Child: [ebbcc1] (diff)

Download this file

searchdata.cpp    976 lines (891 with data), 32.5 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
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
/* Copyright (C) 2006 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.
*/
// Handle translation from rcl's SearchData structures to Xapian Queries
#include <stdio.h>
#include <fnmatch.h>
#include <string>
#include <vector>
#include <algorithm>
#include "xapian.h"
#include "rcldb.h"
#include "searchdata.h"
#include "debuglog.h"
#include "smallut.h"
#include "textsplit.h"
#include "unacpp.h"
#include "utf8iter.h"
#include "stoplist.h"
#include "rclconfig.h"
#ifndef NO_NAMESPACES
using namespace std;
namespace Rcl {
#endif
typedef vector<SearchDataClause *>::iterator qlist_it_t;
typedef vector<SearchDataClause *>::const_iterator qlist_cit_t;
static const int original_term_wqf_booster = 10;
/* The dates-to-query routine is is lifted quasi-verbatim but
* modified from xapian-omega:date.cc. Copyright info:
*
* Copyright 1999,2000,2001 BrightStation PLC
* Copyright 2001 James Aylett
* Copyright 2001,2002 Ananova Ltd
* Copyright 2002 Intercede 1749 Ltd
* Copyright 2002,2003,2006 Olly Betts
*
* 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., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301
* USA
*/
static Xapian::Query
date_range_filter(int y1, int m1, int d1, int y2, int m2, int d2)
{
// Xapian uses a smallbuf and snprintf. Can't be bothered, we're
// only doing %d's !
char buf[200];
sprintf(buf, "D%04d%02d", y1, m1);
vector<Xapian::Query> v;
int d_last = monthdays(m1, y1);
int d_end = d_last;
if (y1 == y2 && m1 == m2 && d2 < d_last) {
d_end = d2;
}
// Deal with any initial partial month
if (d1 > 1 || d_end < d_last) {
for ( ; d1 <= d_end ; d1++) {
sprintf(buf + 7, "%02d", d1);
v.push_back(Xapian::Query(buf));
}
} else {
buf[0] = 'M';
v.push_back(Xapian::Query(buf));
}
if (y1 == y2 && m1 == m2) {
return Xapian::Query(Xapian::Query::OP_OR, v.begin(), v.end());
}
int m_last = (y1 < y2) ? 12 : m2 - 1;
while (++m1 <= m_last) {
sprintf(buf + 5, "%02d", m1);
buf[0] = 'M';
v.push_back(Xapian::Query(buf));
}
if (y1 < y2) {
while (++y1 < y2) {
sprintf(buf + 1, "%04d", y1);
buf[0] = 'Y';
v.push_back(Xapian::Query(buf));
}
sprintf(buf + 1, "%04d", y2);
buf[0] = 'M';
for (m1 = 1; m1 < m2; m1++) {
sprintf(buf + 5, "%02d", m1);
v.push_back(Xapian::Query(buf));
}
}
sprintf(buf + 5, "%02d", m2);
// Deal with any final partial month
if (d2 < monthdays(m2, y2)) {
buf[0] = 'D';
for (d1 = 1 ; d1 <= d2; d1++) {
sprintf(buf + 7, "%02d", d1);
v.push_back(Xapian::Query(buf));
}
} else {
buf[0] = 'M';
v.push_back(Xapian::Query(buf));
}
return Xapian::Query(Xapian::Query::OP_OR, v.begin(), v.end());
}
// Expand categories and mime type wild card exps
bool SearchData::expandFileTypes(RclConfig *cfg, vector<string>& tps)
{
if (!cfg) {
LOGFATAL(("Db::expandFileTypes: null configuration!!\n"));
return false;
}
vector<string> exptps;
list<string> alltypes = cfg->getAllMimeTypes();
for (vector<string>::iterator it = tps.begin(); it != tps.end(); it++) {
if (cfg->isMimeCategory(*it)) {
list<string>tps;
cfg->getMimeCatTypes(*it, tps);
exptps.insert(exptps.end(), tps.begin(), tps.end());
} else {
for (list<string>::const_iterator ait = alltypes.begin();
ait != alltypes.end(); ait++) {
if (fnmatch(it->c_str(), ait->c_str(), FNM_CASEFOLD)
!= FNM_NOMATCH) {
exptps.push_back(*ait);
}
}
}
}
tps = exptps;
return true;
}
bool SearchData::toNativeQuery(Rcl::Db &db, void *d)
{
Xapian::Query xq;
m_reason.erase();
if (!m_query.size() && !m_haveDates) {
m_reason = "empty query";
return false;
}
// Walk the clause list translating each in turn and building the
// Xapian query tree
for (qlist_it_t it = m_query.begin(); it != m_query.end(); it++) {
Xapian::Query nq;
if (!(*it)->toNativeQuery(db, &nq, m_stemlang)) {
LOGERR(("SearchData::toNativeQuery: failed\n"));
m_reason = (*it)->getReason();
return false;
}
if (nq.empty()) {
LOGDEB(("SearchData::toNativeQuery: skipping empty clause\n"));
continue;
}
// If this structure is an AND list, must use AND_NOT for excl clauses.
