|
a/src/common/textsplit.cpp |
|
b/src/common/textsplit.cpp |
| 1 |
#ifndef lint
|
1 |
#ifndef lint
|
| 2 |
static char rcsid[] = "@(#$Id: textsplit.cpp,v 1.19 2006-01-30 09:28:16 dockes Exp $ (C) 2004 J.F.Dockes";
|
2 |
static char rcsid[] = "@(#$Id: textsplit.cpp,v 1.20 2006-02-01 14:18:20 dockes Exp $ (C) 2004 J.F.Dockes";
|
| 3 |
#endif
|
3 |
#endif
|
| 4 |
/*
|
4 |
/*
|
| 5 |
* This program is free software; you can redistribute it and/or modify
|
5 |
* This program is free software; you can redistribute it and/or modify
|
| 6 |
* it under the terms of the GNU General Public License as published by
|
6 |
* it under the terms of the GNU General Public License as published by
|
| 7 |
* the Free Software Foundation; either version 2 of the License, or
|
7 |
* the Free Software Foundation; either version 2 of the License, or
|
|
... |
|
... |
| 177 |
if (!emitterm(false, word, wordpos, bp-word.length(), bp))
|
177 |
if (!emitterm(false, word, wordpos, bp-word.length(), bp))
|
| 178 |
return false;
|
178 |
return false;
|
| 179 |
|
179 |
|
| 180 |
// Adjust state
|
180 |
// Adjust state
|
| 181 |
wordpos++;
|
181 |
wordpos++;
|
| 182 |
word.clear();
|
182 |
word.erase();
|
| 183 |
if (spanerase) {
|
183 |
if (spanerase) {
|
| 184 |
span.clear();
|
184 |
span.erase();
|
| 185 |
spanpos = wordpos;
|
185 |
spanpos = wordpos;
|
| 186 |
}
|
186 |
}
|
| 187 |
|
187 |
|
| 188 |
return true;
|
188 |
return true;
|
| 189 |
}
|
189 |
}
|
|
... |
|
... |
| 210 |
{
|
210 |
{
|
| 211 |
LOGDEB2(("TextSplit::text_to_words: cb %p\n", cb));
|
211 |
LOGDEB2(("TextSplit::text_to_words: cb %p\n", cb));
|
| 212 |
|
212 |
|
| 213 |
setcharclasses();
|
213 |
setcharclasses();
|
| 214 |
|
214 |
|
| 215 |
span.clear();
|
215 |
span.erase();
|
| 216 |
word.clear(); // Current word: no punctuation at all in there
|
216 |
word.erase(); // Current word: no punctuation at all in there
|
| 217 |
number = false;
|
217 |
number = false;
|
| 218 |
wordpos = spanpos = charpos = 0;
|
218 |
wordpos = spanpos = charpos = 0;
|
| 219 |
|
219 |
|
| 220 |
Utf8Iter it(in);
|
220 |
Utf8Iter it(in);
|
| 221 |
|
221 |
|