third_codes/cppjieba
1
0
Fork 0
mirror of https://github.com/yanyiwu/cppjieba.git synced 2025-07-18 00:00:12 +08:00
Code Issues Packages Projects Releases Wiki Activity

fix bug's in SegmentBase where using filterAscii

Browse Source
This commit is contained in:
wyy 2013-12-07 15:22:00 +08:00
parent 8e8a68352b
commit 45a5df5856
1 changed files with 4 additions and 25 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

29
src/SegmentBase.hpp
View File

@ -40,12 +40,14 @@ namespace CppJieba
Unicode unico; Unicode unico;
while(offset < size) while(offset < size)
{ {
if(-1 == (ret = filterAscii(cstr + offset, size, len))) const char * const nstr = cstr + offset;
uint nsize = size - offset;
if(-1 == (ret = filterAscii(nstr, nsize, len)) || 0 == len || len > nsize)
{ {
LogFatal("str[%s] illegal.", cstr); LogFatal("str[%s] illegal.", cstr);
return false; return false;
} }
subs.assign(cstr + offset, len); subs.assign(nstr, len);
if(!ret) if(!ret)
{ {
res.push_back(subs); res.push_back(subs);
@ -60,32 +62,9 @@ namespace CppJieba
} }
cut(unico.begin(), unico.end(), res); cut(unico.begin(), unico.end(), res);
} }
if(len == 0)
{
LogFatal("str[%s] illegal.", cstr);
return false;
}
offset += len; offset += len;
} }
return true; return true;
//ChineseFilter filter;
//filter.feed(str);
//for(ChineseFilter::iterator it = filter.begin(); it != filter.end(); it++)
//{
// if(it.charType == CHWORD)
// {
// cut(it.begin, it.end, res);
// }
// else
// {
// string tmp;
// if(TransCode::encode(it.begin, it.end, tmp))
// {
// res.push_back(tmp);
// }
// }
//}
//return true;
} }
}; };