Repository navigation
Expand file tree
/
Copy pathindexer.php
More file actions
1544 lines (1421 loc) · 52.8 KB
/
Copy pathindexer.php
File metadata and controls
1544 lines (1421 loc) · 52.8 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
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
<?php
/**
* Functions to create the fulltext search index
*
* @license GPL 2 (http://www.gnu.org/licenses/gpl.html)
* @author Andreas Gohr <andi@splitbrain.org>
* @author Tom N Harris <tnharris@whoopdedo.org>
*/
if(!defined('DOKU_INC')) die('meh.');
// Version tag used to force rebuild on upgrade
define('INDEXER_VERSION', 8);
// set the minimum token length to use in the index (note, this doesn't apply to numeric tokens)
if (!defined('IDX_MINWORDLENGTH')) define('IDX_MINWORDLENGTH',2);
// Asian characters are handled as words. The following regexp defines the
// Unicode-Ranges for Asian characters
// Ranges taken from http://en.wikipedia.org/wiki/Unicode_block
// I'm no language expert. If you think some ranges are wrongly chosen or
// a range is missing, please contact me
define('IDX_ASIAN1','[\x{0E00}-\x{0E7F}]'); // Thai
define('IDX_ASIAN2','['.
'\x{2E80}-\x{3040}'. // CJK -> Hangul
'\x{309D}-\x{30A0}'.
'\x{30FD}-\x{31EF}\x{3200}-\x{D7AF}'.
'\x{F900}-\x{FAFF}'. // CJK Compatibility Ideographs
'\x{FE30}-\x{FE4F}'. // CJK Compatibility Forms
"\xF0\xA0\x80\x80-\xF0\xAA\x9B\x9F". // CJK Extension B
"\xF0\xAA\x9C\x80-\xF0\xAB\x9C\xBF". // CJK Extension C
"\xF0\xAB\x9D\x80-\xF0\xAB\xA0\x9F". // CJK Extension D
"\xF0\xAF\xA0\x80-\xF0\xAF\xAB\xBF". // CJK Compatibility Supplement
']');
define('IDX_ASIAN3','['. // Hiragana/Katakana (can be two characters)
'\x{3042}\x{3044}\x{3046}\x{3048}'.
'\x{304A}-\x{3062}\x{3064}-\x{3082}'.
'\x{3084}\x{3086}\x{3088}-\x{308D}'.
'\x{308F}-\x{3094}'.
'\x{30A2}\x{30A4}\x{30A6}\x{30A8}'.
'\x{30AA}-\x{30C2}\x{30C4}-\x{30E2}'.
'\x{30E4}\x{30E6}\x{30E8}-\x{30ED}'.
'\x{30EF}-\x{30F4}\x{30F7}-\x{30FA}'.
']['.
'\x{3041}\x{3043}\x{3045}\x{3047}\x{3049}'.
'\x{3063}\x{3083}\x{3085}\x{3087}\x{308E}\x{3095}-\x{309C}'.
'\x{30A1}\x{30A3}\x{30A5}\x{30A7}\x{30A9}'.
'\x{30C3}\x{30E3}\x{30E5}\x{30E7}\x{30EE}\x{30F5}\x{30F6}\x{30FB}\x{30FC}'.
'\x{31F0}-\x{31FF}'.
']?');
define('IDX_ASIAN', '(?:'.IDX_ASIAN1.'|'.IDX_ASIAN2.'|'.IDX_ASIAN3.')');
/**
* Version of the indexer taking into consideration the external tokenizer.
* The indexer is only compatible with data written by the same version.
