1
|
|
|
<?php |
2
|
|
|
|
3
|
|
|
namespace Foolz\SphinxQL; |
4
|
|
|
|
5
|
|
|
/** |
6
|
|
|
* Query Builder class for Match statements. |
7
|
|
|
*/ |
8
|
|
|
class MatchBuilder |
9
|
|
|
{ |
10
|
|
|
/** |
11
|
|
|
* The last compiled query. |
12
|
|
|
* |
13
|
|
|
* @var string |
14
|
|
|
*/ |
15
|
|
|
protected $last_compiled; |
16
|
|
|
|
17
|
|
|
/** |
18
|
|
|
* List of match operations. |
19
|
|
|
* |
20
|
|
|
* @var array |
21
|
|
|
*/ |
22
|
|
|
protected $tokens = array(); |
23
|
|
|
|
24
|
|
|
/** |
25
|
|
|
* The owning SphinxQL object; used for escaping text. |
26
|
|
|
* |
27
|
|
|
* @var SphinxQL |
28
|
|
|
*/ |
29
|
|
|
protected $sphinxql; |
30
|
|
|
|
31
|
|
|
/** |
32
|
|
|
* @param SphinxQL $sphinxql |
33
|
|
|
*/ |
34
|
|
|
public function __construct(SphinxQL $sphinxql) |
35
|
|
|
{ |
36
|
|
|
$this->sphinxql = $sphinxql; |
37
|
|
|
} |
38
|
|
|
|
39
|
|
|
/** |
40
|
|
|
* Match text or sub expression. |
41
|
|
|
* |
42
|
|
|
* Examples: |
43
|
|
|
* $match->match('test'); |
44
|
|
|
* // test |
45
|
|
|
* |
46
|
|
|
* $match->match('test case'); |
47
|
|
|
* // (test case) |
48
|
|
|
* |
49
|
|
|
* $match->match(function ($m) { |
50
|
|
|
* $m->match('a')->orMatch('b'); |
51
|
|
|
* }); |
52
|
|
|
* // (a | b) |
53
|
|
|
* |
54
|
|
|
* $sub = new MatchBuilder($sphinxql); |
55
|
|
|
* $sub->match('a')->orMatch('b'); |
56
|
|
|
* $match->match($sub); |
57
|
|
|
* // (a | b) |
58
|
|
|
* |
59
|
|
|
* @param string|MatchBuilder|\Closure $keywords The text or expression to match. |
60
|
|
|
* |
61
|
|
|
* @return $this |
62
|
|
|
*/ |
63
|
|
|
public function match($keywords = null) |
64
|
|
|
{ |
65
|
|
|
if ($keywords !== null) { |
66
|
|
|
$this->tokens[] = array('MATCH' => $keywords); |
67
|
|
|
} |
68
|
|
|
|
69
|
|
|
return $this; |
70
|
|
|
} |
71
|
|
|
|
72
|
|
|
/** |
73
|
|
|
* Provide an alternation match. |
74
|
|
|
* |
75
|
|
|
* Examples: |
76
|
|
|
* $match->match('test')->orMatch(); |
77
|
|
|
* // test | |
78
|
|
|
* |
79
|
|
|
* $match->match('test')->orMatch('case'); |
80
|
|
|
* // test | case |
81
|
|
|
* |
82
|
|
|
* @param string|MatchBuilder|\Closure $keywords The text or expression to alternatively match. |
83
|
|
|
* |
84
|
|
|
* @return $this |
85
|
|
|
*/ |
86
|
|
|
public function orMatch($keywords = null) |
87
|
|
|
{ |
88
|
|
|
$this->tokens[] = array('OPERATOR' => '| '); |
89
|
|
|
$this->match($keywords); |
90
|
|
|
|
91
|
|
|
return $this; |
92
|
|
|
} |
93
|
|
|
|
94
|
|
|
/** |
95
|
|
|
* Provide an optional match. |
96
|
|
|
* |
97
|
|
|
* Examples: |
98
|
|
|
* $match->match('test')->maybe(); |
99
|
|
|
* // test MAYBE |
100
|
|
|
* |
101
|
|
|
* $match->match('test')->maybe('case'); |
102
|
|
|
* // test MAYBE case |
103
|
|
|
* |
104
|
|
|
* @param string|MatchBuilder|\Closure $keywords The text or expression to optionally match. |
105
|
|
|
* |
106
|
|
|
* @return $this |
107
|
|
|
*/ |
108
|
|
|
public function maybe($keywords = null) |