// Else this is an OR list, and there can't be excl clauses (checked by
// addClause())
Xapian::Query::op op;
if (m_tp == SCLT_AND) {
if ((*it)->m_tp == SCLT_EXCL) {
op = Xapian::Query::OP_AND_NOT;
} else {
op = Xapian::Query::OP_AND;
}
} else {
op = Xapian::Query::OP_OR;
}
if (xq.empty()) {
if (op == Xapian::Query::OP_AND_NOT)
xq = Xapian::Query(op, Xapian::Query::MatchAll, nq);
else
xq = nq;
} else {
xq = Xapian::Query(op, xq, nq);
}
}
if (m_haveDates) {
// If one of the extremities is unset, compute db extremas
if (m_dates.y1 == 0 || m_dates.y2 == 0) {
int minyear = 1970, maxyear = 2100;
if (!db.maxYearSpan(&minyear, &maxyear)) {
LOGERR(("Can't retrieve index min/max dates\n"));
//whatever, go on.
}
if (m_dates.y1 == 0) {
m_dates.y1 = minyear;
m_dates.m1 = 1;
m_dates.d1 = 1;
}
if (m_dates.y2 == 0) {
m_dates.y2 = maxyear;
m_dates.m2 = 12;
m_dates.d2 = 31;
}
}
LOGDEB(("Db::toNativeQuery: date interval: %d-%d-%d/%d-%d-%d\n",
m_dates.y1, m_dates.m1, m_dates.d1,
m_dates.y2, m_dates.m2, m_dates.d2));
Xapian::Query dq = date_range_filter(m_dates.y1, m_dates.m1, m_dates.d1,
m_dates.y2, m_dates.m2, m_dates.d2);
if (dq.empty()) {
LOGINFO(("Db::toNativeQuery: date filter is empty\n"));
}
// If no probabilistic query is provided then promote the daterange
// filter to be THE query instead of filtering an empty query.
if (xq.empty()) {
LOGINFO(("Db::toNativeQuery: proba query is empty\n"));
xq = dq;
} else {
xq = Xapian::Query(Xapian::Query::OP_FILTER, xq, dq);
}
}
// Add the file type filtering clause if any
if (!m_filetypes.empty()) {
expandFileTypes(db.getConf(), m_filetypes);
Xapian::Query tq;
for (vector<string>::iterator it = m_filetypes.begin();
it != m_filetypes.end(); it++) {
string term = "T" + *it;
LOGDEB0(("Adding file type term: [%s]\n", term.c_str()));
tq = tq.empty() ? Xapian::Query(term) :
Xapian::Query(Xapian::Query::OP_OR, tq, Xapian::Query(term));
}
xq = xq.empty() ? tq : Xapian::Query(Xapian::Query::OP_FILTER, xq, tq);
}
// Add the neg file type filtering clause if any
if (!m_nfiletypes.empty()) {
expandFileTypes(db.getConf(), m_nfiletypes);
Xapian::Query tq;
for (vector<string>::iterator it = m_nfiletypes.begin();
it != m_nfiletypes.end(); it++) {
string term = "T" + *it;
LOGDEB0(("Adding negative file type term: [%s]\n", term.c_str()));
tq = tq.empty() ? Xapian::Query(term) :
Xapian::Query(Xapian::Query::OP_OR, tq, Xapian::Query(term));
}
xq = xq.empty() ? tq : Xapian::Query(Xapian::Query::OP_AND_NOT, xq, tq);
}
// Add the directory filtering clause. This is a phrase of terms
// prefixed with the pathelt prefix XP
if (!m_topdir.empty()) {
vector<string> vpath;
stringToTokens(m_topdir, vpath, "/");
vector<string> pvpath;
pvpath.push_back(pathelt_prefix);
for (vector<string>::const_iterator it = vpath.begin();
it != vpath.end(); it++){
pvpath.push_back(pathelt_prefix + *it);
}
Xapian::Query::op tdop;
if (m_topdirweight == 1.0) {
tdop = m_topdirexcl ?