*
* @triggers INDEXER_VERSION_GET
* Plugins that modify what gets indexed should hook this event and
* add their version info to the event data like so:
* $data[$plugin_name] = $plugin_version;
*
* @author Tom N Harris <tnharris@whoopdedo.org>
* @author Michael Hamann <michael@content-space.de>
*/
function idx_get_version(){
static $indexer_version = null;
if ($indexer_version == null) {
$version = INDEXER_VERSION;
// DokuWiki version is included for the convenience of plugins
$data = array('dokuwiki'=>$version);
trigger_event('INDEXER_VERSION_GET', $data, null, false);
unset($data['dokuwiki']); // this needs to be first
ksort($data);
foreach ($data as $plugin=>$vers)
$version .= '+'.$plugin.'='.$vers;
$indexer_version = $version;
}
return $indexer_version;
}
/**
* Measure the length of a string.
* Differs from strlen in handling of asian characters.
*
* @author Tom N Harris <tnharris@whoopdedo.org>
*/
function wordlen($w){
$l = strlen($w);
// If left alone, all chinese "words" will get put into w3.idx
// So the "length" of a "word" is faked
if(preg_match_all('/[\xE2-\xEF]/',$w,$leadbytes)) {
foreach($leadbytes[0] as $b)
$l += ord($b) - 0xE1;
}
return $l;
}
/**
* Class that encapsulates operations on the indexer database.
*
* @author Tom N Harris <tnharris@whoopdedo.org>
*/
class Doku_Indexer {
/**
* @var array $pidCache Cache for getPID()
*/
protected $pidCache = array();
/**
* Adds the contents of a page to the fulltext index
*
* The added text replaces previous words for the same page.
* An empty value erases the page.
*
* @param string $page a page name
* @param string $text the body of the page
* @return boolean the function completed successfully
* @author Tom N Harris <tnharris@whoopdedo.org>
* @author Andreas Gohr <andi@splitbrain.org>
*/
public function addPageWords($page, $text) {
if (!$this->lock())
return "locked";
// load known documents
$pid = $this->getPIDNoLock($page);
if ($pid === false) {
$this->unlock();
return false;
}
$pagewords = array();
// get word usage in page
$words = $this->getPageWords($text);
if ($words === false) {
$this->unlock();
return false;
}
if (!empty($words)) {
foreach (array_keys($words) as $wlen) {
$index = $this->getIndex('i', $wlen);
foreach ($words[$wlen] as $wid => $freq) {
$idx = ($wid<count($index)) ? $index[$wid] : '';
$index[$wid] = $this->updateTuple($idx, $pid, $freq);
$pagewords[] = "$wlen*$wid";
}
if (!$this->saveIndex('i', $wlen, $index)) {
$this->unlock();
return false;
}
}
}
// Remove obsolete index entries
$pageword_idx = $this->getIndexKey('pageword', '', $pid);
if ($pageword_idx !== '') {
$oldwords = explode(':',$pageword_idx);
$delwords = array_diff($oldwords, $pagewords);
$upwords = array();
foreach ($delwords as $word) {
if ($word != '') {
list($wlen,$wid) = explode('*', $word);
$wid = (int)$wid;
$upwords[$wlen][] = $wid;
}
}
foreach ($upwords as $wlen => $widx) {
$index = $this->getIndex('i', $wlen);
foreach ($widx as $wid) {
$index[$wid] = $this->updateTuple($index[$wid], $pid, 0);
}
$this->saveIndex('i', $wlen, $index);
}
}
// Save the reverse index
$pageword_idx = join(':', $pagewords);
if (!$this->saveIndexKey('pageword', '', $pid, $pageword_idx)) {
$this->unlock();
return false;
}
$this->unlock();
return true;
}
/**
* Split the words in a page and add them to the index.
