flags_hi_bn; // locator for flags, count, indexes, calls bands + int attrc; // ATTR_CONTEXT_CLASS, etc. + uint32_t flag_limit; // 32 or 63, depending on archive_options bit + uint64_t predef; // mask of built-in definitions + uint64_t redef; // mask of local flag definitions or redefinitions + ptrlist layouts; // local (compressor-defined) defs, in index order + int flag_count[X_ATTR_LIMIT_FLAGS_HI]; + intlist overflow_count; + ptrlist strip_names; // what attribute names are being stripped? + ptrlist band_stack; // Temp., used during layout parsing. + ptrlist calls_to_link; // (ditto) + int bands_made; // (ditto) + + void free() + { + layouts.free(); + overflow_count.free(); + strip_names.free(); + band_stack.free(); + calls_to_link.free(); + } + + // Locate the five fixed bands. + band &xxx_flags_hi(); + band &xxx_flags_lo(); + band &xxx_attr_count(); + band &xxx_attr_indexes(); + band &xxx_attr_calls(); + band &fixed_band(int e_class_xxx); + + // Register a new layout, and make bands for it. + layout_definition *defineLayout(int idx, const char *name, const char *layout); + layout_definition *defineLayout(int idx, entry *nameEntry, const char *layout); + band **buildBands(layout_definition *lo); + + // Parse a layout string or part of one, recursively if necessary. + const char *parseLayout(const char *lp, band **&res, int curCble); + const char *parseNumeral(const char *lp, int &res); + const char *parseIntLayout(const char *lp, band *&res, byte le_kind, + bool can_be_signed = false); + band **popBody(int band_stack_base); // pops a body off band_stack + + // Read data into the bands of the idx-th layout. + void readBandData(int idx); // parse layout, make bands, read data + void readBandData(band **body, uint32_t count); // recursive helper + + layout_definition *getLayout(uint32_t idx) + { + if (idx >= (uint32_t)layouts.length()) + return nullptr; + return (layout_definition *)layouts.get(idx); + } + + void setHaveLongFlags(bool z) + { + assert(flag_limit == 0); // not set up yet + flag_limit = (z ? X_ATTR_LIMIT_FLAGS_HI : X_ATTR_LIMIT_NO_FLAGS_HI); + } + bool haveLongFlags() + { + assert(flag_limit == X_ATTR_LIMIT_NO_FLAGS_HI || + flag_limit == X_ATTR_LIMIT_FLAGS_HI); + return flag_limit == X_ATTR_LIMIT_FLAGS_HI; + } + + // Return flag_count if idx is predef and not redef, else zero. + int predefCount(uint32_t idx); + + bool isRedefined(uint32_t idx) + { + if (idx >= flag_limit) + return false; + return (bool)((redef >> idx) & 1); + } + bool isPredefined(uint32_t idx) + { + if (idx >= flag_limit) + return false; + return (bool)(((predef & ~redef) >> idx) & 1); + } + uint64_t flagIndexMask() + { + return (predef | redef); + } + bool isIndex(uint32_t idx) + { + assert(flag_limit != 0); // must be set up already + if (idx < flag_limit) + return (bool)(((predef | redef) >> idx) & 1); + else + return (idx - flag_limit < (uint32_t)overflow_count.length()); + } + int &getCount(uint32_t idx) + { + assert(isIndex(idx)); + if (idx < flag_limit) + return flag_count[idx]; + else + return overflow_count.get(idx - flag_limit); + } + }; + + attr_definitions attr_defs[ATTR_CONTEXT_LIMIT]; + + // Initialization + void init(read_input_fn_t input_fn = nullptr); + // Resets to a known sane state + void reset(); + // Deallocates all storage. + void free(); + // Deallocates temporary storage (volatile after next client call). + void free_temps() + { + tsmallbuf.init(); + tmallocs.freeAll(); + } + + // Option management methods + bool set_option(const char *option, const char *value); + const char *get_option(const char *option); + + // Fetching input. + bool ensure_input(int64_t more); + byte *input_scan() + { + return rp; + } + size_t input_remaining() + { + return rplimit - rp; + } + size_t input_consumed() + { + return rp - input.base(); + } + + // Entry points to the unpack engine + static int run(int argc, char **argv); // Unix-style entry point. + void check_options(); + void start(void *packptr = nullptr, size_t len = 0); + void write_file_to_jar(file *f); + void finish(); + + // Public post unpack methods + int get_files_remaining() + { + return files_remaining; + } + int get_segments_remaining() + { + return archive_next_count; + } + file *get_next_file(); // returns nullptr on last file + + // General purpose methods + void *alloc(size_t size) + { + return alloc_heap(size, true); + } + void *temp_alloc(size_t size) + { + return alloc_heap(size, true, true); + } + void *alloc_heap(size_t size, bool smallOK = false, bool temp = false); + void saveTo(bytes &b, const char *str) + { + saveTo(b, (byte *)str, strlen(str)); + } + void saveTo(bytes &b, bytes &data) + { + saveTo(b, data.ptr, data.len); + } + void saveTo(bytes &b, byte *ptr, size_t len); //{ b.ptr = U_NEW...} + const char *saveStr(const char *str) + { + bytes buf; + saveTo(buf, str); + return buf.strval(); + } + const char *saveIntStr(int num) + { + char buf[30]; + sprintf(buf, "%d", num); + return saveStr(buf); + } + static unpacker *current(); // find current instance + + // Output management + void set_output(fillbytes *which) + { + assert(wp == nullptr); + which->ensureSize(1 << 12); // covers the average classfile + wpbase = which->base(); + wp = which->limit(); + wplimit = which->end(); + } + fillbytes *close_output(fillbytes *which = nullptr); // inverse of set_output + + // These take an implicit parameter of wp/wplimit, and resize as necessary: + byte *put_space(size_t len); // allocates space at wp, returns pointer + size_t put_empty(size_t s) + { + byte *p = put_space(s); + return p - wpbase; + } + void ensure_put_space(size_t len); + void put_bytes(bytes &b) + { + b.writeTo(put_space(b.len)); + } + void putu1(int n) + { + putu1_at(put_space(1), n); + } + void putu1_fast(int n) + { + putu1_at(wp++, n); + } + void putu2(int n); // { putu2_at(put_space(2), n); } + void putu4(int n); // { putu4_at(put_space(4), n); } + void putu8(int64_t n); // { putu8_at(put_space(8), n); } + void putref(entry *e); // { putu2_at(put_space(2), putref_index(e, 2)); } + void putu1ref(entry *e); // { putu1_at(put_space(1), putref_index(e, 1)); } + int putref_index(entry *e, int size); // size in [1..2] + void put_label(int curIP, int size); // size in {2,4} + void putlayout(band **body); + void put_stackmap_type(); + + size_t wpoffset() + { + return (size_t)(wp - wpbase); + } // (unvariant across overflow) + byte *wp_at(size_t offset) + { + return wpbase + offset; + } + uint32_t to_bci(uint32_t bii); + void get_code_header(int &max_stack, int &max_na_locals, int &handler_count, int &cflags); + band *ref_band_for_self_op(int bc, bool &isAloadVar, int &origBCVar); + band *ref_band_for_op(int bc); + + // Definitions of standard classfile int formats: + static void putu1_at(byte *wp, int n) + { + assert(n == (n & 0xFF)); + wp[0] = n; + } + static void putu2_at(byte *wp, int n); + static void putu4_at(byte *wp, int n); + static void putu8_at(byte *wp, int64_t n); + + // Private stuff + void reset_cur_classfile(); + void write_classfile_tail(); + void write_classfile_head(); + void write_code(); + void write_bc_ops(); + void write_members(int num, int attrc); // attrc=ATTR_CONTEXT_FIELD/METHOD + int write_attrs(int attrc, uint64_t indexBits); + + // The readers + void read_bands(); + void read_file_header(); + void read_cp(); + void read_cp_counts(value_stream &hdr); + void read_attr_defs(); + void read_ics(); + void read_attrs(int attrc, int obj_count); + void read_classes(); + void read_code_headers(); + void read_bcs(); + void read_bc_ops(); + void read_files(); + void read_Utf8_values(entry *cpMap, int len); + void read_single_words(band &cp_band, entry *cpMap, int len); + void read_double_words(band &cp_bands, entry *cpMap, int len); + void read_single_refs(band &cp_band, byte refTag, entry *cpMap, int len); + void read_double_refs(band &cp_band, byte ref1Tag, byte ref2Tag, entry *cpMap, int len); + void read_signature_values(entry *cpMap, int len); }; diff --git a/libraries/pack200/src/unpack200.cpp b/libraries/pack200/src/unpack200.cpp index 22b7f3b0..e8826f28 100644 --- a/libraries/pack200/src/unpack200.cpp +++ b/libraries/pack200/src/unpack200.cpp @@ -45,118 +45,118 @@ // Callback for fetching data, Unix style. static int64_t read_input_via_stdio(unpacker *u, void *buf, int64_t minlen, int64_t maxlen) { - assert(u->infileptr != nullptr); - assert(minlen <= maxlen); // don't talk nonsense - int64_t numread = 0; - char *bufptr = (char *)buf; - while (numread < minlen) - { - // read available input, up to buf.length or maxlen - int readlen = (1 << 16); - if (readlen > (maxlen - numread)) - readlen = (int)(maxlen - numread); - int nr = 0; + assert(u->infileptr != nullptr); + assert(minlen <= maxlen); // don't talk nonsense + int64_t numread = 0; + char *bufptr = (char *)buf; + while (numread < minlen) + { + // read available input, up to buf.length or maxlen + int readlen = (1 << 16); + if (readlen > (maxlen - numread)) + readlen = (int)(maxlen - numread); + int nr = 0; - nr = (int)fread(bufptr, 1, readlen, u->infileptr); - if (nr <= 0) - { - if (errno != EINTR) - break; - nr = 0; - } - numread += nr; - bufptr += nr; - assert(numread <= maxlen); - } - return numread; + nr = (int)fread(bufptr, 1, readlen, u->infileptr); + if (nr <= 0) + { + if (errno != EINTR) + break; + nr = 0; + } + numread += nr; + bufptr += nr; + assert(numread <= maxlen); + } + return numread; } enum { - EOF_MAGIC = 0, - BAD_MAGIC = -1 + EOF_MAGIC = 0, + BAD_MAGIC = -1 }; static int read_magic(unpacker *u, char peek[], int peeklen) { - assert(peeklen == 4); // magic numbers are always 4 bytes - int64_t nr = (u->read_input_fn)(u, peek, peeklen, peeklen); - if (nr != peeklen) - { - return (nr == 0) ? EOF_MAGIC : BAD_MAGIC; - } - int magic = 0; - for (int i = 0; i < peeklen; i++) - { - magic <<= 8; - magic += peek[i] & 0xFF; - } - return magic; + assert(peeklen == 4); // magic numbers are always 4 bytes + int64_t nr = (u->read_input_fn)(u, peek, peeklen, peeklen); + if (nr != peeklen) + { + return (nr == 0) ? EOF_MAGIC : BAD_MAGIC; + } + int magic = 0; + for (int i = 0; i < peeklen; i++) + { + magic <<= 8; + magic += peek[i] & 0xFF; + } + return magic; } void unpack_200(FILE *input, FILE *output) { - unpacker u; - u.init(read_input_via_stdio); + unpacker u; + u.init(read_input_via_stdio); - // initialize jar output - // the output takes ownership of the file handle - jar jarout; - jarout.init(&u); - jarout.jarfp = output; + // initialize jar output + // the output takes ownership of the file handle + jar jarout; + jarout.init(&u); + jarout.jarfp = output; - // the input doesn't - u.infileptr = input; + // the input doesn't + u.infileptr = input; - // read the magic! - char peek[4]; - int magic; - magic = read_magic(&u, peek, (int)sizeof(peek)); + // read the magic! + char peek[4]; + int magic; + magic = read_magic(&u, peek, (int)sizeof(peek)); - // if it is a gzip encoded file, we need an extra gzip input filter - if ((magic & GZIP_MAGIC_MASK) == GZIP_MAGIC) - { - gunzip *gzin = NEW(gunzip, 1); - gzin->init(&u); - // FIXME: why the side effects? WHY? - u.gzin->start(magic); - u.start(); - } - else - { - // otherwise, feed the bytes to the unpacker directly - u.start(peek, sizeof(peek)); - } + // if it is a gzip encoded file, we need an extra gzip input filter + if ((magic & GZIP_MAGIC_MASK) == GZIP_MAGIC) + { + gunzip *gzin = NEW(gunzip, 1); + gzin->init(&u); + // FIXME: why the side effects? WHY? + u.gzin->start(magic); + u.start(); + } + else + { + // otherwise, feed the bytes to the unpacker directly + u.start(peek, sizeof(peek)); + } - // Note: The checks to u.aborting() are necessary to gracefully - // terminate processing when the first segment throws an error. - for (;;) - { - // Each trip through this loop unpacks one segment - // and then resets the unpacker. - for (unpacker::file *filep; (filep = u.get_next_file()) != nullptr;) - { - u.write_file_to_jar(filep); - } + // Note: The checks to u.aborting() are necessary to gracefully + // terminate processing when the first segment throws an error. + for (;;) + { + // Each trip through this loop unpacks one segment + // and then resets the unpacker. + for (unpacker::file *filep; (filep = u.get_next_file()) != nullptr;) + { + u.write_file_to_jar(filep); + } - // Peek ahead for more data. - magic = read_magic(&u, peek, (int)sizeof(peek)); - if (magic != (int)JAVA_PACKAGE_MAGIC) - { - // we do not feel strongly about this kind of thing... - /* - if (magic != EOF_MAGIC) - unpack_abort("garbage after end of pack archive"); - */ - break; // all done - } + // Peek ahead for more data. + magic = read_magic(&u, peek, (int)sizeof(peek)); + if (magic != (int)JAVA_PACKAGE_MAGIC) + { + // we do not feel strongly about this kind of thing... + /* + if (magic != EOF_MAGIC) + unpack_abort("garbage after end of pack archive"); + */ + break; // all done + } - // Release all storage from parsing the old segment. - u.reset(); - // Restart, beginning with the peek-ahead. - u.start(peek, sizeof(peek)); - } - u.finish(); - u.free(); // tidy up malloc blocks - fclose(input); + // Release all storage from parsing the old segment. + u.reset(); + // Restart, beginning with the peek-ahead. + u.start(peek, sizeof(peek)); + } + u.finish(); + u.free(); // tidy up malloc blocks + fclose(input); } diff --git a/libraries/pack200/src/utils.cpp b/libraries/pack200/src/utils.cpp index 0b7d91ca..fd6dad60 100644 --- a/libraries/pack200/src/utils.cpp +++ b/libraries/pack200/src/utils.cpp @@ -50,22 +50,22 @@ void *must_malloc(size_t size) { - size_t msize = size; - void *ptr = (msize > PSIZE_MAX) ? nullptr : malloc(msize); - if (ptr != nullptr) - { - memset(ptr, 0, size); - } - else - { - throw std::runtime_error(ERROR_ENOMEM); - } - return ptr; + size_t msize = size; + void *ptr = (msize > PSIZE_MAX) ? nullptr : malloc(msize); + if (ptr != nullptr) + { + memset(ptr, 0, size); + } + else + { + throw std::runtime_error(ERROR_ENOMEM); + } + return ptr; } void unpack_abort(const char *msg) { - if (msg == nullptr) - msg = "corrupt pack file or internal error"; - throw std::runtime_error(msg); + if (msg == nullptr) + msg = "corrupt pack file or internal error"; + throw std::runtime_error(msg); } diff --git a/libraries/pack200/src/utils.h b/libraries/pack200/src/utils.h index 5a3dc8f6..3bd2dae7 100644 --- a/libraries/pack200/src/utils.h +++ b/libraries/pack200/src/utils.h @@ -35,17 +35,17 @@ void *must_malloc(size_t size); inline size_t scale_size(size_t size, size_t scale) { - return (size > PSIZE_MAX / scale) ? OVERFLOW : size * scale; + return (size > PSIZE_MAX / scale) ? OVERFLOW : size * scale; } inline size_t add_size(size_t size1, size_t size2) { - return ((size1 | size2 | (size1 + size2)) > PSIZE_MAX) ? OVERFLOW : size1 + size2; + return ((size1 | size2 | (size1 + size2)) > PSIZE_MAX) ? OVERFLOW : size1 + size2; } inline size_t add_size(size_t size1, size_t size2, int size3) { - return add_size(add_size(size1, size2), size3); + return add_size(add_size(size1, size2), size3); } struct unpacker; diff --git a/libraries/pack200/src/zip.cpp b/libraries/pack200/src/zip.cpp index 32e8bd50..e776510b 100644 --- a/libraries/pack200/src/zip.cpp +++ b/libraries/pack200/src/zip.cpp @@ -52,7 +52,7 @@ inline uint32_t jar::get_crc32(uint32_t c, uchar *ptr, uint32_t len) { - return crc32(c, ptr, len); + return crc32(c, ptr, len); } // FIXME: this is bullshit. Do real endianness detection. @@ -68,175 +68,175 @@ inline uint32_t jar::get_crc32(uint32_t c, uchar *ptr, uint32_t len) void jar::init(unpacker *u_) { - BYTES_OF(*this).clear(); - u = u_; - u->jarout = this; + BYTES_OF(*this).clear(); + u = u_; + u->jarout = this; } // Write data to the ZIP output stream. void jar::write_data(void *buff, int len) { - while (len > 0) - { - int rc = (int)fwrite(buff, 1, len, jarfp); - if (rc <= 0) - { - fprintf(stderr, "Error: write on output file failed err=%d\n", errno); - exit(1); // Called only from the native standalone unpacker - } - output_file_offset += rc; - buff = ((char *)buff) + rc; - len -= rc; - } + while (len > 0) + { + int rc = (int)fwrite(buff, 1, len, jarfp); + if (rc <= 0) + { + fprintf(stderr, "Error: write on output file failed err=%d\n", errno); + exit(1); // Called only from the native standalone unpacker + } + output_file_offset += rc; + buff = ((char *)buff) + rc; + len -= rc; + } } void jar::add_to_jar_directory(const char *fname, bool store, int modtime, int len, int clen, - uint32_t crc) + uint32_t crc) { - uint32_t fname_length = (uint32_t)strlen(fname); - ushort header[23]; - if (modtime == 0) - modtime = default_modtime; - uint32_t dostime = get_dostime(modtime); - - header[0] = (ushort)SWAP_BYTES(0x4B50); - header[1] = (ushort)SWAP_BYTES(0x0201); - header[2] = (ushort)SWAP_BYTES(0xA); - - // required version - header[3] = (ushort)SWAP_BYTES(0xA); - - // flags 02 = maximum sub-compression flag - header[4] = (store) ? 0x0 : SWAP_BYTES(0x2); - - // Compression method 8=deflate. - header[5] = (store) ? 0x0 : SWAP_BYTES(0x08); - - // Last modified date and time. - header[6] = (ushort)GET_INT_LO(dostime); - header[7] = (ushort)GET_INT_HI(dostime); - - // CRC - header[8] = (ushort)GET_INT_LO(crc); - header[9] = (ushort)GET_INT_HI(crc); - - // Compressed length: - header[10] = (ushort)GET_INT_LO(clen); - header[11] = (ushort)GET_INT_HI(clen); - - // Uncompressed length. - header[12] = (ushort)GET_INT_LO(len); - header[13] = (ushort)GET_INT_HI(len); - - // Filename length - header[14] = (ushort)SWAP_BYTES(fname_length); - // So called "extra field" length. - header[15] = 0; - // So called "comment" length. - header[16] = 0; - // Disk number start - header[17] = 0; - // File flags => binary - header[18] = 0; - // More file flags - header[19] = 0; - header[20] = 0; - // Offset within ZIP file. - header[21] = (ushort)GET_INT_LO(output_file_offset); - header[22] = (ushort)GET_INT_HI(output_file_offset); - - // Copy the whole thing into the central directory. - central_directory.append(header, sizeof(header)); - - // Copy the fname to the header. - central_directory.append(fname, fname_length); - - central_directory_count++; + uint32_t fname_length = (uint32_t)strlen(fname); + ushort header[23]; + if (modtime == 0) + modtime = default_modtime; + uint32_t dostime = get_dostime(modtime); + + header[0] = (ushort)SWAP_BYTES(0x4B50); + header[1] = (ushort)SWAP_BYTES(0x0201); + header[2] = (ushort)SWAP_BYTES(0xA); + + // required version + header[3] = (ushort)SWAP_BYTES(0xA); + + // flags 02 = maximum sub-compression flag + header[4] = (store) ? 0x0 : SWAP_BYTES(0x2); + + // Compression method 8=deflate. + header[5] = (store) ? 0x0 : SWAP_BYTES(0x08); + + // Last modified date and time. + header[6] = (ushort)GET_INT_LO(dostime); + header[7] = (ushort)GET_INT_HI(dostime); + + // CRC + header[8] = (ushort)GET_INT_LO(crc); + header[9] = (ushort)GET_INT_HI(crc); + + // Compressed length: + header[10] = (ushort)GET_INT_LO(clen); + header[11] = (ushort)GET_INT_HI(clen); + + // Uncompressed length. + header[12] = (ushort)GET_INT_LO(len); + header[13] = (ushort)GET_INT_HI(len); + + // Filename length + header[14] = (ushort)SWAP_BYTES(fname_length); + // So called "extra field" length. + header[15] = 0; + // So called "comment" length. + header[16] = 0; + // Disk number start + header[17] = 0; + // File flags => binary + header[18] = 0; + // More file flags + header[19] = 0; + header[20] = 0; + // Offset within ZIP file. + header[21] = (ushort)GET_INT_LO(output_file_offset); + header[22] = (ushort)GET_INT_HI(output_file_offset); + + // Copy the whole thing into the central directory. + central_directory.append(header, sizeof(header)); + + // Copy the fname to the header. + central_directory.append(fname, fname_length); + + central_directory_count++; } void jar::write_jar_header(const char *fname, bool store, int modtime, int len, int clen, - uint32_t crc) + uint32_t crc) { - uint32_t fname_length = (uint32_t)strlen(fname); - ushort header[15]; - if (modtime == 0) - modtime = default_modtime; - uint32_t dostime = get_dostime(modtime); + uint32_t fname_length = (uint32_t)strlen(fname); + ushort header[15]; + if (modtime == 0) + modtime = default_modtime; + uint32_t dostime = get_dostime(modtime); - // ZIP LOC magic. - header[0] = (ushort)SWAP_BYTES(0x4B50); - header[1] = (ushort)SWAP_BYTES(0x0403); + // ZIP LOC magic. + header[0] = (ushort)SWAP_BYTES(0x4B50); + header[1] = (ushort)SWAP_BYTES(0x0403); - // Version - header[2] = (ushort)SWAP_BYTES(0xA); + // Version + header[2] = (ushort)SWAP_BYTES(0xA); - // flags 02 = maximum sub-compression flag - header[3] = (store) ? 0x0 : SWAP_BYTES(0x2); + // flags 02 = maximum sub-compression flag + header[3] = (store) ? 0x0 : SWAP_BYTES(0x2); - // Compression method = deflate - header[4] = (store) ? 0x0 : SWAP_BYTES(0x08); + // Compression method = deflate + header[4] = (store) ? 0x0 : SWAP_BYTES(0x08); - // Last modified date and time. - header[5] = (ushort)GET_INT_LO(dostime); - header[6] = (ushort)GET_INT_HI(dostime); + // Last modified date and time. + header[5] = (ushort)GET_INT_LO(dostime); + header[6] = (ushort)GET_INT_HI(dostime); - // CRC - header[7] = (ushort)GET_INT_LO(crc); - header[8] = (ushort)GET_INT_HI(crc); + // CRC + header[7] = (ushort)GET_INT_LO(crc); + header[8] = (ushort)GET_INT_HI(crc); - // Compressed length: - header[9] = (ushort)GET_INT_LO(clen); - header[10] = (ushort)GET_INT_HI(clen); + // Compressed length: + header[9] = (ushort)GET_INT_LO(clen); + header[10] = (ushort)GET_INT_HI(clen); - // Uncompressed length. - header[11] = (ushort)GET_INT_LO(len); - header[12] = (ushort)GET_INT_HI(len); + // Uncompressed length. + header[11] = (ushort)GET_INT_LO(len); + header[12] = (ushort)GET_INT_HI(len); - // Filename length - header[13] = (ushort)SWAP_BYTES(fname_length); - // So called "extra field" length. - header[14] = 0; + // Filename length + header[13] = (ushort)SWAP_BYTES(fname_length); + // So called "extra field" length. + header[14] = 0; - // Write the LOC header to the output file. - write_data(header, (int)sizeof(header)); + // Write the LOC header to the output file. + write_data(header, (int)sizeof(header)); - // Copy the fname to the header. - write_data((char *)fname, (int)fname_length); + // Copy the fname to the header. + write_data((char *)fname, (int)fname_length); } void jar::write_central_directory() { - bytes mc; - mc.set("PACK200"); - - ushort header[11]; - - // Create the End of Central Directory structure. - header[0] = (ushort)SWAP_BYTES(0x4B50); - header[1] = (ushort)SWAP_BYTES(0x0605); - // disk numbers - header[2] = 0; - header[3] = 0; - // Number of entries in central directory. - header[4] = (ushort)SWAP_BYTES(central_directory_count); - header[5] = (ushort)SWAP_BYTES(central_directory_count); - // Size of the central directory} - header[6] = (ushort)GET_INT_LO((int)central_directory.size()); - header[7] = (ushort)GET_INT_HI((int)central_directory.size()); - // Offset of central directory within disk. - header[8] = (ushort)GET_INT_LO(output_file_offset); - header[9] = (ushort)GET_INT_HI(output_file_offset); - // zipfile comment length; - header[10] = (ushort)SWAP_BYTES((int)mc.len); - - // Write the central directory. - write_data(central_directory.b); - - // Write the End of Central Directory structure. - write_data(header, (int)sizeof(header)); - - // Write the comment. - write_data(mc); + bytes mc; + mc.set("PACK200"); + + ushort header[11]; + + // Create the End of Central Directory structure. + header[0] = (ushort)SWAP_BYTES(0x4B50); + header[1] = (ushort)SWAP_BYTES(0x0605); + // disk numbers + header[2] = 0; + header[3] = 0; + // Number of entries in central directory. + header[4] = (ushort)SWAP_BYTES(central_directory_count); + header[5] = (ushort)SWAP_BYTES(central_directory_count); + // Size of the central directory} + header[6] = (ushort)GET_INT_LO((int)central_directory.size()); + header[7] = (ushort)GET_INT_HI((int)central_directory.size()); + // Offset of central directory within disk. + header[8] = (ushort)GET_INT_LO(output_file_offset); + header[9] = (ushort)GET_INT_HI(output_file_offset); + // zipfile comment length; + header[10] = (ushort)SWAP_BYTES((int)mc.len); + + // Write the central directory. + write_data(central_directory.b); + + // Write the End of Central Directory structure. + write_data(header, (int)sizeof(header)); + + // Write the comment. + write_data(mc); } // Public API @@ -244,74 +244,74 @@ void jar::write_central_directory() // Open a Jar file and initialize. void jar::openJarFile(const char *fname) { - if (!jarfp) - { - jarfp = fopen(fname, "wb"); - if (!jarfp) - { - fprintf(stderr, "Error: Could not open jar file: %s\n", fname); - exit(3); // Called only from the native standalone unpacker - } - } + if (!jarfp) + { + jarfp = fopen(fname, "wb"); + if (!jarfp) + { + fprintf(stderr, "Error: Could not open jar file: %s\n", fname); + exit(3); // Called only from the native