Xapian::Query::OP_AND_NOT : Xapian::Query::OP_FILTER;
} else {
tdop = m_topdirexcl ?
Xapian::Query::OP_AND_NOT : Xapian::Query::OP_AND_MAYBE;
}
Xapian::Query tdq = Xapian::Query(Xapian::Query::OP_PHRASE,
pvpath.begin(), pvpath.end());
if (m_topdirweight != 1.0)
tdq = Xapian::Query(Xapian::Query::OP_SCALE_WEIGHT,
tdq, m_topdirweight);
xq = Xapian::Query(tdop, xq, tdq);
}
*((Xapian::Query *)d) = xq;
return true;
}
bool SearchData::maybeAddAutoPhrase()
{
LOGDEB0(("SearchData::maybeAddAutoPhrase()\n"));
if (!m_query.size()) {
LOGDEB2(("SearchData::maybeAddAutoPhrase: empty query\n"));
return false;
}
string field;
string words;
// Walk the clause list. If we find any non simple clause or different
// field names, bail out.
for (qlist_it_t it = m_query.begin(); it != m_query.end(); it++) {
SClType tp = (*it)->m_tp;
if (tp != SCLT_AND && tp != SCLT_OR) {
LOGDEB2(("SearchData::maybeAddAutoPhrase: complex clause\n"));
return false;
}
SearchDataClauseSimple *clp =
dynamic_cast<SearchDataClauseSimple*>(*it);
if (clp == 0) {
LOGDEB2(("SearchData::maybeAddAutoPhrase: dyncast failed\n"));
return false;
}
if (it == m_query.begin()) {
field = clp->getfield();
} else {
if (clp->getfield().compare(field)) {
LOGDEB2(("SearchData::maybeAddAutoPhrase: different fields\n"));
return false;
}
}
if (!words.empty())
words += " ";
words += clp->gettext();
}
if (words.find_first_of("\"*[]?") != string::npos && // has wildcards
TextSplit::countWords(words) <= 1) { // Just one word.
LOGDEB2(("SearchData::maybeAddAutoPhrase: wildcards or single word\n"));
return false;
}
SearchDataClauseDist *nclp =
new SearchDataClauseDist(SCLT_PHRASE, words, 0, field);
if (m_tp == SCLT_OR) {
addClause(nclp);
} else {
// My type is AND. Change it to OR and insert two queries, one
// being the original query as a subquery, the other the
// phrase.
SearchData *sd = new SearchData(m_tp);
sd->m_query = m_query;
m_tp = SCLT_OR;
m_query.clear();
SearchDataClauseSub *oq =
new SearchDataClauseSub(SCLT_OR, RefCntr<SearchData>(sd));
addClause(oq);
addClause(nclp);
}
return true;
}
// Add clause to current list. OR lists cant have EXCL clauses.