*
* @param string $text content of the page
* @return array list of word IDs and number of times used
* @author Andreas Gohr <andi@splitbrain.org>
* @author Christopher Smith <chris@jalakai.co.uk>
* @author Tom N Harris <tnharris@whoopdedo.org>
*/
protected function getPageWords($text) {
$tokens = $this->tokenizer($text);
$tokens = array_count_values($tokens); // count the frequency of each token
$words = array();
foreach ($tokens as $w=>$c) {
$l = wordlen($w);
if (isset($words[$l])){
$words[$l][$w] = $c + (isset($words[$l][$w]) ? $words[$l][$w] : 0);
}else{
$words[$l] = array($w => $c);
}
}
// arrive here with $words = array(wordlen => array(word => frequency))
$word_idx_modified = false;
$index = array(); //resulting index
foreach (array_keys($words) as $wlen) {
$word_idx = $this->getIndex('w', $wlen);
foreach ($words[$wlen] as $word => $freq) {
$word = (string)$word;
$wid = array_search($word, $word_idx, true);
if ($wid === false) {
$wid = count($word_idx);
$word_idx[] = $word;
$word_idx_modified = true;
}
if (!isset($index[$wlen]))
$index[$wlen] = array();
$index[$wlen][$wid] = $freq;
}
// save back the word index
if ($word_idx_modified && !$this->saveIndex('w', $wlen, $word_idx))
return false;
}
return $index;
}
/**
* Add/update keys to/of the metadata index.
*
* Adding new keys does not remove other keys for the page.
* An empty value will erase the key.
* The $key parameter can be an array to add multiple keys. $value will
* not be used if $key is an array.
*
* @param string $page a page name
* @param mixed $key a key string or array of key=>value pairs
* @param mixed $value the value or list of values
* @return boolean the function completed successfully
* @author Tom N Harris <tnharris@whoopdedo.org>
* @author Michael Hamann <michael@content-space.de>
*/
public function addMetaKeys($page, $key, $value=null) {
if (!is_array($key)) {
$key = array($key => $value);
} elseif (!is_null($value)) {
// $key is array, but $value is not null
trigger_error("array passed to addMetaKeys but value is not null", E_USER_WARNING);
}
if (!$this->lock())
return "locked";
// load known documents
$pid = $this->getPIDNoLock($page);
if ($pid === false) {
$this->unlock();
return false;
}
// Special handling for titles so the index file is simpler
if (array_key_exists('title', $key)) {
$value = $key['title'];
if (is_array($value))
$value = $value[0];
$this->saveIndexKey('title', '', $pid, $value);
unset($key['title']);
}
foreach ($key as $name => $values) {
$metaname = idx_cleanName($name);
$this->addIndexKey('metadata', '', $metaname);
$metaidx = $this->getIndex($metaname.'_i', '');
$metawords = $this->getIndex($metaname.'_w', '');
$addwords = false;
if (!is_array($values)) $values = array($values);
$val_idx = $this->getIndexKey($metaname.'_p', '', $pid);
if ($val_idx != '') {
$val_idx = explode(':', $val_idx);
// -1 means remove, 0 keep, 1 add
$val_idx = array_combine($val_idx, array_fill(0, count($val_idx), -1));
} else {
$val_idx = array();
}
foreach ($values as $val) {
$val = (string)$val;
if ($val !== "") {
$id = array_search($val, $metawords, true);
if ($id === false) {
$id = count($metawords);
$metawords[$id] = $val;
$addwords = true;
}
// test if value is already in the index
if (isset($val_idx[$id]) && $val_idx[$id] <= 0)
$val_idx[$id] = 0;
else // else add it
$val_idx[$id] = 1;
}
}
if ($addwords)
$this->saveIndex($metaname.'_w', '', $metawords);
$vals_changed = false;
foreach ($val_idx as $id => $action) {
if ($action == -1) {
$metaidx[$id] = $this->updateTuple($metaidx[$id], $pid, 0);
$vals_changed = true;
unset($val_idx[$id]);
} elseif ($action == 1) {
$metaidx[$id] = $this->updateTuple($metaidx[$id], $pid, 1);
$vals_changed = true;
}
}
if ($vals_changed) {
$this->saveIndex($metaname.'_i', '', $metaidx);
$val_idx = implode(':', array_keys($val_idx));
$this->saveIndexKey($metaname.'_p', '', $pid, $val_idx);
}
unset($metaidx);
unset($metawords);
}
$this->unlock();
return true;
}
/**
* Rename a page in the search index without changing the indexed content. This function doesn't check if the
* old or new name exists in the filesystem. It returns an error if the old page isn't in the page list of the
* indexer and it deletes all previously indexed content of the new page.