standalone unpacker + } + } } // Add a ZIP entry and copy the file data void jar::addJarEntry(const char *fname, bool deflate_hint, int modtime, bytes &head, - bytes &tail) + bytes &tail) { - int len = (int)(head.len + tail.len); - int clen = 0; - - uint32_t crc = get_crc32(0, Z_NULL, 0); - if (head.len != 0) - crc = get_crc32(crc, (uchar *)head.ptr, (uint32_t)head.len); - if (tail.len != 0) - crc = get_crc32(crc, (uchar *)tail.ptr, (uint32_t)tail.len); - - bool deflate = (deflate_hint && len > 0); - - if (deflate) - { - if (deflate_bytes(head, tail) == false) - { - deflate = false; - } - } - clen = (int)((deflate) ? deflated.size() : len); - add_to_jar_directory(fname, !deflate, modtime, len, clen, crc); - write_jar_header(fname, !deflate, modtime, len, clen, crc); - - if (deflate) - { - write_data(deflated.b); - } - else - { - write_data(head); - write_data(tail); - } + int len = (int)(head.len + tail.len); + int clen = 0; + + uint32_t crc = get_crc32(0, Z_NULL, 0); + if (head.len != 0) + crc = get_crc32(crc, (uchar *)head.ptr, (uint32_t)head.len); + if (tail.len != 0) + crc = get_crc32(crc, (uchar *)tail.ptr, (uint32_t)tail.len); + + bool deflate = (deflate_hint && len > 0); + + if (deflate) + { + if (deflate_bytes(head, tail) == false) + { + deflate = false; + } + } + clen = (int)((deflate) ? deflated.size() : len); + add_to_jar_directory(fname, !deflate, modtime, len, clen, crc); + write_jar_header(fname, !deflate, modtime, len, clen, crc); + + if (deflate) + { + write_data(deflated.b); + } + else + { + write_data(head); + write_data(tail); + } } // Add a ZIP entry for a directory name no data void jar::addDirectoryToJarFile(const char *dir_name) { - bool store = true; - add_to_jar_directory((const char *)dir_name, store, default_modtime, 0, 0, 0); - write_jar_header((const char *)dir_name, store, default_modtime, 0, 0, 0); + bool store = true; + add_to_jar_directory((const char *)dir_name, store, default_modtime, 0, 0, 0); + write_jar_header((const char *)dir_name, store, default_modtime, 0, 0, 0); } // Write out the central directory and close the jar file. void jar::closeJarFile(bool central) { - if (jarfp) - { - fflush(jarfp); - if (central) - write_central_directory(); - fflush(jarfp); - fclose(jarfp); - } - reset(); + if (jarfp) + { + fflush(jarfp); + if (central) + write_central_directory(); + fflush(jarfp); + fclose(jarfp); + } + reset(); } /* Convert the date y/n/d and time h:m:s to a four byte DOS date and @@ -320,9 +320,9 @@ void jar::closeJarFile(bool central) */ inline uint32_t jar::dostime(int y, int n, int d, int h, int m, int s) { - return y < 1980 ? dostime(1980, 1, 1, 0, 0, 0) - : (((uint32_t)y - 1980) << 25) | ((uint32_t)n << 21) | ((uint32_t)d << 16) | - ((uint32_t)h << 11) | ((uint32_t)m << 5) | ((uint32_t)s >> 1); + return y < 1980 ? dostime(1980, 1, 1, 0, 0, 0) + : (((uint32_t)y - 1980) << 25) | ((uint32_t)n << 21) | ((uint32_t)d << 16) | + ((uint32_t)h << 11) | ((uint32_t)m << 5) | ((uint32_t)s >> 1); } /* #ifdef _REENTRANT // solaris @@ -336,20 +336,20 @@ extern "C" struct tm *gmtime_r(const time_t *, struct tm *); */ uint32_t jar::get_dostime(int modtime) { - // see defines.h - if (modtime != 0 && modtime == modtime_cache) - return dostime_cache; - if (modtime != 0 && default_modtime == 0) - default_modtime = modtime; // catch a reasonable default - time_t t = modtime; - struct tm sbuf; - (void)memset((void *)&sbuf, 0, sizeof(sbuf)); - struct tm *s = gmtime_r(&t, &sbuf); - modtime_cache = modtime; - dostime_cache = - dostime(s->tm_year + 1900, s->tm_mon + 1, s->tm_mday, s->tm_hour, s->tm_min, s->tm_sec); - // printf("modtime %d => %d\n", modtime_cache, dostime_cache); - return dostime_cache; + // see defines.h + if (modtime != 0 && modtime == modtime_cache) + return dostime_cache; + if (modtime != 0 && default_modtime == 0) + default_modtime = modtime; // catch a reasonable default + time_t t = modtime; + struct tm sbuf; + (void)memset((void *)&sbuf, 0, sizeof(sbuf)); + struct tm *s = gmtime_r(&t, &sbuf); + modtime_cache = modtime; + dostime_cache = + dostime(s->tm_year + 1900, s->tm_mon + 1, s->tm_mday, s->tm_hour, s->tm_min, s->tm_sec); + // printf("modtime %d => %d\n", modtime_cache, dostime_cache); + return dostime_cache; } /* Returns true on success, and will set the clen to the compressed @@ -358,232 +358,232 @@ uint32_t jar::get_dostime(int modtime) */ bool jar::deflate_bytes(bytes &head, bytes &tail) { - int len = (int)(head.len + tail.len); - - z_stream zs; - BYTES_OF(zs).clear(); - - // NOTE: the window size should always be -MAX_WBITS normally -15. - // unzip/zipup.c and java/Deflater.c - - int error = - deflateInit2(&zs, Z_BEST_COMPRESSION, Z_DEFLATED, -MAX_WBITS, 8, Z_DEFAULT_STRATEGY); - if (error != Z_OK) - { - /* - switch (error) - { - case Z_MEM_ERROR: - PRINTCR((2, "Error: deflate error : Out of memory \n")); - break; - case Z_STREAM_ERROR: - PRINTCR((2, "Error: deflate error : Invalid compression level \n")); - break; - case Z_VERSION_ERROR: - PRINTCR((2, "Error: deflate error : Invalid version\n")); - break; - default: - PRINTCR((2, "Error: Internal deflate error error = %d\n", error)); - } - */ - return false; - } - - deflated.empty(); - zs.next_out = (uchar *)deflated.grow(len + (len / 2)); - zs.avail_out = (int)deflated.size(); - - zs.next_in = (uchar *)head.ptr; - zs.avail_in = (int)head.len; - - bytes *first = &head; - bytes *last = &tail; - if (last->len == 0) - { - first = nullptr; - last = &head; - } - else if (first->len == 0) - { - first = nullptr; - } - - if (first != nullptr && error == Z_OK) - { - zs.next_in = (uchar *)first->ptr; - zs.avail_in = (int)first->len; - error = deflate(&zs, Z_NO_FLUSH); - } - if (error == Z_OK) - { - zs.next_in = (uchar *)last->ptr; - zs.avail_in = (int)last->len; - error = deflate(&zs, Z_FINISH); - } - if (error == Z_STREAM_END) - { - if (len > (int)zs.total_out) - { - deflated.b.len = zs.total_out; - deflateEnd(&zs); - return true; - } - deflateEnd(&zs); - return false; - } - - deflateEnd(&zs); - return false; + int len = (int)(head.len + tail.len); + + z_stream zs; + BYTES_OF(zs).clear(); + + // NOTE: the window size should always be -MAX_WBITS normally -15. + // unzip/zipup.c and java/Deflater.c + + int error = + deflateInit2(&zs, Z_BEST_COMPRESSION, Z_DEFLATED, -MAX_WBITS, 8, Z_DEFAULT_STRATEGY); + if (error != Z_OK) + { + /* + switch (error) + { + case Z_MEM_ERROR: + PRINTCR((2, "Error: deflate error : Out of memory \n")); + break; + case Z_STREAM_ERROR: + PRINTCR((2, "Error: deflate error : Invalid compression level \n")); + break; + case Z_VERSION_ERROR: + PRINTCR((2, "Error: deflate error : Invalid version\n")); + break; + default: + PRINTCR((2, "Error: Internal deflate error error = %d\n", error)); + } + */ + return false; + } + + deflated.empty(); + zs.next_out = (uchar *)deflated.grow(len + (len / 2)); + zs.avail_out = (int)deflated.size(); + + zs.next_in = (uchar *)head.ptr; + zs.avail_in = (int)head.len; + + bytes *first = &head; + bytes *last = &tail; + if (last->len == 0) + { + first = nullptr; + last = &head; + } + else if (first->len == 0) + { + first = nullptr; + } + + if (first != nullptr && error == Z_OK) + { + zs.next_in = (uchar *)first->ptr; + zs.avail_in = (int)first->len; + error = deflate(&zs, Z_NO_FLUSH); + } + if (error == Z_OK) + { + zs.next_in = (uchar *)last->ptr; + zs.avail_in = (int)last->len; + error = deflate(&zs, Z_FINISH); + } + if (error == Z_STREAM_END) + { + if (len > (int)zs.total_out) + { + deflated.b.len = zs.total_out; + deflateEnd(&zs); + return true; + } + deflateEnd(&zs); + return false; + } + + deflateEnd(&zs); + return false; } // Callback for fetching data from a GZIP input stream static int64_t read_input_via_gzip(unpacker *u, void *buf, int64_t minlen, int64_t maxlen) { - assert(minlen <= maxlen); // don't talk nonsense - int64_t numread = 0; - char *bufptr = (char *)buf; - char *inbuf = u->gzin->inbuf; - size_t inbuflen = sizeof(u->gzin->inbuf); - unpacker::read_input_fn_t read_gzin_fn = (unpacker::read_input_fn_t)u->gzin->read_input_fn; - z_stream &zs = *(z_stream *)u->gzin->zstream; - while (numread < minlen) - { - int readlen = (1 << 16); // pretty arbitrary - if (readlen > (maxlen - numread)) - readlen = (int)(maxlen - numread); - zs.next_out = (uchar *)bufptr; - zs.avail_out = readlen; - if (zs.avail_in == 0) - { - zs.avail_in = (int)read_gzin_fn(u, inbuf, 1, inbuflen); - zs.next_in = (uchar *)inbuf; - } - int error = inflate(&zs, Z_NO_FLUSH); - if (error != Z_OK && error != Z_STREAM_END) - { - unpack_abort("error inflating input"); - break; - } - int nr = readlen - zs.avail_out; - numread += nr; - bufptr += nr; - assert(numread <= maxlen); - if (error == Z_STREAM_END) - { - enum - { - TRAILER_LEN = 8 - }; - // skip 8-byte trailer - if (zs.avail_in >= TRAILER_LEN) - { - zs.avail_in -= TRAILER_LEN; - } - else - { - // Bug: 5023768,we read past the TRAILER_LEN to see if there is - // any extraneous data, as we dont support concatenated .gz - // files just yet. - int extra = (int)read_gzin_fn(u, inbuf, 1, inbuflen); - zs.avail_in += extra - TRAILER_LEN; - } - // %%% should check final CRC and length here - // %%% should check for concatenated *.gz files here - if (zs.avail_in > 0) - unpack_abort("garbage after end of deflated input stream"); - // pop this filter off: - u->gzin->free(); - break; - } - } - - // fprintf(u->errstrm, "readInputFn(%d,%d) => %d (gunzip)\n", - // (int)minlen, (int)maxlen, (int)numread); - return numread; + assert(minlen <= maxlen); // don't talk nonsense + int64_t numread = 0; + char *bufptr = (char *)buf; + char *inbuf = u->gzin->inbuf; + size_t inbuflen = sizeof(u->gzin->inbuf); + read_input_fn_t read_gzin_fn = u->gzin->read_input_fn; + z_stream &zs = *(z_stream *)u->gzin->zstream; + while (numread < minlen) + { + int readlen = (1 << 16); // pretty arbitrary + if (readlen > (maxlen - numread)) + readlen = (int)(maxlen - numread); + zs.next_out = (uchar *)bufptr; + zs.avail_out = readlen; + if (zs.avail_in == 0) + { + zs.avail_in = (int)read_gzin_fn(u, inbuf, 1, inbuflen); + zs.next_in = (uchar *)inbuf; + } + int error = inflate(&zs, Z_NO_FLUSH); + if (error != Z_OK && error != Z_STREAM_END) + { + unpack_abort("error inflating input"); + break; + } + int nr = readlen - zs.avail_out; + numread += nr; + bufptr += nr; + assert(numread <= maxlen); + if (error == Z_STREAM_END) + { + enum + { + TRAILER_LEN = 8 + }; + // skip 8-byte trailer + if (zs.avail_in >= TRAILER_LEN) + { + zs.avail_in -= TRAILER_LEN; + } + else + { + // Bug: 5023768,we read past the TRAILER_LEN to see if there is + // any extraneous data, as we dont support concatenated .gz + // files just yet. + int extra = (int)read_gzin_fn(u, inbuf, 1, inbuflen); + zs.avail_in += extra - TRAILER_LEN; + } + // %%% should check final CRC and length here + // %%% should check for concatenated *.gz files here + if (zs.avail_in > 0) + unpack_abort("garbage after end of deflated input stream"); + // pop this filter off: + u->gzin->free(); + break; + } + } + + // fprintf(u->errstrm, "readInputFn(%d,%d) => %d (gunzip)\n", + // (int)minlen, (int)maxlen, (int)numread); + return numread; } void gunzip::init(unpacker *u_) { - BYTES_OF(*this).clear(); - u = u_; - assert(u->gzin == nullptr); // once only, please - read_input_fn = (void *)u->read_input_fn; - zstream = NEW(z_stream, 1); - u->gzin = this; - u->read_input_fn = read_input_via_gzip; + BYTES_OF(*this).clear(); + u = u_; + assert(u->gzin == nullptr); // once only, please + read_input_fn = u->read_input_fn; + zstream = NEW(z_stream, 1); + u->gzin = this; + u->read_input_fn = read_input_via_gzip; } void gunzip::start(int magic) { - assert((magic & GZIP_MAGIC_MASK) == GZIP_MAGIC); - int gz_flg = (magic & 0xFF); // keep "flg", discard other 3 bytes - enum - { - FHCRC = (1 << 1), - FEXTRA = (1 << 2), - FNAME = (1 << 3), - FCOMMENT = (1 << 4) - }; - char gz_mtime[4]; - char gz_xfl[1]; - char gz_os[1]; - char gz_extra_len[2]; - char gz_hcrc[2]; - char gz_ignore; - // do not save extra, name, comment - read_fixed_field(gz_mtime, sizeof(gz_mtime)); - read_fixed_field(gz_xfl, sizeof(gz_xfl)); - read_fixed_field(gz_os, sizeof(gz_os)); - if (gz_flg & FEXTRA) - { - read_fixed_field(gz_extra_len, sizeof(gz_extra_len)); - int extra_len = gz_extra_len[0] & 0xFF; - extra_len += (gz_extra_len[1] & 0xFF) << 8; - for (; extra_len > 0; extra_len--) - { - read_fixed_field(&gz_ignore, 1); - } - } - int null_terms = 0; - if (gz_flg & FNAME) - null_terms++; - if (gz_flg & FCOMMENT) - null_terms++; - for (; null_terms; null_terms--) - { - for (;;) - { - gz_ignore = 0; - read_fixed_field(&gz_ignore, 1); - if (gz_ignore == 0) - break; - } - } - if (gz_flg & FHCRC) - read_fixed_field(gz_hcrc, sizeof(gz_hcrc)); - - // now the input stream is ready to read into the inflater - int error = inflateInit2((z_stream *)zstream, -MAX_WBITS); - if (error != Z_OK) - { - unpack_abort("cannot create input"); - } + assert((magic & GZIP_MAGIC_MASK) == GZIP_MAGIC); + int gz_flg = (magic & 0xFF); // keep "flg", discard other 3 bytes + enum + { + FHCRC = (1 << 1), + FEXTRA = (1 << 2), + FNAME = (1 << 3), + FCOMMENT = (1 << 4) + }; + char gz_mtime[4]; + char gz_xfl[1]; + char gz_os[1]; + char gz_extra_len[2]; + char gz_hcrc[2]; + char gz_ignore; + // do not save extra, name, comment + read_fixed_field(gz_mtime, sizeof(gz_mtime)); + read_fixed_field(gz_xfl, sizeof(gz_xfl)); + read_fixed_field(gz_os, sizeof(gz_os)); + if (gz_flg & FEXTRA) + { + read_fixed_field(gz_extra_len, sizeof(gz_extra_len)); + int extra_len = gz_extra_len[0] & 0xFF; + extra_len += (gz_extra_len[1] & 0xFF) << 8; + for (; extra_len > 0; extra_len--) + { + read_fixed_field(&gz_ignore, 1); + } + } + int null_terms = 0; + if (gz_flg & FNAME) + null_terms++; + if (gz_flg & FCOMMENT) + null_terms++; + for (; null_terms; null_terms--) + { + for (;;) + { + gz_ignore = 0; + read_fixed_field(&gz_ignore, 1); + if (gz_ignore == 0) + break; + } + } + if (gz_flg & FHCRC) + read_fixed_field(gz_hcrc, sizeof(gz_hcrc)); + + // now the input stream is ready to read into the inflater + int error = inflateInit2((z_stream *)zstream, -MAX_WBITS); + if (error != Z_OK) + { + unpack_abort("cannot create input"); + } } void gunzip::free() { - assert(u->gzin == this); - u->gzin = nullptr; - u->read_input_fn = (unpacker::read_input_fn_t) this->read_input_fn; - inflateEnd((z_stream *)zstream); - ::free(zstream); - zstream = nullptr; - ::free(this); + assert(u->gzin == this); + u->gzin = nullptr; + u->read_input_fn = this->read_input_fn; + inflateEnd((z_stream *)zstream); + ::free(zstream); + zstream = nullptr; + ::free(this); } void gunzip::read_fixed_field(char *buf, size_t buflen) { - int64_t nr = ((unpacker::read_input_fn_t)read_input_fn)(u, buf, buflen, buflen); - if ((size_t)nr != buflen) - unpack_abort("short stream header"); + int64_t nr = read_input_fn(u, buf, buflen, buflen); + if ((size_t)nr != buflen) + unpack_abort("short stream header"); } diff --git a/libraries/pack200/src/zip.h b/libraries/pack200/src/zip.h index 67ec24da..ff973a86 100644 --- a/libraries/pack200/src/zip.h +++ b/libraries/pack200/src/zip.h @@ -22,89 +22,91 @@ * or visit www.oracle.com if you need additional information or have any * questions. */ +#pragma once + #include <stdint.h> typedef unsigned short ushort; typedef unsigned int uint32_t; typedef unsigned char uchar; -struct unpacker; +#include "unpack.h" struct jar { - // JAR file writer - FILE *jarfp; - int default_modtime; - - // Used by unix2dostime: - int modtime_cache; - uint32_t dostime_cache; - - // Private members - fillbytes central_directory; - ushort central_directory_count; - uint32_t output_file_offset; - fillbytes deflated; // temporary buffer - - // pointer to outer unpacker, for error checks etc. - unpacker *u; - - // Public Methods - void openJarFile(const char *fname); - void addJarEntry(const char *fname, bool deflate_hint, int modtime, bytes &head, - bytes &tail); - void addDirectoryToJarFile(const char *dir_name); - void closeJarFile(bool central); - - void init(unpacker *u_); - - void free() - { - central_directory.free(); - deflated.free(); - } - - void reset() - { - free(); - init(u); - } - - // Private Methods - void write_data(void *ptr, int len); - void write_data(bytes &b) - { - write_data(b.ptr, (int)b.len); - } - void add_to_jar_directory(const char *fname, bool store, int modtime, int len, int clen, - uint32_t crc); - void write_jar_header(const char *fname, bool store, int modtime, int len, int clen, - unsigned int crc); - void write_central_directory(); - uint32_t dostime(int y, int n, int d, int h, int m, int s); - uint32_t get_dostime(int modtime); - - // The definitions of these depend on the NO_ZLIB option: - bool deflate_bytes(bytes &head, bytes &tail); - static uint32_t get_crc32(uint32_t c, unsigned char *ptr, uint32_t len); + // JAR file writer + FILE *jarfp; + int default_modtime; + + // Used by unix2dostime: + int modtime_cache; + uint32_t dostime_cache; + + // Private members + fillbytes central_directory; + ushort central_directory_count; + uint32_t output_file_offset; + fillbytes deflated; // temporary buffer + + // pointer to outer unpacker, for error checks etc. + unpacker *u; + + // Public Methods + void openJarFile(const char *fname); + void addJarEntry(const char *fname, bool deflate_hint, int modtime, bytes &head, + bytes &tail); + void addDirectoryToJarFile(const char *dir_name); + void closeJarFile(bool central); + + void init(unpacker *u_); + + void free() + { + central_directory.free(); + deflated.free(); + } + + void reset() + { + free(); + init(u); + } + + // Private Methods + void write_data(void *ptr, int len); + void write_data(bytes &b) + { + write_data(b.ptr, (int)b.len); + } + void add_to_jar_directory(const char *fname, bool store, int modtime, int len, int clen, + uint32_t crc); + void write_jar_header(const char *fname, bool store, int modtime, int len, int clen, + unsigned int crc); + void write_central_directory(); + uint32_t dostime(int y, int n, int d, int h, int m, int s); + uint32_t get_dostime(int modtime); + + // The definitions of these depend on the NO_ZLIB option: + bool deflate_bytes(bytes &head, bytes &tail); + static uint32_t get_crc32(uint32_t c, unsigned char *ptr, uint32_t len); }; struct gunzip { - // optional gzip input stream control block + // optional gzip input stream control block - // pointer to outer unpacker, for error checks etc. - unpacker *u; + // pointer to outer unpacker, for error checks etc. + unpacker *u; - void *read_input_fn; // underlying \bchar\b stream - void *zstream; // inflater state - char inbuf[1 << 14]; // input buffer + read_input_fn_t read_input_fn; // underlying \bchar\b stream + void *zstream; // inflater state + char inbuf[1 << 14]; // input buffer - void init(unpacker *u_); // pushes new value on u->read_input_fn + void init(unpacker *u_); // pushes new value on u->read_input_fn - void free(); + void free(); - void start(int magic); + void start(int magic); - // private stuff - void read_fixed_field(char *buf, size_t buflen); + // private stuff + void read_fixed_field(char *buf, size_t buflen); }; diff --git a/libraries/quazip b/libraries/quazip -Subproject 683e2ec8ada758d6e48d31ec606840802e6941b +Subproject 3691d57d3af13f49b2be2b62accddefee3c26b9 diff --git a/libraries/rainbow/CMakeLists.txt b/libraries/rainbow/CMakeLists.txt index bc561800..ad806faa 100644 --- a/libraries/rainbow/CMakeLists.txt +++ b/libraries/rainbow/CMakeLists.txt @@ -12,11 +12,11 @@ add_definitions(-DRAINBOW_LIBRARY) add_library(MultiMC_rainbow SHARED ${RAINBOW_SOURCES}) target_include_directories(MultiMC_rainbow PUBLIC "${CMAKE_CURRENT_SOURCE_DIR}/include") -qt5_use_modules(MultiMC_rainbow Core Gui) +target_link_libraries(MultiMC_rainbow Qt5::Core Qt5::Gui) # Install it install( - TARGETS MultiMC_rainbow - RUNTIME DESTINATION ${LIBRARY_DEST_DIR} - LIBRARY DESTINATION ${LIBRARY_DEST_DIR} + TARGETS MultiMC_rainbow + RUNTIME DESTINATION ${LIBRARY_DEST_DIR} + LIBRARY DESTINATION ${LIBRARY_DEST_DIR} ) diff --git a/libraries/rainbow/include/rainbow.h b/libraries/rainbow/include/rainbow.h index b12052b1..67c46300 100644 --- a/libraries/rainbow/include/rainbow.h +++ b/libraries/rainbow/include/rainbow.h @@ -50,7 +50,7 @@ RAINBOW_EXPORT qreal luma(const QColor &); * @since 5.0 */ RAINBOW_EXPORT void getHcy(const QColor &, qreal *hue, qreal *chroma, qreal *luma, - qreal *alpha = 0); + qreal *alpha = 0); /** * Calculate the contrast ratio between two colors, according to the @@ -132,7 +132,7 @@ RAINBOW_EXPORT QColor tint(const QColor &base, const QColor &color, qreal amount /** * Blend two colors into a new color by linear combination. * @code - QColor lighter = Rainbow::mix(myColor, Qt::white) + QColor lighter = Rainbow::mix(myColor, Qt::white) * @endcode * @param c1 first color. * @param c2 second color. @@ -146,9 +146,9 @@ RAINBOW_EXPORT QColor mix(const QColor &c1, const QColor &c2, qreal bias = 0.5); * Blend two colors into a new color by painting the second color over the * first using the specified composition mode. * @code - QColor white(Qt::white); - white.setAlphaF(0.5); - QColor lighter = Rainbow::overlayColors(myColor, white); + QColor white(Qt::white); + white.setAlphaF(0.5); + QColor lighter = Rainbow::overlayColors(myColor, white); @endcode * @param base the base color (alpha channel is ignored). * @param paint the color to be overlayed onto the base color. @@ -156,5 +156,5 @@ RAINBOW_EXPORT QColor mix(const QColor &c1, const QColor &c2, qreal bias = 0.5); */ RAINBOW_EXPORT QColor overlayColors(const QColor &base, const QColor &paint, - QPainter::CompositionMode comp = QPainter::CompositionMode_SourceOver); + QPainter::CompositionMode comp = QPainter::CompositionMode_SourceOver); } diff --git a/libraries/rainbow/include/rainbow_config.h b/libraries/rainbow/include/rainbow_config.h index 9290dc8a..6e467bc7 100644 --- a/libraries/rainbow/include/rainbow_config.h +++ b/libraries/rainbow/include/rainbow_config.h @@ -1,4 +1,4 @@ -/* Copyright 2013-2018 MultiMC Contributors +/* Copyright 2013-2019 MultiMC Contributors * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. @@ -16,11 +16,11 @@ #include <QtCore/QtGlobal> #ifdef RAINBOW_STATIC - #define RAINBOW_EXPORT + #define RAINBOW_EXPORT #else - #ifdef RAINBOW_LIBRARY - #define RAINBOW_EXPORT Q_DECL_EXPORT - #else - #define RAINBOW_EXPORT Q_DECL_IMPORT - #endif + #ifdef RAINBOW_LIBRARY + #define RAINBOW_EXPORT Q_DECL_EXPORT + #else + #define RAINBOW_EXPORT Q_DECL_IMPORT + #endif #endif