bool SearchData::addClause(SearchDataClause* cl)
{
if (m_tp == SCLT_OR && (cl->m_tp == SCLT_EXCL)) {
LOGERR(("SearchData::addClause: cant add EXCL to OR list\n"));
m_reason = "No Negative (AND_NOT) clauses allowed in OR queries";
return false;
}
cl->setParent(this);
m_haveWildCards = m_haveWildCards || cl->m_haveWildCards;
m_query.push_back(cl);
return true;
}
// Make me all new
void SearchData::erase() {
LOGDEB0(("SearchData::erase\n"));
m_tp = SCLT_AND;
for (qlist_it_t it = m_query.begin(); it != m_query.end(); it++)
delete *it;
m_query.clear();
m_filetypes.clear();
m_topdir.erase();
m_topdirexcl = false;
m_description.erase();
m_reason.erase();
m_haveDates = false;
}
// Am I a file name only search ? This is to turn off term highlighting
bool SearchData::fileNameOnly()
{
for (qlist_it_t it = m_query.begin(); it != m_query.end(); it++)
if (!(*it)->isFileName())
return false;
return true;
}
// Extract all terms and term groups
bool SearchData::getTerms(vector<string>& terms,
vector<vector<string> >& groups,
vector<int>& gslks) const
{
for (qlist_cit_t it = m_query.begin(); it != m_query.end(); it++)
(*it)->getTerms(terms, groups, gslks);
return true;
}
// Extract user terms
void SearchData::getUTerms(vector<string>& terms) const
{
for (qlist_cit_t it = m_query.begin(); it != m_query.end(); it++)
(*it)->getUTerms(terms);
sort(terms.begin(), terms.end());
vector<string>::iterator it = unique(terms.begin(), terms.end());
terms.erase(it, terms.end());
}
// Splitter callback for breaking a user string into simple terms and
// phrases. This is for parts of the user entry which would appear as
// a single word because there is no white space inside, but are
// actually multiple terms to rcldb (ie term1,term2)
class TextSplitQ : public TextSplit {
public:
TextSplitQ(Flags flags, const StopList &_stops)
: TextSplit(flags), stops(_stops), alltermcount(0), lastpos(0)
{}
bool takeword(const std::string &interm, int pos, int, int) {
alltermcount++;
lastpos = pos
LOGDEB1(("TextSplitQ::takeword: %s\n", interm.c_str()));
// Check if the first letter is a majuscule in which
// case we do not want to do stem expansion.
bool nostemexp = unaciscapital(interm);
string noaclowterm;
if (!unacmaybefold(interm, noaclowterm, "UTF-8", true)) {
LOGINFO(("SearchData::splitter::takeword: unac failed for [%s]\n",
interm.c_str()));
return true;
}
if (stops.hasStops() && stops.isStop(noaclowterm)) {
LOGDEB1(("TextSplitQ::takeword [%s] in stop list\n",
noaclowterm.c_str()));
return true;
}
terms.push_back(noaclowterm);
nostemexps.push_back(nostemexp);
return true;
}
vector<string> terms;
vector<bool> nostemexps;
const StopList &stops;
// Count of terms including stopwords: this is for adjusting
// phrase/near slack
int alltermcount;
int lastpos;
};
// A class used to translate a user compound string (*not* a query
// language string) as may be entered in any_terms/all_terms search
// entry fields, ex: [term1 "a phrase" term3] into a xapian query
// tree.
// The object keeps track of the query terms and term groups while
// translating.
class StringToXapianQ {
public:
StringToXapianQ(Db& db, const string& field,
const string &stmlng, bool boostUser)
: m_db(db), m_field(field), m_stemlang(stmlng),
m_doBoostUserTerms(boostUser)
{ }
bool processUserString(const string &iq,
string &ermsg,
list<Xapian::Query> &pqueries,
const StopList &stops,
int slack = 0, bool useNear = false);
// After processing the string: return search terms and term
// groups (ie: for highlighting)
bool getTerms(vector<string>& terms, vector<vector<string> >& groups)
{
terms.insert(terms.end(), m_terms.begin(), m_terms.end());
groups.insert(groups.end(), m_groups.begin(), m_groups.end());
return true;
}
bool getUTerms(vector<string>& terms)
{
terms.insert(terms.end(), m_uterms.begin(), m_uterms.end());
return true;
}
private:
void expandTerm(bool dont, const string& term, list<string>& exp,
string& sterm, string *prefix = 0);
// After splitting entry on whitespace: process non-phrase element
void processSimpleSpan(const string& span, bool nostemexp, list<Xapian::Query> &pqueries);
// Process phrase/near element
void processPhraseOrNear(TextSplitQ *splitData,
list<Xapian::Query> &pqueries,
bool useNear, int slack);
Db& m_db;
const string& m_field;
const string& m_stemlang;
bool m_doBoostUserTerms;
// Single terms and phrases resulting from breaking up text;
vector<string> m_uterms;
vector<string> m_terms;
vector<vector<string> > m_groups;
};
#if 0
static void listVector(const string& what, const vector<string>&l)
{
string a;
for (vector<string>::const_iterator it = l.begin(); it != l.end(); it++) {
a = a + *it + " ";
}
LOGDEB(("%s: %s\n", what.c_str(), a.c_str()));
}
#endif
/** Expand stem and wildcards
*
* @param nostemexp don't perform stem expansion. This is mainly used to
* prevent stem expansion inside phrases (because the user probably
* does not expect it). This does NOT prevent wild card expansion.