109
|
|
|
{ |
110
|
|
|
$this->tokens[] = array('OPERATOR' => 'MAYBE '); |
111
|
|
|
$this->match($keywords); |
112
|
|
|
|
113
|
|
|
return $this; |
114
|
|
|
} |
115
|
|
|
|
116
|
|
|
/** |
117
|
|
|
* Do not match a keyword. |
118
|
|
|
* |
119
|
|
|
* Examples: |
120
|
|
|
* $match->not()->match('test'); |
121
|
|
|
* // -test |
122
|
|
|
* |
123
|
|
|
* $match->not('test'); |
124
|
|
|
* // -test |
125
|
|
|
* |
126
|
|
|
* @param string $keyword The word not to match. |
127
|
|
|
* |
128
|
|
|
* @return $this |
129
|
|
|
*/ |
130
|
|
|
public function not($keyword = null) |
131
|
|
|
{ |
132
|
|
|
$this->tokens[] = array('OPERATOR' => '-'); |
133
|
|
|
$this->match($keyword); |
134
|
|
|
|
135
|
|
|
return $this; |
136
|
|
|
} |
137
|
|
|
|
138
|
|
|
/** |
139
|
|
|
* Specify which field(s) to search. |
140
|
|
|
* |
141
|
|
|
* Examples: |
142
|
|
|
* $match->field('*')->match('test'); |
143
|
|
|
* // @* test |
144
|
|
|
* |
145
|
|
|
* $match->field('title')->match('test'); |
146
|
|
|
* // @title test |
147
|
|
|
* |
148
|
|
|
* $match->field('body', 50)->match('test'); |
149
|
|
|
* // @body[50] test |
150
|
|
|
* |
151
|
|
|
* $match->field('title', 'body')->match('test'); |
152
|
|
|
* // @(title,body) test |
153
|
|
|
* |
154
|
|
|
* $match->field(['title', 'body'])->match('test'); |
155
|
|
|
* // @(title,body) test |
156
|
|
|
* |
157
|
|
|
* $match->field('@relaxed')->field('nosuchfield')->match('test'); |
158
|
|
|
* // @@relaxed @nosuchfield test |
159
|
|
|
* |
160
|
|
|
* @param string|array $fields Field or fields to search. |
161
|
|
|
* @param int $limit Maximum position limit in field a match is allowed at. |
162
|
|
|
* |
163
|
|
|
* @return $this |
164
|
|
|
*/ |
165
|
|
|
public function field($fields, $limit = null) |
166
|
|
|
{ |
167
|
|
|
if (is_string($fields)) { |
168
|
|
|
$fields = func_get_args(); |
169
|
|
|
$limit = null; |
170
|
|
|
} |
171
|
|
|
if (!is_string(end($fields))) { |
172
|
|
|
$limit = array_pop($fields); |
173
|
|
|
} |
174
|
|
|
$this->tokens[] = array( |
175
|
|
|
'FIELD' => '@', |
176
|
|
|
'fields' => $fields, |
177
|
|
|
'limit' => $limit, |
178
|
|
|
); |
179
|
|
|
|
180
|
|
|
return $this; |
181
|
|
|
} |
182
|
|
|
|
183
|
|
|
/** |
184
|
|
|
* Specify which field(s) not to search. |
185
|
|
|
* |
186
|
|
|
* Examples: |
187
|
|
|
* $match->ignoreField('title')->match('test'); |
188
|
|
|
* // @!title test |
189
|
|
|
* |
190
|
|
|
* $match->ignoreField('title', 'body')->match('test'); |
191
|
|
|
* // @!(title,body) test |
192
|
|
|
* |
193
|
|
|
* $match->ignoreField(['title', 'body'])->match('test'); |
194
|
|
|
* // @!(title,body) test |
195
|
|
|
* |
196
|
|
|
* @param string|array $fields Field or fields to ignore during search. |
197
|
|
|
* |
198
|
|
|
* @return $this |
199
|
|
|
*/ |
200
|
|
|
public function ignoreField($fields) |
201
|
|
|
{ |
202
|
|
|
if (is_string($fields)) { |
203
|
|
|
