Parent: [2c27cb] (diff)

Download this file

plaintorich.cpp    475 lines (427 with data), 14.5 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
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
#ifndef lint
static char rcsid[] = "@(#$Id: plaintorich.cpp,v 1.34 2008-10-13 11:44:17 dockes Exp $ (C) 2005 J.F.Dockes";
#endif
/*
* 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 <string>
#include <utility>
#include <list>
#include <set>
#include <vector>
#include <map>
#include <algorithm>
#ifndef NO_NAMESPACES
using std::vector;
using std::list;
using std::pair;
using std::set;
#endif /* NO_NAMESPACES */
#include "rcldb.h"
#include "rclconfig.h"
#include "debuglog.h"
#include "textsplit.h"
#include "utf8iter.h"
#include "smallut.h"
#include "plaintorich.h"
#include "cancelcheck.h"
const string PlainToRich::snull = "";
// For debug printing
static string vecStringToString(const vector<string>& t)
{
string sterms;
for (vector<string>::const_iterator it = t.begin(); it != t.end(); it++) {
sterms += "[" + *it + "] ";
}
return sterms;
}
// Text splitter callback used to take note of the position of query terms
// inside the result text. This is then used to insert highlight tags.
class myTextSplitCB : public TextSplitCB {
public:
// Out: begin and end byte positions of query terms/groups in text
vector<pair<int, int> > tboffs;
myTextSplitCB(const vector<string>& its,
const vector<vector<string> >&groups,
const vector<int>& slacks)
: m_wcount(0), m_groups(groups), m_slacks(slacks)
{
for (vector<string>::const_iterator it = its.begin();
it != its.end(); it++) {
m_terms.insert(*it);
}
for (vector<vector<string> >::const_iterator vit = m_groups.begin();
vit != m_groups.end(); vit++) {
for (vector<string>::const_iterator it = (*vit).begin();
it != (*vit).end(); it++) {
m_gterms.insert(*it);
}
}
}
// Callback called by the text-to-words breaker for each word
virtual bool takeword(const std::string& term, int pos, int bts, int bte) {
string dumb;
Rcl::dumb_string(term, dumb);
//LOGDEB2(("Input dumbbed term: '%s' %d %d %d\n", dumb.c_str(),
// pos, bts, bte));
// If this word is a search term, remember its byte-offset span.
if (m_terms.find(dumb) != m_terms.end()) {
tboffs.push_back(pair<int, int>(bts, bte));
}
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 %d: %d %d\n", pos, bts, bte));
}
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:
virtual bool matchGroup(const vector<string>& terms, int dist);
int m_wcount;
// In: user query terms
set<string> m_terms;
// In: user query groups, for near/phrase searches.
const vector<vector<string> >& m_groups;
const vector<int>& m_slacks;
set<string> m_gterms;
// group/near terms word positions.
map<string, vector<int> > m_plists;
map<int, pair<int, int> > m_gpostobytes;
};
/** Sort by shorter comparison class */
class VecIntCmpShorter {
public:
/** Return true if and only if a is strictly shorter than b.
*/
bool operator()(const vector<int> *a, const vector<int> *b) {
return a->size() < b->size();
}
};
#define SETMINMAX(POS, STA, STO) {if ((POS) < (STA)) (STA) = (POS); \
if ((POS) > (STO)) (STO) = (POS);}
// Recursively check that each term is inside the window (which is
// readjusted as the successive terms are found). i is the index for
// the next position list to use (initially 1)
static bool do_proximity_test(int window, vector<vector<int>* >& plists,
unsigned int i, int min, int max,
int *sp, int *ep)
{
int tmp = max + 1;
// take care to avoid underflow
if (window <= tmp)
tmp -= window;
else
tmp = 0;
vector<int>::iterator it = plists[i]->begin();
// Find 1st position bigger than window start
while (it != plists[i]->end() && *it < tmp)
it++;
// Try each position inside window in turn for match with other lists
while (it != plists[i]->end()) {
int pos = *it;
if (pos > min + window - 1)
return false;
if (i + 1 == plists.size()) {
SETMINMAX(pos, *sp, *ep);
return true;
}
if (pos < min) {
min = pos;
} else if (pos > max) {
max = pos;
}
if (do_proximity_test(window, plists, i + 1, min, max, sp, ep)) {
SETMINMAX(pos, *sp, *ep);
return true;
}
it++;
}
return false;
}
// Check if there is a NEAR match for the group of terms
bool myTextSplitCB::matchGroup(const vector<string>& terms, int window)
{
LOGDEB0(("myTextSplitCB::matchGroup:d %d: %s\n", window,
vecStringToString(terms).c_str()));
// The position lists we are going to work with. We extract them from the
// (string->plist) map
vector<vector<int>* > plists;
// A revert plist->term map. This is so that we can find who is who after
// sorting the plists by length.