* Other factors than nostemexp can prevent stem expansion:
* a null stemlang, resulting from a global user preference, a
* capitalized term, or wildcard(s)
* @param term input single word
* @param exp output expansion list
* @param sterm output original input term if there were no wildcards
*/
void StringToXapianQ::expandTerm(bool nostemexp,
const string& term,
list<string>& exp,
string &sterm, string *prefix)
{
LOGDEB2(("expandTerm: field [%s] term [%s] stemlang [%s] nostemexp %d\n",
m_field.c_str(), term.c_str(), m_stemlang.c_str(), nostemexp));
sterm.erase();
exp.clear();
if (term.empty()) {
return;
}
bool haswild = term.find_first_of("*?[") != string::npos;
// No stemming if there are wildcards or prevented globally.
if (haswild || m_stemlang.empty())
nostemexp = true;
if (nostemexp && !haswild) {
// Neither stemming nor wildcard expansion: just the word
string pfx;
if (!m_field.empty())
m_db.fieldToPrefix(m_field, pfx);
sterm = term;
m_uterms.push_back(sterm);
exp.push_front(pfx+term);
exp.resize(1);
if (prefix)
*prefix = pfx;
} else {
TermMatchResult res;
if (haswild) {
m_db.termMatch(Rcl::Db::ET_WILD, m_stemlang, term, res, -1,
m_field, prefix);
} else {
sterm = term;
m_uterms.push_back(sterm);
m_db.termMatch(Rcl::Db::ET_STEM, m_stemlang, term, res, -1, m_field,
prefix);
}
for (list<TermMatchEntry>::const_iterator it = res.entries.begin();
it != res.entries.end(); it++) {
exp.push_back(it->term);
}
}
//listVector("ExpandTerm:uterms now: ", m_uterms);
}
// Do distribution of string vectors: a,b c,d -> a,c a,d b,c b,d
void multiply_groups(vector<vector<string> >::const_iterator vvit,
vector<vector<string> >::const_iterator vvend,
vector<string>& comb,
vector<vector<string> >&allcombs)
{
// Remember my string vector and compute next, for recursive calls.
vector<vector<string> >::const_iterator myvit = vvit++;
// Walk the string vector I'm called upon and, for each string,
// add it to current result, an call myself recursively on the
// next string vector. The last call (last element of the vector of
// vectors), adds the elementary result to the output
// Walk my string vector
for (vector<string>::const_iterator strit = (*myvit).begin();
strit != (*myvit).end(); strit++) {
// Add my current value to the string vector we're building
comb.push_back(*strit);
if (vvit == vvend) {
// Last call: store current result
allcombs.push_back(comb);
} else {
// Call recursively on next string vector
multiply_groups(vvit, vvend, comb, allcombs);
}
// Pop the value I just added (make room for the next element in my
// vector)
comb.pop_back();
}
}
void StringToXapianQ::processSimpleSpan(const string& span, bool nostemexp,
list<Xapian::Query> &pqueries)
{
list<string> exp;
string sterm; // dumb version of user term
string prefix;
expandTerm(nostemexp, span, exp, sterm, &prefix);
// m_terms is used for highlighting, we don't want prefixes in there.