$fields = func_get_args(); |
204
|
|
|
} |
205
|
|
|
$this->tokens[] = array( |
206
|
|
|
'FIELD' => '@!', |
207
|
|
|
'fields' => $fields, |
208
|
|
|
'limit' => null, |
209
|
|
|
); |
210
|
|
|
|
211
|
|
|
return $this; |
212
|
|
|
} |
213
|
|
|
|
214
|
|
|
/** |
215
|
|
|
* Match an exact phrase. |
216
|
|
|
* |
217
|
|
|
* Example: |
218
|
|
|
* $match->phrase('test case'); |
219
|
|
|
* // "test case" |
220
|
|
|
* |
221
|
|
|
* @param string $keywords The phrase to match. |
222
|
|
|
* |
223
|
|
|
* @return $this |
224
|
|
|
*/ |
225
|
|
|
public function phrase($keywords) |
226
|
|
|
{ |
227
|
|
|
$this->tokens[] = array('PHRASE' => $keywords); |
228
|
|
|
|
229
|
|
|
return $this; |
230
|
|
|
} |
231
|
|
|
|
232
|
|
|
/** |
233
|
|
|
* Provide an optional phrase. |
234
|
|
|
* |
235
|
|
|
* Example: |
236
|
|
|
* $match->phrase('test case')->orPhrase('another case'); |
237
|
|
|
* // "test case" | "another case" |
238
|
|
|
* |
239
|
|
|
* @param string $keywords The phrase to match. |
240
|
|
|
* |
241
|
|
|
* @return $this |
242
|
|
|
*/ |
243
|
|
|
public function orPhrase($keywords) |
244
|
|
|
{ |
245
|
|
|
$this->tokens[] = array('OPERATOR' => '| '); |
246
|
|
|
$this->phrase($keywords); |
247
|
|
|
|
248
|
|
|
return $this; |
249
|
|
|
} |
250
|
|
|
|
251
|
|
|
/** |
252
|
|
|
* Match if keywords are close enough. |
253
|
|
|
* |
254
|
|
|
* Example: |
255
|
|
|
* $match->proximity('test case', 5); |
256
|
|
|
* // "test case"~5 |
257
|
|
|
* |
258
|
|
|
* @param string $keywords The words to match. |
259
|
|
|
* @param int $distance The upper limit on separation between words. |
260
|
|
|
* |
261
|
|
|
* @return $this |
262
|
|
|
*/ |
263
|
|
|
public function proximity($keywords, $distance) |
264
|
|
|
{ |
265
|
|
|
$this->tokens[] = array( |
266
|
|
|
'PROXIMITY' => $distance, |
267
|
|
|
'keywords' => $keywords, |
268
|
|
|
); |
269
|
|
|
|
270
|
|
|
return $this; |
271
|
|
|
} |
272
|
|
|
|
273
|
|
|
/** |
274
|
|
|
* Match if enough keywords are present. |
275
|
|
|
* |
276
|
|
|
* Examples: |
277
|
|
|
* $match->quorum('this is a test case', 3); |
278
|
|
|
* // "this is a test case"/3 |
279
|
|
|
* |
280
|
|
|
* $match->quorum('this is a test case', 0.5); |
281
|
|
|
* // "this is a test case"/0.5 |
282
|
|
|
* |
283
|
|
|
* @param string $keywords The words to match. |
284
|
|
|
* @param int|float $threshold The minimum number or percent of words that must match. |
285
|
|
|
* |
286
|
|
|
* @return $this |
287
|
|
|
*/ |
288
|
|
|
public function quorum($keywords, $threshold) |
289
|
|
|
{ |
290
|
|
|
$this->tokens[] = array( |
291
|
|
|
'QUORUM' => $threshold, |
292
|
|
|
'keywords' => $keywords, |
293
|
|
|
); |
294
|
|
|
|
295
|
|
|
return $this; |
296
|
|
|
} |
297
|
|
|
|
298
|
|
|
/** |
299
|
|
|
* Assert keywords or expressions must be matched in order. |
300
|
|
|
* |
301
|
|
|
* Examples: |
302
|
|
|
* $match->match('test')->before(); |
303
|
|
|
* // test << |
304
