Parent: [2be261] (diff)

Download this file

plaintorich.cpp    388 lines (345 with data), 13.1 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
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
/* Copyright (C) 2005 J.F.Dockes
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation; either version 2 of the License, or
* (at your option) any later version.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program; if not, write to the
* Free Software Foundation, Inc.,
* 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
*/
#include <limits.h>
#include <string>
#include <utility>
#include <list>
#include <set>
#include <vector>
#include <map>
#include <algorithm>
#include <regex>
using std::vector;
using std::list;
using std::pair;
using std::set;
#include "rcldb.h"
#include "rclconfig.h"
#include "log.h"
#include "textsplit.h"
#include "utf8iter.h"
#include "smallut.h"
#include "chrono.h"
#include "plaintorich.h"
#include "cancelcheck.h"
#include "unacpp.h"
// Text splitter used to take note of the position of query terms
// inside the result text. This is then used to insert highlight tags.
class TextSplitPTR : public TextSplit {
public:
// Out: begin and end byte positions of query terms/groups in text
vector<GroupMatchEntry> m_tboffs;
TextSplitPTR(const HighlightData& hdata)
: m_wcount(0), m_hdata(hdata) {
// We separate single terms and groups and extract the group
// terms for computing positions list before looking for group
// matches
for (vector<vector<string> >::const_iterator vit = hdata.groups.begin();
vit != hdata.groups.end(); vit++) {
if (vit->size() == 1) {
m_terms[vit->front()] = vit - hdata.groups.begin();
} else if (vit->size() > 1) {
for (vector<string>::const_iterator it = vit->begin();
it != vit->end(); it++) {
m_gterms.insert(*it);
}
}
}
}
// Accept word and its position. If word is search term, add
// highlight zone definition. If word is part of search group
// (phrase or near), update positions list.
virtual bool takeword(const std::string& term, int pos, int bts, int bte) {
string dumb = term;
if (o_index_stripchars) {
if (!unacmaybefold(term, dumb, "UTF-8", UNACOP_UNACFOLD)) {
LOGINFO("PlainToRich::takeword: unac failed for [" << term <<
"]\n");
return true;
}
}
LOGDEB2("Input dumbbed term: '" << dumb << "' " << pos << " " << bts
<< " " << bte << "\n");
// If this word is a search term, remember its byte-offset span.
map<string, size_t>::const_iterator it = m_terms.find(dumb);
if (it != m_terms.end()) {
m_tboffs.push_back(GroupMatchEntry(bts, bte, it->second));
}
// If word is part of a search group, update its positions list
if (m_gterms.find(dumb) != m_gterms.end()) {
// Term group (phrase/near) handling
m_plists[dumb].push_back(pos);
m_gpostobytes[pos] = pair<int,int>(bts, bte);
LOGDEB2("Recorded bpos for " << pos << ": " << bts << " " <<
bte << "\n");
}
// Check for cancellation request
if ((m_wcount++ & 0xfff) == 0)
CancelCheck::instance().checkCancel();
return true;
}
// Must be called after the split to find the phrase/near match positions
virtual bool matchGroups();
private:
// Word count. Used to call checkCancel from time to time.
int m_wcount;
// In: user query terms
map<string, size_t> m_terms;
// m_gterms holds all the terms in m_groups, as a set for quick lookup
set<string> m_gterms;
const HighlightData& m_hdata;
// group/near terms word positions.
map<string, vector<int> > m_plists;
map<int, pair<int, int> > m_gpostobytes;
};
// Look for matches to PHRASE and NEAR term groups and finalize the
// matched regions list (sort it by increasing start then decreasing
// length)
// Actually, we handle all groups as NEAR (ignore order).
bool TextSplitPTR::matchGroups()
{
for (unsigned int i = 0; i < m_hdata.groups.size(); i++) {
if (m_hdata.groups[i].size() <= 1)
continue;
matchGroup(m_hdata, i, m_plists, m_gpostobytes, m_tboffs);
}
// Sort regions by increasing start and decreasing width.
// The output process will skip overlapping entries.
std::sort(m_tboffs.begin(), m_tboffs.end(),
[](const GroupMatchEntry& a, const GroupMatchEntry& b) -> bool {
if (a.offs.first != b.offs.first)
return a.offs.first < b.offs.first;
return a.offs.second > b.offs.second;
}
);
return true;
}
// Replace HTTP(s) urls in text/plain with proper HTML anchors so that
// they become clickable in the preview. We don't make a lot of effort
// for validating, or catching things which are probably urls but miss
// a scheme (e.g. www.xxx.com/index.html), because complicated.
static const string urlRE = "(https?://[[:alnum:]~_/.%?&=,#@]+)[[:space:]|]";
static const string urlRep{"<a href=\"$1\">$1</a>"};
static std::regex url_re(urlRE);
static string activate_urls(const string& in)
{
return std::regex_replace(in, url_re, urlRep);
}
// Fix result text for display inside the gui text window.
//
// We call overridden functions to output header data, beginnings and ends of
// matches etc.