*
* @param string $oldpage The old page name
* @param string $newpage The new page name
* @return string|bool If the page was successfully renamed, can be a message in the case of an error
*/
public function renamePage($oldpage, $newpage) {
if (!$this->lock()) return 'locked';
$pages = $this->getPages();
$id = array_search($oldpage, $pages, true);
if ($id === false) {
$this->unlock();
return 'page is not in index';
}
$new_id = array_search($newpage, $pages, true);
if ($new_id !== false) {
// make sure the page is not in the index anymore
if ($this->deletePageNoLock($newpage) !== true) {
return false;
}
$pages[$new_id] = 'deleted:'.time().rand(0, 9999);
}
$pages[$id] = $newpage;
// update index
if (!$this->saveIndex('page', '', $pages)) {
$this->unlock();
return false;
}
// reset the pid cache
$this->pidCache = array();
$this->unlock();
return true;
}
/**
* Renames a meta value in the index. This doesn't change the meta value in the pages, it assumes that all pages
* will be updated.
*
* @param string $key The metadata key of which a value shall be changed
* @param string $oldvalue The old value that shall be renamed
* @param string $newvalue The new value to which the old value shall be renamed, can exist (then values will be merged)
* @return bool|string If renaming the value has been successful, false or error message on error.
*/
public function renameMetaValue($key, $oldvalue, $newvalue) {
if (!$this->lock()) return 'locked';
// change the relation references index
$metavalues = $this->getIndex($key, '_w');
$oldid = array_search($oldvalue, $metavalues, true);
if ($oldid !== false) {
$newid = array_search($newvalue, $metavalues, true);
if ($newid !== false) {
// free memory
unset ($metavalues);
// okay, now we have two entries for the same value. we need to merge them.
$indexline = $this->getIndexKey($key.'_i', '', $oldid);
if ($indexline != '') {
$newindexline = $this->getIndexKey($key.'_i', '', $newid);
$pagekeys = $this->getIndex($key.'_p', '');
$parts = explode(':', $indexline);
foreach ($parts as $part) {
list($id, $count) = explode('*', $part);
$newindexline = $this->updateTuple($newindexline, $id, $count);
$keyline = explode(':', $pagekeys[$id]);
// remove old meta value
$keyline = array_diff($keyline, array($oldid));
// add new meta value when not already present
if (!in_array($newid, $keyline)) {
array_push($keyline, $newid);
}
$pagekeys[$id] = implode(':', $keyline);
}
$this->saveIndex($key.'_p', '', $pagekeys);
unset($pagekeys);
$this->saveIndexKey($key.'_i', '', $oldid, '');
$this->saveIndexKey($key.'_i', '', $newid, $newindexline);
}
} else {
$metavalues[$oldid] = $newvalue;
if (!$this->saveIndex($key.'_w', '', $metavalues)) {
$this->unlock();
return false;
}
}
}
$this->unlock();
return true;
}
/**
* Remove a page from the index
*
* Erases entries in all known indexes.
*
* @param string $page a page name
* @return boolean the function completed successfully
* @author Tom N Harris <tnharris@whoopdedo.org>
*/
public function deletePage($page) {
if (!$this->lock())
return "locked";
$result = $this->deletePageNoLock($page);
$this->unlock();
return $result;
}
/**
* Remove a page from the index without locking the index, only use this function if the index is already locked
*
* Erases entries in all known indexes.
