coffee/uchardet
1
0
Fork 0
mirror of https://gitlab.freedesktop.org/uchardet/uchardet.git synced 2025-12-06 08:46:40 +08:00
Code Issues Packages Projects Releases Wiki Activity

Stop detection early on control characters

Browse Source
This commit is contained in:
Ophir LOJKINE 2015-11-24 22:07:41 +03:00
parent d082704fec
commit 5ef60164fc
2 changed files with 7 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
src/LangHebrewModel.cpp
View File

@ -54,7 +54,7 @@ static const unsigned char win1255_CharToOrderMap[] =
{
255,255,255,255,255,255,255,255,255,255,254,255,255,254,255,255, //00
255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255, //10
+253,253,253,253,253,253,253,253,253,253,253,253,253,253,253,253, //20
253,253,253,253,253,253,253,253,253,253,253,253,253,253,253,253, //20
252,252,252,252,252,252,252,252,252,252,253,253,253,253,253,253, //30
253, 69, 91, 79, 80, 92, 89, 97, 90, 68,111,112, 82, 73, 95, 85, //40
78,121, 86, 71, 67,102,107, 84,114,103,115,253,253,253,253,253, //50
@ -67,7 +67,7 @@ static const unsigned char win1255_CharToOrderMap[] =
30, 59, 41, 88, 33, 37, 36, 31, 29, 35,235, 62, 28,236,126,237,
238, 38, 45,239,240,241,242,243,127,244,245,246,247,248,249,250,
9, 8, 20, 16, 3, 2, 24, 14, 22, 1, 25, 15, 4, 11, 6, 23,
12, 19, 13, 26, 18, 27, 21, 17, 7, 10, 5,251,252,128, 96,253,
12, 19, 13, 26, 18, 27, 21, 17, 7, 10, 5,255,255,128, 96,255,
};
//Model Table:

5
src/nsSBCharSetProber.cpp
View File

@ -46,6 +46,11 @@ nsProbingState nsSingleByteCharSetProber::HandleData(const char* aBuf, PRUint32
{
order = mModel->charToOrderMap[(unsigned char)aBuf[i]];
if (order == 255)
{
mState = eNotMe;
break;
}
if (order < SYMBOL_CAT_ORDER)
mTotalChar++;
if (order < SAMPLE_SIZE)