summaryrefslogtreecommitdiffstats
path: root/extensions/universalchardet/src/base/nsCharSetProber.cpp
diff options
context:
space:
mode:
authorMatt A. Tobin <mattatobin@localhost.localdomain>2018-02-02 04:16:08 -0500
committerMatt A. Tobin <mattatobin@localhost.localdomain>2018-02-02 04:16:08 -0500
commit5f8de423f190bbb79a62f804151bc24824fa32d8 (patch)
tree10027f336435511475e392454359edea8e25895d /extensions/universalchardet/src/base/nsCharSetProber.cpp
parent49ee0794b5d912db1f95dce6eb52d781dc210db5 (diff)
downloadUXP-5f8de423f190bbb79a62f804151bc24824fa32d8.tar
UXP-5f8de423f190bbb79a62f804151bc24824fa32d8.tar.gz
UXP-5f8de423f190bbb79a62f804151bc24824fa32d8.tar.lz
UXP-5f8de423f190bbb79a62f804151bc24824fa32d8.tar.xz
UXP-5f8de423f190bbb79a62f804151bc24824fa32d8.zip
Add m-esr52 at 52.6.0
Diffstat (limited to 'extensions/universalchardet/src/base/nsCharSetProber.cpp')
-rw-r--r--extensions/universalchardet/src/base/nsCharSetProber.cpp92
1 files changed, 92 insertions, 0 deletions
diff --git a/extensions/universalchardet/src/base/nsCharSetProber.cpp b/extensions/universalchardet/src/base/nsCharSetProber.cpp
new file mode 100644
index 000000000..cd15a10c3
--- /dev/null
+++ b/extensions/universalchardet/src/base/nsCharSetProber.cpp
@@ -0,0 +1,92 @@
+/* -*- Mode: C; tab-width: 4; indent-tabs-mode: nil; c-basic-offset: 2 -*- */
+/* This Source Code Form is subject to the terms of the Mozilla Public
+ * License, v. 2.0. If a copy of the MPL was not distributed with this
+ * file, You can obtain one at http://mozilla.org/MPL/2.0/. */
+
+#include "nsCharSetProber.h"
+#include "prmem.h"
+
+//This filter applies to all scripts which do not use English characters
+bool nsCharSetProber::FilterWithoutEnglishLetters(const char* aBuf, uint32_t aLen, char** newBuf, uint32_t& newLen)
+{
+ char *newptr;
+ char *prevPtr, *curPtr;
+
+ bool meetMSB = false;
+ newptr = *newBuf = (char*)PR_Malloc(aLen);
+ if (!newptr)
+ return false;
+
+ for (curPtr = prevPtr = (char*)aBuf; curPtr < aBuf+aLen; curPtr++)
+ {
+ if (*curPtr & 0x80)
+ {
+ meetMSB = true;
+ }
+ else if (*curPtr < 'A' || (*curPtr > 'Z' && *curPtr < 'a') || *curPtr > 'z')
+ {
+ //current char is a symbol, most likely a punctuation. we treat it as segment delimiter
+ if (meetMSB && curPtr > prevPtr)
+ //this segment contains more than single symbol, and it has upper ASCII, we need to keep it
+ {
+ while (prevPtr < curPtr) *newptr++ = *prevPtr++;
+ prevPtr++;
+ *newptr++ = ' ';
+ meetMSB = false;
+ }
+ else //ignore current segment. (either because it is just a symbol or just an English word)
+ prevPtr = curPtr+1;
+ }
+ }
+ if (meetMSB && curPtr > prevPtr)
+ while (prevPtr < curPtr) *newptr++ = *prevPtr++;
+
+ newLen = newptr - *newBuf;
+
+ return true;
+}
+
+//This filter applies to all scripts which contain both English characters and upper ASCII characters.
+bool nsCharSetProber::FilterWithEnglishLetters(const char* aBuf, uint32_t aLen, char** newBuf, uint32_t& newLen)
+{
+ //do filtering to reduce load to probers
+ char *newptr;
+ char *prevPtr, *curPtr;
+ bool isInTag = false;
+
+ newptr = *newBuf = (char*)PR_Malloc(aLen);
+ if (!newptr)
+ return false;
+
+ for (curPtr = prevPtr = (char*)aBuf; curPtr < aBuf+aLen; curPtr++)
+ {
+ if (*curPtr == '>')
+ isInTag = false;
+ else if (*curPtr == '<')
+ isInTag = true;
+
+ if (!(*curPtr & 0x80) &&
+ (*curPtr < 'A' || (*curPtr > 'Z' && *curPtr < 'a') || *curPtr > 'z') )
+ {
+ if (curPtr > prevPtr && !isInTag) // Current segment contains more than just a symbol
+ // and it is not inside a tag, keep it.
+ {
+ while (prevPtr < curPtr) *newptr++ = *prevPtr++;
+ prevPtr++;
+ *newptr++ = ' ';
+ }
+ else
+ prevPtr = curPtr+1;
+ }
+ }
+
+ // If the current segment contains more than just a symbol
+ // and it is not inside a tag then keep it.
+ if (!isInTag)
+ while (prevPtr < curPtr)
+ *newptr++ = *prevPtr++;
+
+ newLen = newptr - *newBuf;
+
+ return true;
+}