//
// If the input is text, we output the result in chunks, arranging not
// to cut in the middle of a tag, which would confuse qtextedit. If
// the input is html, the body is always a single output chunk.
bool PlainToRich::plaintorich(const string& in,
list<string>& out, // Output chunk list
const HighlightData& hdata,
int chunksize)
{
Chrono chron;
bool ret = true;
LOGDEB1("plaintorichich: in: [" << in << "]\n");
m_hdata = &hdata;
// Compute the positions for the query terms. We use the text
// splitter to break the text into words, and compare the words to
// the search terms,
TextSplitPTR splitter(hdata);
// Note: the splitter returns the term locations in byte, not
// character, offsets.
splitter.text_to_words(in);
LOGDEB2("plaintorich: split done " << chron.millis() << " mS\n");
// Compute the positions for NEAR and PHRASE groups.
splitter.matchGroups();
LOGDEB2("plaintorich: group match done " << chron.millis() << " mS\n");
out.clear();
out.push_back("");
list<string>::iterator olit = out.begin();
// Rich text output
*olit = header();
// No term matches. Happens, for example on a snippet selected for
// a term match when we are actually looking for a group match
// (the snippet generator does this...).
if (splitter.m_tboffs.empty()) {
LOGDEB1("plaintorich: no term matches\n");
ret = false;
}
// Iterator for the list of input term positions. We use it to
// output highlight tags and to compute term positions in the
// output text
vector<GroupMatchEntry>::iterator tPosIt = splitter.m_tboffs.begin();
vector<GroupMatchEntry>::iterator tPosEnd = splitter.m_tboffs.end();
#if 0
for (vector<pair<int, int> >::const_iterator it = splitter.m_tboffs.begin();
it != splitter.m_tboffs.end(); it++) {
LOGDEB2("plaintorich: region: " << it->first << " "<<it->second<< "\n");
}
#endif
// Input character iterator
Utf8Iter chariter(in);
// State variables used to limit the number of consecutive empty lines,
// convert all eol to '\n', and preserve some indentation
int eol = 0;
int hadcr = 0;
int inindent = 1;
// HTML state
bool intag = false, inparamvalue = false;
// My tag state
int inrcltag = 0;
string::size_type headend = 0;
if (m_inputhtml) {
headend = in.find("</head>");
if (headend == string::npos)
headend = in.find("</HEAD>");
if (headend != string::npos)
headend += 7;
}
for (string::size_type pos = 0; pos != string::npos; pos = chariter++) {
// Check from time to time if we need to stop
if ((pos & 0xfff) == 0) {
CancelCheck::instance().checkCancel();
}
// If we still have terms positions, check (byte) position. If
// we are at or after a term match, mark.
if (tPosIt != tPosEnd) {
int ibyteidx = int(chariter.getBpos());
if (ibyteidx == tPosIt->offs.first) {
if (!intag && ibyteidx >= (int)headend) {
*olit += startMatch((unsigned int)(tPosIt->grpidx));
}
inrcltag = 1;
} else if (ibyteidx == tPosIt->offs.second) {
// Output end of match region tags
if (!intag && ibyteidx > (int)headend) {
*olit += endMatch();
}
// Skip all highlight areas that would overlap this one
int crend = tPosIt->offs.second;
while (tPosIt != splitter.m_tboffs.end() &&
tPosIt->offs.first < crend)
tPosIt++;
inrcltag = 0;
}
}
unsigned int car = *chariter;
if (car == '\n') {
if (!hadcr)
eol++;
hadcr = 0;
continue;
} else if (car == '\r') {
hadcr++;
eol++;
continue;
} else if (eol) {
// Got non eol char in line break state. Do line break;
inindent = 1;
hadcr = 0;
if (eol > 2)
eol = 2;
while (eol) {
if (!m_inputhtml && m_eolbr)
*olit += "<br>";
*olit += "\n";
eol--;
}
// Maybe end this chunk, begin next. Don't do it on html
// there is just no way to do it right (qtextedit cant grok
// chunks cut in the middle of <a></a> for example).
if (!m_inputhtml && !inrcltag &&
olit->size() > (unsigned int)chunksize) {
if (m_activatelinks) {
*olit = activate_urls(*olit);
}
out.push_back(string(startChunk()));
olit++;
}
}
switch (car) {
case '<':
inindent = 0;
if (m_inputhtml) {
if (!inparamvalue)
intag = true;
chariter.appendchartostring(*olit);
} else {
*olit += "&lt;";
}
break;
case '>':
inindent = 0;
if (m_inputhtml) {
if (!inparamvalue)
intag = false;
}
chariter.appendchartostring(*olit);
break;
case '&':
inindent = 0;
if (m_inputhtml) {
chariter.appendchartostring(*olit);
} else {
*olit += "&amp;";
}
break;
case '"':
inindent = 0;
if (m_inputhtml && intag) {
inparamvalue = !inparamvalue;
}
chariter.appendchartostring(*olit);
break;
case ' ':
if (m_eolbr && inindent) {
*olit += "&nbsp;";
} else {
chariter.appendchartostring(*olit);
}
break;
case '\t':
if (m_eolbr && inindent) {
*olit += "&nbsp;&nbsp;&nbsp;&nbsp;";
} else {
chariter.appendchartostring(*olit);
}
break;
default:
inindent = 0;
chariter.appendchartostring(*olit);
}
} // End chariter loop
#if 0
{
FILE *fp = fopen("/tmp/debugplaintorich", "a");
fprintf(fp, "BEGINOFPLAINTORICHOUTPUT\n");
for (list<string>::iterator it = out.begin();
it != out.end(); it++) {
fprintf(fp, "BEGINOFPLAINTORICHCHUNK\n");
fprintf(fp, "%s", it->c_str());
fprintf(fp, "ENDOFPLAINTORICHCHUNK\n");
}
fprintf(fp, "ENDOFPLAINTORICHOUTPUT\n");
fclose(fp);
}
#endif
LOGDEB2("plaintorich: done " << chron.millis() << " mS\n");
if (!m_inputhtml && m_activatelinks) {
out.back() = activate_urls(out.back());
}
return ret;
}