map<vector<int>*, string> plistToTerm;
// For traces
vector<string> realgroup;
// Find the position list for each term in the group. Not all
// necessarily exist (esp for NEAR where terms have been
// stem-expanded: we don't know which matched)
for (vector<string>::const_iterator it = terms.begin();
it != terms.end(); it++) {
map<string, vector<int> >::iterator pl = m_plists.find(*it);
if (pl == m_plists.end()) {
LOGDEB0(("myTextSplitCB::matchGroup: [%s] not found in m_plists\n",
(*it).c_str()));
continue;
}
plists.push_back(&(pl->second));
plistToTerm[&(pl->second)] = *it;
realgroup.push_back(*it);
}
LOGDEB0(("myTextSplitCB::matchGroup:d %d:real group after expansion %s\n",
window, vecStringToString(realgroup).c_str()));
if (plists.size() < 2) {
LOGDEB0(("myTextSplitCB::matchGroup: no actual groups found\n"));
return false;
}
// Sort the positions lists so that the shorter is first
std::sort(plists.begin(), plists.end(), VecIntCmpShorter());
{ // Debug
map<vector<int>*, string>::iterator it;
it = plistToTerm.find(plists[0]);
if (it == plistToTerm.end()) {
// SuperWeird
LOGERR(("matchGroup: term for first list not found !?!\n"));
return false;
}
LOGDEB0(("matchGroup: walking the shortest plist. Term [%s], len %d\n",
it->second.c_str(), plists[0]->size()));
}
// Walk the shortest plist and look for matches
for (vector<int>::iterator it = plists[0]->begin();
it != plists[0]->end(); it++) {
int pos = *it;
int sta = int(10E9), sto = 0;
LOGDEB0(("MatchGroup: Testing at pos %d\n", pos));
if (do_proximity_test(window, plists, 1, pos, pos, &sta, &sto)) {
LOGDEB0(("myTextSplitCB::matchGroup: MATCH termpos [%d,%d]\n",
sta, sto));
// Maybe extend the window by 1st term position, this was not
// done by do_prox..
SETMINMAX(pos, sta, sto);
// Translate the position window into a byte offset window
int bs = 0;
map<int, pair<int, int> >::iterator i1 = m_gpostobytes.find(sta);
map<int, pair<int, int> >::iterator i2 = m_gpostobytes.find(sto);
if (i1 != m_gpostobytes.end() && i2 != m_gpostobytes.end()) {
LOGDEB0(("myTextSplitCB::matchGroup: pushing bpos %d %d\n",
i1->second.first, i2->second.second));
tboffs.push_back(pair<int, int>(i1->second.first,
i2->second.second));
bs = i1->second.first;
} else {
LOGDEB(("matchGroup: no bpos found for %d or %d\n", sta, sto));
}
}
}
return true;
}
/** Sort integer pairs by increasing first value and decreasing width */
class PairIntCmpFirst {
public:
bool operator()(pair<int,int> a, pair<int, int>b) {
if (a.first != b.first)
return a.first < b.first;
return a.second > b.second;
}
};
// Do the phrase match thing, then merge the highlight lists
bool myTextSplitCB::matchGroups()
{
vector<vector<string> >::const_iterator vit = m_groups.begin();
vector<int>::const_iterator sit = m_slacks.begin();
for (; vit != m_groups.end() && sit != m_slacks.end(); vit++, sit++) {
matchGroup(*vit, *sit + (*vit).size());
}
// Sort by start and end offsets. The merging of overlapping entries
// will be handled during output.
std::sort(tboffs.begin(), tboffs.end(), PairIntCmpFirst());
return true;
}
// Fix result text for display inside the gui text window.