\ No newline at end of file diff --git a/libraries/rainbow/src/rainbow.cpp b/libraries/rainbow/src/rainbow.cpp index 8502fcd0..9054d71d 100644 --- a/libraries/rainbow/src/rainbow.cpp +++ b/libraries/rainbow/src/rainbow.cpp @@ -33,8 +33,8 @@ static inline qreal wrap(qreal a, qreal d = 1.0) { - qreal r = fmod(a, d); - return (r < 0.0 ? d + r : (r > 0.0 ? r : 0.0)); + qreal r = fmod(a, d); + return (r < 0.0 ? d + r : (r > 0.0 ? r : 0.0)); } // normalize: like qBound(a, 0.0, 1.0) but without needing the args and with @@ -60,306 +60,306 @@ static const qreal yc[3] = {0.34375, 0.5, 0.15625}; class KHCY { public: - explicit KHCY(const QColor &color) - { - qreal r = gamma(color.redF()); - qreal g = gamma(color.greenF()); - qreal b = gamma(color.blueF()); - a = color.alphaF(); - - // luma component - y = lumag(r, g, b); - - // hue component - qreal p = qMax(qMax(r, g), b); - qreal n = qMin(qMin(r, g), b); - qreal d = 6.0 * (p - n); - if (n == p) - { - h = 0.0; - } - else if (r == p) - { - h = ((g - b) / d); - } - else if (g == p) - { - h = ((b - r) / d) + (1.0 / 3.0); - } - else - { - h = ((r - g) / d) + (2.0 / 3.0); - } - - // chroma component - if (r == g && g == b) - { - c = 0.0; - } - else - { - c = qMax((y - n) / y, (p - y) / (1 - y)); - } - } - explicit KHCY(qreal h_, qreal c_, qreal y_, qreal a_ = 1.0) - { - h = h_; - c = c_; - y = y_; - a = a_; - } - - QColor qColor() const - { - // start with sane component values - qreal _h = wrap(h); - qreal _c = normalize(c); - qreal _y = normalize(y); - - // calculate some needed variables - qreal _hs = _h * 6.0, th, tm; - if (_hs < 1.0) - { - th = _hs; - tm = yc[0] + yc[1] * th; - } - else if (_hs < 2.0) - { - th = 2.0 - _hs; - tm = yc[1] + yc[0] * th; - } - else if (_hs < 3.0) - { - th = _hs - 2.0; - tm = yc[1] + yc[2] * th; - } - else if (_hs < 4.0) - { - th = 4.0 - _hs; - tm = yc[2] + yc[1] * th; - } - else if (_hs < 5.0) - { - th = _hs - 4.0; - tm = yc[2] + yc[0] * th; - } - else - { - th = 6.0 - _hs; - tm = yc[0] + yc[2] * th; - } - - // calculate RGB channels in sorted order - qreal tn, to, tp; - if (tm >= _y) - { - tp = _y + _y * _c * (1.0 - tm) / tm; - to = _y + _y * _c * (th - tm) / tm; - tn = _y - (_y * _c); - } - else - { - tp = _y + (1.0 - _y) * _c; - to = _y + (1.0 - _y) * _c * (th - tm) / (1.0 - tm); - tn = _y - (1.0 - _y) * _c * tm / (1.0 - tm); - } - - // return RGB channels in appropriate order - if (_hs < 1.0) - { - return QColor::fromRgbF(igamma(tp), igamma(to), igamma(tn), a); - } - else if (_hs < 2.0) - { - return QColor::fromRgbF(igamma(to), igamma(tp), igamma(tn), a); - } - else if (_hs < 3.0) - { - return QColor::fromRgbF(igamma(tn), igamma(tp), igamma(to), a); - } - else if (_hs < 4.0) - { - return QColor::fromRgbF(igamma(tn), igamma(to), igamma(tp), a); - } - else if (_hs < 5.0) - { - return QColor::fromRgbF(igamma(to), igamma(tn), igamma(tp), a); - } - else - { - return QColor::fromRgbF(igamma(tp), igamma(tn), igamma(to), a); - } - } - - qreal h, c, y, a; - static qreal luma(const QColor &color) - { - return lumag(gamma(color.redF()), gamma(color.greenF()), gamma(color.blueF())); - } + explicit KHCY(const QColor &color) + { + qreal r = gamma(color.redF()); + qreal g = gamma(color.greenF()); + qreal b = gamma(color.blueF()); + a = color.alphaF(); + + // luma component + y = lumag(r, g, b); + + // hue component + qreal p = qMax(qMax(r, g), b); + qreal n = qMin(qMin(r, g), b); + qreal d = 6.0 * (p - n); + if (n == p) + { + h = 0.0; + } + else if (r == p) + { + h = ((g - b) / d); + } + else if (g == p) + { + h = ((b - r) / d) + (1.0 / 3.0); + } + else + { + h = ((r - g) / d) + (2.0 / 3.0); + } + + // chroma component + if (r == g && g == b) + { + c = 0.0; + } + else + { + c = qMax((y - n) / y, (p - y) / (1 - y)); + } + } + explicit KHCY(qreal h_, qreal c_, qreal y_, qreal a_ = 1.0) + { + h = h_; + c = c_; + y = y_; + a = a_; + } + + QColor qColor() const + { + // start with sane component values + qreal _h = wrap(h); + qreal _c = normalize(c); + qreal _y = normalize(y); + + // calculate some needed variables + qreal _hs = _h * 6.0, th, tm; + if (_hs < 1.0) + { + th = _hs; + tm = yc[0] + yc[1] * th; + } + else if (_hs < 2.0) + { + th = 2.0 - _hs; + tm = yc[1] + yc[0] * th; + } + else if (_hs < 3.0) + { + th = _hs - 2.0; + tm = yc[1] + yc[2] * th; + } + else if (_hs < 4.0) + { + th = 4.0 - _hs; + tm = yc[2] + yc[1] * th; + } + else if (_hs < 5.0) + { + th = _hs - 4.0; + tm = yc[2] + yc[0] * th; + } + else + { + th = 6.0 - _hs; + tm = yc[0] + yc[2] * th; + } + + // calculate RGB channels in sorted order + qreal tn, to, tp; + if (tm >= _y) + { + tp = _y + _y * _c * (1.0 - tm) / tm; + to = _y + _y * _c * (th - tm) / tm; + tn = _y - (_y * _c); + } + else + { + tp = _y + (1.0 - _y) * _c; + to = _y + (1.0 - _y) * _c * (th - tm) / (1.0 - tm); + tn = _y - (1.0 - _y) * _c * tm / (1.0 - tm); + } + + // return RGB channels in appropriate order + if (_hs < 1.0) + { + return QColor::fromRgbF(igamma(tp), igamma(to), igamma(tn), a); + } + else if (_hs < 2.0) + { + return QColor::fromRgbF(igamma(to), igamma(tp), igamma(tn), a); + } + else if (_hs < 3.0) + { + return QColor::fromRgbF(igamma(tn), igamma(tp), igamma(to), a); + } + else if (_hs < 4.0) + { + return QColor::fromRgbF(igamma(tn), igamma(to), igamma(tp), a); + } + else if (_hs < 5.0) + { + return QColor::fromRgbF(igamma(to), igamma(tn), igamma(tp), a); + } + else + { + return QColor::fromRgbF(igamma(tp), igamma(tn), igamma(to), a); + } + } + + qreal h, c, y, a; + static qreal luma(const QColor &color) + { + return lumag(gamma(color.redF()), gamma(color.greenF()), gamma(color.blueF())); + } private: - static qreal gamma(qreal n) - { - return pow(normalize(n), 2.2); - } - static qreal igamma(qreal n) - { - return pow(normalize(n), 1.0 / 2.2); - } - static qreal lumag(qreal r, qreal g, qreal b) - { - return r * yc[0] + g * yc[1] + b * yc[2]; - } + static qreal gamma(qreal n) + { + return pow(normalize(n), 2.2); + } + static qreal igamma(qreal n) + { + return pow(normalize(n), 1.0 / 2.2); + } + static qreal lumag(qreal r, qreal g, qreal b) + { + return r * yc[0] + g * yc[1] + b * yc[2]; + } }; static inline qreal mixQreal(qreal a, qreal b, qreal bias) { - return a + (b - a) * bias; + return a + (b - a) * bias; } //END internal helper functions qreal Rainbow::luma(const QColor &color) { - return KHCY::luma(color); + return KHCY::luma(color); } void Rainbow::getHcy(const QColor &color, qreal *h, qreal *c, qreal *y, qreal *a) { - if (!c || !h || !y) - { - return; - } - KHCY khcy(color); - *c = khcy.c; - *h = khcy.h; - *y = khcy.y; - if (a) - { - *a = khcy.a; - } + if (!c || !h || !y) + { + return; + } + KHCY khcy(color); + *c = khcy.c; + *h = khcy.h; + *y = khcy.y; + if (a) + { + *a = khcy.a; + } } static qreal contrastRatioForLuma(qreal y1, qreal y2) { - if (y1 > y2) - { - return (y1 + 0.05) / (y2 + 0.05); - } - else - { - return (y2 + 0.05) / (y1 + 0.05); - } + if (y1 > y2) + { + return (y1 + 0.05) / (y2 + 0.05); + } + else + { + return (y2 + 0.05) / (y1 + 0.05); + } } qreal Rainbow::contrastRatio(const QColor &c1, const QColor &c2) { - return contrastRatioForLuma(luma(c1), luma(c2)); + return contrastRatioForLuma(luma(c1), luma(c2)); } QColor Rainbow::lighten(const QColor &color, qreal ky, qreal kc) { - KHCY c(color); - c.y = 1.0 - normalize((1.0 - c.y) * (1.0 - ky)); - c.c = 1.0 - normalize((1.0 - c.c) * kc); - return c.qColor(); + KHCY c(color); + c.y = 1.0 - normalize((1.0 - c.y) * (1.0 - ky)); + c.c = 1.0 - normalize((1.0 - c.c) * kc); + return c.qColor(); } QColor Rainbow::darken(const QColor &color, qreal ky, qreal kc) { - KHCY c(color); - c.y = normalize(c.y * (1.0 - ky)); - c.c = normalize(c.c * kc); - return c.qColor(); + KHCY c(color); + c.y = normalize(c.y * (1.0 - ky)); + c.c = normalize(c.c * kc); + return c.qColor(); } QColor Rainbow::shade(const QColor &color, qreal ky, qreal kc) { - KHCY c(color); - c.y = normalize(c.y + ky); - c.c = normalize(c.c + kc); - return c.qColor(); + KHCY c(color); + c.y = normalize(c.y + ky); + c.c = normalize(c.c + kc); + return c.qColor(); } static QColor tintHelper(const QColor &base, qreal baseLuma, const QColor &color, qreal amount) { - KHCY result(Rainbow::mix(base, color, pow(amount, 0.3))); - result.y = mixQreal(baseLuma, result.y, amount); + KHCY result(Rainbow::mix(base, color, pow(amount, 0.3))); + result.y = mixQreal(baseLuma, result.y, amount); - return result.qColor(); + return result.qColor(); } QColor Rainbow::tint(const QColor &base, const QColor &color, qreal amount) { - if (amount <= 0.0) - { - return base; - } - if (amount >= 1.0) - { - return color; - } - if (qIsNaN(amount)) - { - return base; - } - - qreal baseLuma = luma(base); // cache value because luma call is expensive - double ri = contrastRatioForLuma(baseLuma, luma(color)); - double rg = 1.0 + ((ri + 1.0) * amount * amount * amount); - double u = 1.0, l = 0.0; - QColor result; - for (int i = 12; i; --i) - { - double a = 0.5 * (l + u); - result = tintHelper(base, baseLuma, color, a); - double ra = contrastRatioForLuma(baseLuma, luma(result)); - if (ra > rg) - { - u = a; - } - else - { - l = a; - } - } - return result; + if (amount <= 0.0) + { + return base; + } + if (amount >= 1.0) + { + return color; + } + if (qIsNaN(amount)) + { + return base; + } + + qreal baseLuma = luma(base); // cache value because luma call is expensive + double ri = contrastRatioForLuma(baseLuma, luma(color)); + double rg = 1.0 + ((ri + 1.0) * amount * amount * amount); + double u = 1.0, l = 0.0; + QColor result; + for (int i = 12; i; --i) + { + double a = 0.5 * (l + u); + result = tintHelper(base, baseLuma, color, a); + double ra = contrastRatioForLuma(baseLuma, luma(result)); + if (ra > rg) + { + u = a; + } + else + { + l = a; + } + } + return result; } QColor Rainbow::mix(const QColor &c1, const QColor &c2, qreal bias) { - if (bias <= 0.0) - { - return c1; - } - if (bias >= 1.0) - { - return c2; - } - if (qIsNaN(bias)) - { - return c1; - } - - qreal r = mixQreal(c1.redF(), c2.redF(), bias); - qreal g = mixQreal(c1.greenF(), c2.greenF(), bias); - qreal b = mixQreal(c1.blueF(), c2.blueF(), bias); - qreal a = mixQreal(c1.alphaF(), c2.alphaF(), bias); - - return QColor::fromRgbF(r, g, b, a); + if (bias <= 0.0) + { + return c1; + } + if (bias >= 1.0) + { + return c2; + } + if (qIsNaN(bias)) + { + return c1; + } + + qreal r = mixQreal(c1.redF(), c2.redF(), bias); + qreal g = mixQreal(c1.greenF(), c2.greenF(), bias); + qreal b = mixQreal(c1.blueF(), c2.blueF(), bias); + qreal a = mixQreal(c1.alphaF(), c2.alphaF(), bias); + + return QColor::fromRgbF(r, g, b, a); } QColor Rainbow::overlayColors(const QColor &base, const QColor &paint, - QPainter::CompositionMode comp) + QPainter::CompositionMode comp) { - // This isn't the fastest way, but should be "fast enough". - // It's also the only safe way to use QPainter::CompositionMode - QImage img(1, 1, QImage::Format_ARGB32_Premultiplied); - QPainter p(&img); - QColor start = base; - start.setAlpha(255); // opaque - p.fillRect(0, 0, 1, 1, start); - p.setCompositionMode(comp); - p.fillRect(0, 0, 1, 1, paint); - p.end(); - return img.pixel(0, 0); + // This isn't the fastest way, but should be "fast enough". + // It's also the only safe way to use QPainter::CompositionMode + QImage img(1, 1, QImage::Format_ARGB32_Premultiplied); + QPainter p(&img); + QColor start = base; + start.setAlpha(255); // opaque + p.fillRect(0, 0, 1, 1, start); + p.setCompositionMode(comp); + p.fillRect(0, 0, 1, 1, paint); + p.end(); + return img.pixel(0, 0); } diff --git a/libraries/systeminfo/CMakeLists.txt b/libraries/systeminfo/CMakeLists.txt index 393b5318..548a589c 100644 --- a/libraries/systeminfo/CMakeLists.txt +++ b/libraries/systeminfo/CMakeLists.txt @@ -9,21 +9,21 @@ src/distroutils.cpp ) if (WIN32) - list(APPEND systeminfo_SOURCES src/sys_win32.cpp) + list(APPEND systeminfo_SOURCES src/sys_win32.cpp) elseif (UNIX) - if(APPLE) - list(APPEND systeminfo_SOURCES src/sys_apple.cpp) - else() - list(APPEND systeminfo_SOURCES src/sys_unix.cpp) - endif() + if(APPLE) + list(APPEND systeminfo_SOURCES src/sys_apple.cpp) + else() + list(APPEND systeminfo_SOURCES src/sys_unix.cpp) + endif() endif() add_library(systeminfo STATIC ${systeminfo_SOURCES}) -qt5_use_modules(systeminfo Core Gui Network) +target_link_libraries(systeminfo Qt5::Core Qt5::Gui Qt5::Network) target_include_directories(systeminfo PUBLIC include) include (UnitTest) add_unit_test(sys - SOURCES src/sys_test.cpp - LIBS systeminfo + SOURCES src/sys_test.cpp + LIBS systeminfo ) diff --git a/libraries/systeminfo/include/distroutils.h b/libraries/systeminfo/include/distroutils.h index 5ff8d591..2e85f433 100644 --- a/libraries/systeminfo/include/distroutils.h +++ b/libraries/systeminfo/include/distroutils.h @@ -4,10 +4,10 @@ namespace Sys { struct LsbInfo { - QString distributor; - QString version; - QString description; - QString codename; + QString distributor; + QString version; + QString description; + QString codename; }; bool main_lsb_info(LsbInfo & out); diff --git a/libraries/systeminfo/include/sys.h b/libraries/systeminfo/include/sys.h index c573eb53..7980dfdf 100644 --- a/libraries/systeminfo/include/sys.h +++ b/libraries/systeminfo/include/sys.h @@ -6,37 +6,37 @@ namespace Sys const uint64_t megabyte = 1024ull * 1024ull; struct KernelInfo { - QString kernelName; - QString kernelVersion; + QString kernelName; + QString kernelVersion; }; KernelInfo getKernelInfo(); struct DistributionInfo { - DistributionInfo operator+(const DistributionInfo& rhs) const - { - DistributionInfo out; - if(!distributionName.isEmpty()) - { - out.distributionName = distributionName; - } - else - { - out.distributionName = rhs.distributionName; - } - if(!distributionVersion.isEmpty()) - { - out.distributionVersion = distributionVersion; - } - else - { - out.distributionVersion = rhs.distributionVersion; - } - return out; - } - QString distributionName; - QString distributionVersion; + DistributionInfo operator+(const DistributionInfo& rhs) const + { + DistributionInfo out; + if(!distributionName.isEmpty()) + { + out.distributionName = distributionName; + } + else + { + out.distributionName = rhs.distributionName; + } + if(!distributionVersion.isEmpty()) + { + out.distributionVersion = distributionVersion; + } + else + { + out.distributionVersion = rhs.distributionVersion; + } + return out; + } + QString distributionName; + QString distributionVersion; }; DistributionInfo getDistributionInfo(); diff --git a/libraries/systeminfo/src/distroutils.cpp b/libraries/systeminfo/src/distroutils.cpp index cdba05d0..fb9ae25d 100644 --- a/libraries/systeminfo/src/distroutils.cpp +++ b/libraries/systeminfo/src/distroutils.cpp @@ -41,243 +41,243 @@ SOFTWARE. Sys::DistributionInfo Sys::read_os_release() { - Sys::DistributionInfo out; - QStringList files = { "/etc/os-release", "/usr/lib/os-release" }; - QString name; - QString version; - for (auto &file: files) - { - if(!QFile::exists(file)) - { - continue; - } - QSettings settings(file, QSettings::IniFormat); - if(settings.contains("ID")) - { - name = settings.value("ID").toString().toLower(); - } - else if (settings.contains("NAME")) - { - name = settings.value("NAME").toString().toLower(); - } - else - { - continue; - } + Sys::DistributionInfo out; + QStringList files = { "/etc/os-release", "/usr/lib/os-release" }; + QString name; + QString version; + for (auto &file: files) + { + if(!QFile::exists(file)) + { + continue; + } + QSettings settings(file, QSettings::IniFormat); + if(settings.contains("ID")) + { + name = settings.value("ID").toString().toLower(); + } + else if (settings.contains("NAME")) + { + name = settings.value("NAME").toString().toLower(); + } + else + { + continue; + } - if(settings.contains("VERSION_ID")) - { - version = settings.value("VERSION_ID").toString().toLower(); - } - else if(settings.contains("VERSION")) - { - version = settings.value("VERSION").toString().toLower(); - } - break; - } - if(name.isEmpty()) - { - return out; - } - out.distributionName = name; - out.distributionVersion = version; - return out; + if(settings.contains("VERSION_ID")) + { + version = settings.value("VERSION_ID").toString().toLower(); + } + else if(settings.contains("VERSION")) + { + version = settings.value("VERSION").toString().toLower(); + } + break; + } + if(name.isEmpty()) + { + return out; + } + out.distributionName = name; + out.distributionVersion = version; + return out; } bool Sys::main_lsb_info(Sys::LsbInfo & out) { - int status=0; - QProcess lsbProcess; - lsbProcess.start("lsb_release -a"); - lsbProcess.waitForFinished(); - status = lsbProcess.exitStatus(); - QString output = lsbProcess.readAllStandardOutput(); - qDebug() << output; - lsbProcess.close(); - if(status == 0) - { - auto lines = output.split('\n'); - for(auto line:lines) - { - int index = line.indexOf(':'); - auto key = line.left(index).trimmed(); - auto value = line.mid(index + 1).toLower().trimmed(); - if(key == "Distributor ID") - out.distributor = value; - else if(key == "Release") - out.version = value; - else if(key == "Description") - out.description = value; - else if(key == "Codename") - out.codename = value; - } - return !out.distributor.isEmpty(); - } - return false; + int status=0; + QProcess lsbProcess; + lsbProcess.start("lsb_release -a"); + lsbProcess.waitForFinished(); + status = lsbProcess.exitStatus(); + QString output = lsbProcess.readAllStandardOutput(); + qDebug() << output; + lsbProcess.close(); + if(status == 0) + { + auto lines = output.split('\n'); + for(auto line:lines) + { + int index = line.indexOf(':'); + auto key = line.left(index).trimmed(); + auto value = line.mid(index + 1).toLower().trimmed(); + if(key == "Distributor ID") + out.distributor = value; + else if(key == "Release") + out.version = value; + else if(key == "Description") + out.description = value; + else if(key == "Codename") + out.codename = value; + } + return !out.distributor.isEmpty(); + } + return false; } bool Sys::fallback_lsb_info(Sys::LsbInfo & out) { - // running lsb_release failed, try to read the file instead - // /etc/lsb-release format, if the file even exists, is non-standard. - // Only the `lsb_release` command is specified by LSB. Nonetheless, some - // distributions install an /etc/lsb-release as part of the base - // distribution, but `lsb_release` remains optional. - QString file = "/etc/lsb-release"; - if (QFile::exists(file)) - { - QSettings settings(file, QSettings::IniFormat); - if(settings.contains("DISTRIB_ID")) - { - out.distributor = settings.value("DISTRIB_ID").toString().toLower(); - } - if(settings.contains("DISTRIB_RELEASE")) - { - out.version = settings.value("DISTRIB_RELEASE").toString().toLower(); - } - return !out.distributor.isEmpty(); - } - return false; + // running lsb_release failed, try to read the file instead + // /etc/lsb-release format, if the file even exists, is non-standard. + // Only the `lsb_release` command is specified by LSB. Nonetheless, some + // distributions install an /etc/lsb-release as part of the base + // distribution, but `lsb_release` remains optional. + QString file = "/etc/lsb-release"; + if (QFile::exists(file)) + { + QSettings settings(file, QSettings::IniFormat); + if(settings.contains("DISTRIB_ID")) + { + out.distributor = settings.value("DISTRIB_ID").toString().toLower(); + } + if(settings.contains("DISTRIB_RELEASE")) + { + out.version = settings.value("DISTRIB_RELEASE").toString().toLower(); + } + return !out.distributor.isEmpty(); + } + return false; } void Sys::lsb_postprocess(Sys::LsbInfo & lsb, Sys::DistributionInfo & out) { - QString dist = lsb.distributor; - QString vers = lsb.version; - if(dist.startsWith("redhatenterprise")) - { - dist = "rhel"; - } - else if(dist == "archlinux") - { - dist = "arch"; - } - else if (dist.startsWith("suse")) - { - if(lsb.description.startsWith("opensuse")) - { - dist = "opensuse"; - } - else if (lsb.description.startsWith("suse linux enterprise")) - { - dist = "sles"; - } - } - else if (dist == "debian" and vers == "testing") - { - vers = lsb.codename; - } - else - { - // ubuntu, debian, gentoo, scientific, slackware, ... ? - auto parts = dist.split(QRegExp("\\s+"), QString::SkipEmptyParts); - if(parts.size()) - { - dist = parts[0]; - } - } - if(!dist.isEmpty()) - { - out.distributionName = dist; - out.distributionVersion = vers; - } + QString dist = lsb.distributor; + QString vers = lsb.version; + if(dist.startsWith("redhatenterprise")) + { + dist = "rhel"; + } + else if(dist == "archlinux") + { + dist = "arch"; + } + else if (dist.startsWith("suse")) + { + if(lsb.description.startsWith("opensuse")) + { + dist = "opensuse"; + } + else if (lsb.description.startsWith("suse linux enterprise")) + { + dist = "sles"; + } + } + else if (dist == "debian" and vers == "testing") + { + vers = lsb.codename; + } + else + { + // ubuntu, debian, gentoo, scientific, slackware, ... ? + auto parts = dist.split(QRegExp("\\s+"), QString::SkipEmptyParts); + if(parts.size()) + { + dist = parts[0]; + } + } + if(!dist.isEmpty()) + { + out.distributionName = dist; + out.distributionVersion = vers; + } } Sys::DistributionInfo Sys::read_lsb_release() { - LsbInfo lsb; - if(!main_lsb_info(lsb)) - { - if(!fallback_lsb_info(lsb)) - { - return Sys::DistributionInfo(); - } - } - Sys::DistributionInfo out; - lsb_postprocess(lsb, out); - return out; + LsbInfo lsb; + if(!main_lsb_info(lsb)) + { + if(!fallback_lsb_info(lsb)) + { + return Sys::DistributionInfo(); + } + } + Sys::DistributionInfo out; + lsb_postprocess(lsb, out); + return out; } QString Sys::_extract_distribution(const QString & x) { - QString release = x.toLower(); - if (release.startsWith("red hat enterprise")) - { - return "rhel"; - } - if (release.startsWith("suse linux enterprise")) - { - return "sles"; - } - QStringList list = release.split(QRegExp("\\s+"), QString::SkipEmptyParts); - if(list.size()) - { - return list[0]; - } - return QString(); + QString release = x.toLower(); + if (release.startsWith("red hat enterprise")) + { + return "rhel"; + } + if (release.startsWith("suse linux enterprise")) + { + return "sles"; + } + QStringList list = release.split(QRegExp("\\s+"), QString::SkipEmptyParts); + if(list.size()) + { + return list[0]; + } + return QString(); } QString Sys::_extract_version(const QString & x) { - QRegExp versionish_string("\\d+(?:\\.\\d+)*$"); - QStringList list = x.split(QRegExp("\\s+"), QString::SkipEmptyParts); - for(int i = list.size() - 1; i >= 0; --i) - { - QString chunk = list[i]; - if(versionish_string.exactMatch(chunk)) - { - return chunk; - } - } - return QString(); + QRegExp versionish_string("\\d+(?:\\.\\d+)*$"); + QStringList list = x.split(QRegExp("\\s+"), QString::SkipEmptyParts); + for(int i = list.size() - 1; i >= 0; --i) + { + QString chunk = list[i]; + if(versionish_string.exactMatch(chunk)) + { + return chunk; + } + } + return QString(); } Sys::DistributionInfo Sys::read_legacy_release() { - struct checkEntry - { - QString file; - std::function<QString(const QString &)> extract_distro; - std::function<QString(const QString &)> extract_version; - }; - QList<checkEntry> checks = - { - {"/etc/arch-release", [](const QString &){ return "arch";}, [](const QString &){ return "rolling";}}, - {"/etc/slackware-version", &Sys::_extract_distribution, &Sys::_extract_version}, - {QString(), &Sys::_extract_distribution, &Sys::_extract_version}, - {"/etc/debian_version", [](const QString &){ return "debian";}, [](const QString & x){ return x;}}, - }; - for(auto & check: checks) - { - QStringList files; - if(check.file.isNull()) - { - QDir etcDir("/etc"); - etcDir.setNameFilters({"*-release"}); - etcDir.setFilter(QDir::Files | QDir::NoDot | QDir::NoDotDot | QDir::Readable | QDir::Hidden); - files = etcDir.entryList(); - } - else - { - files.append(check.file); - } - for (auto file : files) - { - QFile relfile(file); - if(!relfile.open(QIODevice::ReadOnly | QIODevice::Text)) - continue; - QString contents = QString::fromUtf8(relfile.readLine()).trimmed(); - QString dist = check.extract_distro(contents); - QString vers = check.extract_version(contents); - if(!dist.isEmpty()) - { - Sys::DistributionInfo out; - out.distributionName = dist; - out.distributionVersion = vers; - return out; - } - } - } - return Sys::DistributionInfo(); + struct checkEntry + { + QString file; + std::function<QString(const QString &)> extract_distro; + std::function<QString(const QString &)> extract_version; + }; + QList<checkEntry> checks = + { + {"/etc/arch-release", [](const QString &){ return "arch";}, [](const QString &){ return "rolling";}}, + {"/etc/slackware-version", &Sys::_extract_distribution, &Sys::_extract_version}, + {QString(), &Sys::_extract_distribution, &Sys::_extract_version}, + {"/etc/debian_version", [](const QString &){ return "debian";}, [](const QString & x){ return x;}}, + }; + for(auto & check: checks) + { + QStringList files; + if(check.file.isNull()) + { + QDir etcDir("/etc"); + etcDir.setNameFilters({"*-release"}); + etcDir.setFilter(QDir::Files | QDir::NoDot | QDir::NoDotDot | QDir::Readable | QDir::Hidden); + files = etcDir.entryList(); + } + else + { + files.append(check.file); + } + for (auto file : files) + { + QFile relfile(file); + if(!relfile.open(QIODevice::ReadOnly | QIODevice::Text)) + continue; + QString contents = QString::fromUtf8(relfile.readLine()).trimmed(); + QString dist = check.extract_distro(contents); + QString vers = check.extract_version(contents); + if(!dist.isEmpty()) + { + Sys::DistributionInfo out; + out.distributionName = dist; + out.distributionVersion = vers; + return out; + } + } + } + return Sys::DistributionInfo(); } diff --git a/libraries/systeminfo/src/sys_apple.cpp b/libraries/systeminfo/src/sys_apple.cpp index 62e6037d..4bcffae4 100644 --- a/libraries/systeminfo/src/sys_apple.cpp +++ b/libraries/systeminfo/src/sys_apple.cpp @@ -4,44 +4,44 @@ Sys::KernelInfo Sys::getKernelInfo() { - Sys::KernelInfo out; - struct utsname buf; - uname(&buf); - out.kernelName = buf.sysname; - out.kernelVersion = buf.release; - return out; + Sys::KernelInfo out; + struct utsname buf; + uname(&buf); + out.kernelName = buf.sysname; + out.kernelVersion = buf.release; + return out; } #include <sys/sysctl.h> uint64_t Sys::getSystemRam() { - uint64_t memsize; - size_t memsizesize = sizeof(memsize); - if(!sysctlbyname("hw.memsize", &memsize, &memsizesize, NULL, 0)) - { - return memsize; - } - else - { - return 0; - } + uint64_t memsize; + size_t memsizesize = sizeof(memsize); + if(!sysctlbyname("hw.memsize", &memsize, &memsizesize, NULL, 0)) + { + return memsize; + } + else + { + return 0; + } } bool Sys::isCPU64bit() { - // not even going to pretend I'm going to support anything else - return true; + // not even going to pretend I'm going to support anything else + return true; } bool Sys::isSystem64bit() { - // yep. maybe when we have 128bit CPUs on consumer devices. - return true; + // yep. maybe when we have 128bit CPUs on consumer devices. + return true; } Sys::DistributionInfo Sys::getDistributionInfo() { - DistributionInfo result; - return result; + DistributionInfo result; + return result; } diff --git a/libraries/systeminfo/src/sys_test.cpp b/libraries/systeminfo/src/sys_test.cpp index 5888e02b..315050d2 100644 --- a/libraries/systeminfo/src/sys_test.cpp +++ b/libraries/systeminfo/src/sys_test.cpp @@ -5,24 +5,24 @@ class SysTest : public QObject { - Q_OBJECT + Q_OBJECT private slots: - void test_kernelNotNull() - { - auto kinfo = Sys::getKernelInfo(); - QVERIFY(!kinfo.kernelName.isEmpty()); - QVERIFY(kinfo.kernelVersion != "0.0"); - } + void test_kernelNotNull() + { + auto kinfo = Sys::getKernelInfo(); + QVERIFY(!kinfo.kernelName.isEmpty()); + QVERIFY(kinfo.kernelVersion != "0.0"); + } /* - void test_systemDistroNotNull() - { - auto kinfo = Sys::getDistributionInfo(); - QVERIFY(!kinfo.distributionName.isEmpty()); - QVERIFY(!kinfo.distributionVersion.isEmpty()); - qDebug() << "Distro: " << kinfo.distributionName << "version" << kinfo.distributionVersion; - } + void test_systemDistroNotNull() + { + auto kinfo = Sys::getDistributionInfo(); + QVERIFY(!kinfo.distributionName.isEmpty()); + QVERIFY(!kinfo.distributionVersion.isEmpty()); + qDebug() << "Distro: " << kinfo.distributionName << "version" << kinfo.distributionVersion; + } */ }; diff --git a/libraries/systeminfo/src/sys_unix.cpp b/libraries/systeminfo/src/sys_unix.cpp index 313908f3..ab3f302e 100644 --- a/libraries/systeminfo/src/sys_unix.cpp +++ b/libraries/systeminfo/src/sys_unix.cpp @@ -7,69 +7,69 @@ Sys::KernelInfo Sys::getKernelInfo() { - Sys::KernelInfo out; - struct utsname buf; - uname(&buf); - out.kernelName = buf.sysname; - out.kernelVersion = buf.release; - return out; + Sys::KernelInfo out; + struct utsname buf; + uname(&buf); + out.kernelName = buf.sysname; + out.kernelVersion = buf.release; + return out; } uint64_t Sys::getSystemRam() { - std::string token; - std::ifstream file("/proc/meminfo"); - while(file >> token) - { - if(token == "MemTotal:") - { - uint64_t mem; - if(file >> mem) - { - return mem * 1024ull; - } - else - { - return 0; - } - } - // ignore rest of the line - file.ignore(std::numeric_limits<std::streamsize>::max(), '\n'); - } - return 0; // nothing found + std::string token; + std::ifstream file("/proc/meminfo"); + while(file >> token) + { + if(token == "MemTotal:") + { + uint64_t mem; + if(file >> mem) + { + return mem * 1024ull; + } + else + { + return 0; + } + } + // ignore rest of the line + file.ignore(std::numeric_limits<std::streamsize>::max(), '\n'); + } + return 0; // nothing found } bool Sys::isCPU64bit() { - return isSystem64bit(); + return isSystem64bit(); } bool Sys::isSystem64bit() { - // kernel build arch on linux - return QSysInfo::currentCpuArchitecture() == "x86_64"; + // kernel build arch on linux + return QSysInfo::currentCpuArchitecture() == "x86_64"; } Sys::DistributionInfo Sys::getDistributionInfo() { - DistributionInfo systemd_info = read_os_release(); - DistributionInfo lsb_info = read_lsb_release(); - DistributionInfo legacy_info = read_legacy_release(); - DistributionInfo result = systemd_info + lsb_info + legacy_info; - if(result.distributionName.isNull()) - { - result.distributionName = "unknown"; - } - if(result.distributionVersion.isNull()) - { - if(result.distributionName == "arch") - { - result.distributionVersion = "rolling"; - } - else - { - result.distributionVersion = "unknown"; - } - } - return result; + DistributionInfo systemd_info = read_os_release(); + DistributionInfo lsb_info = read_lsb_release(); + DistributionInfo legacy_info = read_legacy_release(); + DistributionInfo result = systemd_info + lsb_info + legacy_info; + if(result.distributionName.isNull()) + { + result.distributionName = "unknown"; + } + if(result.distributionVersion.isNull()) + { + if(result.distributionName == "arch") + { + result.distributionVersion = "rolling"; + } + else + { + result.distributionVersion = "unknown"; + } + } + return result; } diff --git a/libraries/systeminfo/src/sys_win32.cpp b/libraries/systeminfo/src/sys_win32.cpp index cc1d61c1..a750b3a7 100644 --- a/libraries/systeminfo/src/sys_win32.cpp +++ b/libraries/systeminfo/src/sys_win32.cpp @@ -4,49 +4,49 @@ Sys::KernelInfo Sys::getKernelInfo() { - Sys::KernelInfo out; - out.kernelName = "Windows"; - OSVERSIONINFOW osvi; - ZeroMemory(&osvi, sizeof(OSVERSIONINFOW)); - osvi.dwOSVersionInfoSize = sizeof(OSVERSIONINFOW); - GetVersionExW(&osvi); - out.kernelVersion = QString("%1.%2").arg(osvi.dwMajorVersion).arg(osvi.dwMinorVersion); - return out; + Sys::KernelInfo out; + out.kernelName = "Windows"; + OSVERSIONINFOW osvi; + ZeroMemory(&osvi, sizeof(OSVERSIONINFOW)); + osvi.dwOSVersionInfoSize = sizeof(OSVERSIONINFOW); + GetVersionExW(&osvi); + out.kernelVersion = QString("%1.%2").arg(osvi.dwMajorVersion).arg(osvi.dwMinorVersion); + return out; } uint64_t Sys::getSystemRam() { - MEMORYSTATUSEX status; - status.dwLength = sizeof(status); - GlobalMemoryStatusEx( &status ); - // bytes - return (uint64_t)status.ullTotalPhys; + MEMORYSTATUSEX status; + status.dwLength = sizeof(status); + GlobalMemoryStatusEx( &status ); + // bytes + return (uint64_t)status.ullTotalPhys; } bool Sys::isSystem64bit() { #if defined(_WIN64) - return true; + return true; #elif defined(_WIN32) - BOOL f64 = false; - return IsWow64Process(GetCurrentProcess(), &f64) && f64; + BOOL f64 = false; + return IsWow64Process(GetCurrentProcess(), &f64) && f64; #else - // it's some other kind of system... - return false; + // it's some other kind of system... + return false; #endif } bool Sys::isCPU64bit() { - SYSTEM_INFO info; - ZeroMemory(&info, sizeof(SYSTEM_INFO)); - GetNativeSystemInfo(&info); - auto arch = info.wProcessorArchitecture; - return arch == PROCESSOR_ARCHITECTURE_AMD64 || arch == PROCESSOR_ARCHITECTURE_IA64; + SYSTEM_INFO info; + ZeroMemory(&info, sizeof(SYSTEM_INFO)); + GetNativeSystemInfo(&info); + auto arch = info.wProcessorArchitecture; + return arch == PROCESSOR_ARCHITECTURE_AMD64 || arch == PROCESSOR_ARCHITECTURE_IA64; } Sys::DistributionInfo Sys::getDistributionInfo() { - DistributionInfo result; - return result; + DistributionInfo result; + return result; } diff --git a/libraries/xz-embedded/CMakeLists.txt b/libraries/xz-embedded/CMakeLists.txt index 5f744671..86ac60c8 100644 --- a/libraries/xz-embedded/CMakeLists.txt +++ b/libraries/xz-embedded/CMakeLists.txt @@ -9,18 +9,18 @@ option(XZ_BUILD_MINIDEC "Build a tiny utility that decompresses xz streams" OFF) # tweak this list and xz.h to fit your needs set(XZ_SOURCES - src/xz_crc32.c - src/xz_crc64.c - src/xz_dec_lzma2.c - src/xz_dec_stream.c -# src/xz_dec_bcj.c + src/xz_crc32.c + src/xz_crc64.c + src/xz_dec_lzma2.c + src/xz_dec_stream.c +# src/xz_dec_bcj.c ) add_library(xz-embedded STATIC ${XZ_SOURCES}) target_include_directories(xz-embedded PUBLIC "${CMAKE_CURRENT_SOURCE_DIR}/include") set_property(TARGET xz-embedded PROPERTY C_STANDARD 99) if(${XZ_BUILD_MINIDEC}) - add_executable(xzminidec xzminidec.c) - target_link_libraries(xzminidec xz-embedded) - set_property(TARGET xzminidec PROPERTY C_STANDARD 99) + add_executable(xzminidec xzminidec.c) + target_link_libraries(xzminidec xz-embedded) + set_property(TARGET xzminidec PROPERTY C_STANDARD 99) endif() diff --git a/libraries/xz-embedded/include/xz.h b/libraries/xz-embedded/include/xz.h index eef8ef69..3779124c 100644 --- a/libraries/xz-embedded/include/xz.h +++ b/libraries/xz-embedded/include/xz.h @@ -69,9 +69,9 @@ extern "C" { */ enum xz_mode { - XZ_SINGLE, - XZ_PREALLOC, - XZ_DYNALLOC + XZ_SINGLE, + XZ_PREALLOC, + XZ_DYNALLOC }; /** @@ -126,15 +126,15 @@ enum xz_mode */ enum xz_ret { - XZ_OK, - XZ_STREAM_END, - XZ_UNSUPPORTED_CHECK, - XZ_MEM_ERROR, - XZ_MEMLIMIT_ERROR, - XZ_FORMAT_ERROR, - XZ_OPTIONS_ERROR, - XZ_DATA_ERROR, - XZ_BUF_ERROR + XZ_OK, + XZ_STREAM_END, + XZ_UNSUPPORTED_CHECK, + XZ_MEM_ERROR, + XZ_MEMLIMIT_ERROR, + XZ_FORMAT_ERROR, + XZ_OPTIONS_ERROR, + XZ_DATA_ERROR, + XZ_BUF_ERROR }; /** @@ -155,13 +155,13 @@ enum xz_ret */ struct xz_buf { - const uint8_t *in; - size_t in_pos; - size_t in_size; + const uint8_t *in; + size_t in_pos; + size_t in_size; - uint8_t *out; - size_t out_pos; - size_t out_size; + uint8_t *out; + size_t out_pos; + size_t out_size; }; /** diff --git a/libraries/xz-embedded/src/xz_config.h b/libraries/xz-embedded/src/xz_config.h index 40805b75..effdb1bd 100644 --- a/libraries/xz-embedded/src/xz_config.h +++ b/libraries/xz-embedded/src/xz_config.h @@ -74,36 +74,36 @@ typedef unsigned char bool; #ifndef get_unaligned_le32 static inline uint32_t get_unaligned_le32(const uint8_t *buf) { - return (uint32_t)buf[0] | ((uint32_t)buf[1] << 8) | ((uint32_t)buf[2] << 16) | - ((uint32_t)buf[3] << 24); + return (uint32_t)buf[0] | ((uint32_t)buf[1] << 8) | ((uint32_t)buf[2] << 16) | + ((uint32_t)buf[3] << 24); } #endif #ifndef get_unaligned_be32 static inline uint32_t get_unaligned_be32(const uint8_t *buf) { - return (uint32_t)(buf[0] << 24) | ((uint32_t)buf[1] << 16) | ((uint32_t)buf[2] << 8) | - (uint32_t)buf[3]; + return (uint32_t)(buf[0] << 24) | ((uint32_t)buf[1] << 16) | ((uint32_t)buf[2] << 8) | + (uint32_t)buf[3]; } #endif #ifndef put_unaligned_le32 static inline void put_unaligned_le32(uint32_t val, uint8_t *buf) { - buf[0] = (uint8_t)val; - buf[1] = (uint8_t)(val >> 8); - buf[2] = (uint8_t)(val >> 16); - buf[3] = (uint8_t)(val >> 24); + buf[0] = (uint8_t)val; + buf[1] = (uint8_t)(val >> 8); + buf[2] = (uint8_t)(val >> 16); + buf[3] = (uint8_t)(val >> 24); } #endif #ifndef put_unaligned_be32 static inline void put_unaligned_be32(uint32_t val, uint8_t *buf) { - buf[0] = (uint8_t)(val >> 24); - buf[1] = (uint8_t)(val >> 16); - buf[2] = (uint8_t)(val >> 8); - buf[3] = (uint8_t)val; + buf[0] = (uint8_t)(val >> 24); + buf[1] = (uint8_t)(val >> 16); + buf[2] = (uint8_t)(val >> 8); + buf[3] = (uint8_t)val; } #endif diff --git a/libraries/xz-embedded/src/xz_crc32.c b/libraries/xz-embedded/src/xz_crc32.c index c412662b..65d9d5b8 100644 --- a/libraries/xz-embedded/src/xz_crc32.c +++ b/libraries/xz-embedded/src/xz_crc32.c @@ -29,33 +29,33 @@ STATIC_RW_DATA uint32_t xz_crc32_table[256]; XZ_EXTERN void xz_crc32_init(void) { - const uint32_t poly = 0xEDB88320; + const uint32_t poly = 0xEDB88320; - uint32_t i; - uint32_t j; - uint32_t r; + uint32_t i; + uint32_t j; + uint32_t r; - for (i = 0; i < 256; ++i) - { - r = i; - for (j = 0; j < 8; ++j) - r = (r >> 1) ^ (poly & ~((r & 1) - 1)); + for (i = 0; i < 256; ++i) + { + r = i; + for (j = 0; j < 8; ++j) + r = (r >> 1) ^ (poly & ~((r & 1) - 1)); - xz_crc32_table[i] = r; - } + xz_crc32_table[i] = r; + } - return; + return; } XZ_EXTERN uint32_t xz_crc32(const uint8_t *buf, size_t size, uint32_t crc) { - crc = ~crc; + crc = ~crc; - while (size != 0) - { - crc = xz_crc32_table[*buf++ ^ (crc & 0xFF)] ^ (crc >> 8); - --size; - } + while (size != 0) + { + crc = xz_crc32_table[*buf++ ^ (crc & 0xFF)] ^ (crc >> 8); + --size; + } - return ~crc; + return ~crc; } diff --git a/libraries/xz-embedded/src/xz_crc64.c b/libraries/xz-embedded/src/xz_crc64.c index 4794b9d3..0f711d8d 100644 --- a/libraries/xz-embedded/src/xz_crc64.c +++ b/libraries/xz-embedded/src/xz_crc64.c @@ -20,33 +20,33 @@ STATIC_RW_DATA uint64_t xz_crc64_table[256]; XZ_EXTERN void xz_crc64_init(void) { - const uint64_t poly = 0xC96C5795D7870F42; + const uint64_t poly = 0xC96C5795D7870F42; - uint32_t i; - uint32_t j; - uint64_t r; + uint32_t i; + uint32_t j; + uint64_t r; - for (i = 0; i < 256; ++i) - { - r = i; - for (j = 0; j < 8; ++j) - r = (r >> 1) ^ (poly & ~((r & 1) - 1)); + for (i = 0; i < 256; ++i) + { + r = i; + for (j = 0; j < 8; ++j) + r = (r >> 1) ^ (poly & ~((r & 1) - 1)); - xz_crc64_table[i] = r; - } + xz_crc64_table[i] = r; + } - return; + return; } XZ_EXTERN uint64_t xz_crc64(const uint8_t *buf, size_t size, uint64_t crc) { - crc = ~crc; + crc = ~crc; - while (size != 0) - { - crc = xz_crc64_table[*buf++ ^ (crc & 0xFF)] ^ (crc >> 8); - --size; - } + while (size != 0) + { + crc = xz_crc64_table[*buf++ ^ (crc & 0xFF)] ^ (crc >> 8); + --size; + } - return ~crc; + return ~crc; } diff --git a/libraries/xz-embedded/src/xz_dec_bcj.c b/libraries/xz-embedded/src/xz_dec_bcj.c index 9ffda3bd..a79fa76d 100644 --- a/libraries/xz-embedded/src/xz_dec_bcj.c +++ b/libraries/xz-embedded/src/xz_dec_bcj.c @@ -18,64 +18,64 @@ struct xz_dec_bcj { - /* Type of the BCJ filter being used */ - enum - { - BCJ_X86 = 4, /* x86 or x86-64 */ - BCJ_POWERPC = 5, /* Big endian only */ - BCJ_IA64 = 6, /* Big or little endian */ - BCJ_ARM = 7, /* Little endian only */ - BCJ_ARMTHUMB = 8, /* Little endian only */ - BCJ_SPARC = 9 /* Big or little endian */ - } type; - - /* - * Return value of the next filter in the chain. We need to preserve - * this information across calls, because we must not call the next - * filter anymore once it has returned XZ_STREAM_END. - */ - enum xz_ret ret; - - /* True if we are operating in single-call mode. */ - bool single_call; - - /* - * Absolute position relative to the beginning of the uncompressed - * data (in a single .xz Block). We care only about the lowest 32 - * bits so this doesn't need to be uint64_t even with big files. - */ - uint32_t pos; - - /* x86 filter state */ - uint32_t x86_prev_mask; - - /* Temporary space to hold the variables from struct xz_buf */ - uint8_t *out; - size_t out_pos; - size_t out_size; - - struct - { - /* Amount of already filtered data in the beginning of buf */ - size_t filtered; - - /* Total amount of data currently stored in buf */ - size_t size; - - /* - * Buffer to hold a mix of filtered and unfiltered data. This - * needs to be big enough to hold Alignment + 2 * Look-ahead: - * - * Type Alignment Look-ahead - * x86 1 4 - * PowerPC 4 0 - * IA-64 16 0 - * ARM 4 0 - * ARM-Thumb 2 2 - * SPARC 4 0 - */ - uint8_t buf[16]; - } temp; + /* Type of the BCJ filter being used */ + enum + { + BCJ_X86 = 4, /* x86 or x86-64 */ + BCJ_POWERPC = 5, /* Big endian only */ + BCJ_IA64 = 6, /* Big or little endian */ + BCJ_ARM = 7, /* Little endian only */ + BCJ_ARMTHUMB = 8, /* Little endian only */ + BCJ_SPARC = 9 /* Big or little endian */ + } type; + + /* + * Return value of the next filter in the chain. We need to preserve + * this information across calls, because we must not call the next + * filter anymore once it has returned XZ_STREAM_END. + */ + enum xz_ret ret; + + /* True if we are operating in single-call mode. */ + bool single_call; + + /* + * Absolute position relative to the beginning of the uncompressed + * data (in a single .xz Block). We care only about the lowest 32 + * bits so this doesn't need to be uint64_t even with big files. + */ + uint32_t pos; + + /* x86 filter state */ + uint32_t x86_prev_mask; + + /* Temporary space to hold the variables from struct xz_buf */ + uint8_t *out; + size_t out_pos; + size_t out_size; + + struct + { + /* Amount of already filtered data in the beginning of buf */ + size_t filtered; + + /* Total amount of data currently stored in buf */ + size_t size; + + /* + * Buffer to hold a mix of filtered and unfiltered data. This + * needs to be big enough to hold Alignment + 2 * Look-ahead: + * + * Type Alignment Look-ahead + * x86 1 4 + * PowerPC 4 0 + * IA-64 16 0 + * ARM 4 0 + * ARM-Thumb 2 2 + * SPARC 4 0 + */ + uint8_t buf[16]; + } temp; }; #ifdef XZ_DEC_X86 @@ -85,264 +85,264 @@ struct xz_dec_bcj */ static inline int bcj_x86_test_msbyte(uint8_t b) { - return b == 0x00 || b == 0xFF; + return b == 0x00 || b == 0xFF; } static size_t bcj_x86(struct xz_dec_bcj *s, uint8_t *buf, size_t size) { - static const bool mask_to_allowed_status[8] = {true, true, true, false, - true, false, false, false}; - - static const uint8_t mask_to_bit_num[8] = {0, 1, 2, 2, 3, 3, 3, 3}; - - size_t i; - size_t prev_pos = (size_t) - 1; - uint32_t prev_mask = s->x86_prev_mask; - uint32_t src; - uint32_t dest; - uint32_t j; - uint8_t b; - - if (size <= 4) - return 0; - - size -= 4; - for (i = 0; i < size; ++i) - { - if ((buf[i] & 0xFE) != 0xE8) - continue; - - prev_pos = i - prev_pos; - if (prev_pos > 3) - { - prev_mask = 0; - } - else - { - prev_mask = (prev_mask << (prev_pos - 1)) & 7; - if (prev_mask != 0) - { - b = buf[i + 4 - mask_to_bit_num[prev_mask]]; - if (!mask_to_allowed_status[prev_mask] || bcj_x86_test_msbyte(b)) - { - prev_pos = i; - prev_mask = (prev_mask << 1) | 1; - continue; - } - } - } - - prev_pos = i; - - if (bcj_x86_test_msbyte(buf[i + 4])) - { - src = get_unaligned_le32(buf + i + 1); - while (true) - { - dest = src - (s->pos + (uint32_t)i + 5); - if (prev_mask == 0) - break; - - j = mask_to_bit_num[prev_mask] * 8; - b = (uint8_t)(dest >> (24 - j)); - if (!bcj_x86_test_msbyte(b)) - break; - - src = dest ^ (((uint32_t)1 << (32 - j)) - 1); - } - - dest &= 0x01FFFFFF; - dest |= (uint32_t)0 - (dest & 0x01000000); - put_unaligned_le32(dest, buf + i + 1); - i += 4; - } - else - { - prev_mask = (prev_mask << 1) | 1; - } - } - - prev_pos = i - prev_pos; - s->x86_prev_mask = prev_pos > 3 ? 0 : prev_mask << (prev_pos - 1); - return i; + static const bool mask_to_allowed_status[8] = {true, true, true, false, + true, false, false, false}; + + static const uint8_t mask_to_bit_num[8] = {0, 1, 2, 2, 3, 3, 3, 3}; + + size_t i; + size_t prev_pos = (size_t) - 1; + uint32_t prev_mask = s->x86_prev_mask; + uint32_t src; + uint32_t dest; + uint32_t j; + uint8_t b; + + if (size <= 4) + return 0; + + size -= 4; + for (i = 0; i < size; ++i) + { + if ((buf[i] & 0xFE) != 0xE8) + continue; + + prev_pos = i - prev_pos; + if (prev_pos > 3) + { + prev_mask = 0; + } + else + { + prev_mask = (prev_mask << (prev_pos - 1)) & 7; + if (prev_mask != 0) + { + b = buf[i + 4 - mask_to_bit_num[prev_mask]]; + if (!mask_to_allowed_status[prev_mask] || bcj_x86_test_msbyte(b)) + { + prev_pos = i; + prev_mask = (prev_mask << 1) | 1; + continue; + } + } + } + + prev_pos = i; + + if (bcj_x86_test_msbyte(buf[i + 4])) + { + src = get_unaligned_le32(buf + i + 1); + while (true) + { + dest = src - (s->pos + (uint32_t)i + 5); + if (prev_mask == 0) + break; + + j = mask_to_bit_num[prev_mask] * 8; + b = (uint8_t)(dest >> (24 - j)); + if (!bcj_x86_test_msbyte(b)) + break; + + src = dest ^ (((uint32_t)1 << (32 - j)) - 1); + } + + dest &= 0x01FFFFFF; + dest |= (uint32_t)0 - (dest & 0x01000000); + put_unaligned_le32(dest, buf + i + 1); + i += 4; + } + else + { + prev_mask = (prev_mask << 1) | 1; + } + } + + prev_pos = i - prev_pos; + s->x86_prev_mask = prev_pos > 3 ? 0 : prev_mask << (prev_pos - 1); + return i; } #endif #ifdef XZ_DEC_POWERPC static size_t bcj_powerpc(struct xz_dec_bcj *s, uint8_t *buf, size_t size) { - size_t i; - uint32_t instr; - - for (i = 0; i + 4 <= size; i += 4) - { - instr = get_unaligned_be32(buf + i); - if ((instr & 0xFC000003) == 0x48000001) - { - instr &= 0x03FFFFFC; - instr -= s->pos + (uint32_t)i; - instr &= 0x03FFFFFC; - instr |= 0x48000001; - put_unaligned_be32(instr, buf + i); - } - } - - return i; + size_t i; + uint32_t instr; + + for (i = 0; i + 4 <= size; i += 4) + { + instr = get_unaligned_be32(buf + i); + if ((instr & 0xFC000003) == 0x48000001) + { + instr &= 0x03FFFFFC; + instr -= s->pos + (uint32_t)i; + instr &= 0x03FFFFFC; + instr |= 0x48000001; + put_unaligned_be32(instr, buf + i); + } + } + + return i; } #endif #ifdef XZ_DEC_IA64 static size_t bcj_ia64(struct xz_dec_bcj *s, uint8_t *buf, size_t size) { - static const uint8_t branch_table[32] = {0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, - 4, 4, 6, 6, 0, 0, 7, 7, 4, 4, 0, 0, 4, 4, 0, 0}; - - /* - * The local variables take a little bit stack space, but it's less - * than what LZMA2 decoder takes, so it doesn't make sense to reduce - * stack usage here without doing that for the LZMA2 decoder too. - */ - - /* Loop counters */ - size_t i; - size_t j; - - /* Instruction slot (0, 1, or 2) in the 128-bit instruction word */ - uint32_t slot; - - /* Bitwise offset of the instruction indicated by slot */ - uint32_t bit_pos; - - /* bit_pos split into byte and bit parts */ - uint32_t byte_pos; - uint32_t bit_res; - - /* Address part of an instruction */ - uint32_t addr; - - /* Mask used to detect which instructions to convert */ - uint32_t mask; - - /* 41-bit instruction stored somewhere in the lowest 48 bits */ - uint64_t instr; - - /* Instruction normalized with bit_res for easier manipulation */ - uint64_t norm; - - for (i = 0; i + 16 <= size; i += 16) - { - mask = branch_table[buf[i] & 0x1F]; - for (slot = 0, bit_pos = 5; slot < 3; ++slot, bit_pos += 41) - { - if (((mask >> slot) & 1) == 0) - continue; - - byte_pos = bit_pos >> 3; - bit_res = bit_pos & 7; - instr = 0; - for (j = 0; j < 6; ++j) - instr |= (uint64_t)(buf[i + j + byte_pos]) << (8 * j); - - norm = instr >> bit_res; - - if (((norm >> 37) & 0x0F) == 0x05 && ((norm >> 9) & 0x07) == 0) - { - addr = (norm >> 13) & 0x0FFFFF; - addr |= ((uint32_t)(norm >> 36) & 1) << 20; - addr <<= 4; - addr -= s->pos + (uint32_t)i; - addr >>= 4; - - norm &= ~((uint64_t)0x8FFFFF << 13); - norm |= (uint64_t)(addr & 0x0FFFFF) << 13; - norm |= (uint64_t)(addr & 0x100000) << (36 - 20); - - instr &= (1 << bit_res) - 1; - instr |= norm << bit_res; - - for (j = 0; j < 6; j++) - buf[i + j + byte_pos] = (uint8_t)(instr >> (8 * j)); - } - } - } - - return i; + static const uint8_t branch_table[32] = {0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 4, 4, 6, 6, 0, 0, 7, 7, 4, 4, 0, 0, 4, 4, 0, 0}; + + /* + * The local variables take a little bit stack space, but it's less + * than what LZMA2 decoder takes, so it doesn't make sense to reduce + * stack usage here without doing that for the LZMA2 decoder too. + */ + + /* Loop counters */ + size_t i; + size_t j; + + /* Instruction slot (0, 1, or 2) in the 128-bit instruction word */ + uint32_t slot; + + /* Bitwise offset of the instruction indicated by slot */ + uint32_t bit_pos; + + /* bit_pos split into byte and bit parts */ + uint32_t byte_pos; + uint32_t bit_res; + + /* Address part of an instruction */ + uint32_t addr; + + /* Mask used to detect which instructions to convert */ + uint32_t mask; + + /* 41-bit instruction stored somewhere in the lowest 48 bits */ + uint64_t instr; + + /* Instruction normalized with bit_res for easier manipulation */ + uint64_t norm; + + for (i = 0; i + 16 <= size; i += 16) + { + mask = branch_table[buf[i] & 0x1F]; + for (slot = 0, bit_pos = 5; slot < 3; ++slot, bit_pos += 41) + { + if (((mask >> slot) & 1) == 0) + continue; + + byte_pos = bit_pos >> 3; + bit_res = bit_pos & 7; + instr = 0; + for (j = 0; j < 6; ++j) + instr |= (uint64_t)(buf[i + j + byte_pos]) << (8 * j); + + norm = instr >> bit_res; + + if (((norm >> 37) & 0x0F) == 0x05 && ((norm >> 9) & 0x07) == 0) + { + addr = (norm >> 13) & 0x0FFFFF; + addr |= ((uint32_t)(norm >> 36) & 1) << 20; + addr <<= 4; + addr -= s->pos + (uint32_t)i; + addr >>= 4; + + norm &= ~((uint64_t)0x8FFFFF << 13); + norm |= (uint64_t)(addr & 0x0FFFFF) << 13; + norm |= (uint64_t)(addr & 0x100000) << (36 - 20); + + instr &= (1 << bit_res) - 1; + instr |= norm << bit_res; + + for (j = 0; j < 6; j++) + buf[i + j + byte_pos] = (uint8_t)(instr >> (8 * j)); + } + } + } + + return i; } #endif #ifdef XZ_DEC_ARM static size_t bcj_arm(struct xz_dec_bcj *s, uint8_t *buf, size_t size) { - size_t i; - uint32_t addr; - - for (i = 0; i + 4 <= size; i += 4) - { - if (buf[i + 3] == 0xEB) - { - addr = - (uint32_t)buf[i] | ((uint32_t)buf[i + 1] << 8) | ((uint32_t)buf[i + 2] << 16); - addr <<= 2; - addr -= s->pos + (uint32_t)i + 8; - addr >>= 2; - buf[i] = (uint8_t)addr; - buf[i + 1] = (uint8_t)(addr >> 8); - buf[i + 2] = (uint8_t)(addr >> 16); - } - } - - return i; + size_t i; + uint32_t addr; + + for (i = 0; i + 4 <= size; i += 4) + { + if (buf[i + 3] == 0xEB) + { + addr = + (uint32_t)buf[i] | ((uint32_t)buf[i + 1] << 8) | ((uint32_t)buf[i + 2] << 16); + addr <<= 2; + addr -= s->pos + (uint32_t)i + 8; + addr >>= 2; + buf[i] = (uint8_t)addr; + buf[i + 1] = (uint8_t)(addr >> 8); + buf[i + 2] = (uint8_t)(addr >> 16); + } + } + + return i; } #endif #ifdef XZ_DEC_ARMTHUMB static size_t bcj_armthumb(struct xz_dec_bcj *s, uint8_t *buf, size_t size) { - size_t i; - uint32_t addr; - - for (i = 0; i + 4 <= size; i += 2) - { - if ((buf[i + 1] & 0xF8) == 0xF0 && (buf[i + 3] & 0xF8) == 0xF8) - { - addr = (((uint32_t)buf[i + 1] & 0x07) << 19) | ((uint32_t)buf[i] << 11) | - (((uint32_t)buf[i + 3] & 0x07) << 8) | (uint32_t)buf[i + 2]; - addr <<= 1; - addr -= s->pos + (uint32_t)i + 4; - addr >>= 1; - buf[i + 1] = (uint8_t)(0xF0 | ((addr >> 19) & 0x07)); - buf[i] = (uint8_t)(addr >> 11); - buf[i + 3] = (uint8_t)(0xF8 | ((addr >> 8) & 0x07)); - buf[i + 2] = (uint8_t)addr; - i += 2; - } - } - - return i; + size_t i; + uint32_t addr; + + for (i = 0; i + 4 <= size; i += 2) + { + if ((buf[i + 1] & 0xF8) == 0xF0 && (buf[i + 3] & 0xF8) == 0xF8) + { + addr = (((uint32_t)buf[i + 1] & 0x07) << 19) | ((uint32_t)buf[i] << 11) | + (((uint32_t)buf[i + 3] & 0x07) << 8) | (uint32_t)buf[i + 2]; + addr <<= 1; + addr -= s->pos + (uint32_t)i + 4; + addr >>= 1; + buf[i + 1] = (uint8_t)(0xF0 | ((addr >> 19) & 0x07)); + buf[i] = (uint8_t)(addr >> 11); + buf[i + 3] = (uint8_t)(0xF8 | ((addr >> 8) & 0x07)); + buf[i + 2] = (uint8_t)addr; + i += 2; + } + } + + return i; } #endif #ifdef XZ_DEC_SPARC static size_t bcj_sparc(struct xz_dec_bcj *s, uint8_t *buf, size_t size) { - size_t i; - uint32_t instr; - - for (i = 0; i + 4 <= size; i += 4) - { - instr = get_unaligned_be32(buf + i); - if ((instr >> 22) == 0x100 || (instr >> 22) == 0x1FF) - { - instr <<= 2; - instr -= s->pos + (uint32_t)i; - instr >>= 2; - instr = - ((uint32_t)0x40000000 - (instr & 0x400000)) | 0x40000000 | (instr & 0x3FFFFF); - put_unaligned_be32(instr, buf + i); - } - } - - return i; + size_t i; + uint32_t instr; + + for (i = 0; i + 4 <= size; i += 4) + { + instr = get_unaligned_be32(buf + i); + if ((instr >> 22) == 0x100 || (instr >> 22) == 0x1FF) + { + instr <<= 2; + instr -= s->pos + (uint32_t)i; + instr >>= 2; + instr = + ((uint32_t)0x40000000 - (instr & 0x400000)) | 0x40000000 | (instr & 0x3FFFFF); + put_unaligned_be32(instr, buf + i); + } + } + + return i; } #endif @@ -356,51 +356,51 @@ static size_t bcj_sparc(struct xz_dec_bcj *s, uint8_t *buf, size_t size) */ static void bcj_apply(struct xz_dec_bcj *s, uint8_t *buf, size_t *pos, size_t size) { - size_t filtered; + size_t filtered; - buf += *pos; - size -= *pos; + buf += *pos; + size -= *pos; - switch (s->type) - { + switch (s->type) + { #ifdef XZ_DEC_X86 - case BCJ_X86: - filtered = bcj_x86(s, buf, size); - break; + case BCJ_X86: + filtered = bcj_x86(s, buf, size); + break; #endif #ifdef XZ_DEC_POWERPC - case BCJ_POWERPC: - filtered = bcj_powerpc(s, buf, size); - break; + case BCJ_POWERPC: + filtered = bcj_powerpc(s, buf, size); + break; #endif #ifdef XZ_DEC_IA64 - case BCJ_IA64: - filtered = bcj_ia64(s, buf, size); - break; + case BCJ_IA64: + filtered = bcj_ia64(s, buf, size); + break; #endif #ifdef XZ_DEC_ARM - case BCJ_ARM: - filtered = bcj_arm(s, buf, size); - break; + case BCJ_ARM: + filtered = bcj_arm(s, buf, size); + break; #endif #ifdef XZ_DEC_ARMTHUMB - case BCJ_ARMTHUMB: - filtered = bcj_armthumb(s, buf, size); - break; + case BCJ_ARMTHUMB: + filtered = bcj_armthumb(s, buf, size); + break; #endif #ifdef XZ_DEC_SPARC - case BCJ_SPARC: - filtered = bcj_sparc(s, buf, size); - break; + case BCJ_SPARC: + filtered = bcj_sparc(s, buf, size); + break; #endif - default: - /* Never reached but silence compiler warnings. */ - filtered = 0; - break; - } - - *pos += filtered; - s->pos += filtered; + default: + /* Never reached but silence compiler warnings. */ + filtered = 0; + break; + } + + *pos += filtered; + s->pos += filtered; } /* @@ -410,15 +410,15 @@ static void bcj_apply(struct xz_dec_bcj *s, uint8_t *buf, size_t *pos, size_t si */ static void bcj_flush(struct xz_dec_bcj *s, struct xz_buf *b) { - size_t copy_size; + size_t copy_size; - copy_size = min_t(size_t, s->temp.filtered, b->out_size - b->out_pos); - memcpy(b->out + b->out_pos, s->temp.buf, copy_size); - b->out_pos += copy_size; + copy_size = min_t(size_t, s->temp.filtered, b->out_size - b->out_pos); + memcpy(b->out + b->out_pos, s->temp.buf, copy_size); + b->out_pos += copy_size; - s->temp.filtered -= copy_size; - s->temp.size -= copy_size; - memmove(s->temp.buf, s->temp.buf + copy_size, s->temp.size); + s->temp.filtered -= copy_size; + s->temp.size -= copy_size; + memmove(s->temp.buf, s->temp.buf + copy_size, s->temp.size); } /* @@ -427,162 +427,162 @@ static void bcj_flush(struct xz_dec_bcj *s, struct xz_buf *b) * some buffering. */ XZ_EXTERN enum xz_ret xz_dec_bcj_run(struct xz_dec_bcj *s, struct xz_dec_lzma2 *lzma2, - struct xz_buf *b) + struct xz_buf *b) { - size_t out_start; - - /* - * Flush pending already filtered data to the output buffer. Return - * immediatelly if we couldn't flush everything, or if the next - * filter in the chain had already returned XZ_STREAM_END. - */ - if (s->temp.filtered > 0) - { - bcj_flush(s, b); - if (s->temp.filtered > 0) - return XZ_OK; - - if (s->ret == XZ_STREAM_END) - return XZ_STREAM_END; - } - - /* - * If we have more output space than what is currently pending in - * temp, copy the unfiltered data from temp to the output buffer - * and try to fill the output buffer by decoding more data from the - * next filter in the chain. Apply the BCJ filter on the new data - * in the output buffer. If everything cannot be filtered, copy it - * to temp and rewind the output buffer position accordingly. - * - * This needs to be always run when temp.size == 0 to handle a special - * case where the output buffer is full and the next filter has no - * more output coming but hasn't returned XZ_STREAM_END yet. - */ - if (s->temp.size < b->out_size - b->out_pos || s->temp.size == 0) - { - out_start = b->out_pos; - memcpy(b->out + b->out_pos, s->temp.buf, s->temp.size); - b->out_pos += s->temp.size; - - s->ret = xz_dec_lzma2_run(lzma2, b); - if (s->ret != XZ_STREAM_END && (s->ret != XZ_OK || s->single_call)) - return s->ret; - - bcj_apply(s, b->out, &out_start, b->out_pos); - - /* - * As an exception, if the next filter returned XZ_STREAM_END, - * we can do that too, since the last few bytes that remain - * unfiltered are meant to remain unfiltered. - */ - if (s->ret == XZ_STREAM_END) - return XZ_STREAM_END; - - s->temp.size = b->out_pos - out_start; - b->out_pos -= s->temp.size; - memcpy(s->temp.buf, b->out + b->out_pos, s->temp.size); - - /* - * If there wasn't enough input to the next filter to fill - * the output buffer with unfiltered data, there's no point - * to try decoding more data to temp. - */ - if (b->out_pos + s->temp.size < b->out_size) - return XZ_OK; - } - - /* - * We have unfiltered data in temp. If the output buffer isn't full - * yet, try to fill the temp buffer by decoding more data from the - * next filter. Apply the BCJ filter on temp. Then we hopefully can - * fill the actual output buffer by copying filtered data from temp. - * A mix of filtered and unfiltered data may be left in temp; it will - * be taken care on the next call to this function. - */ - if (b->out_pos < b->out_size) - { - /* Make b->out{,_pos,_size} temporarily point to s->temp. */ - s->out = b->out; - s->out_pos = b->out_pos; - s->out_size = b->out_size; - b->out = s->temp.buf; - b->out_pos = s->temp.size; - b->out_size = sizeof(s->temp.buf); - - s->ret = xz_dec_lzma2_run(lzma2, b); - - s->temp.size = b->out_pos; - b->out = s->out; - b->out_pos = s->out_pos; - b->out_size = s->out_size; - - if (s->ret != XZ_OK && s->ret != XZ_STREAM_END) - return s->ret; - - bcj_apply(s, s->temp.buf, &s->temp.filtered, s->temp.size); - - /* - * If the next filter returned XZ_STREAM_END, we mark that - * everything is filtered, since the last unfiltered bytes - * of the stream are meant to be left as is. - */ - if (s->ret == XZ_STREAM_END) - s->temp.filtered = s->temp.size; - - bcj_flush(s, b); - if (s->temp.filtered > 0) - return XZ_OK; - } - - return s->ret; + size_t out_start; + + /* + * Flush pending already filtered data to the output buffer. Return + * immediatelly if we couldn't flush everything, or if the next + * filter in the chain had already returned XZ_STREAM_END. + */ + if (s->temp.filtered > 0) + { + bcj_flush(s, b); + if (s->temp.filtered > 0) + return XZ_OK; + + if (s->ret == XZ_STREAM_END) + return XZ_STREAM_END; + } + + /* + * If we have more output space than what is currently pending in + * temp, copy the unfiltered data from temp to the output buffer + * and try to fill the output buffer by decoding more data from the + * next filter in the chain. Apply the BCJ filter on the new data + * in the output buffer. If everything cannot be filtered, copy it + * to temp and rewind the output buffer position accordingly. + * + * This needs to be always run when temp.size == 0 to handle a special + * case where the output buffer is full and the next filter has no + * more output coming but hasn't returned XZ_STREAM_END yet. + */ + if (s->temp.size < b->out_size - b->out_pos || s->temp.size == 0) + { + out_start = b->out_pos; + memcpy(b->out + b->out_pos, s->temp.buf, s->temp.size); + b->out_pos += s->temp.size; + + s->ret = xz_dec_lzma2_run(lzma2, b); + if (s->ret != XZ_STREAM_END && (s->ret != XZ_OK || s->single_call)) + return s->ret; + + bcj_apply(s, b->out, &out_start, b->out_pos); + + /* + * As an exception, if the next filter returned XZ_STREAM_END, + * we can do that too, since the last few bytes that remain + * unfiltered are meant to remain unfiltered. + */ + if (s->ret == XZ_STREAM_END) + return XZ_STREAM_END; + + s->temp.size = b->out_pos - out_start; + b->out_pos -= s->temp.size; + memcpy(s->temp.buf, b->out + b->out_pos, s->temp.size); + + /* + * If there wasn't enough input to the next filter to fill + * the output buffer with unfiltered data, there's no point + * to try decoding more data to temp. + */ + if (b->out_pos + s->temp.size < b->out_size) + return XZ_OK; + } + + /* + * We have unfiltered data in temp. If the output buffer isn't full + * yet, try to fill the temp buffer by decoding more data from the + * next filter. Apply the BCJ filter on temp. Then we hopefully can + * fill the actual output buffer by copying filtered data from temp. + * A mix of filtered and unfiltered data may be left in temp; it will + * be taken care on the next call to this function. + */ + if (b->out_pos < b->out_size) + { + /* Make b->out{,_pos,_size} temporarily point to s->temp. */ + s->out = b->out; + s->out_pos = b->out_pos; + s->out_size = b->out_size; + b->out = s->temp.buf; + b->out_pos = s->temp.size; + b->out_size = sizeof(s->temp.buf); + + s->ret = xz_dec_lzma2_run(lzma2, b); + + s->temp.size = b->out_pos; + b->out = s->out; + b->out_pos = s->out_pos; + b->out_size = s->out_size; + + if (s->ret != XZ_OK && s->ret != XZ_STREAM_END) + return s->ret; + + bcj_apply(s, s->temp.buf, &s->temp.filtered, s->temp.size); + + /* + * If the next filter returned XZ_STREAM_END, we mark that + * everything is filtered, since the last unfiltered bytes + * of the stream are meant to be left as is. + */ + if (s->ret == XZ_STREAM_END) + s->temp.filtered = s->temp.size; + + bcj_flush(s, b); + if (s->temp.filtered > 0) + return XZ_OK; + } + + return s->ret; } XZ_EXTERN struct xz_dec_bcj *xz_dec_bcj_create(bool single_call) { - struct xz_dec_bcj *s = kmalloc(sizeof(*s), GFP_KERNEL); - if (s != NULL) - s->single_call = single_call; + struct xz_dec_bcj *s = kmalloc(sizeof(*s), GFP_KERNEL); + if (s != NULL) + s->single_call = single_call; - return s; + return s; } XZ_EXTERN enum xz_ret xz_dec_bcj_reset(struct xz_dec_bcj *s, uint8_t id) { - switch (id) - { + switch (id) + { #ifdef XZ_DEC_X86 - case BCJ_X86: + case BCJ_X86: #endif #ifdef XZ_DEC_POWERPC - case BCJ_POWERPC: + case BCJ_POWERPC: #endif #ifdef XZ_DEC_IA64 - case BCJ_IA64: + case BCJ_IA64: #endif #ifdef XZ_DEC_ARM - case BCJ_ARM: + case BCJ_ARM: #endif #ifdef XZ_DEC_ARMTHUMB - case BCJ_ARMTHUMB: + case BCJ_ARMTHUMB: #endif #ifdef XZ_DEC_SPARC - case BCJ_SPARC: + case BCJ_SPARC: #endif - break; + break; - default: - /* Unsupported Filter ID */ - return XZ_OPTIONS_ERROR; - } + default: + /* Unsupported Filter ID */ + return XZ_OPTIONS_ERROR; + } - s->type = id; - s->ret = XZ_OK; - s->pos = 0; - s->x86_prev_mask = 0; - s->temp.filtered = 0; - s->temp.size = 0; + s->type = id; + s->ret = XZ_OK; + s->pos = 0; + s->x86_prev_mask = 0; + s->temp.filtered = 0; + s->temp.size = 0; - return XZ_OK; + return XZ_OK; } #endif diff --git a/libraries/xz-embedded/src/xz_dec_lzma2.c b/libraries/xz-embedded/src/xz_dec_lzma2.c index 3d7b9a2e..365ace2b 100644 --- a/libraries/xz-embedded/src/xz_dec_lzma2.c +++ b/libraries/xz-embedded/src/xz_dec_lzma2.c @@ -43,244 +43,244 @@ */ struct dictionary { - /* Beginning of the history buffer */ - uint8_t *buf; - - /* Old position in buf (before decoding more data) */ - size_t start; - - /* Position in buf */ - size_t pos; - - /* - * How full dictionary is. This is used to detect corrupt input that - * would read beyond the beginning of the uncompressed stream. - */ - size_t full; - - /* Write limit; we don't write to buf[limit] or later bytes. */ - size_t limit; - - /* - * End of the dictionary buffer. In multi-call mode, this is - * the same as the dictionary size. In single-call mode, this - * indicates the size of the output buffer. - */ - size_t end; - - /* - * Size of the dictionary as specified in Block Header. This is used - * together with "full" to detect corrupt input that would make us - * read beyond the beginning of the uncompressed stream. - */ - uint32_t size; - - /* - * Maximum allowed dictionary size in multi-call mode. - * This is ignored in single-call mode. - */ - uint32_t size_max; - - /* - * Amount of memory currently allocated for the dictionary. - * This is used only with XZ_DYNALLOC. (With XZ_PREALLOC, - * size_max is always the same as the allocated size.) - */ - uint32_t allocated; - - /* Operation mode */ - enum xz_mode mode; + /* Beginning of the history buffer */ + uint8_t *buf; + + /* Old position in buf (before decoding more data) */ + size_t start; + + /* Position in buf */ + size_t pos; + + /* + * How full dictionary is. This is used to detect corrupt input that + * would read beyond the beginning of the uncompressed stream. + */ + size_t full; + + /* Write limit; we don't write to buf[limit] or later bytes. */ + size_t limit; + + /* + * End of the dictionary buffer. In multi-call mode, this is + * the same as the dictionary size. In single-call mode, this + * indicates the size of the output buffer. + */ + size_t end; + + /* + * Size of the dictionary as specified in Block Header. This is used + * together with "full" to detect corrupt input that would make us + * read beyond the beginning of the uncompressed stream. + */ + uint32_t size; + + /* + * Maximum allowed dictionary size in multi-call mode. + * This is ignored in single-call mode. + */ + uint32_t size_max; + + /* + * Amount of memory currently allocated for the dictionary. + * This is used only with XZ_DYNALLOC. (With XZ_PREALLOC, + * size_max is always the same as the allocated size.) + */ + uint32_t allocated; + + /* Operation mode */ + enum xz_mode mode; }; /* Range decoder */ struct rc_dec { - uint32_t range; - uint32_t code; - - /* - * Number of initializing bytes remaining to be read - * by rc_read_init(). - */ - uint32_t init_bytes_left; - - /* - * Buffer from which we read our input. It can be either - * temp.buf or the caller-provided input buffer. - */ - const uint8_t *in; - size_t in_pos; - size_t in_limit; + uint32_t range; + uint32_t code; + + /* + * Number of initializing bytes remaining to be read + * by rc_read_init(). + */ + uint32_t init_bytes_left; + + /* + * Buffer from which we read our input. It can be either + * temp.buf or the caller-provided input buffer. + */ + const uint8_t *in; + size_t in_pos; + size_t in_limit; }; /* Probabilities for a length decoder. */ struct lzma_len_dec { - /* Probability of match length being at least 10 */ - uint16_t choice; + /* Probability of match length being at least 10 */ + uint16_t choice; - /* Probability of match length being at least 18 */ - uint16_t choice2; + /* Probability of match length being at least 18 */ + uint16_t choice2; - /* Probabilities for match lengths 2-9 */ - uint16_t low[POS_STATES_MAX][LEN_LOW_SYMBOLS]; + /* Probabilities for match lengths 2-9 */ + uint16_t low[POS_STATES_MAX][LEN_LOW_SYMBOLS]; - /* Probabilities for match lengths 10-17 */ - uint16_t mid[POS_STATES_MAX][LEN_MID_SYMBOLS]; + /* Probabilities for match lengths 10-17 */ + uint16_t mid[POS_STATES_MAX][LEN_MID_SYMBOLS]; - /* Probabilities for match lengths 18-273 */ - uint16_t high[LEN_HIGH_SYMBOLS]; + /* Probabilities for match lengths 18-273 */ + uint16_t high[LEN_HIGH_SYMBOLS]; }; struct lzma_dec { - /* Distances of latest four matches */ - uint32_t rep0; - uint32_t rep1; - uint32_t rep2; - uint32_t rep3; - - /* Types of the most recently seen LZMA symbols */ - enum lzma_state state; - - /* - * Length of a match. This is updated so that dict_repeat can - * be called again to finish repeating the whole match. - */ - uint32_t len; - - /* - * LZMA properties or related bit masks (number of literal - * context bits, a mask dervied from the number of literal - * position bits, and a mask dervied from the number - * position bits) - */ - uint32_t lc; - uint32_t literal_pos_mask; /* (1 << lp) - 1 */ - uint32_t pos_mask; /* (1 << pb) - 1 */ - - /* If 1, it's a match. Otherwise it's a single 8-bit literal. */ - uint16_t is_match[STATES][POS_STATES_MAX]; - - /* If 1, it's a repeated match. The distance is one of rep0 .. rep3. */ - uint16_t is_rep[STATES]; - - /* - * If 0, distance of a repeated match is rep0. - * Otherwise check is_rep1. - */ - uint16_t is_rep0[STATES]; - - /* - * If 0, distance of a repeated match is rep1. - * Otherwise check is_rep2. - */ - uint16_t is_rep1[STATES]; - - /* If 0, distance of a repeated match is rep2. Otherwise it is rep3. */ - uint16_t is_rep2[STATES]; - - /* - * If 1, the repeated match has length of one byte. Otherwise - * the length is decoded from rep_len_decoder. - */ - uint16_t is_rep0_long[STATES][POS_STATES_MAX]; - - /* - * Probability tree for the highest two bits of the match - * distance. There is a separate probability tree for match - * lengths of 2 (i.e. MATCH_LEN_MIN), 3, 4, and [5, 273]. - */ - uint16_t dist_slot[DIST_STATES][DIST_SLOTS]; - - /* - * Probility trees for additional bits for match distance - * when the distance is in the range [4, 127]. - */ - uint16_t dist_special[FULL_DISTANCES - DIST_MODEL_END]; - - /* - * Probability tree for the lowest four bits of a match - * distance that is equal to or greater than 128. - */ - uint16_t dist_align[ALIGN_SIZE]; - - /* Length of a normal match */ - struct lzma_len_dec match_len_dec; - - /* Length of a repeated match */ - struct lzma_len_dec rep_len_dec; - - /* Probabilities of literals */ - uint16_t literal[LITERAL_CODERS_MAX][LITERAL_CODER_SIZE]; + /* Distances of latest four matches */ + uint32_t rep0; + uint32_t rep1; + uint32_t rep2; + uint32_t rep3; + + /* Types of the most recently seen LZMA symbols */ + enum lzma_state state; + + /* + * Length of a match. This is updated so that dict_repeat can + * be called again to finish repeating the whole match. + */ + uint32_t len; + + /* + * LZMA properties or related bit masks (number of literal + * context bits, a mask dervied from the number of literal + * position bits, and a mask dervied from the number + * position bits) + */ + uint32_t lc; + uint32_t literal_pos_mask; /* (1 << lp) - 1 */ + uint32_t pos_mask; /* (1 << pb) - 1 */ + + /* If 1, it's a match. Otherwise it's a single 8-bit literal. */ + uint16_t is_match[STATES][POS_STATES_MAX]; + + /* If 1, it's a repeated match. The distance is one of rep0 .. rep3. */ + uint16_t is_rep[STATES]; + + /* + * If 0, distance of a repeated match is rep0. + * Otherwise check is_rep1. + */ + uint16_t is_rep0[STATES]; + + /* + * If 0, distance of a repeated match is rep1. + * Otherwise check is_rep2. + */ + uint16_t is_rep1[STATES]; + + /* If 0, distance of a repeated match is rep2. Otherwise it is rep3. */ + uint16_t is_rep2[STATES]; + + /* + * If 1, the repeated match has length of one byte. Otherwise + * the length is decoded from rep_len_decoder. + */ + uint16_t is_rep0_long[STATES][POS_STATES_MAX]; + + /* + * Probability tree for the highest two bits of the match + * distance. There is a separate probability tree for match + * lengths of 2 (i.e. MATCH_LEN_MIN), 3, 4, and [5, 273]. + */ + uint16_t dist_slot[DIST_STATES][DIST_SLOTS]; + + /* + * Probility trees for additional bits for match distance + * when the distance is in the range [4, 127]. + */ + uint16_t dist_special[FULL_DISTANCES - DIST_MODEL_END]; + + /* + * Probability tree for the lowest four bits of a match + * distance that is equal to or greater than 128. + */ + uint16_t dist_align[ALIGN_SIZE]; + + /* Length of a normal match */ + struct lzma_len_dec match_len_dec; + + /* Length of a repeated match */ + struct lzma_len_dec rep_len_dec; + + /* Probabilities of literals */ + uint16_t literal[LITERAL_CODERS_MAX][LITERAL_CODER_SIZE]; }; struct lzma2_dec { - /* Position in xz_dec_lzma2_run(). */ - enum lzma2_seq - { - SEQ_CONTROL, - SEQ_UNCOMPRESSED_1, - SEQ_UNCOMPRESSED_2, - SEQ_COMPRESSED_0, - SEQ_COMPRESSED_1, - SEQ_PROPERTIES, - SEQ_LZMA_PREPARE, - SEQ_LZMA_RUN, - SEQ_COPY - } sequence; - - /* Next position after decoding the compressed size of the chunk. */ - enum lzma2_seq next_sequence; - - /* Uncompressed size of LZMA chunk (2 MiB at maximum) */ - uint32_t uncompressed; - - /* - * Compressed size of LZMA chunk or compressed/uncompressed - * size of uncompressed chunk (64 KiB at maximum) - */ - uint32_t compressed; - - /* - * True if dictionary reset is needed. This is false before - * the first chunk (LZMA or uncompressed). - */ - bool need_dict_reset; - - /* - * True if new LZMA properties are needed. This is false - * before the first LZMA chunk. - */ - bool need_props; + /* Position in xz_dec_lzma2_run(). */ + enum lzma2_seq + { + SEQ_CONTROL, + SEQ_UNCOMPRESSED_1, + SEQ_UNCOMPRESSED_2, + SEQ_COMPRESSED_0, + SEQ_COMPRESSED_1, + SEQ_PROPERTIES, + SEQ_LZMA_PREPARE, + SEQ_LZMA_RUN, + SEQ_COPY + } sequence; + + /* Next position after decoding the compressed size of the chunk. */ + enum lzma2_seq next_sequence; + + /* Uncompressed size of LZMA chunk (2 MiB at maximum) */ + uint32_t uncompressed; + + /* + * Compressed size of LZMA chunk or compressed/uncompressed + * size of uncompressed chunk (64 KiB at maximum) + */ + uint32_t compressed; + + /* + * True if dictionary reset is needed. This is false before + * the first chunk (LZMA or uncompressed). + */ + bool need_dict_reset; + + /* + * True if new LZMA properties are needed. This is false + * before the first LZMA chunk. + */ + bool need_props; }; struct xz_dec_lzma2 { - /* - * The order below is important on x86 to reduce code size and - * it shouldn't hurt on other platforms. Everything up to and - * including lzma.pos_mask are in the first 128 bytes on x86-32, - * which allows using smaller instructions to access those - * variables. On x86-64, fewer variables fit into the first 128 - * bytes, but this is still the best order without sacrificing - * the readability by splitting the structures. - */ - struct rc_dec rc; - struct dictionary dict; - struct lzma2_dec lzma2; - struct lzma_dec lzma; - - /* - * Temporary buffer which holds small number of input bytes between - * decoder calls. See lzma2_lzma() for details. - */ - struct - { - uint32_t size; - uint8_t buf[3 * LZMA_IN_REQUIRED]; - } temp; + /* + * The order below is important on x86 to reduce code size and + * it shouldn't hurt on other platforms. Everything up to and + * including lzma.pos_mask are in the first 128 bytes on x86-32, + * which allows using smaller instructions to access those + * variables. On x86-64, fewer variables fit into the first 128 + * bytes, but this is still the best order without sacrificing + * the readability by splitting the structures. + */ + struct rc_dec rc; + struct dictionary dict; + struct lzma2_dec lzma2; + struct lzma_dec lzma; + + /* + * Temporary buffer which holds small number of input bytes between + * decoder calls. See lzma2_lzma() for details. + */ + struct + { + uint32_t size; + uint8_t buf[3 * LZMA_IN_REQUIRED]; + } temp; }; /************** @@ -293,31 +293,31 @@ struct xz_dec_lzma2 */ static void dict_reset(struct dictionary *dict, struct xz_buf *b) { - if (DEC_IS_SINGLE(dict->mode)) - { - dict->buf = b->out + b->out_pos; - dict->end = b->out_size - b->out_pos; - } - - dict->start = 0; - dict->pos = 0; - dict->limit = 0; - dict->full = 0; + if (DEC_IS_SINGLE(dict->mode)) + { + dict->buf = b->out + b->out_pos; + dict->end = b->out_size - b->out_pos; + } + + dict->start = 0; + dict->pos = 0; + dict->limit = 0; + dict->full = 0; } /* Set dictionary write limit */ static void dict_limit(struct dictionary *dict, size_t out_max) { - if (dict->end - dict->pos <= out_max) - dict->limit = dict->end; - else - dict->limit = dict->pos + out_max; + if (dict->end - dict->pos <= out_max) + dict->limit = dict->end; + else + dict->limit = dict->pos + out_max; } /* Return true if at least one byte can be written into the dictionary. */ static inline bool dict_has_space(const struct dictionary *dict) { - return dict->pos < dict->limit; + return dict->pos < dict->limit; } /* @@ -328,12 +328,12 @@ static inline bool dict_has_space(const struct dictionary *dict) */ static inline uint32_t dict_get(const struct dictionary *dict, uint32_t dist) { - size_t offset = dict->pos - dist - 1; + size_t offset = dict->pos - dist - 1; - if (dist >= dict->pos) - offset += dict->end; + if (dist >= dict->pos) + offset += dict->end; - return dict->full > 0 ? dict->buf[offset] : 0; + return dict->full > 0 ? dict->buf[offset] : 0; } /* @@ -341,10 +341,10 @@ static inline uint32_t dict_get(const struct dictionary *dict, uint32_t dist) */ static inline void dict_put(struct dictionary *dict, uint8_t byte) { - dict->buf[dict->pos++] = byte; + dict->buf[dict->pos++] = byte; - if (dict->full < dict->pos) - dict->full = dict->pos; + if (dict->full < dict->pos) + dict->full = dict->pos; } /* @@ -354,66 +354,66 @@ static inline void dict_put(struct dictionary *dict, uint8_t