*
* @param string $page a page name
* @return boolean the function completed successfully
* @author Tom N Harris <tnharris@whoopdedo.org>
*/
protected function deletePageNoLock($page) {
// load known documents
$pid = $this->getPIDNoLock($page);
if ($pid === false) {
return false;
}
// Remove obsolete index entries
$pageword_idx = $this->getIndexKey('pageword', '', $pid);
if ($pageword_idx !== '') {
$delwords = explode(':',$pageword_idx);
$upwords = array();
foreach ($delwords as $word) {
if ($word != '') {
list($wlen,$wid) = explode('*', $word);
$wid = (int)$wid;
$upwords[$wlen][] = $wid;
}
}
foreach ($upwords as $wlen => $widx) {
$index = $this->getIndex('i', $wlen);
foreach ($widx as $wid) {
$index[$wid] = $this->updateTuple($index[$wid], $pid, 0);
}
$this->saveIndex('i', $wlen, $index);
}
}
// Save the reverse index
if (!$this->saveIndexKey('pageword', '', $pid, "")) {
return false;
}
$this->saveIndexKey('title', '', $pid, "");
$keyidx = $this->getIndex('metadata', '');
foreach ($keyidx as $metaname) {
$val_idx = explode(':', $this->getIndexKey($metaname.'_p', '', $pid));
$meta_idx = $this->getIndex($metaname.'_i', '');
foreach ($val_idx as $id) {
if ($id === '') continue;
$meta_idx[$id] = $this->updateTuple($meta_idx[$id], $pid, 0);
}
$this->saveIndex($metaname.'_i', '', $meta_idx);
$this->saveIndexKey($metaname.'_p', '', $pid, '');
}
return true;
}
/**
* Clear the whole index
*
* @return bool If the index has been cleared successfully
*/
public function clear() {
global $conf;
if (!$this->lock()) return false;
@unlink($conf['indexdir'].'/page.idx');
@unlink($conf['indexdir'].'/title.idx');
@unlink($conf['indexdir'].'/pageword.idx');
@unlink($conf['indexdir'].'/metadata.idx');
$dir = @opendir($conf['indexdir']);
if($dir!==false){
while(($f = readdir($dir)) !== false){
if(substr($f,-4)=='.idx' &&
(substr($f,0,1)=='i' || substr($f,0,1)=='w'
|| substr($f,-6)=='_w.idx' || substr($f,-6)=='_i.idx' || substr($f,-6)=='_p.idx'))
@unlink($conf['indexdir']."/$f");
}
}
@unlink($conf['indexdir'].'/lengths.idx');
// clear the pid cache
$this->pidCache = array();
$this->unlock();
return true;
}
/**
* Split the text into words for fulltext search
*
* TODO: does this also need &$stopwords ?
*
* @triggers INDEXER_TEXT_PREPARE
* This event allows plugins to modify the text before it gets tokenized.
* Plugins intercepting this event should also intercept INDEX_VERSION_GET
*
* @param string $text plain text
* @param boolean $wc are wildcards allowed?
* @return array list of words in the text
* @author Tom N Harris <tnharris@whoopdedo.org>
* @author Andreas Gohr <andi@splitbrain.org>
*/
public function tokenizer($text, $wc=false) {
$wc = ($wc) ? '' : '\*';
$stopwords =& idx_get_stopwords();
// prepare the text to be tokenized
$evt = new Doku_Event('INDEXER_TEXT_PREPARE', $text);
if ($evt->advise_before(true)) {
if (preg_match('/[^0-9A-Za-z ]/u', $text)) {
// handle asian chars as single words (may fail on older PHP version)
$asia = @preg_replace('/('.IDX_ASIAN.')/u', ' \1 ', $text);
if (!is_null($asia)) $text = $asia; // recover from regexp falure
}
}
$evt->advise_after();
unset($evt);
$text = strtr($text,
array(
"\r" => ' ',
"\n" => ' ',
"\t" => ' ',
"\xC2\xAD" => '', //soft-hyphen
)
);
if (preg_match('/[^0-9A-Za-z ]/u', $text))
$text = utf8_stripspecials($text, ' ', '\._\-:'.$wc);
$wordlist = explode(' ', $text);
foreach ($wordlist as $i => $word) {
$wordlist[$i] = (preg_match('/[^0-9A-Za-z]/u', $word)) ?