//
// To compute the term character positions in the output text, we used
// to emulate how qt's textedit counts chars (ignoring tags and
// duplicate whitespace etc...). This was tricky business, dependant
// on qtextedit internals, and we don't do it any more, so we finally
// don't know the term par/car positions in the editor text.
// Instead, we now mark the search term positions with html anchors
//
// We output the result in chunks, arranging not to cut in the middle of
// a tag, which would confuse qtextedit.
bool PlainToRich::plaintorich(const string& in,
list<string>& out, // Output chunk list
const HiliteData& hdata,
int chunksize)
{
Chrono chron;
const vector<string>& terms(hdata.terms);
const vector<vector<string> >& groups(hdata.groups);
const vector<int>& slacks(hdata.gslks);
if (DebugLog::getdbl()->getlevel() >= DEBDEB0) {
LOGDEB0(("plaintorich: terms: \n"));
string sterms = vecStringToString(terms);
LOGDEB0((" %s\n", sterms.c_str()));
sterms = "\n";
LOGDEB0(("plaintorich: groups: \n"));
for (vector<vector<string> >::const_iterator vit = groups.begin();
vit != groups.end(); vit++) {
sterms += "GROUP: ";
sterms += vecStringToString(*vit);
sterms += "\n";
}
LOGDEB0((" %s", sterms.c_str()));
}
// 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,
myTextSplitCB cb(terms, groups, slacks);
TextSplit splitter(&cb);
// Note: the splitter returns the term locations in byte, not
// character, offsets.
splitter.text_to_words(in);
LOGDEB0(("plaintorich: split done %d mS\n", chron.millis()));
// Compute the positions for NEAR and PHRASE groups.
cb.matchGroups();
out.clear();
out.push_back("");
list<string>::iterator olit = out.begin();
// Rich text output
*olit = header();
// 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<pair<int, int> >::iterator tPosIt = cb.tboffs.begin();
vector<pair<int, int> >::iterator tPosEnd = cb.tboffs.end();
#if 0
for (vector<pair<int, int> >::const_iterator it = cb.tboffs.begin();
it != cb.tboffs.end(); it++) {
LOGDEB2(("plaintorich: region: %d %d\n", it->first, it->second));
}
#endif
// Input character iterator
Utf8Iter chariter(in);
// State variable used to limit the number of consecutive empty lines
int ateol = 0;
// Value for numbered anchors at each term match
int anchoridx = 1;
// html state
bool intag = false, inparamvalue = false;
unsigned int 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 = chariter.getBpos();
if (ibyteidx == tPosIt->first) {
if (!intag && ibyteidx > (int)headend) {
*olit += startAnchor(anchoridx);
*olit += startMatch();
}
anchoridx++;
} else if (ibyteidx == tPosIt->second) {
// Output end or match region tags
if (!intag && ibyteidx > (int)headend) {
*olit += endMatch();
*olit += endAnchor();
}
// Skip all highlight areas that would overlap this one
int crend = tPosIt->second;
while (tPosIt != cb.tboffs.end() && tPosIt->first < crend)
tPosIt++;
// 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 && olit->size() > (unsigned int)chunksize) {
out.push_back("");
olit++;
}
}
}
if (m_inputhtml) {
switch (*chariter) {
case '<':
if (!inparamvalue)
intag = true;
break;
case '>':
if (!inparamvalue)
intag = false;
break;
case '"':
if (intag) {
inparamvalue = !inparamvalue;
}
break;
}
chariter.appendchartostring(*olit);
} else switch (*chariter) {
case '\n':
if (ateol < 2) {
*olit += "<br>\n";
ateol++;
}
break;
case '\r':
break;
case '<':
ateol = 0;
*olit += "&lt;";
break;
case '&':
ateol = 0;
*olit += "&amp;";
break;
default:
// We don't change the eol status for whitespace, want
// a real line
if (!(*chariter == ' ' || *chariter == '\t')) {
ateol = 0;
}
chariter.appendchartostring(*olit);
}
}
#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
LOGDEB0(("plaintorich: done %d mS\n", chron.millis()));
return true;
}