|
|
|
* |
305
|
|
|
* $match->match('test')->before('case'); |
306
|
|
|
* // test << case |
307
|
|
|
* |
308
|
|
|
* @param string|Match|\Closure $keywords The text or expression that must come after. |
309
|
|
|
* |
310
|
|
|
* @return $this |
311
|
|
|
*/ |
312
|
|
|
public function before($keywords = null) |
313
|
|
|
{ |
314
|
|
|
$this->tokens[] = array('OPERATOR' => '<< '); |
315
|
|
|
$this->match($keywords); |
316
|
|
|
|
317
|
|
|
return $this; |
318
|
|
|
} |
319
|
|
|
|
320
|
|
|
/** |
321
|
|
|
* Assert a keyword must be matched exactly as written. |
322
|
|
|
* |
323
|
|
|
* Examples: |
324
|
|
|
* $match->match('test')->exact('cases'); |
325
|
|
|
* // test =cases |
326
|
|
|
* |
327
|
|
|
* $match->match('test')->exact()->phrase('specific cases'); |
328
|
|
|
* // test ="specific cases" |
329
|
|
|
* |
330
|
|
|
* @param string $keyword The word that must be matched exactly. |
331
|
|
|
* |
332
|
|
|
* @return $this |
333
|
|
|
*/ |
334
|
|
|
public function exact($keyword = null) |
335
|
|
|
{ |
336
|
|
|
$this->tokens[] = array('OPERATOR' => '='); |
337
|
|
|
$this->match($keyword); |
338
|
|
|
|
339
|
|
|
return $this; |
340
|
|
|
} |
341
|
|
|
|
342
|
|
|
/** |
343
|
|
|
* Boost the IDF score of a keyword. |
344
|
|
|
* |
345
|
|
|
* Examples: |
346
|
|
|
* $match->match('test')->boost(1.2); |
347
|
|
|
* // test^1.2 |
348
|
|
|
* |
349
|
|
|
* $match->match('test')->boost('case', 1.2); |
350
|
|
|
* // test case^1.2 |
351
|
|
|
* |
352
|
|
|
* @param string $keyword The word to modify the score of. |
353
|
|
|
* @param float $amount The amount to boost the score. |
354
|
|
|
* |
355
|
|
|
* @return $this |
356
|
|
|
*/ |
357
|
|
|
public function boost($keyword, $amount = null) |
358
|
|
|
{ |
359
|
|
|
if ($amount === null) { |
360
|
|
|
$amount = $keyword; |
361
|
|
|
} else { |
362
|
|
|
$this->match($keyword); |
363
|
|
|
} |
364
|
|
|
$this->tokens[] = array('BOOST' => $amount); |
365
|
|
|
|
366
|
|
|
return $this; |
367
|
|
|
} |
368
|
|
|
|
369
|
|
|
/** |
370
|
|
|
* Assert keywords or expressions must be matched close to each other. |
371
|
|
|
* |
372
|
|
|
* Examples: |
373
|
|
|
* $match->match('test')->near(3); |
374
|
|
|
* // test NEAR/3 |
375
|
|
|
* |
376
|
|
|
* $match->match('test')->near('case', 3); |
377
|
|
|
* // test NEAR/3 case |
378
|
|
|
* |
379
|
|
|
* @param string|Match|\Closure $keywords The text or expression to match nearby. |
380
|
|
|
* @param int $distance Maximum distance to the match. |
381
|
|
|
* |
382
|
|
|
* @return $this |
383
|
|
|
*/ |
384
|
|
|
public function near($keywords, $distance = null) |
385
|
|
|
{ |
386
|
|
|
$this->tokens[] = array('NEAR' => $distance ?: $keywords); |
387
|
|
|
if ($distance !== null) { |
388
|
|
|
$this->match($keywords); |
389
|
|
|
} |
390
|
|
|
|
391
|
|
|
return $this; |
392
|
|
|
} |
393
|
|
|
|
394
|
|
|
/** |
395
|
|
|
* Assert matches must be in the same sentence. |
396
|
|
|
* |
