Parent: [ce740a] (diff)

Child: [293619] (diff)

Download this file

plaintorich.cpp    169 lines (152 with data), 4.4 kB

  1
  2
  3
  4
  5
  6
  7
  8
  9
 10
 11
 12
 13
 14
 15
 16
 17
 18
 19
 20
 21
 22
 23
 24
 25
 26
 27
 28
 29
 30
 31
 32
 33
 34
 35
 36
 37
 38
 39
 40
 41
 42
 43
 44
 45
 46
 47
 48
 49
 50
 51
 52
 53
 54
 55
 56
 57
 58
 59
 60
 61
 62
 63
 64
 65
 66
 67
 68
 69
 70
 71
 72
 73
 74
 75
 76
 77
 78
 79
 80
 81
 82
 83
 84
 85
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
#ifndef lint
static char rcsid[] = "@(#$Id: plaintorich.cpp,v 1.4 2005-10-20 08:34:51 dockes Exp $ (C) 2005 J.F.Dockes";
#endif
#include <string>
#include <utility>
#include <list>
using std::list;
using std::pair;
#include "rcldb.h"
#include "rclconfig.h"
#include "debuglog.h"
#include "textsplit.h"
#include "utf8iter.h"
#include "transcode.h"
#include "smallut.h"
// Text splitter callback used to take note of the position of query terms
// inside the result text. This is then used to post highlight tags.
class myTextSplitCB : public TextSplitCB {
public:
const list<string> *terms; // in: query terms
list<pair<int, int> > tboffs; // out: begin and end positions of
// query terms in text
myTextSplitCB(const list<string>& terms)
: terms(&terms) {
}
// Callback called by the text-to-words breaker for each word
virtual bool takeword(const std::string& term, int, int bts, int bte) {
string dumb;
Rcl::dumb_string(term, dumb);
//LOGDEB(("Input dumbbed term: '%s' %d %d %d\n", dumb.c_str(),
// pos, bts, bte));
for (list<string>::const_iterator it = terms->begin();
it != terms->end(); it++) {
if (!stringlowercmp(*it, dumb)) {
tboffs.push_back(pair<int, int>(bts, bte));
break;
}
}
return true;
}
};
// Just strip/escape things that would look like markup
string stripMarkup(const string &in)
{
string out;
for (string::size_type pos = 0; pos < in.length(); pos++) {
switch(in.at(pos)) {
case '<':
out += "&lt;";
break;
case '&':
out += "&amp;";
break;
default:
out += in.at(pos);
}
}
return out;
}
// Fix result text for display inside the gui text window.
//
// To compute the term character positions in the output text, we have
// to emulate how qt's textedit counts chars (ignoring tags and
// duplicate whitespace etc...). This is tricky business and it might
// be better to insert the text char by char, taking note of where qt
// thinks it is at each term.
string plaintorich(const string &in, const list<string>& terms,
list<pair<int, int> >&termoffsets)
{
LOGDEB(("plaintorich: terms: %s\n",
stringlistdisp(terms).c_str()));
termoffsets.erase(termoffsets.begin(), termoffsets.end());
myTextSplitCB cb(terms);
TextSplit splitter(&cb, true);
// Note that splitter returns the term locations in byte, not
// character offset
splitter.text_to_words(in);
// Rich text output
string out = "<qt><head><title></title></head><body><p>";
// Iterator for the list of input term positions. We use it to
// output highlight tags and to compute term positions in the
// output text
list<pair<int, int> >::iterator it = cb.tboffs.begin();
// Storage for the current term _character_ position in output.
pair<int, int> otermcpos;
// Current char position in output, excluding tags
int outcpos=0;
// Input character iterator
Utf8Iter chariter(in);
// State variable used to limitate the number of consecutive empty lines
int ateol = 0;
// State variable to update the char pos only for the first of
// consecutive blank chars
int atblank = 0;
for (string::size_type pos = 0; pos != string::npos; pos = chariter++) {
// If we still have terms, check (byte) position
if (it != cb.tboffs.end()) {
int ibyteidx = chariter.getBpos();
if (ibyteidx == it->first) {
out += "<termtag>";
otermcpos.first = outcpos;
} else if (ibyteidx == it->second) {
if (it != cb.tboffs.end())
it++;
otermcpos.second = outcpos;
termoffsets.push_back(otermcpos);
out += "</termtag>";
}
}
switch(*chariter) {
case '\n':
if (ateol < 2) {
out += "<br>\n";
ateol++;
outcpos++;
}
break;
case '\r':
break;
case '<':
ateol = 0;
out += "&lt;";
outcpos++;
break;
case '&':
ateol = 0;
out += "&amp;";
outcpos++;
break;
default:
// We don't change the eol status for whitespace, want a real line
if (*chariter == ' ' || *chariter == ' ') {
if (!atblank)
outcpos++;
atblank = 1;
} else {
ateol = 0;
atblank = 0;
outcpos++;
}
chariter.appendchartostring(out);
}
}
#if 0
{
FILE *fp = fopen("/tmp/debugplaintorich", "w");
fprintf(fp, "%s\n", out.c_str());
fclose(fp);
}
#endif
return out;
}