1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
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;
}
|
