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

src: handle long sequences of characters.

Browse Source 
Actually my previous commit was not handling all cases, though it was
taking care of the buffer overflow triggered by the provided byte
sequence. Yet I believe it was still possible to craft special input
sequences too long for codePointBuffer.
This additional commit would handle these other cases by processing the
input in manageable sub-strings.
This commit is contained in:
Jehan 2023-07-17 20:09:10 +02:00
parent 9910941387
commit ab1d2f1120
1 changed files with 21 additions and 10 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

31
src/nsMBCSGroupProber.cpp
View File

@ -310,17 +310,28 @@ nsProbingState nsMBCSGroupProber::HandleData(const char* aBuf, PRUint32 aLen,
} }
if (codePointBuffer[i]) if (codePointBuffer[i])
st = mProbers[i]->HandleData(aBuf + start, sequenceLength, {
&(codePointBuffer[i]), &(codePointBufferIdx[i])); while (sequenceLength > 0)
else {
st = mProbers[i]->HandleData(aBuf + start, sequenceLength, NULL, NULL); int subLength = (sequenceLength > codePointBufferSize[i]) ? codePointBufferSize[i] : sequenceLength;
if (codePointBufferIdx[i] > 0 && codePointBuffer[i]) st = mProbers[i]->HandleData(aBuf + start, subLength,
{ &(codePointBuffer[i]), &(codePointBufferIdx[i]));
for (PRUint32 j = 0; j < NUM_OF_LANGUAGES; j++)
langDetectors[i][j]->HandleData(codePointBuffer[i], codePointBufferIdx[i]); if (codePointBufferIdx[i] > 0)
codePointBufferIdx[i] = 0; {
} for (PRUint32 j = 0; j < NUM_OF_LANGUAGES; j++)
langDetectors[i][j]->HandleData(codePointBuffer[i], codePointBufferIdx[i]);
codePointBufferIdx[i] = 0;
}
sequenceLength -= subLength;
}
}
else
{
st = mProbers[i]->HandleData(aBuf + start, sequenceLength, NULL, NULL);
}
if (st == eFoundIt) if (st == eFoundIt)
{ {