397
|
|
|
* Examples: |
398
|
|
|
* $match->match('test')->sentence(); |
399
|
|
|
* // test SENTENCE |
400
|
|
|
* |
401
|
|
|
* $match->match('test')->sentence('case'); |
402
|
|
|
* // test SENTENCE case |
403
|
|
|
* |
404
|
|
|
* @param string|Match|\Closure $keywords The text or expression that must be in the sentence. |
405
|
|
|
* |
406
|
|
|
* @return $this |
407
|
|
|
*/ |
408
|
|
|
public function sentence($keywords = null) |
409
|
|
|
{ |
410
|
|
|
$this->tokens[] = array('OPERATOR' => 'SENTENCE '); |
411
|
|
|
$this->match($keywords); |
412
|
|
|
|
413
|
|
|
return $this; |
414
|
|
|
} |
415
|
|
|
|
416
|
|
|
/** |
417
|
|
|
* Assert matches must be in the same paragraph. |
418
|
|
|
* |
419
|
|
|
* Examples: |
420
|
|
|
* $match->match('test')->paragraph(); |
421
|
|
|
* // test PARAGRAPH |
422
|
|
|
* |
423
|
|
|
* $match->match('test')->paragraph('case'); |
424
|
|
|
* // test PARAGRAPH case |
425
|
|
|
* |
426
|
|
|
* @param string|Match|\Closure $keywords The text or expression that must be in the paragraph. |
427
|
|
|
* |
428
|
|
|
* @return $this |
429
|
|
|
*/ |
430
|
|
|
public function paragraph($keywords = null) |
431
|
|
|
{ |
432
|
|
|
$this->tokens[] = array('OPERATOR' => 'PARAGRAPH '); |
433
|
|
|
$this->match($keywords); |
434
|
|
|
|
435
|
|
|
return $this; |
436
|
|
|
} |
437
|
|
|
|
438
|
|
|
/** |
439
|
|
|
* Assert matches must be in the specified zone(s). |
440
|
|
|
* |
441
|
|
|
* Examples: |
442
|
|
|
* $match->zone('th'); |
443
|
|
|
* // ZONE:(th) |
444
|
|
|
* |
445
|
|
|
* $match->zone(['h3', 'h4']); |
446
|
|
|
* // ZONE:(h3,h4) |
447
|
|
|
* |
448
|
|
|
* $match->zone('th', 'test'); |
449
|
|
|
* // ZONE:(th) test |
450
|
|
|
* |
451
|
|
|
* @param string|array $zones The zone or zones to search. |
452
|
|
|
* @param string|Match|\Closure $keywords The text or expression that must be in these zones. |
453
|
|
|
* |
454
|
|
|
* @return $this |
455
|
|
|
*/ |
456
|
|
|
public function zone($zones, $keywords = null) |
457
|
|
|
{ |
458
|
|
|
if (is_string($zones)) { |
459
|
|
|
$zones = array($zones); |
460
|
|
|
} |
461
|
|
|
$this->tokens[] = array('ZONE' => $zones); |
462
|
|
|
$this->match($keywords); |
463
|
|
|
|
464
|
|
|
return $this; |
465
|
|
|
} |
466
|
|
|
|
467
|
|
|
|
468
|
|
|
/** |
469
|
|
|
* Assert matches must be in the same instance of the specified zone. |
470
|
|
|
* |
471
|
|
|
* Examples: |
472
|
|
|
* $match->zonespan('th'); |
473
|
|
|
* // ZONESPAN:(th) |
474
|
|
|
* |
475
|
|
|
* $match->zonespan('th', 'test'); |
476
|
|
|
* // ZONESPAN:(th) test |
477
|
|
|
* |
478
|
|
|
* @param string $zone The zone to search. |
479
|
|
|
* @param string|Match|\Closure $keywords The text or expression that must be in this zone. |
480
|
|
|
* |
481
|
|
|
* @return $this |
482
|
|
|
*/ |
483
|
|
|
public function zonespan($zone, $keywords = null) |
484
|
|
|
{ |
485
|
|
|
$this->tokens[] = array('ZONESPAN' => $zone); |
486
|
|
|