for (list<string>::const_iterator it = exp.begin();
it != exp.end(); it++) {
m_terms.push_back(it->substr(prefix.size()));
}
// Push either term or OR of stem-expanded set
Xapian::Query xq(Xapian::Query::OP_OR, exp.begin(), exp.end());
// If sterm (simplified original user term) is not null, give it a
// relevance boost. We do this even if no expansion occurred (else
// the non-expanded terms in a term list would end-up with even
// less wqf). This does not happen if there are wildcards anywhere
// in the search.
if (m_doBoostUserTerms && !sterm.empty()) {
xq = Xapian::Query(Xapian::Query::OP_OR,
xq,
Xapian::Query(prefix+sterm,
original_term_wqf_booster));
}
pqueries.push_back(xq);
}
// User entry element had several terms: transform into a PHRASE or
// NEAR xapian query, the elements of which can themselves be OR
// queries if the terms get expanded by stemming or wildcards (we
// don't do stemming for PHRASE though)
void StringToXapianQ::processPhraseOrNear(TextSplitQ *splitData,
list<Xapian::Query> &pqueries,
bool useNear, int slack)
{
Xapian::Query::op op = useNear ? Xapian::Query::OP_NEAR :
Xapian::Query::OP_PHRASE;
list<Xapian::Query> orqueries;
bool hadmultiple = false;
vector<vector<string> >groups;
// Go through the list and perform stem/wildcard expansion for each element
vector<bool>::iterator nxit = splitData->nostemexps.begin();
for (vector<string>::iterator it = splitData->terms.begin();
it != splitData->terms.end(); it++, nxit++) {
// Adjust when we do stem expansion. Not inside phrases, and
// some versions of xapian will accept only one OR clause
// inside NEAR, all others must be leafs.
bool nostemexp = *nxit || (op == Xapian::Query::OP_PHRASE) || hadmultiple;
string sterm;
list<string>exp;
string prefix;
expandTerm(nostemexp, *it, exp, sterm, &prefix);
// groups is used for highlighting, we don't want prefixes in there.
vector<string> noprefs;
for (list<string>::const_iterator it = exp.begin();
it != exp.end(); it++) {
noprefs.push_back(it->substr(prefix.size()));
}
groups.push_back(noprefs);
orqueries.push_back(Xapian::Query(Xapian::Query::OP_OR,
exp.begin(), exp.end()));
#ifdef XAPIAN_NEAR_EXPAND_SINGLE_BUF
if (exp.size() > 1)
hadmultiple = true;
#endif
}
// Generate an appropriate PHRASE/NEAR query with adjusted slack
// For phrases, give a relevance boost like we do for original terms
LOGDEB2(("PHRASE/NEAR: alltermcount %d lastpos %d\n",
splitData->alltermcount, splitData->lastpos));
Xapian::Query xq(op, orqueries.begin(), orqueries.end(),
splitData->lastpos + 1 + slack);
if (op == Xapian::Query::OP_PHRASE)
xq = Xapian::Query(Xapian::Query::OP_SCALE_WEIGHT, xq,
original_term_wqf_booster);
pqueries.push_back(xq);
// Add all combinations of NEAR/PHRASE groups to the highlighting data.
vector<vector<string> > allcombs;
vector<string> comb;
multiply_groups(groups.begin(), groups.end(), comb, allcombs);
m_groups.insert(m_groups.end(), allcombs.begin(), allcombs.end());
}
/**
* Turn user entry string (NOT query language) into a list of xapian queries.
* We just separate words and phrases, and do wildcard and stem expansion,
*
* This is used to process data entered into an OR/AND/NEAR/PHRASE field of
* the GUI.
*
* The final list contains one query for each term or phrase
* - Elements corresponding to a stem-expanded part are an OP_OR
* composition of the stem-expanded terms (or a single term query).
* - Elements corresponding to phrase/near are an OP_PHRASE/NEAR
* composition of the phrase terms (no stem expansion in this case)
* @return the subquery count (either or'd stem-expanded terms or phrase word
* count)
*/
bool StringToXapianQ::processUserString(const string &iq,
string &ermsg,
list<Xapian::Query> &pqueries,
const StopList& stops,
int slack,
bool useNear
)
{
LOGDEB(("StringToXapianQ:: query string: [%s]\n", iq.c_str()));
ermsg.erase();
m_uterms.clear();
m_terms.clear();
m_groups.clear();
// Simple whitespace-split input into user-level words and
// double-quoted phrases: word1 word2 "this is a phrase".