utf8_strtolower($word) : strtolower($word);
}
foreach ($wordlist as $i => $word) {
if ((!is_numeric($word) && strlen($word) < IDX_MINWORDLENGTH)
|| array_search($word, $stopwords, true) !== false)
unset($wordlist[$i]);
}
return array_values($wordlist);
}
/**
* Get the numeric PID of a page
*
* @param string $page The page to get the PID for
* @return bool|int The page id on success, false on error
*/
public function getPID($page) {
// return PID without locking when it is in the cache
if (isset($this->pidCache[$page])) return $this->pidCache[$page];
if (!$this->lock())
return false;
// load known documents
$pid = $this->getPIDNoLock($page);
if ($pid === false) {
$this->unlock();
return false;
}
$this->unlock();
return $pid;
}
/**
* Get the numeric PID of a page without locking the index.
* Only use this function when the index is already locked.
*
* @param string $page The page to get the PID for
* @return bool|int The page id on success, false on error
*/
protected function getPIDNoLock($page) {
// avoid expensive addIndexKey operation for the most recently requested pages by using a cache
if (isset($this->pidCache[$page])) return $this->pidCache[$page];
$pid = $this->addIndexKey('page', '', $page);
// limit cache to 10 entries by discarding the oldest element as in DokuWiki usually only the most recently
// added item will be requested again
if (count($this->pidCache) > 10) array_shift($this->pidCache);
$this->pidCache[$page] = $pid;
return $pid;
}
/**
* Get the page id of a numeric PID
*
* @param int $pid The PID to get the page id for
* @return string The page id
*/
public function getPageFromPID($pid) {
return $this->getIndexKey('page', '', $pid);
}
/**
* Find pages in the fulltext index containing the words,
*
* The search words must be pre-tokenized, meaning only letters and
* numbers with an optional wildcard
*
* The returned array will have the original tokens as key. The values
* in the returned list is an array with the page names as keys and the
* number of times that token appears on the page as value.
*
* @param array $tokens list of words to search for
* @return array list of page names with usage counts
* @author Tom N Harris <tnharris@whoopdedo.org>
* @author Andreas Gohr <andi@splitbrain.org>
*/
public function lookup(&$tokens) {
$result = array();
$wids = $this->getIndexWords($tokens, $result);
if (empty($wids)) return array();
// load known words and documents
$page_idx = $this->getIndex('page', '');
$docs = array();
foreach (array_keys($wids) as $wlen) {
$wids[$wlen] = array_unique($wids[$wlen]);
$index = $this->getIndex('i', $wlen);
foreach($wids[$wlen] as $ixid) {
if ($ixid < count($index))
$docs["$wlen*$ixid"] = $this->parseTuples($page_idx, $index[$ixid]);
}
}
// merge found pages into final result array
$final = array();
foreach ($result as $word => $res) {
$final[$word] = array();
foreach ($res as $wid) {
// handle the case when ($ixid < count($index)) has been false
// and thus $docs[$wid] hasn't been set.
if (!isset($docs[$wid])) continue;
$hits = &$docs[$wid];
foreach ($hits as $hitkey => $hitcnt) {
// make sure the document still exists
if (!page_exists($hitkey, '', false)) continue;
if (!isset($final[$word][$hitkey]))
$final[$word][$hitkey] = $hitcnt;
else
$final[$word][$hitkey] += $hitcnt;
}
}
}
return $final;
}
/**
* Find pages containing a metadata key.
*
* The metadata values are compared as case-sensitive strings. Pass a
* callback function that returns true or false to use a different
* comparison function. The function will be called with the $value being
* searched for as the first argument, and the word in the index as the
* second argument. The function preg_match can be used directly if the
* values are regexes.