byte) */ static bool dict_repeat(struct dictionary *dict, uint32_t *len, uint32_t dist) { - size_t back; - uint32_t left; + size_t back; + uint32_t left; - if (dist >= dict->full || dist >= dict->size) - return false; + if (dist >= dict->full || dist >= dict->size) + return false; - left = min_t(size_t, dict->limit - dict->pos, *len); - *len -= left; + left = min_t(size_t, dict->limit - dict->pos, *len); + *len -= left; - back = dict->pos - dist - 1; - if (dist >= dict->pos) - back += dict->end; + back = dict->pos - dist - 1; + if (dist >= dict->pos) + back += dict->end; - do - { - dict->buf[dict->pos++] = dict->buf[back++]; - if (back == dict->end) - back = 0; - } while (--left > 0); + do + { + dict->buf[dict->pos++] = dict->buf[back++]; + if (back == dict->end) + back = 0; + } while (--left > 0); - if (dict->full < dict->pos) - dict->full = dict->pos; + if (dict->full < dict->pos) + dict->full = dict->pos; - return true; + return true; } /* Copy uncompressed data as is from input to dictionary and output buffers. */ static void dict_uncompressed(struct dictionary *dict, struct xz_buf *b, uint32_t *left) { - size_t copy_size; + size_t copy_size; - while (*left > 0 && b->in_pos < b->in_size && b->out_pos < b->out_size) - { - copy_size = min(b->in_size - b->in_pos, b->out_size - b->out_pos); - if (copy_size > dict->end - dict->pos) - copy_size = dict->end - dict->pos; - if (copy_size > *left) - copy_size = *left; + while (*left > 0 && b->in_pos < b->in_size && b->out_pos < b->out_size) + { + copy_size = min(b->in_size - b->in_pos, b->out_size - b->out_pos); + if (copy_size > dict->end - dict->pos) + copy_size = dict->end - dict->pos; + if (copy_size > *left) + copy_size = *left; - *left -= copy_size; + *left -= copy_size; - memcpy(dict->buf + dict->pos, b->in + b->in_pos, copy_size); - dict->pos += copy_size; + memcpy(dict->buf + dict->pos, b->in + b->in_pos, copy_size); + dict->pos += copy_size; - if (dict->full < dict->pos) - dict->full = dict->pos; + if (dict->full < dict->pos) + dict->full = dict->pos; - if (DEC_IS_MULTI(dict->mode)) - { - if (dict->pos == dict->end) - dict->pos = 0; + if (DEC_IS_MULTI(dict->mode)) + { + if (dict->pos == dict->end) + dict->pos = 0; - memcpy(b->out + b->out_pos, b->in + b->in_pos, copy_size); - } + memcpy(b->out + b->out_pos, b->in + b->in_pos, copy_size); + } - dict->start = dict->pos; + dict->start = dict->pos; - b->out_pos += copy_size; - b->in_pos += copy_size; - } + b->out_pos += copy_size; + b->in_pos += copy_size; + } } /* @@ -423,19 +423,19 @@ static void dict_uncompressed(struct dictionary *dict, struct xz_buf *b, uint32_ */ static uint32_t dict_flush(struct dictionary *dict, struct xz_buf *b) { - size_t copy_size = dict->pos - dict->start; + size_t copy_size = dict->pos - dict->start; - if (DEC_IS_MULTI(dict->mode)) - { - if (dict->pos == dict->end) - dict->pos = 0; + if (DEC_IS_MULTI(dict->mode)) + { + if (dict->pos == dict->end) + dict->pos = 0; - memcpy(b->out + b->out_pos, dict->buf + dict->start, copy_size); - } + memcpy(b->out + b->out_pos, dict->buf + dict->start, copy_size); + } - dict->start = dict->pos; - b->out_pos += copy_size; - return copy_size; + dict->start = dict->pos; + b->out_pos += copy_size; + return copy_size; } /***************** @@ -445,9 +445,9 @@ static uint32_t dict_flush(struct dictionary *dict, struct xz_buf *b) /* Reset the range decoder. */ static void rc_reset(struct rc_dec *rc) { - rc->range = (uint32_t) - 1; - rc->code = 0; - rc->init_bytes_left = RC_INIT_BYTES; + rc->range = (uint32_t) - 1; + rc->code = 0; + rc->init_bytes_left = RC_INIT_BYTES; } /* @@ -456,22 +456,22 @@ static void rc_reset(struct rc_dec *rc) */ static bool rc_read_init(struct rc_dec *rc, struct xz_buf *b) { - while (rc->init_bytes_left > 0) - { - if (b->in_pos == b->in_size) - return false; + while (rc->init_bytes_left > 0) + { + if (b->in_pos == b->in_size) + return false; - rc->code = (rc->code << 8) + b->in[b->in_pos++]; - --rc->init_bytes_left; - } + rc->code = (rc->code << 8) + b->in[b->in_pos++]; + --rc->init_bytes_left; + } - return true; + return true; } /* Return true if there may not be enough input for the next decoding loop. */ static inline bool rc_limit_exceeded(const struct rc_dec *rc) { - return rc->in_pos > rc->in_limit; + return rc->in_pos > rc->in_limit; } /* @@ -480,17 +480,17 @@ static inline bool rc_limit_exceeded(const struct rc_dec *rc) */ static inline bool rc_is_finished(const struct rc_dec *rc) { - return rc->code == 0; + return rc->code == 0; } /* Read the next input byte if needed. */ static __always_inline void rc_normalize(struct rc_dec *rc) { - if (rc->range < RC_TOP_VALUE) - { - rc->range <<= RC_SHIFT_BITS; - rc->code = (rc->code << RC_SHIFT_BITS) + rc->in[rc->in_pos++]; - } + if (rc->range < RC_TOP_VALUE) + { + rc->range <<= RC_SHIFT_BITS; + rc->code = (rc->code << RC_SHIFT_BITS) + rc->in[rc->in_pos++]; + } } /* @@ -506,79 +506,79 @@ static __always_inline void rc_normalize(struct rc_dec *rc) */ static __always_inline int rc_bit(struct rc_dec *rc, uint16_t *prob) { - uint32_t bound; - int bit; - - rc_normalize(rc); - bound = (rc->range >> RC_BIT_MODEL_TOTAL_BITS) * *prob; - if (rc->code < bound) - { - rc->range = bound; - *prob += (RC_BIT_MODEL_TOTAL - *prob) >> RC_MOVE_BITS; - bit = 0; - } - else - { - rc->range -= bound; - rc->code -= bound; - *prob -= *prob >> RC_MOVE_BITS; - bit = 1; - } - - return bit; + uint32_t bound; + int bit; + + rc_normalize(rc); + bound = (rc->range >> RC_BIT_MODEL_TOTAL_BITS) * *prob; + if (rc->code < bound) + { + rc->range = bound; + *prob += (RC_BIT_MODEL_TOTAL - *prob) >> RC_MOVE_BITS; + bit = 0; + } + else + { + rc->range -= bound; + rc->code -= bound; + *prob -= *prob >> RC_MOVE_BITS; + bit = 1; + } + + return bit; } /* Decode a bittree starting from the most significant bit. */ static __always_inline uint32_t rc_bittree(struct rc_dec *rc, uint16_t *probs, uint32_t limit) { - uint32_t symbol = 1; + uint32_t symbol = 1; - do - { - if (rc_bit(rc, &probs[symbol])) - symbol = (symbol << 1) + 1; - else - symbol <<= 1; - } while (symbol < limit); + do + { + if (rc_bit(rc, &probs[symbol])) + symbol = (symbol << 1) + 1; + else + symbol <<= 1; + } while (symbol < limit); - return symbol; + return symbol; } /* Decode a bittree starting from the least significant bit. */ static __always_inline void rc_bittree_reverse(struct rc_dec *rc, uint16_t *probs, - uint32_t *dest, uint32_t limit) + uint32_t *dest, uint32_t limit) { - uint32_t symbol = 1; - uint32_t i = 0; - - do - { - if (rc_bit(rc, &probs[symbol])) - { - symbol = (symbol << 1) + 1; - *dest += 1 << i; - } - else - { - symbol <<= 1; - } - } while (++i < limit); + uint32_t symbol = 1; + uint32_t i = 0; + + do + { + if (rc_bit(rc, &probs[symbol])) + { + symbol = (symbol << 1) + 1; + *dest += 1 << i; + } + else + { + symbol <<= 1; + } + } while (++i < limit); } /* Decode direct bits (fixed fifty-fifty probability) */ static inline void rc_direct(struct rc_dec *rc, uint32_t *dest, uint32_t limit) { - uint32_t mask; - - do - { - rc_normalize(rc); - rc->range >>= 1; - rc->code -= rc->range; - mask = (uint32_t)0 - (rc->code >> 31); - rc->code += rc->range & mask; - *dest = (*dest << 1) + (mask + 1); - } while (--limit > 0); + uint32_t mask; + + do + { + rc_normalize(rc); + rc->range >>= 1; + rc->code -= rc->range; + mask = (uint32_t)0 - (rc->code >> 31); + rc->code += rc->range & mask; + *dest = (*dest << 1) + (mask + 1); + } while (--limit > 0); } /******** @@ -588,128 +588,128 @@ static inline void rc_direct(struct rc_dec *rc, uint32_t *dest, uint32_t limit) /* Get pointer to literal coder probability array. */ static uint16_t *lzma_literal_probs(struct xz_dec_lzma2 *s) { - uint32_t prev_byte = dict_get(&s->dict, 0); - uint32_t low = prev_byte >> (8 - s->lzma.lc); - uint32_t high = (s->dict.pos & s->lzma.literal_pos_mask) << s->lzma.lc; - return s->lzma.literal[low + high]; + uint32_t prev_byte = dict_get(&s->dict, 0); + uint32_t low = prev_byte >> (8 - s->lzma.lc); + uint32_t high = (s->dict.pos & s->lzma.literal_pos_mask) << s->lzma.lc; + return s->lzma.literal[low + high]; } /* Decode a literal (one 8-bit byte) */ static void lzma_literal(struct xz_dec_lzma2 *s) { - uint16_t *probs; - uint32_t symbol; - uint32_t match_byte; - uint32_t match_bit; - uint32_t offset; - uint32_t i; - - probs = lzma_literal_probs(s); - - if (lzma_state_is_literal(s->lzma.state)) - { - symbol = rc_bittree(&s->rc, probs, 0x100); - } - else - { - symbol = 1; - match_byte = dict_get(&s->dict, s->lzma.rep0) << 1; - offset = 0x100; - - do - { - match_bit = match_byte & offset; - match_byte <<= 1; - i = offset + match_bit + symbol; - - if (rc_bit(&s->rc, &probs[i])) - { - symbol = (symbol << 1) + 1; - offset &= match_bit; - } - else - { - symbol <<= 1; - offset &= ~match_bit; - } - } while (symbol < 0x100); - } - - dict_put(&s->dict, (uint8_t)symbol); - lzma_state_literal(&s->lzma.state); + uint16_t *probs; + uint32_t symbol; + uint32_t match_byte; + uint32_t match_bit; + uint32_t offset; + uint32_t i; + + probs = lzma_literal_probs(s); + + if (lzma_state_is_literal(s->lzma.state)) + { + symbol = rc_bittree(&s->rc, probs, 0x100); + } + else + { + symbol = 1; + match_byte = dict_get(&s->dict, s->lzma.rep0) << 1; + offset = 0x100; + + do + { + match_bit = match_byte & offset; + match_byte <<= 1; + i = offset + match_bit + symbol; + + if (rc_bit(&s->rc, &probs[i])) + { + symbol = (symbol << 1) + 1; + offset &= match_bit; + } + else + { + symbol <<= 1; + offset &= ~match_bit; + } + } while (symbol < 0x100); + } + + dict_put(&s->dict, (uint8_t)symbol); + lzma_state_literal(&s->lzma.state); } /* Decode the length of the match into s->lzma.len. */ static void lzma_len(struct xz_dec_lzma2 *s, struct lzma_len_dec *l, uint32_t pos_state) { - uint16_t *probs; - uint32_t limit; - - if (!rc_bit(&s->rc, &l->choice)) - { - probs = l->low[pos_state]; - limit = LEN_LOW_SYMBOLS; - s->lzma.len = MATCH_LEN_MIN; - } - else - { - if (!rc_bit(&s->rc, &l->choice2)) - { - probs = l->mid[pos_state]; - limit = LEN_MID_SYMBOLS; - s->lzma.len = MATCH_LEN_MIN + LEN_LOW_SYMBOLS; - } - else - { - probs = l->high; - limit = LEN_HIGH_SYMBOLS; - s->lzma.len = MATCH_LEN_MIN + LEN_LOW_SYMBOLS + LEN_MID_SYMBOLS; - } - } - - s->lzma.len += rc_bittree(&s->rc, probs, limit) - limit; + uint16_t *probs; + uint32_t limit; + + if (!rc_bit(&s->rc, &l->choice)) + { + probs = l->low[pos_state]; + limit = LEN_LOW_SYMBOLS; + s->lzma.len = MATCH_LEN_MIN; + } + else + { + if (!rc_bit(&s->rc, &l->choice2)) + { + probs = l->mid[pos_state]; + limit = LEN_MID_SYMBOLS; + s->lzma.len = MATCH_LEN_MIN + LEN_LOW_SYMBOLS; + } + else + { + probs = l->high; + limit = LEN_HIGH_SYMBOLS; + s->lzma.len = MATCH_LEN_MIN + LEN_LOW_SYMBOLS + LEN_MID_SYMBOLS; + } + } + + s->lzma.len += rc_bittree(&s->rc, probs, limit) - limit; } /* Decode a match. The distance will be stored in s->lzma.rep0. */ static void lzma_match(struct xz_dec_lzma2 *s, uint32_t pos_state) { - uint16_t *probs; - uint32_t dist_slot; - uint32_t limit; - - lzma_state_match(&s->lzma.state); - - s->lzma.rep3 = s->lzma.rep2; - s->lzma.rep2 = s->lzma.rep1; - s->lzma.rep1 = s->lzma.rep0; - - lzma_len(s, &s->lzma.match_len_dec, pos_state); - - probs = s->lzma.dist_slot[lzma_get_dist_state(s->lzma.len)]; - dist_slot = rc_bittree(&s->rc, probs, DIST_SLOTS) - DIST_SLOTS; - - if (dist_slot < DIST_MODEL_START) - { - s->lzma.rep0 = dist_slot; - } - else - { - limit = (dist_slot >> 1) - 1; - s->lzma.rep0 = 2 + (dist_slot & 1); - - if (dist_slot < DIST_MODEL_END) - { - s->lzma.rep0 <<= limit; - probs = s->lzma.dist_special + s->lzma.rep0 - dist_slot - 1; - rc_bittree_reverse(&s->rc, probs, &s->lzma.rep0, limit); - } - else - { - rc_direct(&s->rc, &s->lzma.rep0, limit - ALIGN_BITS); - s->lzma.rep0 <<= ALIGN_BITS; - rc_bittree_reverse(&s->rc, s->lzma.dist_align, &s->lzma.rep0, ALIGN_BITS); - } - } + uint16_t *probs; + uint32_t dist_slot; + uint32_t limit; + + lzma_state_match(&s->lzma.state); + + s->lzma.rep3 = s->lzma.rep2; + s->lzma.rep2 = s->lzma.rep1; + s->lzma.rep1 = s->lzma.rep0; + + lzma_len(s, &s->lzma.match_len_dec, pos_state); + + probs = s->lzma.dist_slot[lzma_get_dist_state(s->lzma.len)]; + dist_slot = rc_bittree(&s->rc, probs, DIST_SLOTS) - DIST_SLOTS; + + if (dist_slot < DIST_MODEL_START) + { + s->lzma.rep0 = dist_slot; + } + else + { + limit = (dist_slot >> 1) - 1; + s->lzma.rep0 = 2 + (dist_slot & 1); + + if (dist_slot < DIST_MODEL_END) + { + s->lzma.rep0 <<= limit; + probs = s->lzma.dist_special + s->lzma.rep0 - dist_slot - 1; + rc_bittree_reverse(&s->rc, probs, &s->lzma.rep0, limit); + } + else + { + rc_direct(&s->rc, &s->lzma.rep0, limit - ALIGN_BITS); + s->lzma.rep0 <<= ALIGN_BITS; + rc_bittree_reverse(&s->rc, s->lzma.dist_align, &s->lzma.rep0, ALIGN_BITS); + } + } } /* @@ -718,89 +718,89 @@ static void lzma_match(struct xz_dec_lzma2 *s, uint32_t pos_state) */ static void lzma_rep_match(struct xz_dec_lzma2 *s, uint32_t pos_state) { - uint32_t tmp; - - if (!rc_bit(&s->rc, &s->lzma.is_rep0[s->lzma.state])) - { - if (!rc_bit(&s->rc, &s->lzma.is_rep0_long[s->lzma.state][pos_state])) - { - lzma_state_short_rep(&s->lzma.state); - s->lzma.len = 1; - return; - } - } - else - { - if (!rc_bit(&s->rc, &s->lzma.is_rep1[s->lzma.state])) - { - tmp = s->lzma.rep1; - } - else - { - if (!rc_bit(&s->rc, &s->lzma.is_rep2[s->lzma.state])) - { - tmp = s->lzma.rep2; - } - else - { - tmp = s->lzma.rep3; - s->lzma.rep3 = s->lzma.rep2; - } - - s->lzma.rep2 = s->lzma.rep1; - } - - s->lzma.rep1 = s->lzma.rep0; - s->lzma.rep0 = tmp; - } - - lzma_state_long_rep(&s->lzma.state); - lzma_len(s, &s->lzma.rep_len_dec, pos_state); + uint32_t tmp; + + if (!rc_bit(&s->rc, &s->lzma.is_rep0[s->lzma.state])) + { + if (!rc_bit(&s->rc, &s->lzma.is_rep0_long[s->lzma.state][pos_state])) + { + lzma_state_short_rep(&s->lzma.state); + s->lzma.len = 1; + return; + } + } + else + { + if (!rc_bit(&s->rc, &s->lzma.is_rep1[s->lzma.state])) + { + tmp = s->lzma.rep1; + } + else + { + if (!rc_bit(&s->rc, &s->lzma.is_rep2[s->lzma.state])) + { + tmp = s->lzma.rep2; + } + else + { + tmp = s->lzma.rep3; + s->lzma.rep3 = s->lzma.rep2; + } + + s->lzma.rep2 = s->lzma.rep1; + } + + s->lzma.rep1 = s->lzma.rep0; + s->lzma.rep0 = tmp; + } + + lzma_state_long_rep(&s->lzma.state); + lzma_len(s, &s->lzma.rep_len_dec, pos_state); } /* LZMA decoder core */ static bool lzma_main(struct xz_dec_lzma2 *s) { - uint32_t pos_state; - - /* - * If the dictionary was reached during the previous call, try to - * finish the possibly pending repeat in the dictionary. - */ - if (dict_has_space(&s->dict) && s->lzma.len > 0) - dict_repeat(&s->dict, &s->lzma.len, s->lzma.rep0); - - /* - * Decode more LZMA symbols. One iteration may consume up to - * LZMA_IN_REQUIRED - 1 bytes. - */ - while (dict_has_space(&s->dict) && !rc_limit_exceeded(&s->rc)) - { - pos_state = s->dict.pos & s->lzma.pos_mask; - - if (!rc_bit(&s->rc, &s->lzma.is_match[s->lzma.state][pos_state])) - { - lzma_literal(s); - } - else - { - if (rc_bit(&s->rc, &s->lzma.is_rep[s->lzma.state])) - lzma_rep_match(s, pos_state); - else - lzma_match(s, pos_state); - - if (!dict_repeat(&s->dict, &s->lzma.len, s->lzma.rep0)) - return false; - } - } - - /* - * Having the range decoder always normalized when we are outside - * this function makes it easier to correctly handle end of the chunk. - */ - rc_normalize(&s->rc); - - return true; + uint32_t pos_state; + + /* + * If the dictionary was reached during the previous call, try to + * finish the possibly pending repeat in the dictionary. + */ + if (dict_has_space(&s->dict) && s->lzma.len > 0) + dict_repeat(&s->dict, &s->lzma.len, s->lzma.rep0); + + /* + * Decode more LZMA symbols. One iteration may consume up to + * LZMA_IN_REQUIRED - 1 bytes. + */ + while (dict_has_space(&s->dict) && !rc_limit_exceeded(&s->rc)) + { + pos_state = s->dict.pos & s->lzma.pos_mask; + + if (!rc_bit(&s->rc, &s->lzma.is_match[s->lzma.state][pos_state])) + { + lzma_literal(s); + } + else + { + if (rc_bit(&s->rc, &s->lzma.is_rep[s->lzma.state])) + lzma_rep_match(s, pos_state); + else + lzma_match(s, pos_state); + + if (!dict_repeat(&s->dict, &s->lzma.len, s->lzma.rep0)) + return false; + } + } + + /* + * Having the range decoder always normalized when we are outside + * this function makes it easier to correctly handle end of the chunk. + */ + rc_normalize(&s->rc); + + return true; } /* @@ -809,29 +809,29 @@ static bool lzma_main(struct xz_dec_lzma2 *s) */ static void lzma_reset(struct xz_dec_lzma2 *s) { - uint16_t *probs; - size_t i; - - s->lzma.state = STATE_LIT_LIT; - s->lzma.rep0 = 0; - s->lzma.rep1 = 0; - s->lzma.rep2 = 0; - s->lzma.rep3 = 0; - - /* - * All probabilities are initialized to the same value. This hack - * makes the code smaller by avoiding a separate loop for each - * probability array. - * - * This could be optimized so that only that part of literal - * probabilities that are actually required. In the common case - * we would write 12 KiB less. - */ - probs = s->lzma.is_match[0]; - for (i = 0; i < PROBS_TOTAL; ++i) - probs[i] = RC_BIT_MODEL_TOTAL / 2; - - rc_reset(&s->rc); + uint16_t *probs; + size_t i; + + s->lzma.state = STATE_LIT_LIT; + s->lzma.rep0 = 0; + s->lzma.rep1 = 0; + s->lzma.rep2 = 0; + s->lzma.rep3 = 0; + + /* + * All probabilities are initialized to the same value. This hack + * makes the code smaller by avoiding a separate loop for each + * probability array. + * + * This could be optimized so that only that part of literal + * probabilities that are actually required. In the common case + * we would write 12 KiB less. + */ + probs = s->lzma.is_match[0]; + for (i = 0; i < PROBS_TOTAL; ++i) + probs[i] = RC_BIT_MODEL_TOTAL / 2; + + rc_reset(&s->rc); } /* @@ -841,35 +841,35 @@ static void lzma_reset(struct xz_dec_lzma2 *s) */ static bool lzma_props(struct xz_dec_lzma2 *s, uint8_t props) { - if (props > (4 * 5 + 4) * 9 + 8) - return false; + if (props > (4 * 5 + 4) * 9 + 8) + return false; - s->lzma.pos_mask = 0; - while (props >= 9 * 5) - { - props -= 9 * 5; - ++s->lzma.pos_mask; - } + s->lzma.pos_mask = 0; + while (props >= 9 * 5) + { + props -= 9 * 5; + ++s->lzma.pos_mask; + } - s->lzma.pos_mask = (1 << s->lzma.pos_mask) - 1; + s->lzma.pos_mask = (1 << s->lzma.pos_mask) - 1; - s->lzma.literal_pos_mask = 0; - while (props >= 9) - { - props -= 9; - ++s->lzma.literal_pos_mask; - } + s->lzma.literal_pos_mask = 0; + while (props >= 9) + { + props -= 9; + ++s->lzma.literal_pos_mask; + } - s->lzma.lc = props; + s->lzma.lc = props; - if (s->lzma.lc + s->lzma.literal_pos_mask > 4) - return false; + if (s->lzma.lc + s->lzma.literal_pos_mask > 4) + return false; - s->lzma.literal_pos_mask = (1 << s->lzma.literal_pos_mask) - 1; + s->lzma.literal_pos_mask = (1 << s->lzma.literal_pos_mask) - 1; - lzma_reset(s); + lzma_reset(s); - return true; + return true; } /********* @@ -890,89 +890,89 @@ static bool lzma_props(struct xz_dec_lzma2 *s, uint8_t props) */ static bool lzma2_lzma(struct xz_dec_lzma2 *s, struct xz_buf *b) { - size_t in_avail; - uint32_t tmp; - - in_avail = b->in_size - b->in_pos; - if (s->temp.size > 0 || s->lzma2.compressed == 0) - { - tmp = 2 * LZMA_IN_REQUIRED - s->temp.size; - if (tmp > s->lzma2.compressed - s->temp.size) - tmp = s->lzma2.compressed - s->temp.size; - if (tmp > in_avail) - tmp = in_avail; - - memcpy(s->temp.buf + s->temp.size, b->in + b->in_pos, tmp); - - if (s->temp.size + tmp == s->lzma2.compressed) - { - memzero(s->temp.buf + s->temp.size + tmp, sizeof(s->temp.buf) - s->temp.size - tmp); - s->rc.in_limit = s->temp.size + tmp; - } - else if (s->temp.size + tmp < LZMA_IN_REQUIRED) - { - s->temp.size += tmp; - b->in_pos += tmp; - return true; - } - else - { - s->rc.in_limit = s->temp.size + tmp - LZMA_IN_REQUIRED; - } - - s->rc.in = s->temp.buf; - s->rc.in_pos = 0; - - if (!lzma_main(s) || s->rc.in_pos > s->temp.size + tmp) - return false; - - s->lzma2.compressed -= s->rc.in_pos; - - if (s->rc.in_pos < s->temp.size) - { - s->temp.size -= s->rc.in_pos; - memmove(s->temp.buf, s->temp.buf + s->rc.in_pos, s->temp.size); - return true; - } - - b->in_pos += s->rc.in_pos - s->temp.size; - s->temp.size = 0; - } - - in_avail = b->in_size - b->in_pos; - if (in_avail >= LZMA_IN_REQUIRED) - { - s->rc.in = b->in; - s->rc.in_pos = b->in_pos; - - if (in_avail >= s->lzma2.compressed + LZMA_IN_REQUIRED) - s->rc.in_limit = b->in_pos + s->lzma2.compressed; - else - s->rc.in_limit = b->in_size - LZMA_IN_REQUIRED; - - if (!lzma_main(s)) - return false; - - in_avail = s->rc.in_pos - b->in_pos; - if (in_avail > s->lzma2.compressed) - return false; - - s->lzma2.compressed -= in_avail; - b->in_pos = s->rc.in_pos; - } - - in_avail = b->in_size - b->in_pos; - if (in_avail < LZMA_IN_REQUIRED) - { - if (in_avail > s->lzma2.compressed) - in_avail = s->lzma2.compressed; - - memcpy(s->temp.buf, b->in + b->in_pos, in_avail); - s->temp.size = in_avail; - b->in_pos += in_avail; - } - - return true; + size_t in_avail; + uint32_t tmp; + + in_avail = b->in_size - b->in_pos; + if (s->temp.size > 0 || s->lzma2.compressed == 0) + { + tmp = 2 * LZMA_IN_REQUIRED - s->temp.size; + if (tmp > s->lzma2.compressed - s->temp.size) + tmp = s->lzma2.compressed - s->temp.size; + if (tmp > in_avail) + tmp = in_avail; + + memcpy(s->temp.buf + s->temp.size, b->in + b->in_pos, tmp); + + if (s->temp.size + tmp == s->lzma2.compressed) + { + memzero(s->temp.buf + s->temp.size + tmp, sizeof(s->temp.buf) - s->temp.size - tmp); + s->rc.in_limit = s->temp.size + tmp; + } + else if (s->temp.size + tmp < LZMA_IN_REQUIRED) + { + s->temp.size += tmp; + b->in_pos += tmp; + return true; + } + else + { + s->rc.in_limit = s->temp.size + tmp - LZMA_IN_REQUIRED; + } + + s->rc.in = s->temp.buf; + s->rc.in_pos = 0; + + if (!lzma_main(s) || s->rc.in_pos > s->temp.size + tmp) + return false; + + s->lzma2.compressed -= s->rc.in_pos; + + if (s->rc.in_pos < s->temp.size) + { + s->temp.size -= s->rc.in_pos; + memmove(s->temp.buf, s->temp.buf + s->rc.in_pos, s->temp.size); + return true; + } + + b->in_pos += s->rc.in_pos - s->temp.size; + s->temp.size = 0; + } + + in_avail = b->in_size - b->in_pos; + if (in_avail >= LZMA_IN_REQUIRED) + { + s->rc.in = b->in; + s->rc.in_pos = b->in_pos; + + if (in_avail >= s->lzma2.compressed + LZMA_IN_REQUIRED) + s->rc.in_limit = b->in_pos + s->lzma2.compressed; + else + s->rc.in_limit = b->in_size - LZMA_IN_REQUIRED; + + if (!lzma_main(s)) + return false; + + in_avail = s->rc.in_pos - b->in_pos; + if (in_avail > s->lzma2.compressed) + return false; + + s->lzma2.compressed -= in_avail; + b->in_pos = s->rc.in_pos; + } + + in_avail = b->in_size - b->in_pos; + if (in_avail < LZMA_IN_REQUIRED) + { + if (in_avail > s->lzma2.compressed) + in_avail = s->lzma2.compressed; + + memcpy(s->temp.buf, b->in + b->in_pos, in_avail); + s->temp.size = in_avail; + b->in_pos += in_avail; + } + + return true; } /* @@ -981,251 +981,251 @@ static bool lzma2_lzma(struct xz_dec_lzma2 *s, struct xz_buf *b) */ XZ_EXTERN enum xz_ret xz_dec_lzma2_run(struct xz_dec_lzma2 *s, struct xz_buf *b) { - uint32_t tmp; - - while (b->in_pos < b->in_size || s->lzma2.sequence == SEQ_LZMA_RUN) - { - switch (s->lzma2.sequence) - { - case SEQ_CONTROL: - /* - * LZMA2 control byte - * - * Exact values: - * 0x00 End marker - * 0x01 Dictionary reset followed by - * an uncompressed chunk - * 0x02 Uncompressed chunk (no dictionary reset) - * - * Highest three bits (s->control & 0xE0): - * 0xE0 Dictionary reset, new properties and state - * reset, followed by LZMA compressed chunk - * 0xC0 New properties and state reset, followed - * by LZMA compressed chunk (no dictionary - * reset) - * 0xA0 State reset using old properties, - * followed by LZMA compressed chunk (no - * dictionary reset) - * 0x80 LZMA chunk (no dictionary or state reset) - * - * For LZMA compressed chunks, the lowest five bits - * (s->control & 1F) are the highest bits of the - * uncompressed size (bits 16-20). - * - * A new LZMA2 stream must begin with a dictionary - * reset. The first LZMA chunk must set new - * properties and reset the LZMA state. - * - * Values that don't match anything described above - * are invalid and we return XZ_DATA_ERROR. - */ - tmp = b->in[b->in_pos++]; - - if (tmp == 0x00) - return XZ_STREAM_END; - - if (tmp >= 0xE0 || tmp == 0x01) - { - s->lzma2.need_props = true; - s->lzma2.need_dict_reset = false; - dict_reset(&s->dict, b); - } - else if (s->lzma2.need_dict_reset) - { - return XZ_DATA_ERROR; - } - - if (tmp >= 0x80) - { - s->lzma2.uncompressed = (tmp & 0x1F) << 16; - s->lzma2.sequence = SEQ_UNCOMPRESSED_1; - - if (tmp >= 0xC0) - { - /* - * When there are new properties, - * state reset is done at - * SEQ_PROPERTIES. - */ - s->lzma2.need_props = false; - s->lzma2.next_sequence = SEQ_PROPERTIES; - } - else if (s->lzma2.need_props) - { - return XZ_DATA_ERROR; - } - else - { - s->lzma2.next_sequence = SEQ_LZMA_PREPARE; - if (tmp >= 0xA0) - lzma_reset(s); - } - } - else - { - if (tmp > 0x02) - return