$this->match($keywords); |
487
|
|
|
|
488
|
|
|
return $this; |
489
|
|
|
} |
490
|
|
|
|
491
|
|
|
/** |
492
|
|
|
* Build the match expression. |
493
|
|
|
* |
494
|
|
|
* @return $this |
495
|
|
|
*/ |
496
|
|
|
public function compile() |
497
|
|
|
{ |
498
|
|
|
$query = ''; |
499
|
|
|
foreach ($this->tokens as $token) { |
500
|
|
|
if (key($token) == 'MATCH') { |
501
|
|
|
if ($token['MATCH'] instanceof Expression) { |
502
|
|
|
$query .= $token['MATCH']->value().' '; |
503
|
|
|
} elseif ($token['MATCH'] instanceof MatchBuilder) { |
504
|
|
|
$query .= '('.$token['MATCH']->compile()->getCompiled().') '; |
505
|
|
|
} elseif ($token['MATCH'] instanceof \Closure) { |
506
|
|
|
$sub = new static($this->sphinxql); |
507
|
|
|
call_user_func($token['MATCH'], $sub); |
508
|
|
|
$query .= '('.$sub->compile()->getCompiled().') '; |
509
|
|
|
} elseif (strpos($token['MATCH'], ' ') === false) { |
510
|
|
|
$query .= $this->sphinxql->escapeMatch($token['MATCH']).' '; |
511
|
|
|
} else { |
512
|
|
|
$query .= '('.$this->sphinxql->escapeMatch($token['MATCH']).') '; |
513
|
|
|
} |
514
|
|
|
} elseif (key($token) == 'OPERATOR') { |
515
|
|
|
$query .= $token['OPERATOR']; |
516
|
|
|
} elseif (key($token) == 'FIELD') { |
517
|
|
|
$query .= $token['FIELD']; |
518
|
|
|
if (count($token['fields']) == 1) { |
519
|
|
|
$query .= $token['fields'][0]; |
520
|
|
|
} else { |
521
|
|
|
$query .= '('.implode(',', $token['fields']).')'; |
522
|
|
|
} |
523
|
|
|
if ($token['limit']) { |
524
|
|
|
$query .= '['.$token['limit'].']'; |
525
|
|
|
} |
526
|
|
|
$query .= ' '; |
527
|
|
|
} elseif (key($token) == 'PHRASE') { |
528
|
|
|
$query .= '"'.$this->sphinxql->escapeMatch($token['PHRASE']).'" '; |
529
|
|
|
} elseif (key($token) == 'PROXIMITY') { |
530
|
|
|
$query .= '"'.$this->sphinxql->escapeMatch($token['keywords']).'"~'; |
531
|
|
|
$query .= $token['PROXIMITY'].' '; |
532
|
|
|
} elseif (key($token) == 'QUORUM') { |
533
|
|
|
$query .= '"'.$this->sphinxql->escapeMatch($token['keywords']).'"/'; |
534
|
|
|
$query .= $token['QUORUM'].' '; |
535
|
|
|
} elseif (key($token) == 'BOOST') { |
536
|
|
|
$query = rtrim($query).'^'.$token['BOOST'].' '; |
537
|
|
|
} elseif (key($token) == 'NEAR') { |
538
|
|
|
$query .= 'NEAR/'.$token['NEAR'].' '; |
539
|
|
|
} elseif (key($token) == 'ZONE') { |
540
|
|
|
$query .= 'ZONE:('.implode(',', $token['ZONE']).') '; |
541
|
|
|
} elseif (key($token) == 'ZONESPAN') { |
542
|
|
|
$query .= 'ZONESPAN:('.$token['ZONESPAN'].') '; |
543
|
|
|
} |
544
|
|
|
} |
545
|
|
|
$this->last_compiled = trim($query); |
546
|
|
|
|
547
|
|
|
return $this; |
548
|
|
|
} |
549
|
|
|
|
550
|
|
|
/** |
551
|
|
|
* Returns the latest compiled match expression. |
552
|
|
|
* |
553
|
|
|
* @return string The last compiled match expression. |
554
|
|
|
*/ |
555
|
|
|
public function getCompiled() |
556
|
|
|
{ |
557
|
|
|
return $this->last_compiled; |
558
|
|
|
} |
559
|
|
|
} |
560
|
|
|
|