//
// The text splitter may further still decide that the resulting
// "words" are really phrases, this depends on separators:
// [paul@dom.net] would still be a word (span), but [about:me]
// will probably be handled as a phrase.
list<string> phrases;
TextSplit::stringToStrings(iq, phrases);
// Process each element: textsplit into terms, handle stem/wildcard
// expansion and transform into an appropriate Xapian::Query
try {
for (list<string>::iterator it = phrases.begin();
it != phrases.end(); it++) {
LOGDEB0(("strToXapianQ: phrase/word: [%s]\n", it->c_str()));
// If there are multiple spans in this element, including
// at least one composite, we have to increase the slack
// else a phrase query including a span would fail.
// Ex: "term0@term1 term2" is onlyspans-split as:
// 0 term0@term1 0 12
// 2 term2 13 18
// The position of term2 is 2, not 1, so a phrase search
// would fail.
// We used to do word split, searching for
// "term0 term1 term2" instead, which may have worse
// performance, but will succeed.
// We now adjust the phrase/near slack by the term count
// difference (this is mainly better for cjk where this is a very
// common occurrence because of the ngrams thing.
TextSplitQ splitterS(TextSplit::Flags(TextSplit::TXTS_ONLYSPANS |
TextSplit::TXTS_KEEPWILD),
stops);
splitterS.text_to_words(*it);
TextSplitQ splitterW(TextSplit::Flags(TextSplit::TXTS_NOSPANS |
TextSplit::TXTS_KEEPWILD),
stops);
splitterW.text_to_words(*it);
TextSplitQ *splitter = &splitterS;
if (splitterS.terms.size() > 1 &&
splitterS.terms.size() != splitterW.terms.size()) {
slack += splitterW.terms.size() - splitterS.terms.size();
// used to: splitData = &splitDataW;
}
LOGDEB0(("strToXapianQ: termcount: %d\n", splitter->terms.size()));
switch (splitter->terms.size()) {
case 0:
continue;// ??
case 1:
processSimpleSpan(splitter->terms.front(),
splitter->nostemexps.front(), pqueries);
break;
default:
processPhraseOrNear(splitter, pqueries, useNear, slack);
}
}
} catch (const Xapian::Error &e) {
ermsg = e.get_msg();
} catch (const string &s) {
ermsg = s;
} catch (const char *s) {
ermsg = s;
} catch (...) {
ermsg = "Caught unknown exception";
}
if (!ermsg.empty()) {
LOGERR(("stringToXapianQueries: %s\n", ermsg.c_str()));
return false;
}
return true;
}
static const string nullstemlang;
// Translate a simple OR, AND, or EXCL search clause.
bool SearchDataClauseSimple::toNativeQuery(Rcl::Db &db, void *p,
const string& stemlang)
{
const string& l_stemlang = (m_modifiers&SDCM_NOSTEMMING)? nullstemlang:
stemlang;
m_terms.clear();
m_groups.clear();
Xapian::Query *qp = (Xapian::Query *)p;
*qp = Xapian::Query();
Xapian::Query::op op;
switch (m_tp) {
case SCLT_AND: op = Xapian::Query::OP_AND; break;
// EXCL will be set with AND_NOT in the list. So it's an OR list here
case SCLT_OR:
case SCLT_EXCL: op = Xapian::Query::OP_OR; break;
default:
LOGERR(("SearchDataClauseSimple: bad m_tp %d\n", m_tp));
return false;
}
list<Xapian::Query> pqueries;
// We normally boost the original term in the stem expansion list. Don't
// do it if there are wildcards anywhere, this would skew the results.