XZ_DATA_ERROR; - - s->lzma2.sequence = SEQ_COMPRESSED_0; - s->lzma2.next_sequence = SEQ_COPY; - } - - break; - - case SEQ_UNCOMPRESSED_1: - s->lzma2.uncompressed += (uint32_t)b->in[b->in_pos++] << 8; - s->lzma2.sequence = SEQ_UNCOMPRESSED_2; - break; - - case SEQ_UNCOMPRESSED_2: - s->lzma2.uncompressed += (uint32_t)b->in[b->in_pos++] + 1; - s->lzma2.sequence = SEQ_COMPRESSED_0; - break; - - case SEQ_COMPRESSED_0: - s->lzma2.compressed = (uint32_t)b->in[b->in_pos++] << 8; - s->lzma2.sequence = SEQ_COMPRESSED_1; - break; - - case SEQ_COMPRESSED_1: - s->lzma2.compressed += (uint32_t)b->in[b->in_pos++] + 1; - s->lzma2.sequence = s->lzma2.next_sequence; - break; - - case SEQ_PROPERTIES: - if (!lzma_props(s, b->in[b->in_pos++])) - return XZ_DATA_ERROR; - - s->lzma2.sequence = SEQ_LZMA_PREPARE; - - case SEQ_LZMA_PREPARE: - if (s->lzma2.compressed < RC_INIT_BYTES) - return XZ_DATA_ERROR; - - if (!rc_read_init(&s->rc, b)) - return XZ_OK; - - s->lzma2.compressed -= RC_INIT_BYTES; - s->lzma2.sequence = SEQ_LZMA_RUN; - - case SEQ_LZMA_RUN: - /* - * Set dictionary limit to indicate how much we want - * to be encoded at maximum. Decode new data into the - * dictionary. Flush the new data from dictionary to - * b->out. Check if we finished decoding this chunk. - * In case the dictionary got full but we didn't fill - * the output buffer yet, we may run this loop - * multiple times without changing s->lzma2.sequence. - */ - dict_limit(&s->dict, - min_t(size_t, b->out_size - b->out_pos, s->lzma2.uncompressed)); - if (!lzma2_lzma(s, b)) - return XZ_DATA_ERROR; - - s->lzma2.uncompressed -= dict_flush(&s->dict, b); - - if (s->lzma2.uncompressed == 0) - { - if (s->lzma2.compressed > 0 || s->lzma.len > 0 || !rc_is_finished(&s->rc)) - return XZ_DATA_ERROR; - - rc_reset(&s->rc); - s->lzma2.sequence = SEQ_CONTROL; - } - else if (b->out_pos == b->out_size || - (b->in_pos == b->in_size && s->temp.size < s->lzma2.compressed)) - { - return XZ_OK; - } - - break; - - case SEQ_COPY: - dict_uncompressed(&s->dict, b, &s->lzma2.compressed); - if (s->lzma2.compressed > 0) - return XZ_OK; - - s->lzma2.sequence = SEQ_CONTROL; - break; - } - } - - return XZ_OK; + uint32_t tmp; + + while (b->in_pos < b->in_size || s->lzma2.sequence == SEQ_LZMA_RUN) + { + switch (s->lzma2.sequence) + { + case SEQ_CONTROL: + /* + * LZMA2 control byte + * + * Exact values: + * 0x00 End marker + * 0x01 Dictionary reset followed by + * an uncompressed chunk + * 0x02 Uncompressed chunk (no dictionary reset) + * + * Highest three bits (s->control & 0xE0): + * 0xE0 Dictionary reset, new properties and state + * reset, followed by LZMA compressed chunk + * 0xC0 New properties and state reset, followed + * by LZMA compressed chunk (no dictionary + * reset) + * 0xA0 State reset using old properties, + * followed by LZMA compressed chunk (no + * dictionary reset) + * 0x80 LZMA chunk (no dictionary or state reset) + * + * For LZMA compressed chunks, the lowest five bits + * (s->control & 1F) are the highest bits of the + * uncompressed size (bits 16-20). + * + * A new LZMA2 stream must begin with a dictionary + * reset. The first LZMA chunk must set new + * properties and reset the LZMA state. + * + * Values that don't match anything described above + * are invalid and we return XZ_DATA_ERROR. + */ + tmp = b->in[b->in_pos++]; + + if (tmp == 0x00) + return XZ_STREAM_END; + + if (tmp >= 0xE0 || tmp == 0x01) + { + s->lzma2.need_props = true; + s->lzma2.need_dict_reset = false; + dict_reset(&s->dict, b); + } + else if (s->lzma2.need_dict_reset) + { + return XZ_DATA_ERROR; + } + + if (tmp >= 0x80) + { + s->lzma2.uncompressed = (tmp & 0x1F) << 16; + s->lzma2.sequence = SEQ_UNCOMPRESSED_1; + + if (tmp >= 0xC0) + { + /* + * When there are new properties, + * state reset is done at + * SEQ_PROPERTIES. + */ + s->lzma2.need_props = false; + s->lzma2.next_sequence = SEQ_PROPERTIES; + } + else if (s->lzma2.need_props) + { + return XZ_DATA_ERROR; + } + else + { + s->lzma2.next_sequence = SEQ_LZMA_PREPARE; + if (tmp >= 0xA0) + lzma_reset(s); + } + } + else + { + if (tmp > 0x02) + return XZ_DATA_ERROR; + + s->lzma2.sequence = SEQ_COMPRESSED_0; + s->lzma2.next_sequence = SEQ_COPY; + } + + break; + + case SEQ_UNCOMPRESSED_1: + s->lzma2.uncompressed += (uint32_t)b->in[b->in_pos++] << 8; + s->lzma2.sequence = SEQ_UNCOMPRESSED_2; + break; + + case SEQ_UNCOMPRESSED_2: + s->lzma2.uncompressed += (uint32_t)b->in[b->in_pos++] + 1; + s->lzma2.sequence = SEQ_COMPRESSED_0; + break; + + case SEQ_COMPRESSED_0: + s->lzma2.compressed = (uint32_t)b->in[b->in_pos++] << 8; + s->lzma2.sequence = SEQ_COMPRESSED_1; + break; + + case SEQ_COMPRESSED_1: + s->lzma2.compressed += (uint32_t)b->in[b->in_pos++] + 1; + s->lzma2.sequence = s->lzma2.next_sequence; + break; + + case SEQ_PROPERTIES: + if (!lzma_props(s, b->in[b->in_pos++])) + return XZ_DATA_ERROR; + + s->lzma2.sequence = SEQ_LZMA_PREPARE; + + case SEQ_LZMA_PREPARE: + if (s->lzma2.compressed < RC_INIT_BYTES) + return XZ_DATA_ERROR; + + if (!rc_read_init(&s->rc, b)) + return XZ_OK; + + s->lzma2.compressed -= RC_INIT_BYTES; + s->lzma2.sequence = SEQ_LZMA_RUN; + + case SEQ_LZMA_RUN: + /* + * Set dictionary limit to indicate how much we want + * to be encoded at maximum. Decode new data into the + * dictionary. Flush the new data from dictionary to + * b->out. Check if we finished decoding this chunk. + * In case the dictionary got full but we didn't fill + * the output buffer yet, we may run this loop + * multiple times without changing s->lzma2.sequence. + */ + dict_limit(&s->dict, + min_t(size_t, b->out_size - b->out_pos, s->lzma2.uncompressed)); + if (!lzma2_lzma(s, b)) + return XZ_DATA_ERROR; + + s->lzma2.uncompressed -= dict_flush(&s->dict, b); + + if (s->lzma2.uncompressed == 0) + { + if (s->lzma2.compressed > 0 || s->lzma.len > 0 || !rc_is_finished(&s->rc)) + return XZ_DATA_ERROR; + + rc_reset(&s->rc); + s->lzma2.sequence = SEQ_CONTROL; + } + else if (b->out_pos == b->out_size || + (b->in_pos == b->in_size && s->temp.size < s->lzma2.compressed)) + { + return XZ_OK; + } + + break; + + case SEQ_COPY: + dict_uncompressed(&s->dict, b, &s->lzma2.compressed); + if (s->lzma2.compressed > 0) + return XZ_OK; + + s->lzma2.sequence = SEQ_CONTROL; + break; + } + } + + return XZ_OK; } XZ_EXTERN struct xz_dec_lzma2 *xz_dec_lzma2_create(enum xz_mode mode, uint32_t dict_max) { - struct xz_dec_lzma2 *s = kmalloc(sizeof(*s), GFP_KERNEL); - if (s == NULL) - return NULL; - - s->dict.mode = mode; - s->dict.size_max = dict_max; - - if (DEC_IS_PREALLOC(mode)) - { - s->dict.buf = vmalloc(dict_max); - if (s->dict.buf == NULL) - { - kfree(s); - return NULL; - } - } - else if (DEC_IS_DYNALLOC(mode)) - { - s->dict.buf = NULL; - s->dict.allocated = 0; - } - - return s; + struct xz_dec_lzma2 *s = kmalloc(sizeof(*s), GFP_KERNEL); + if (s == NULL) + return NULL; + + s->dict.mode = mode; + s->dict.size_max = dict_max; + + if (DEC_IS_PREALLOC(mode)) + { + s->dict.buf = vmalloc(dict_max); + if (s->dict.buf == NULL) + { + kfree(s); + return NULL; + } + } + else if (DEC_IS_DYNALLOC(mode)) + { + s->dict.buf = NULL; + s->dict.allocated = 0; + } + + return s; } XZ_EXTERN enum xz_ret xz_dec_lzma2_reset(struct xz_dec_lzma2 *s, uint8_t props) { - /* This limits dictionary size to 3 GiB to keep parsing simpler. */ - if (props > 39) - return XZ_OPTIONS_ERROR; - - s->dict.size = 2 + (props & 1); - s->dict.size <<= (props >> 1) + 11; - - if (DEC_IS_MULTI(s->dict.mode)) - { - if (s->dict.size > s->dict.size_max) - return XZ_MEMLIMIT_ERROR; - - s->dict.end = s->dict.size; - - if (DEC_IS_DYNALLOC(s->dict.mode)) - { - if (s->dict.allocated < s->dict.size) - { - vfree(s->dict.buf); - s->dict.buf = vmalloc(s->dict.size); - if (s->dict.buf == NULL) - { - s->dict.allocated = 0; - return XZ_MEM_ERROR; - } - } - } - } - - s->lzma.len = 0; - - s->lzma2.sequence = SEQ_CONTROL; - s->lzma2.need_dict_reset = true; - - s->temp.size = 0; - - return XZ_OK; + /* This limits dictionary size to 3 GiB to keep parsing simpler. */ + if (props > 39) + return XZ_OPTIONS_ERROR; + + s->dict.size = 2 + (props & 1); + s->dict.size <<= (props >> 1) + 11; + + if (DEC_IS_MULTI(s->dict.mode)) + { + if (s->dict.size > s->dict.size_max) + return XZ_MEMLIMIT_ERROR; + + s->dict.end = s->dict.size; + + if (DEC_IS_DYNALLOC(s->dict.mode)) + { + if (s->dict.allocated < s->dict.size) + { + vfree(s->dict.buf); + s->dict.buf = vmalloc(s->dict.size); + if (s->dict.buf == NULL) + { + s->dict.allocated = 0; + return XZ_MEM_ERROR; + } + } + } + } + + s->lzma.len = 0; + + s->lzma2.sequence = SEQ_CONTROL; + s->lzma2.need_dict_reset = true; + + s->temp.size = 0; + + return XZ_OK; } XZ_EXTERN void xz_dec_lzma2_end(struct xz_dec_lzma2 *s) { - if (DEC_IS_MULTI(s->dict.mode)) - vfree(s->dict.buf); + if (DEC_IS_MULTI(s->dict.mode)) + vfree(s->dict.buf); - kfree(s); + kfree(s); } diff --git a/libraries/xz-embedded/src/xz_dec_stream.c b/libraries/xz-embedded/src/xz_dec_stream.c index 6e935ded..f8d7be05 100644 --- a/libraries/xz-embedded/src/xz_dec_stream.c +++ b/libraries/xz-embedded/src/xz_dec_stream.c @@ -19,146 +19,146 @@ /* Hash used to validate the Index field */ struct xz_dec_hash { - vli_type unpadded; - vli_type uncompressed; - uint32_t crc32; + vli_type unpadded; + vli_type uncompressed; + uint32_t crc32; }; struct xz_dec { - /* Position in dec_main() */ - enum - { - SEQ_STREAM_HEADER, - SEQ_BLOCK_START, - SEQ_BLOCK_HEADER, - SEQ_BLOCK_UNCOMPRESS, - SEQ_BLOCK_PADDING, - SEQ_BLOCK_CHECK, - SEQ_INDEX, - SEQ_INDEX_PADDING, - SEQ_INDEX_CRC32, - SEQ_STREAM_FOOTER - } sequence; - - /* Position in variable-length integers and Check fields */ - uint32_t pos; - - /* Variable-length integer decoded by dec_vli() */ - vli_type vli; - - /* Saved in_pos and out_pos */ - size_t in_start; - size_t out_start; + /* Position in dec_main() */ + enum + { + SEQ_STREAM_HEADER, + SEQ_BLOCK_START, + SEQ_BLOCK_HEADER, + SEQ_BLOCK_UNCOMPRESS, + SEQ_BLOCK_PADDING, + SEQ_BLOCK_CHECK, + SEQ_INDEX, + SEQ_INDEX_PADDING, + SEQ_INDEX_CRC32, + SEQ_STREAM_FOOTER + } sequence; + + /* Position in variable-length integers and Check fields */ + uint32_t pos; + + /* Variable-length integer decoded by dec_vli() */ + vli_type vli; + + /* Saved in_pos and out_pos */ + size_t in_start; + size_t out_start; #ifdef XZ_USE_CRC64 - /* CRC32 or CRC64 value in Block or CRC32 value in Index */ - uint64_t crc; + /* CRC32 or CRC64 value in Block or CRC32 value in Index */ + uint64_t crc; #else - /* CRC32 value in Block or Index */ - uint32_t crc; + /* CRC32 value in Block or Index */ + uint32_t crc; #endif - /* Type of the integrity check calculated from uncompressed data */ - enum xz_check check_type; - - /* Operation mode */ - enum xz_mode mode; - - /* - * True if the next call to xz_dec_run() is allowed to return - * XZ_BUF_ERROR. - */ - bool allow_buf_error; - - /* Information stored in Block Header */ - struct - { - /* - * Value stored in the Compressed Size field, or - * VLI_UNKNOWN if Compressed Size is not present. - */ - vli_type compressed; - - /* - * Value stored in the Uncompressed Size field, or - * VLI_UNKNOWN if Uncompressed Size is not present. - */ - vli_type uncompressed; - - /* Size of the Block Header field */ - uint32_t size; - } block_header; - - /* Information collected when decoding Blocks */ - struct - { - /* Observed compressed size of the current Block */ - vli_type compressed; - - /* Observed uncompressed size of the current Block */ - vli_type uncompressed; - - /* Number of Blocks decoded so far */ - vli_type count; - - /* - * Hash calculated from the Block sizes. This is used to - * validate the Index field. - */ - struct xz_dec_hash hash; - } block; - - /* Variables needed when verifying the Index field */ - struct - { - /* Position in dec_index() */ - enum - { - SEQ_INDEX_COUNT, - SEQ_INDEX_UNPADDED, - SEQ_INDEX_UNCOMPRESSED - } sequence; - - /* Size of the Index in bytes */ - vli_type size; - - /* Number of Records (matches block.count in valid files) */ - vli_type count; - - /* - * Hash calculated from the Records (matches block.hash in - * valid files). - */ - struct xz_dec_hash hash; - } index; - - /* - * Temporary buffer needed to hold Stream Header, Block Header, - * and Stream Footer. The Block Header is the biggest (1 KiB) - * so we reserve space according to that. buf[] has to be aligned - * to a multiple of four bytes; the size_t variables before it - * should guarantee this. - */ - struct - { - size_t pos; - size_t size; - uint8_t buf[1024]; - } temp; - - struct xz_dec_lzma2 *lzma2; + /* Type of the integrity check calculated from uncompressed data */ + enum xz_check check_type; + + /* Operation mode */ + enum xz_mode mode; + + /* + * True if the next call to xz_dec_run() is allowed to return + * XZ_BUF_ERROR. + */ + bool allow_buf_error; + + /* Information stored in Block Header */ + struct + { + /* + * Value stored in the Compressed Size field, or + * VLI_UNKNOWN if Compressed Size is not present. + */ + vli_type compressed; + + /* + * Value stored in the Uncompressed Size field, or + * VLI_UNKNOWN if Uncompressed Size is not present. + */ + vli_type uncompressed; + + /* Size of the Block Header field */ + uint32_t size; + } block_header; + + /* Information collected when decoding Blocks */ + struct + { + /* Observed compressed size of the current Block */ + vli_type compressed; + + /* Observed uncompressed size of the current Block */ + vli_type uncompressed; + + /* Number of Blocks decoded so far */ + vli_type count; + + /* + * Hash calculated from the Block sizes. This is used to + * validate the Index field. + */ + struct xz_dec_hash hash; + } block; + + /* Variables needed when verifying the Index field */ + struct + { + /* Position in dec_index() */ + enum + { + SEQ_INDEX_COUNT, + SEQ_INDEX_UNPADDED, + SEQ_INDEX_UNCOMPRESSED + } sequence; + + /* Size of the Index in bytes */ + vli_type size; + + /* Number of Records (matches block.count in valid files) */ + vli_type count; + + /* + * Hash calculated from the Records (matches block.hash in + * valid files). + */ + struct xz_dec_hash hash; + } index; + + /* + * Temporary buffer needed to hold Stream Header, Block Header, + * and Stream Footer. The Block Header is the biggest (1 KiB) + * so we reserve space according to that. buf[] has to be aligned + * to a multiple of four bytes; the size_t variables before it + * should guarantee this. + */ + struct + { + size_t pos; + size_t size; + uint8_t buf[1024]; + } temp; + + struct xz_dec_lzma2 *lzma2; #ifdef XZ_DEC_BCJ - struct xz_dec_bcj *bcj; - bool bcj_active; + struct xz_dec_bcj *bcj; + bool bcj_active; #endif }; #ifdef XZ_DEC_ANY_CHECK /* Sizes of the Check field with different Check IDs */ static const uint8_t check_sizes[16] = {0, 4, 4, 4, 8, 8, 8, 16, - 16, 16, 32, 32, 32, 64, 64, 64}; + 16, 16, 32, 32, 32, 64, 64, 64}; #endif /* @@ -169,52 +169,52 @@ static const uint8_t check_sizes[16] = {0, 4, 4, 4, 8, 8, 8, 16, */ static bool fill_temp(struct xz_dec *s, struct xz_buf *b) { - size_t copy_size = min_t(size_t, b->in_size - b->in_pos, s->temp.size - s->temp.pos); + size_t copy_size = min_t(size_t, b->in_size - b->in_pos, s->temp.size - s->temp.pos); - memcpy(s->temp.buf + s->temp.pos, b->in + b->in_pos, copy_size); - b->in_pos += copy_size; - s->temp.pos += copy_size; + memcpy(s->temp.buf + s->temp.pos, b->in + b->in_pos, copy_size); + b->in_pos += copy_size; + s->temp.pos += copy_size; - if (s->temp.pos == s->temp.size) - { - s->temp.pos = 0; - return true; - } + if (s->temp.pos == s->temp.size) + { + s->temp.pos = 0; + return true; + } - return false; + return false; } /* Decode a variable-length integer (little-endian base-128 encoding) */ static enum xz_ret dec_vli(struct xz_dec *s, const uint8_t *in, size_t *in_pos, size_t in_size) { - uint8_t byte; + uint8_t byte; - if (s->pos == 0) - s->vli = 0; + if (s->pos == 0) + s->vli = 0; - while (*in_pos < in_size) - { - byte = in[*in_pos]; - ++*in_pos; + while (*in_pos < in_size) + { + byte = in[*in_pos]; + ++*in_pos; - s->vli |= (vli_type)(byte & 0x7F) << s->pos; + s->vli |= (vli_type)(byte & 0x7F) << s->pos; - if ((byte & 0x80) == 0) - { - /* Don't allow non-minimal encodings. */ - if (byte == 0 && s->pos != 0) - return XZ_DATA_ERROR; + if ((byte & 0x80) == 0) + { + /* Don't allow non-minimal encodings. */ + if (byte == 0 && s->pos != 0) + return XZ_DATA_ERROR; - s->pos = 0; - return XZ_STREAM_END; - } + s->pos = 0; + return XZ_STREAM_END; + } - s->pos += 7; - if (s->pos == 7 * VLI_BYTES_MAX) - return XZ_DATA_ERROR; - } + s->pos += 7; + if (s->pos == 7 * VLI_BYTES_MAX) + return XZ_DATA_ERROR; + } - return XZ_OK; + return XZ_OK; } /* @@ -231,73 +231,73 @@ static enum xz_ret dec_vli(struct xz_dec *s, const uint8_t *in, size_t *in_pos, */ static enum xz_ret dec_block(struct xz_dec *s, struct xz_buf *b) { - enum xz_ret ret; + enum xz_ret ret; - s->in_start = b->in_pos; - s->out_start = b->out_pos; + s->in_start = b->in_pos; + s->out_start = b->out_pos; #ifdef XZ_DEC_BCJ - if (s->bcj_active) - ret = xz_dec_bcj_run(s->bcj, s->lzma2, b); - else + if (s->bcj_active) + ret = xz_dec_bcj_run(s->bcj, s->lzma2, b); + else #endif - ret = xz_dec_lzma2_run(s->lzma2, b); + ret = xz_dec_lzma2_run(s->lzma2, b); - s->block.compressed += b->in_pos - s->in_start; - s->block.uncompressed += b->out_pos - s->out_start; + s->block.compressed += b->in_pos - s->in_start; + s->block.uncompressed += b->out_pos - s->out_start; - /* - * There is no need to separately check for VLI_UNKNOWN, since - * the observed sizes are always smaller than VLI_UNKNOWN. - */ - if (s->block.compressed > s->block_header.compressed || - s->block.uncompressed > s->block_header.uncompressed) - return XZ_DATA_ERROR; + /* + * There is no need to separately check for VLI_UNKNOWN, since + * the observed sizes are always smaller than VLI_UNKNOWN. + */ + if (s->block.compressed > s->block_header.compressed || + s->block.uncompressed > s->block_header.uncompressed) + return XZ_DATA_ERROR; - if (s->check_type == XZ_CHECK_CRC32) - s->crc = xz_crc32(b->out + s->out_start, b->out_pos - s->out_start, s->crc); + if (s->check_type == XZ_CHECK_CRC32) + s->crc = xz_crc32(b->out + s->out_start, b->out_pos - s->out_start, s->crc); #ifdef XZ_USE_CRC64 - else if (s->check_type == XZ_CHECK_CRC64) - s->crc = xz_crc64(b->out + s->out_start, b->out_pos - s->out_start, s->crc); + else if (s->check_type == XZ_CHECK_CRC64) + s->crc = xz_crc64(b->out + s->out_start, b->out_pos - s->out_start, s->crc); #endif - if (ret == XZ_STREAM_END) - { - if (s->block_header.compressed != VLI_UNKNOWN && - s->block_header.compressed != s->block.compressed) - return XZ_DATA_ERROR; + if (ret == XZ_STREAM_END) + { + if (s->block_header.compressed != VLI_UNKNOWN && + s->block_header.compressed != s->block.compressed) + return XZ_DATA_ERROR; - if (s->block_header.uncompressed != VLI_UNKNOWN && - s->block_header.uncompressed != s->block.uncompressed) - return XZ_DATA_ERROR; + if (s->block_header.uncompressed != VLI_UNKNOWN && + s->block_header.uncompressed != s->block.uncompressed) + return XZ_DATA_ERROR; - s->block.hash.unpadded += s->block_header.size + s->block.compressed; + s->block.hash.unpadded += s->block_header.size + s->block.compressed; #ifdef XZ_DEC_ANY_CHECK - s->block.hash.unpadded += check_sizes[s->check_type]; + s->block.hash.unpadded += check_sizes[s->check_type]; #else - if (s->check_type == XZ_CHECK_CRC32) - s->block.hash.unpadded += 4; - else if (IS_CRC64(s->check_type)) - s->block.hash.unpadded += 8; + if (s->check_type == XZ_CHECK_CRC32) + s->block.hash.unpadded += 4; + else if (IS_CRC64(s->check_type)) + s->block.hash.unpadded += 8; #endif - s->block.hash.uncompressed += s->block.uncompressed; - s->block.hash.crc32 = xz_crc32((const uint8_t *)&s->block.hash, sizeof(s->block.hash), - s->block.hash.crc32); + s->block.hash.uncompressed += s->block.uncompressed; + s->block.hash.crc32 = xz_crc32((const uint8_t *)&s->block.hash, sizeof(s->block.hash), + s->block.hash.crc32); - ++s->block.count; - } + ++s->block.count; + } - return ret; + return ret; } /* Update the Index size and the CRC32 value. */ static void index_update(struct xz_dec *s, const struct xz_buf *b) { - size_t in_used = b->in_pos - s->in_start; - s->index.size += in_used; - s->crc = xz_crc32(b->in + s->in_start, in_used, s->crc); + size_t in_used = b->in_pos - s->in_start; + s->index.size += in_used; + s->crc = xz_crc32(b->in + s->in_start, in_used, s->crc); } /* @@ -310,49 +310,49 @@ static void index_update(struct xz_dec *s, const struct xz_buf *b) */ static enum xz_ret dec_index(struct xz_dec *s, struct xz_buf *b) { - enum xz_ret ret; - - do - { - ret = dec_vli(s, b->in, &b->in_pos, b->in_size); - if (ret != XZ_STREAM_END) - { - index_update(s, b); - return ret; - } - - switch (s->index.sequence) - { - case SEQ_INDEX_COUNT: - s->index.count = s->vli; - - /* - * Validate that the Number of Records field - * indicates the same number of Records as - * there were Blocks in the Stream. - */ - if (s->index.count != s->block.count) - return XZ_DATA_ERROR; - - s->index.sequence = SEQ_INDEX_UNPADDED; - break; - - case SEQ_INDEX_UNPADDED: - s->index.hash.unpadded += s->vli; - s->index.sequence = SEQ_INDEX_UNCOMPRESSED; - break; - - case SEQ_INDEX_UNCOMPRESSED: - s->index.hash.uncompressed += s->vli; - s->index.hash.crc32 = xz_crc32((const uint8_t *)&s->index.hash, - sizeof(s->index.hash), s->index.hash.crc32); - --s->index.count; - s->index.sequence = SEQ_INDEX_UNPADDED; - break; - } - } while (s->index.count > 0); - - return XZ_STREAM_END; + enum xz_ret ret; + + do + { + ret = dec_vli(s, b->in, &b->in_pos, b->in_size); + if (ret != XZ_STREAM_END) + { + index_update(s, b); + return ret; + } + + switch (s->index.sequence) + { + case SEQ_INDEX_COUNT: + s->index.count = s->vli; + + /* + * Validate that the Number of Records field + * indicates the same number of Records as + * there were Blocks in the Stream. + */ + if (s->index.count != s->block.count) + return XZ_DATA_ERROR; + + s->index.sequence = SEQ_INDEX_UNPADDED; + break; + + case SEQ_INDEX_UNPADDED: + s->index.hash.unpadded += s->vli; + s->index.sequence = SEQ_INDEX_UNCOMPRESSED; + break; + + case SEQ_INDEX_UNCOMPRESSED: + s->index.hash.uncompressed += s->vli; + s->index.hash.crc32 = xz_crc32((const uint8_t *)&s->index.hash, + sizeof(s->index.hash), s->index.hash.crc32); + --s->index.count; + s->index.sequence = SEQ_INDEX_UNPADDED; + break; + } + } while (s->index.count > 0); + + return XZ_STREAM_END; } /* @@ -362,22 +362,22 @@ static enum xz_ret dec_index(struct xz_dec *s, struct xz_buf *b) */ static enum xz_ret crc_validate(struct xz_dec *s, struct xz_buf *b, uint32_t bits) { - do - { - if (b->in_pos == b->in_size) - return XZ_OK; + do + { + if (b->in_pos == b->in_size) + return XZ_OK; - if (((s->crc >> s->pos) & 0xFF) != b->in[b->in_pos++]) - return XZ_DATA_ERROR; + if (((s->crc >> s->pos) & 0xFF) != b->in[b->in_pos++]) + return XZ_DATA_ERROR; - s->pos += 8; + s->pos += 8; - } while (s->pos < bits); + } while (s->pos < bits); - s->crc = 0; - s->pos = 0; + s->crc = 0; + s->pos = 0; - return XZ_STREAM_END; + return XZ_STREAM_END; } #ifdef XZ_DEC_ANY_CHECK @@ -387,358 +387,358 @@ static enum xz_ret crc_validate(struct xz_dec *s, struct xz_buf *b, uint32_t bit */ static bool check_skip(struct xz_dec *s, struct xz_buf *b) { - while (s->pos < check_sizes[s->check_type]) - { - if (b->in_pos == b->in_size) - return false; + while (s->pos < check_sizes[s->check_type]) + { + if (b->in_pos == b->in_size) + return false; - ++b->in_pos; - ++s->pos; - } + ++b->in_pos; + ++s->pos; + } - s->pos = 0; + s->pos = 0; - return true; + return true; } #endif /* Decode the Stream Header field (the first 12 bytes of the .xz Stream). */ static enum xz_ret dec_stream_header(struct xz_dec *s) { - if (!memeq(s->temp.buf, HEADER_MAGIC, HEADER_MAGIC_SIZE)) - return XZ_FORMAT_ERROR; + if (!memeq(s->temp.buf, HEADER_MAGIC, HEADER_MAGIC_SIZE)) + return XZ_FORMAT_ERROR; - if (xz_crc32(s->temp.buf + HEADER_MAGIC_SIZE, 2, 0) != - get_le32(s->temp.buf + HEADER_MAGIC_SIZE + 2)) - return XZ_DATA_ERROR; + if (xz_crc32(s->temp.buf + HEADER_MAGIC_SIZE, 2, 0) != + get_le32(s->temp.buf + HEADER_MAGIC_SIZE + 2)) + return XZ_DATA_ERROR; - if (s->temp.buf[HEADER_MAGIC_SIZE] != 0) - return XZ_OPTIONS_ERROR; + if (s->temp.buf[HEADER_MAGIC_SIZE] != 0) + return XZ_OPTIONS_ERROR; - /* - * Of integrity checks, we support none (Check ID = 0), - * CRC32 (Check ID = 1), and optionally CRC64 (Check ID = 4). - * However, if XZ_DEC_ANY_CHECK is defined, we will accept other - * check types too, but then the check won't be verified and - * a warning (XZ_UNSUPPORTED_CHECK) will be given. - */ - s->check_type = s->temp.buf[HEADER_MAGIC_SIZE + 1]; + /* + * Of integrity checks, we support none (Check ID = 0), + * CRC32 (Check ID = 1), and optionally CRC64 (Check ID = 4). + * However, if XZ_DEC_ANY_CHECK is defined, we will accept other + * check types too, but then the check won't be verified and + * a warning (XZ_UNSUPPORTED_CHECK) will