*
* @param string $key name of the metadata key to look for
* @param string $value search term to look for, must be a string or array of strings
* @param callback $func comparison function
* @return array lists with page names, keys are query values if $value is array
* @author Tom N Harris <tnharris@whoopdedo.org>
* @author Michael Hamann <michael@content-space.de>
*/
public function lookupKey($key, &$value, $func=null) {
if (!is_array($value))
$value_array = array($value);
else
$value_array =& $value;
// the matching ids for the provided value(s)
$value_ids = array();
$metaname = idx_cleanName($key);
// get all words in order to search the matching ids
if ($key == 'title') {
$words = $this->getIndex('title', '');
} else {
$words = $this->getIndex($metaname.'_w', '');
}
if (!is_null($func)) {
foreach ($value_array as $val) {
foreach ($words as $i => $word) {
if (call_user_func_array($func, array($val, $word)))
$value_ids[$i][] = $val;
}
}
} else {
foreach ($value_array as $val) {
$xval = $val;
$caret = '^';
$dollar = '$';
// check for wildcards
if (substr($xval, 0, 1) == '*') {
$xval = substr($xval, 1);
$caret = '';
}
if (substr($xval, -1, 1) == '*') {
$xval = substr($xval, 0, -1);
$dollar = '';
}
if (!$caret || !$dollar) {
$re = $caret.preg_quote($xval, '/').$dollar;
foreach(array_keys(preg_grep('/'.$re.'/', $words)) as $i)
$value_ids[$i][] = $val;
} else {
if (($i = array_search($val, $words, true)) !== false)
$value_ids[$i][] = $val;
}
}
}
unset($words); // free the used memory
// initialize the result so it won't be null
$result = array();
foreach ($value_array as $val) {
$result[$val] = array();
}
$page_idx = $this->getIndex('page', '');
// Special handling for titles
if ($key == 'title') {
foreach ($value_ids as $pid => $val_list) {
$page = $page_idx[$pid];
foreach ($val_list as $val) {
$result[$val][] = $page;
}
}
} else {
// load all lines and pages so the used lines can be taken and matched with the pages
$lines = $this->getIndex($metaname.'_i', '');
foreach ($value_ids as $value_id => $val_list) {
// parse the tuples of the form page_id*1:page2_id*1 and so on, return value
// is an array with page_id => 1, page2_id => 1 etc. so take the keys only
$pages = array_keys($this->parseTuples($page_idx, $lines[$value_id]));
foreach ($val_list as $val) {
$result[$val] = array_merge($result[$val], $pages);
}
}
}
if (!is_array($value)) $result = $result[$value];
return $result;
}
/**
* Find the index ID of each search term.
*
* The query terms should only contain valid characters, with a '*' at
* either the beginning or end of the word (or both).
* The $result parameter can be used to merge the index locations with
* the appropriate query term.
*
* @param array $words The query terms.
* @param array $result Set to word => array("length*id" ...)
* @return array Set to length => array(id ...)
* @author Tom N Harris <tnharris@whoopdedo.org>
*/
protected function getIndexWords(&$words, &$result) {
$tokens = array();
$tokenlength = array();
$tokenwild = array();
foreach ($words as $word) {
$result[$word] = array();
$caret = '^';
$dollar = '$';
$xword = $word;
$wlen = wordlen($word);
// check for wildcards
if (substr($xword, 0, 1) == '*') {
$xword = substr($xword, 1);
$caret = '';
$wlen -= 1;
}
if (substr($xword, -1, 1) == '*') {
$xword = substr($xword, 0, -1);
$dollar = '';
$wlen -= 1;
}
if ($wlen < IDX_MINWORDLENGTH && $caret && $dollar && !is_numeric($xword))
continue;
if (!isset($tokens[$xword]))
$tokenlength[$wlen][] = $xword;
if (!$caret || !$dollar) {
$re = $caret.preg_quote($xword, '/').$dollar;
$tokens[$xword][] = array($word, '/'.$re.'/');
if (!isset($tokenwild[$xword]))
$tokenwild[$xword] = $wlen;
} else {
$tokens[$xword][] = array($word, null);
}
}
asort($tokenwild);
// $tokens = array( base word => array( [ query term , regexp ] ... ) ... )
// $tokenlength = array( base word length => base word ... )
// $tokenwild = array( base word => base word length ... )
$length_filter = empty($tokenwild) ? $tokenlength : min(array_keys($tokenlength));
$indexes_known = $this->indexLengths($length_filter);
if (!empty($tokenwild)) sort($indexes_known);
// get word IDs
$wids = array();
foreach ($indexes_known as $ixlen) {
$word_idx = $this->getIndex('w', $ixlen);
// handle exact search
if (isset($tokenlength[$ixlen])) {
foreach ($tokenlength[$ixlen] as $xword) {
$wid = array_search($xword, $word_idx, true);
if ($wid !== false) {
$wids[$ixlen][] = $wid;
foreach ($tokens[$xword] as $w)
$result[$w[0]][] = "$ixlen*$wid";
}
}
}
// handle wildcard search
foreach ($tokenwild as $xword => $wlen) {
if ($wlen >= $ixlen) break;
foreach ($tokens[$xword] as $w) {
if (is_null($w[1])) continue;
foreach(array_keys(preg_grep($w[1], $word_idx)) as $wid) {
$wids[$ixlen][] = $wid;
$result[$w[0]][] = "$ixlen*$wid";
}
}
}
}
return $wids;
}
/**
* Return a list of all pages
* Warning: pages may not exist!