bool doBoostUserTerm =
(m_parentSearch && !m_parentSearch->haveWildCards()) ||
(m_parentSearch == 0 && !m_haveWildCards);
StringToXapianQ tr(db, m_field, l_stemlang, doBoostUserTerm);
if (!tr.processUserString(m_text, m_reason, pqueries, db.getStopList()))
return false;
if (pqueries.empty()) {
LOGERR(("SearchDataClauseSimple: resolved to null query\n"));
return true;
}
tr.getTerms(m_terms, m_groups);
tr.getUTerms(m_uterms);
//listVector("SearchDataClauseSimple: Uterms: ", m_uterms);
*qp = Xapian::Query(op, pqueries.begin(), pqueries.end());
if (m_weight != 1.0) {
*qp = Xapian::Query(Xapian::Query::OP_SCALE_WEIGHT, *qp, m_weight);
}
return true;
}
// Translate a FILENAME search clause. This mostly (or always) comes
// from a "filename" search from the gui or recollq. A query language
// "filename:"-prefixed field will not go through here, but through
// the generic field-processing code.
//
// In the case of multiple space-separated fragments, we generate an
// AND of OR queries. Each OR query comes from the expansion of a
// fragment. We used to generate a single OR with all expanded terms,
// which did not make much sense.
bool SearchDataClauseFilename::toNativeQuery(Rcl::Db &db, void *p,
const string&)
{
Xapian::Query *qp = (Xapian::Query *)p;
*qp = Xapian::Query();
list<string> patterns;
TextSplit::stringToStrings(m_text, patterns);
list<string> names;
for (list<string>::iterator it = patterns.begin();
it != patterns.end(); it++) {
list<string> more;
db.filenameWildExp(*it, more);
Xapian::Query tq = Xapian::Query(Xapian::Query::OP_OR, more.begin(),
more.end());
*qp = qp->empty() ? tq : Xapian::Query(Xapian::Query::OP_AND, *qp, tq);
}
if (m_weight != 1.0) {
*qp = Xapian::Query(Xapian::Query::OP_SCALE_WEIGHT, *qp, m_weight);
}
return true;
}
// Translate NEAR or PHRASE clause.
bool SearchDataClauseDist::toNativeQuery(Rcl::Db &db, void *p,
const string& stemlang)
{
const string& l_stemlang = (m_modifiers&SDCM_NOSTEMMING)? nullstemlang:
stemlang;
LOGDEB(("SearchDataClauseDist::toNativeQuery\n"));
m_terms.clear();
m_groups.clear();
Xapian::Query *qp = (Xapian::Query *)p;
*qp = Xapian::Query();
list<Xapian::Query> pqueries;
Xapian::Query nq;
// We normally boost the original term in the stem expansion list. Don't
// do it if there are wildcards anywhere, this would skew the results.
bool doBoostUserTerm =
(m_parentSearch && !m_parentSearch->haveWildCards()) ||
(m_parentSearch == 0 && !m_haveWildCards);
// We produce a single phrase out of the user entry then use
// stringToXapianQueries() to lowercase and simplify the phrase
// terms etc. This will result into a single (complex)
// Xapian::Query.
if (m_text.find_first_of("\"") != string::npos) {
m_text = neutchars(m_text, "\"");
}
string s = string("\"") + m_text + string("\"");
bool useNear = (m_tp == SCLT_NEAR);
StringToXapianQ tr(db, m_field, l_stemlang, doBoostUserTerm);
if (!tr.processUserString(s, m_reason, pqueries, db.getStopList(),
m_slack, useNear))
return false;
if (pqueries.empty()) {
LOGERR(("SearchDataClauseDist: resolved to null query\n"));
return true;
}
tr.getTerms(m_terms, m_groups);
tr.getUTerms(m_uterms);
*qp = *pqueries.begin();
if (m_weight != 1.0) {
*qp = Xapian::Query(Xapian::Query::OP_SCALE_WEIGHT, *qp, m_weight);
}
return true;
}
// Translate subquery
bool SearchDataClauseSub::toNativeQuery(Rcl::Db &db, void *p, const string&)
{
return m_sub->toNativeQuery(db, p);
}
bool SearchDataClauseSub::getTerms(vector<string>& terms,
vector<vector<string> >& groups,
vector<int>& gslks) const
{
return m_sub.getconstptr()->getTerms(terms, groups, gslks);
}
void SearchDataClauseSub::getUTerms(vector<string>& terms) const
{
m_sub.getconstptr()->getUTerms(terms);
}
} // Namespace Rcl