be given. + */ + s->check_type = s->temp.buf[HEADER_MAGIC_SIZE + 1]; #ifdef XZ_DEC_ANY_CHECK - if (s->check_type > XZ_CHECK_MAX) - return XZ_OPTIONS_ERROR; + if (s->check_type > XZ_CHECK_MAX) + return XZ_OPTIONS_ERROR; - if (s->check_type > XZ_CHECK_CRC32 && !IS_CRC64(s->check_type)) - return XZ_UNSUPPORTED_CHECK; + if (s->check_type > XZ_CHECK_CRC32 && !IS_CRC64(s->check_type)) + return XZ_UNSUPPORTED_CHECK; #else - if (s->check_type > XZ_CHECK_CRC32 && !IS_CRC64(s->check_type)) - return XZ_OPTIONS_ERROR; + if (s->check_type > XZ_CHECK_CRC32 && !IS_CRC64(s->check_type)) + return XZ_OPTIONS_ERROR; #endif - return XZ_OK; + return XZ_OK; } /* Decode the Stream Footer field (the last 12 bytes of the .xz Stream) */ static enum xz_ret dec_stream_footer(struct xz_dec *s) { - if (!memeq(s->temp.buf + 10, FOOTER_MAGIC, FOOTER_MAGIC_SIZE)) - return XZ_DATA_ERROR; - - if (xz_crc32(s->temp.buf + 4, 6, 0) != get_le32(s->temp.buf)) - return XZ_DATA_ERROR; - - /* - * Validate Backward Size. Note that we never added the size of the - * Index CRC32 field to s->index.size, thus we use s->index.size / 4 - * instead of s->index.size / 4 - 1. - */ - if ((s->index.size >> 2) != get_le32(s->temp.buf + 4)) - return XZ_DATA_ERROR; - - if (s->temp.buf[8] != 0 || s->temp.buf[9] != s->check_type) - return XZ_DATA_ERROR; - - /* - * Use XZ_STREAM_END instead of XZ_OK to be more convenient - * for the caller. - */ - return XZ_STREAM_END; + if (!memeq(s->temp.buf + 10, FOOTER_MAGIC, FOOTER_MAGIC_SIZE)) + return XZ_DATA_ERROR; + + if (xz_crc32(s->temp.buf + 4, 6, 0) != get_le32(s->temp.buf)) + return XZ_DATA_ERROR; + + /* + * Validate Backward Size. Note that we never added the size of the + * Index CRC32 field to s->index.size, thus we use s->index.size / 4 + * instead of s->index.size / 4 - 1. + */ + if ((s->index.size >> 2) != get_le32(s->temp.buf + 4)) + return XZ_DATA_ERROR; + + if (s->temp.buf[8] != 0 || s->temp.buf[9] != s->check_type) + return XZ_DATA_ERROR; + + /* + * Use XZ_STREAM_END instead of XZ_OK to be more convenient + * for the caller. + */ + return XZ_STREAM_END; } /* Decode the Block Header and initialize the filter chain. */ static enum xz_ret dec_block_header(struct xz_dec *s) { - enum xz_ret ret; + enum xz_ret ret; - /* - * Validate the CRC32. We know that the temp buffer is at least - * eight bytes so this is safe. - */ - s->temp.size -= 4; - if (xz_crc32(s->temp.buf, s->temp.size, 0) != get_le32(s->temp.buf + s->temp.size)) - return XZ_DATA_ERROR; + /* + * Validate the CRC32. We know that the temp buffer is at least + * eight bytes so this is safe. + */ + s->temp.size -= 4; + if (xz_crc32(s->temp.buf, s->temp.size, 0) != get_le32(s->temp.buf + s->temp.size)) + return XZ_DATA_ERROR; - s->temp.pos = 2; + s->temp.pos = 2; /* * Catch unsupported Block Flags. We support only one or two filters * in the chain, so we catch that with the same test. */ #ifdef XZ_DEC_BCJ - if (s->temp.buf[1] & 0x3E) + if (s->temp.buf[1] & 0x3E) #else - if (s->temp.buf[1] & 0x3F) + if (s->temp.buf[1] & 0x3F) #endif - return XZ_OPTIONS_ERROR; - - /* Compressed Size */ - if (s->temp.buf[1] & 0x40) - { - if (dec_vli(s, s->temp.buf, &s->temp.pos, s->temp.size) != XZ_STREAM_END) - return XZ_DATA_ERROR; - - s->block_header.compressed = s->vli; - } - else - { - s->block_header.compressed = VLI_UNKNOWN; - } - - /* Uncompressed Size */ - if (s->temp.buf[1] & 0x80) - { - if (dec_vli(s, s->temp.buf, &s->temp.pos, s->temp.size) != XZ_STREAM_END) - return XZ_DATA_ERROR; - - s->block_header.uncompressed = s->vli; - } - else - { - s->block_header.uncompressed = VLI_UNKNOWN; - } + return XZ_OPTIONS_ERROR; + + /* Compressed Size */ + if (s->temp.buf[1] & 0x40) + { + if (dec_vli(s, s->temp.buf, &s->temp.pos, s->temp.size) != XZ_STREAM_END) + return XZ_DATA_ERROR; + + s->block_header.compressed = s->vli; + } + else + { + s->block_header.compressed = VLI_UNKNOWN; + } + + /* Uncompressed Size */ + if (s->temp.buf[1] & 0x80) + { + if (dec_vli(s, s->temp.buf, &s->temp.pos, s->temp.size) != XZ_STREAM_END) + return XZ_DATA_ERROR; + + s->block_header.uncompressed = s->vli; + } + else + { + s->block_header.uncompressed = VLI_UNKNOWN; + } #ifdef XZ_DEC_BCJ - /* If there are two filters, the first one must be a BCJ filter. */ - s->bcj_active = s->temp.buf[1] & 0x01; - if (s->bcj_active) - { - if (s->temp.size - s->temp.pos < 2) - return XZ_OPTIONS_ERROR; - - ret = xz_dec_bcj_reset(s->bcj, s->temp.buf[s->temp.pos++]); - if (ret != XZ_OK) - return ret; - - /* - * We don't support custom start offset, - * so Size of Properties must be zero. - */ - if (s->temp.buf[s->temp.pos++] != 0x00) - return XZ_OPTIONS_ERROR; - } + /* If there are two filters, the first one must be a BCJ filter. */ + s->bcj_active = s->temp.buf[1] & 0x01; + if (s->bcj_active) + { + if (s->temp.size - s->temp.pos < 2) + return XZ_OPTIONS_ERROR; + + ret = xz_dec_bcj_reset(s->bcj, s->temp.buf[s->temp.pos++]); + if (ret != XZ_OK) + return ret; + + /* + * We don't support custom start offset, + * so Size of Properties must be zero. + */ + if (s->temp.buf[s->temp.pos++] != 0x00) + return XZ_OPTIONS_ERROR; + } #endif - /* Valid Filter Flags always take at least two bytes. */ - if (s->temp.size - s->temp.pos < 2) - return XZ_DATA_ERROR; + /* Valid Filter Flags always take at least two bytes. */ + if (s->temp.size - s->temp.pos < 2) + return XZ_DATA_ERROR; - /* Filter ID = LZMA2 */ - if (s->temp.buf[s->temp.pos++] != 0x21) - return XZ_OPTIONS_ERROR; + /* Filter ID = LZMA2 */ + if (s->temp.buf[s->temp.pos++] != 0x21) + return XZ_OPTIONS_ERROR; - /* Size of Properties = 1-byte Filter Properties */ - if (s->temp.buf[s->temp.pos++] != 0x01) - return XZ_OPTIONS_ERROR; + /* Size of Properties = 1-byte Filter Properties */ + if (s->temp.buf[s->temp.pos++] != 0x01) + return XZ_OPTIONS_ERROR; - /* Filter Properties contains LZMA2 dictionary size. */ - if (s->temp.size - s->temp.pos < 1) - return XZ_DATA_ERROR; + /* Filter Properties contains LZMA2 dictionary size. */ + if (s->temp.size - s->temp.pos < 1) + return XZ_DATA_ERROR; - ret = xz_dec_lzma2_reset(s->lzma2, s->temp.buf[s->temp.pos++]); - if (ret != XZ_OK) - return ret; + ret = xz_dec_lzma2_reset(s->lzma2, s->temp.buf[s->temp.pos++]); + if (ret != XZ_OK) + return ret; - /* The rest must be Header Padding. */ - while (s->temp.pos < s->temp.size) - if (s->temp.buf[s->temp.pos++] != 0x00) - return XZ_OPTIONS_ERROR; + /* The rest must be Header Padding. */ + while (s->temp.pos < s->temp.size) + if (s->temp.buf[s->temp.pos++] != 0x00) + return XZ_OPTIONS_ERROR; - s->temp.pos = 0; - s->block.compressed = 0; - s->block.uncompressed = 0; + s->temp.pos = 0; + s->block.compressed = 0; + s->block.uncompressed = 0; - return XZ_OK; + return XZ_OK; } static enum xz_ret dec_main(struct xz_dec *s, struct xz_buf *b) { - enum xz_ret ret; - - /* - * Store the start position for the case when we are in the middle - * of the Index field. - */ - s->in_start = b->in_pos; - - while (true) - { - switch (s->sequence) - { - case SEQ_STREAM_HEADER: - /* - * Stream Header is copied to s->temp, and then - * decoded from there. This way if the caller - * gives us only little input at a time, we can - * still keep the Stream Header decoding code - * simple. Similar approach is used in many places - * in this file. - */ - if (!fill_temp(s, b)) - return XZ_OK; - - /* - * If dec_stream_header() returns - * XZ_UNSUPPORTED_CHECK, it is still possible - * to continue decoding if working in multi-call - * mode. Thus, update s->sequence before calling - * dec_stream_header(). - */ - s->sequence = SEQ_BLOCK_START; - - ret = dec_stream_header(s); - if (ret != XZ_OK) - return ret; - - case SEQ_BLOCK_START: - /* We need one byte of input to continue. */ - if (b->in_pos == b->in_size) - return XZ_OK; - - /* See if this is the beginning of the Index field. */ - if (b->in[b->in_pos] == 0) - { - s->in_start = b->in_pos++; - s->sequence = SEQ_INDEX; - break; - } - - /* - * Calculate the size of the Block Header and - * prepare to decode it. - */ - s->block_header.size = ((uint32_t)b->in[b->in_pos] + 1) * 4; - - s->temp.size = s->block_header.size; - s->temp.pos = 0; - s->sequence = SEQ_BLOCK_HEADER; - - case SEQ_BLOCK_HEADER: - if (!fill_temp(s, b)) - return XZ_OK; - - ret = dec_block_header(s); - if (ret != XZ_OK) - return ret; - - s->sequence = SEQ_BLOCK_UNCOMPRESS; - - case SEQ_BLOCK_UNCOMPRESS: - ret = dec_block(s, b); - if (ret != XZ_STREAM_END) - return ret; - - s->sequence = SEQ_BLOCK_PADDING; - - case SEQ_BLOCK_PADDING: - /* - * Size of Compressed Data + Block Padding - * must be a multiple of four. We don't need - * s->block.compressed for anything else - * anymore, so we use it here to test the size - * of the Block Padding field. - */ - while (s->block.compressed & 3) - { - if (b->in_pos == b->in_size) - return XZ_OK; - - if (b->in[b->in_pos++] != 0) - return XZ_DATA_ERROR; - - ++s->block.compressed; - } - - s->sequence = SEQ_BLOCK_CHECK; - - case SEQ_BLOCK_CHECK: - if (s->check_type == XZ_CHECK_CRC32) - { - ret = crc_validate(s, b, 32); - if (ret != XZ_STREAM_END) - return ret; - } - else if (IS_CRC64(s->check_type)) - { - ret = crc_validate(s, b, 64); - if (ret != XZ_STREAM_END) - return ret; - } + enum xz_ret ret; + + /* + * Store the start position for the case when we are in the middle + * of the Index field. + */ + s->in_start = b->in_pos; + + while (true) + { + switch (s->sequence) + { + case SEQ_STREAM_HEADER: + /* + * Stream Header is copied to s->temp, and then + * decoded from there. This way if the caller + * gives us only little input at a time, we can + * still keep the Stream Header decoding code + * simple. Similar approach is used in many places + * in this file. + */ + if (!fill_temp(s, b)) + return XZ_OK; + + /* + * If dec_stream_header() returns + * XZ_UNSUPPORTED_CHECK, it is still possible + * to continue decoding if working in multi-call + * mode. Thus, update s->sequence before calling + * dec_stream_header(). + */ + s->sequence = SEQ_BLOCK_START; + + ret = dec_stream_header(s); + if (ret != XZ_OK) + return ret; + + case SEQ_BLOCK_START: + /* We need one byte of input to continue. */ + if (b->in_pos == b->in_size) + return XZ_OK; + + /* See if this is the beginning of the Index field. */ + if (b->in[b->in_pos] == 0) + { + s->in_start = b->in_pos++; + s->sequence = SEQ_INDEX; + break; + } + + /* + * Calculate the size of the Block Header and + * prepare to decode it. + */ + s->block_header.size = ((uint32_t)b->in[b->in_pos] + 1) * 4; + + s->temp.size = s->block_header.size; + s->temp.pos = 0; + s->sequence = SEQ_BLOCK_HEADER; + + case SEQ_BLOCK_HEADER: + if (!fill_temp(s, b)) + return XZ_OK; + + ret = dec_block_header(s); + if (ret != XZ_OK) + return ret; + + s->sequence = SEQ_BLOCK_UNCOMPRESS; + + case SEQ_BLOCK_UNCOMPRESS: + ret = dec_block(s, b); + if (ret != XZ_STREAM_END) + return ret; + + s->sequence = SEQ_BLOCK_PADDING; + + case SEQ_BLOCK_PADDING: + /* + * Size of Compressed Data + Block Padding + * must be a multiple of four. We don't need + * s->block.compressed for anything else + * anymore, so we use it here to test the size + * of the Block Padding field. + */ + while (s->block.compressed & 3) + { + if (b->in_pos == b->in_size) + return XZ_OK; + + if (b->in[b->in_pos++] != 0) + return XZ_DATA_ERROR; + + ++s->block.compressed; + } + + s->sequence = SEQ_BLOCK_CHECK; + + case SEQ_BLOCK_CHECK: + if (s->check_type == XZ_CHECK_CRC32) + { + ret = crc_validate(s, b, 32); + if (ret != XZ_STREAM_END) + return ret; + } + else if (IS_CRC64(s->check_type)) + { + ret = crc_validate(s, b, 64); + if (ret != XZ_STREAM_END) + return ret; + } #ifdef XZ_DEC_ANY_CHECK - else if (!check_skip(s, b)) - { - return XZ_OK; - } + else if (!check_skip(s, b)) + { + return XZ_OK; + } #endif - s->sequence = SEQ_BLOCK_START; - break; + s->sequence = SEQ_BLOCK_START; + break; - case SEQ_INDEX: - ret = dec_index(s, b); - if (ret != XZ_STREAM_END) - return ret; + case SEQ_INDEX: + ret = dec_index(s, b); + if (ret != XZ_STREAM_END) + return ret; - s->sequence = SEQ_INDEX_PADDING; + s->sequence = SEQ_INDEX_PADDING; - case SEQ_INDEX_PADDING: - while ((s->index.size + (b->in_pos - s->in_start)) & 3) - { - if (b->in_pos == b->in_size) - { - index_update(s, b); - return XZ_OK; - } + case SEQ_INDEX_PADDING: + while ((s->index.size + (b->in_pos - s->in_start)) & 3) + { + if (b->in_pos == b->in_size) + { + index_update(s, b); + return XZ_OK; + } - if (b->in[b->in_pos++] != 0) - return XZ_DATA_ERROR; - } + if (b->in[b->in_pos++] != 0) + return XZ_DATA_ERROR; + } - /* Finish the CRC32 value and Index size. */ - index_update(s, b); + /* Finish the CRC32 value and Index size. */ + index_update(s, b); - /* Compare the hashes to validate the Index field. */ - if (!memeq(&s->block.hash, &s->index.hash, sizeof(s->block.hash))) - return XZ_DATA_ERROR; + /* Compare the hashes to validate the Index field. */ + if (!memeq(&s->block.hash, &s->index.hash, sizeof(s->block.hash))) + return XZ_DATA_ERROR; - s->sequence = SEQ_INDEX_CRC32; + s->sequence = SEQ_INDEX_CRC32; - case SEQ_INDEX_CRC32: - ret = crc_validate(s, b, 32); - if (ret != XZ_STREAM_END) - return ret; + case SEQ_INDEX_CRC32: + ret = crc_validate(s, b, 32); + if (ret != XZ_STREAM_END) + return ret; - s->temp.size = STREAM_HEADER_SIZE; - s->sequence = SEQ_STREAM_FOOTER; + s->temp.size = STREAM_HEADER_SIZE; + s->sequence = SEQ_STREAM_FOOTER; - case SEQ_STREAM_FOOTER: - if (!fill_temp(s, b)) - return XZ_OK; + case SEQ_STREAM_FOOTER: + if (!fill_temp(s, b)) + return XZ_OK; - return dec_stream_footer(s); - } - } + return dec_stream_footer(s); + } + } - /* Never reached */ + /* Never reached */ } /* @@ -768,93 +768,93 @@ static enum xz_ret dec_main(struct xz_dec *s, struct xz_buf *b) */ XZ_EXTERN enum xz_ret xz_dec_run(struct xz_dec *s, struct xz_buf *b) { - size_t in_start; - size_t out_start; - enum xz_ret ret; - - if (DEC_IS_SINGLE(s->mode)) - xz_dec_reset(s); - - in_start = b->in_pos; - out_start = b->out_pos; - ret = dec_main(s, b); - - if (DEC_IS_SINGLE(s->mode)) - { - if (ret == XZ_OK) - ret = b->in_pos == b->in_size ? XZ_DATA_ERROR : XZ_BUF_ERROR; - - if (ret != XZ_STREAM_END) - { - b->in_pos = in_start; - b->out_pos = out_start; - } - } - else if (ret == XZ_OK && in_start == b->in_pos && out_start == b->out_pos) - { - if (s->allow_buf_error) - ret = XZ_BUF_ERROR; - - s->allow_buf_error = true; - } - else - { - s->allow_buf_error = false; - } - - return ret; + size_t in_start; + size_t out_start; + enum xz_ret ret; + + if (DEC_IS_SINGLE(s->mode)) + xz_dec_reset(s); + + in_start = b->in_pos; + out_start = b->out_pos; + ret = dec_main(s, b); + + if (DEC_IS_SINGLE(s->mode)) + { + if (ret == XZ_OK) + ret = b->in_pos == b->in_size ? XZ_DATA_ERROR : XZ_BUF_ERROR; + + if (ret != XZ_STREAM_END) + { + b->in_pos = in_start; + b->out_pos = out_start; + } + } + else if (ret == XZ_OK && in_start == b->in_pos && out_start == b->out_pos) + { + if (s->allow_buf_error) + ret = XZ_BUF_ERROR; + + s->allow_buf_error = true; + } + else + { + s->allow_buf_error = false; + } + + return ret; } XZ_EXTERN struct xz_dec *xz_dec_init(enum xz_mode mode, uint32_t dict_max) { - struct xz_dec *s = kmalloc(sizeof(*s), GFP_KERNEL); - if (s == NULL) - return NULL; + struct xz_dec *s = kmalloc(sizeof(*s), GFP_KERNEL); + if (s == NULL) + return NULL; - s->mode = mode; + s->mode = mode; #ifdef XZ_DEC_BCJ - s->bcj = xz_dec_bcj_create(DEC_IS_SINGLE(mode)); - if (s->bcj == NULL) - goto error_bcj; + s->bcj = xz_dec_bcj_create(DEC_IS_SINGLE(mode)); + if (s->bcj == NULL) + goto error_bcj; #endif - s->lzma2 = xz_dec_lzma2_create(mode, dict_max); - if (s->lzma2 == NULL) - goto error_lzma2; + s->lzma2 = xz_dec_lzma2_create(mode, dict_max); + if (s->lzma2 == NULL) + goto error_lzma2; - xz_dec_reset(s); - return s; + xz_dec_reset(s); + return s; error_lzma2: #ifdef XZ_DEC_BCJ - xz_dec_bcj_end(s->bcj); + xz_dec_bcj_end(s->bcj); error_bcj: #endif - kfree(s); - return NULL; + kfree(s); + return NULL; } XZ_EXTERN void xz_dec_reset(struct xz_dec *s) { - s->sequence = SEQ_STREAM_HEADER; - s->allow_buf_error = false; - s->pos = 0; - s->crc = 0; - memzero(&s->block, sizeof(s->block)); - memzero(&s->index, sizeof(s->index)); - s->temp.pos = 0; - s->temp.size = STREAM_HEADER_SIZE; + s->sequence = SEQ_STREAM_HEADER; + s->allow_buf_error = false; + s->pos = 0; + s->crc = 0; + memzero(&s->block, sizeof(s->block)); + memzero(&s->index, sizeof(s->index)); + s->temp.pos = 0; + s->temp.size = STREAM_HEADER_SIZE; } XZ_EXTERN void xz_dec_end(struct xz_dec *s) { - if (s != NULL) - { - xz_dec_lzma2_end(s->lzma2); + if (s != NULL) + { + xz_dec_lzma2_end(s->lzma2); #ifdef XZ_DEC_BCJ - xz_dec_bcj_end(s->bcj); + xz_dec_bcj_end(s->bcj); #endif - kfree(s); - } + kfree(s); + } } diff --git a/libraries/xz-embedded/src/xz_lzma2.h b/libraries/xz-embedded/src/xz_lzma2.h index 3976033a..82a425f2 100644 --- a/libraries/xz-embedded/src/xz_lzma2.h +++ b/libraries/xz-embedded/src/xz_lzma2.h @@ -41,18 +41,18 @@ */ enum lzma_state { - STATE_LIT_LIT, - STATE_MATCH_LIT_LIT, - STATE_REP_LIT_LIT, - STATE_SHORTREP_LIT_LIT, - STATE_MATCH_LIT, - STATE_REP_LIT, - STATE_SHORTREP_LIT, - STATE_LIT_MATCH, - STATE_LIT_LONGREP, - STATE_LIT_SHORTREP, - STATE_NONLIT_MATCH, - STATE_NONLIT_REP + STATE_LIT_LIT, + STATE_MATCH_LIT_LIT, + STATE_REP_LIT_LIT, + STATE_SHORTREP_LIT_LIT, + STATE_MATCH_LIT, + STATE_REP_LIT, + STATE_SHORTREP_LIT, + STATE_LIT_MATCH, + STATE_LIT_LONGREP, + STATE_LIT_SHORTREP, + STATE_NONLIT_MATCH, + STATE_NONLIT_REP }; /* Total number of states */ @@ -64,36 +64,36 @@ enum lzma_state /* Indicate that the latest symbol was a literal. */ static inline void lzma_state_literal(enum lzma_state *state) { - if (*state <= STATE_SHORTREP_LIT_LIT) - *state = STATE_LIT_LIT; - else if (*state <= STATE_LIT_SHORTREP) - *state -= 3; - else - *state -= 6; + if (*state <= STATE_SHORTREP_LIT_LIT) + *state = STATE_LIT_LIT; + else if (*state <= STATE_LIT_SHORTREP) + *state -= 3; + else + *state -= 6; } /* Indicate that the latest symbol was a match. */ static inline void lzma_state_match(enum lzma_state *state) { - *state = *state < LIT_STATES ? STATE_LIT_MATCH : STATE_NONLIT_MATCH; + *state = *state < LIT_STATES ? STATE_LIT_MATCH : STATE_NONLIT_MATCH; } /* Indicate that the latest state was a long repeated match. */ static inline void lzma_state_long_rep(enum lzma_state *state) { - *state = *state < LIT_STATES ? STATE_LIT_LONGREP : STATE_NONLIT_REP; + *state = *state < LIT_STATES ? STATE_LIT_LONGREP : STATE_NONLIT_REP; } /* Indicate that the latest symbol was a short match. */ static inline void lzma_state_short_rep(enum lzma_state *state) { - *state = *state < LIT_STATES ? STATE_LIT_SHORTREP : STATE_NONLIT_REP; + *state = *state < LIT_STATES ? STATE_LIT_SHORTREP : STATE_NONLIT_REP; } /* Test if the previous symbol was a literal. */ static inline bool lzma_state_is_literal(enum lzma_state state) { - return state < LIT_STATES; + return state < LIT_STATES; } /* Each literal coder is divided in three sections: @@ -147,7 +147,7 @@ static inline bool lzma_state_is_literal(enum lzma_state state) */ static inline uint32_t lzma_get_dist_state(uint32_t len) { - return len < DIST_STATES + MATCH_LEN_MIN ? len - MATCH_LEN_MIN : DIST_STATES - 1; + return len < DIST_STATES + MATCH_LEN_MIN ? len - MATCH_LEN_MIN : DIST_STATES - 1; } /* diff --git a/libraries/xz-embedded/src/xz_private.h b/libraries/xz-embedded/src/xz_private.h index 55a3af1c..1b616430 100644 --- a/libraries/xz-embedded/src/xz_private.h +++ b/libraries/xz-embedded/src/xz_private.h @@ -94,8 +94,8 @@ */ #ifndef XZ_DEC_BCJ #if defined(XZ_DEC_X86) || defined(XZ_DEC_POWERPC) || defined(XZ_DEC_IA64) || \ - defined(XZ_DEC_ARM) || defined(XZ_DEC_ARM) || defined(XZ_DEC_ARMTHUMB) || \ - defined(XZ_DEC_SPARC) + defined(XZ_DEC_ARM) || defined(XZ_DEC_ARM) || defined(XZ_DEC_ARMTHUMB) || \ + defined(XZ_DEC_SPARC) #define XZ_DEC_BCJ #endif #endif @@ -141,7 +141,7 @@ XZ_EXTERN enum xz_ret xz_dec_bcj_reset(struct xz_dec_bcj *s, uint8_t id); * must be called directly. */ XZ_EXTERN enum xz_ret xz_dec_bcj_run(struct xz_dec_bcj *s, struct xz_dec_lzma2 *lzma2, - struct xz_buf *b); + struct xz_buf *b); /* Free the memory allocated for the BCJ filters. */ #define xz_dec_bcj_end(s) kfree(s) diff --git a/libraries/xz-embedded/src/xz_stream.h b/libraries/xz-embedded/src/xz_stream.h index c0e191e6..b3d2c9fd 100644 --- a/libraries/xz-embedded/src/xz_stream.h +++ b/libraries/xz-embedded/src/xz_stream.h @@ -50,10 +50,10 @@ typedef uint64_t vli_type; /* Integrity Check types */ enum xz_check { - XZ_CHECK_NONE = 0, - XZ_CHECK_CRC32 = 1, - XZ_CHECK_CRC64 = 4, - XZ_CHECK_SHA256 = 10 + XZ_CHECK_NONE = 0, + XZ_CHECK_CRC32 = 1, + XZ_CHECK_CRC64 = 4, + XZ_CHECK_SHA256 = 10 }; /* Maximum possible Check ID */ diff --git a/libraries/xz-embedded/xzminidec.c b/libraries/xz-embedded/xzminidec.c index bb62c3ac..44f60602 100644 --- a/libraries/xz-embedded/xzminidec.c +++ b/libraries/xz-embedded/xzminidec.c @@ -24,121 +24,121 @@ static uint8_t out[BUFSIZ]; int main(int argc, char **argv) { - struct xz_buf b; - struct xz_dec *s; - enum xz_ret ret; - const char *msg; - - if (argc >= 2 && strcmp(argv[1], "--help") == 0) - { - fputs("Uncompress a .xz file from stdin to stdout.\n" - "Arguments other than `--help' are ignored.\n", - stdout); - return 0; - } - - xz_crc32_init(); + struct xz_buf b; + struct xz_dec *s; + enum xz_ret ret; + const char *msg; + + if (argc >= 2 && strcmp(argv[1], "--help") == 0) + { + fputs("Uncompress a .xz file from stdin to stdout.\n" + "Arguments other than `--help' are ignored.\n", + stdout); + return 0; + } + + xz_crc32_init(); #ifdef XZ_USE_CRC64 - xz_crc64_init(); + xz_crc64_init(); #endif - /* - * Support up to 64 MiB dictionary. The actually needed memory - * is allocated once the headers have been parsed. - */ - s = xz_dec_init(XZ_DYNALLOC, 1 << 26); - if (s == NULL) - { - msg = "Memory allocation failed\n"; - goto error; - } - - b.in = in; - b.in_pos = 0; - b.in_size = 0; - b.out = out; - b.out_pos = 0; - b.out_size = BUFSIZ; - - while (true) - { - if (b.in_pos == b.in_size) - { - b.in_size = fread(in, 1, sizeof(in), stdin); - b.in_pos = 0; - } - - ret = xz_dec_run(s, &b); - - if (b.out_pos == sizeof(out)) - { - if (fwrite(out, 1, b.out_pos, stdout) != b.out_pos) - { - msg = "Write error\n"; - goto error; - } - - b.out_pos = 0; - } - - if (ret == XZ_OK) - continue; + /* + * Support up to 64 MiB dictionary. The actually needed memory + * is allocated once the headers have been parsed. + */ + s = xz_dec_init(XZ_DYNALLOC, 1 << 26); + if (s == NULL) + { + msg = "Memory allocation failed\n"; + goto error; + } + + b.in = in; + b.in_pos = 0; + b.in_size = 0; + b.out = out; + b.out_pos = 0; + b.out_size = BUFSIZ; + + while (true) + { + if (b.in_pos == b.in_size) + { + b.in_size = fread(in, 1, sizeof(in), stdin); + b.in_pos = 0; + } + + ret = xz_dec_run(s, &b); + + if (b.out_pos == sizeof(out)) + { + if (fwrite(out, 1, b.out_pos, stdout) != b.out_pos) + { + msg = "Write error\n"; + goto error; + } + + b.out_pos = 0; + } + + if (ret == XZ_OK) + continue; #ifdef XZ_DEC_ANY_CHECK - if (ret == XZ_UNSUPPORTED_CHECK) - { - fputs(argv[0], stderr); - fputs(": ", stderr); - fputs("Unsupported check; not verifying " - "file integrity\n", - stderr); - continue; - } + if (ret == XZ_UNSUPPORTED_CHECK) + { + fputs(argv[0], stderr); + fputs(": ", stderr); + fputs("Unsupported check; not verifying " + "file integrity\n", + stderr); + continue; + } #endif - if (fwrite(out, 1, b.out_pos, stdout) != b.out_pos || fclose(stdout)) - { - msg = "Write error\n"; - goto error; - } - - switch (ret) - { - case XZ_STREAM_END: - xz_dec_end(s); - return 0; - - case XZ_MEM_ERROR: - msg = "Memory allocation failed\n"; - goto error; - - case XZ_MEMLIMIT_ERROR: - msg = "Memory usage limit reached\n"; - goto error; - - case XZ_FORMAT_ERROR: - msg = "Not a .xz file\n"; - goto error; - - case XZ_OPTIONS_ERROR: - msg = "Unsupported options in the .xz headers\n"; - goto error; - - case XZ_DATA_ERROR: - case XZ_BUF_ERROR: - msg = "File is corrupt\n"; - goto error; - - default: - msg = "Bug!\n"; - goto error; - } - } + if (fwrite(out, 1, b.out_pos, stdout) != b.out_pos || fclose(stdout)) + { + msg = "Write error\n"; + goto error; + } + + switch (ret) + { + case XZ_STREAM_END: + xz_dec_end(s); + return 0; + + case XZ_MEM_ERROR: + msg = "Memory allocation failed\n"; + goto error; + + case XZ_MEMLIMIT_ERROR: + msg = "Memory usage limit reached\n"; + goto error; + + case XZ_FORMAT_ERROR: + msg = "Not a .xz file\n"; + goto error; + + case XZ_OPTIONS_ERROR: + msg = "Unsupported options in the .xz headers\n"; + goto error; + + case XZ_DATA_ERROR: + case XZ_BUF_ERROR: + msg = "File is corrupt\n"; + goto error; + + default: + msg = "Bug!\n"; + goto error; + } + } error: - xz_dec_end(s); - fputs(argv[0], stderr); - fputs(": ", stderr); - fputs(msg, stderr); - return 1; + xz_dec_end(s); + fputs(argv[0], stderr); + fputs(": ", stderr); + fputs(msg, stderr); + return 1; } |