*
* @param string $key list only pages co
8FCC
ntaining the metadata key (optional)
* @return array list of page names
* @author Tom N Harris <tnharris@whoopdedo.org>
*/
public function getPages($key=null) {
$page_idx = $this->getIndex('page', '');
if (is_null($key)) return $page_idx;
$metaname = idx_cleanName($key);
// Special handling for titles
if ($key == 'title') {
$title_idx = $this->getIndex('title', '');
array_splice($page_idx, count($title_idx));
foreach ($title_idx as $i => $title)
if ($title === "") unset($page_idx[$i]);
return array_values($page_idx);
}
$pages = array();
$lines = $this->getIndex($metaname.'_i', '');
foreach ($lines as $line) {
$pages = array_merge($pages, $this->parseTuples($page_idx, $line));
}
return array_keys($pages);
}
/**
* Return a list of words sorted by number of times used
*
* @param int $min bottom frequency threshold
* @param int $max upper frequency limit. No limit if $max<$min
* @param int $minlen minimum length of words to count
* @param string $key metadata key to list. Uses the fulltext index if not given
* @return array list of words as the keys and frequency as values
* @author Tom N Harris <tnharris@whoopdedo.org>
*/
public function histogram($min=1, $max=0, $minlen=3, $key=null) {
if ($min < 1)
$min = 1;
if ($max < $min)
$max = 0;
$result = array();
if ($key == 'title') {
$index = $this->getIndex('title', '');
$index = array_count_values($index);
foreach ($index as $val => $cnt) {
if ($cnt >= $min && (!$max || $cnt <= $max) && strlen($val) >= $minlen)
$result[$val] = $cnt;
}
}
elseif (!is_null($key)) {
$metaname = idx_cleanName($key);
$index = $this->getIndex($metaname.'_i', '');
$val_idx = array();
foreach ($index as $wid => $line) {
$freq = $this->countTuples($line);
if ($freq >= $min && (!$max || $freq <= $max))
$val_idx[$wid] = $freq;
}
if (!empty($val_idx)) {
$words = $this->getIndex($metaname.'_w', '');
foreach ($val_idx as $wid => $freq) {
if (strlen($words[$wid]) >= $minlen)
$result[$words[$wid]] = $freq;
}
}
}
else {
$lengths = idx_listIndexLengths();
foreach ($lengths as $length) {
if ($length < $minlen) continue;
$index = $this->getIndex('i', $length);
$words = null;
foreach ($index as $wid => $line) {
$freq = $this->countTuples($line);
if ($freq >= $min && (!$max || $freq <= $max)) {
if ($words === null)
$words = $this->getIndex('w', $length);
$result[$words[$wid]] = $freq;
}
}
}
}
arsort($result);
return $result;
}
/**
* Lock the indexer.
*
* @author Tom N Harris <tnharris@whoopdedo.org>
*/