|
a/src/qtgui/plaintorich.cpp |
|
b/src/qtgui/plaintorich.cpp |
| 1 |
#ifndef lint
|
1 |
#ifndef lint
|
| 2 |
static char rcsid[] = "@(#$Id: plaintorich.cpp,v 1.6 2005-11-24 07:16:16 dockes Exp $ (C) 2005 J.F.Dockes";
|
2 |
static char rcsid[] = "@(#$Id: plaintorich.cpp,v 1.7 2006-01-19 12:01:42 dockes Exp $ (C) 2005 J.F.Dockes";
|
| 3 |
#endif
|
3 |
#endif
|
| 4 |
|
4 |
|
| 5 |
|
5 |
|
| 6 |
#include <string>
|
6 |
#include <string>
|
| 7 |
#include <utility>
|
7 |
#include <utility>
|
|
... |
|
... |
| 47 |
|
47 |
|
| 48 |
return true;
|
48 |
return true;
|
| 49 |
}
|
49 |
}
|
| 50 |
};
|
50 |
};
|
| 51 |
|
51 |
|
| 52 |
// Just strip/escape things that would look like markup
|
|
|
| 53 |
string stripMarkup(const string &in)
|
|
|
| 54 |
{
|
|
|
| 55 |
string out;
|
|
|
| 56 |
for (string::size_type pos = 0; pos < in.length(); pos++) {
|
|
|
| 57 |
switch(in.at(pos)) {
|
|
|
| 58 |
case '<':
|
|
|
| 59 |
out += "<";
|
|
|
| 60 |
break;
|
|
|
| 61 |
case '&':
|
|
|
| 62 |
out += "&";
|
|
|
| 63 |
break;
|
|
|
| 64 |
default:
|
|
|
| 65 |
out += in.at(pos);
|
|
|
| 66 |
}
|
|
|
| 67 |
}
|
|
|
| 68 |
return out;
|
|
|
| 69 |
}
|
|
|
| 70 |
|
|
|
| 71 |
// Fix result text for display inside the gui text window.
|
52 |
// Fix result text for display inside the gui text window.
|
| 72 |
//
|
53 |
//
|
| 73 |
// To compute the term character positions in the output text, we have
|
54 |
// To compute the term character positions in the output text, we have
|
| 74 |
// to emulate how qt's textedit counts chars (ignoring tags and
|
55 |
// to emulate how qt's textedit counts chars (ignoring tags and
|
| 75 |
// duplicate whitespace etc...). This is tricky business and it might
|
56 |
// duplicate whitespace etc...). This is tricky business and it might
|
|
... |
|
... |
| 81 |
LOGDEB(("plaintorich: terms: %s\n",
|
62 |
LOGDEB(("plaintorich: terms: %s\n",
|
| 82 |
stringlistdisp(terms).c_str()));
|
63 |
stringlistdisp(terms).c_str()));
|
| 83 |
|
64 |
|
| 84 |
termoffsets.erase(termoffsets.begin(), termoffsets.end());
|
65 |
termoffsets.erase(termoffsets.begin(), termoffsets.end());
|
| 85 |
|
66 |
|
|
|
67 |
// We first use the text splitter to break the text into words,
|
|
|
68 |
// and compare the words to the search terms, which yields the
|
|
|
69 |
// query terms positions inside the text
|
| 86 |
myTextSplitCB cb(terms);
|
70 |
myTextSplitCB cb(terms);
|
| 87 |
TextSplit splitter(&cb, true);
|
71 |
TextSplit splitter(&cb, true);
|
| 88 |
// Note that splitter returns the term locations in byte, not
|
72 |
// Note that splitter returns the term locations in byte, not
|
| 89 |
// character offset
|
73 |
// character offset
|
| 90 |
splitter.text_to_words(in);
|
74 |
splitter.text_to_words(in);
|