tris/recoll
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

m_words_in_span was always properly reset between invocations (if discardspan() was not called for some reason), resulting in crashes

Browse Source
This commit is contained in:
Jean-Francois Dockes 2017-05-15 10:26:38 +02:00
parent 4671f5ea25
commit 15ea565e9f
2 changed files with 18 additions and 9 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

19
src/common/textsplit.cpp
View File

@ -468,8 +468,8 @@ inline bool TextSplit::doemit(bool spanerase, size_t _bp)
void TextSplit::discardspan()
{
m_span.clear();
m_words_in_span.clear();
m_span.erase();
m_spanpos = m_wordpos;
m_wordStart = 0;
m_wordLen = m_wordChars = 0;
@ -513,10 +513,9 @@ bool TextSplit::text_to_words(const string &in)
if (in.empty())
return true;
m_span.erase();
m_inNumber = false;
m_wordStart = m_wordLen = m_wordChars = m_prevpos = m_prevlen = m_wordpos
= m_spanpos = 0;
// Reset the data members relative to splitting state
clearsplitstate();
bool pagepending = false;
bool softhyphenpending = false;
@ -935,10 +934,12 @@ bool TextSplit::cjk_to_words(Utf8Iter *itp, unsigned int *cp)
}
}
m_span.erase();
m_inNumber = false;
m_wordStart = m_wordLen = m_wordChars = m_prevpos = m_prevlen = 0;
m_spanpos = m_wordpos;
// Reset state, saving term position, and return the found non-cjk
// unicode character value. The current input byte offset is kept
// in the utf8Iter
int pos = m_wordpos;
clearsplitstate();
m_spanpos = m_wordpos = pos;
*cp = c;
return true;
}

8
src/common/textsplit.h
View File

@ -213,6 +213,14 @@ private:
// Word length in characters. Declared but not updated if !TEXTSPLIT_STATS
unsigned int m_wordChars;
void clearsplitstate() {
m_span.clear();
m_words_in_span.clear();
m_inNumber = false;
m_wordStart = m_wordLen = m_wordpos = m_spanpos = m_prevpos =
m_prevlen = m_wordChars = 0;
}
// This processes cjk text:
bool cjk_to_words(Utf8Iter *it, unsigned int *cp);