MediaWiki REL1_30
SearchMySQL.php
Go to the documentation of this file.
1<?php
32 protected $strictMatching = true;
33
34 private static $mMinSearchLength;
35
45 function parseQuery( $filteredText, $fulltext ) {
46 global $wgContLang;
47
48 $lc = $this->legalSearchChars( self::CHARS_NO_SYNTAX ); // Minus syntax chars (" and *)
49 $searchon = '';
50 $this->searchTerms = [];
51
52 # @todo FIXME: This doesn't handle parenthetical expressions.
53 $m = [];
54 if ( preg_match_all( '/([-+<>~]?)(([' . $lc . ']+)(\*?)|"[^"]*")/',
55 $filteredText, $m, PREG_SET_ORDER ) ) {
56 foreach ( $m as $bits ) {
57 MediaWiki\suppressWarnings();
58 list( /* all */, $modifier, $term, $nonQuoted, $wildcard ) = $bits;
59 MediaWiki\restoreWarnings();
60
61 if ( $nonQuoted != '' ) {
62 $term = $nonQuoted;
63 $quote = '';
64 } else {
65 $term = str_replace( '"', '', $term );
66 $quote = '"';
67 }
68
69 if ( $searchon !== '' ) {
70 $searchon .= ' ';
71 }
72 if ( $this->strictMatching && ( $modifier == '' ) ) {
73 // If we leave this out, boolean op defaults to OR which is rarely helpful.
74 $modifier = '+';
75 }
76
77 // Some languages such as Serbian store the input form in the search index,
78 // so we may need to search for matches in multiple writing system variants.
79 $convertedVariants = $wgContLang->autoConvertToAllVariants( $term );
80 if ( is_array( $convertedVariants ) ) {
81 $variants = array_unique( array_values( $convertedVariants ) );
82 } else {
83 $variants = [ $term ];
84 }
85
86 // The low-level search index does some processing on input to work
87 // around problems with minimum lengths and encoding in MySQL's
88 // fulltext engine.
89 // For Chinese this also inserts spaces between adjacent Han characters.
90 $strippedVariants = array_map(
91 [ $wgContLang, 'normalizeForSearch' ],
92 $variants );
93
94 // Some languages such as Chinese force all variants to a canonical
95 // form when stripping to the low-level search index, so to be sure
96 // let's check our variants list for unique items after stripping.
97 $strippedVariants = array_unique( $strippedVariants );
98
99 $searchon .= $modifier;
100 if ( count( $strippedVariants ) > 1 ) {
101 $searchon .= '(';
102 }
103 foreach ( $strippedVariants as $stripped ) {
104 $stripped = $this->normalizeText( $stripped );
105 if ( $nonQuoted && strpos( $stripped, ' ' ) !== false ) {
106 // Hack for Chinese: we need to toss in quotes for
107 // multiple-character phrases since normalizeForSearch()
108 // added spaces between them to make word breaks.
109 $stripped = '"' . trim( $stripped ) . '"';
110 }
111 $searchon .= "$quote$stripped$quote$wildcard ";
112 }
113 if ( count( $strippedVariants ) > 1 ) {
114 $searchon .= ')';
115 }
116
117 // Match individual terms or quoted phrase in result highlighting...
118 // Note that variants will be introduced in a later stage for highlighting!
119 $regexp = $this->regexTerm( $term, $wildcard );
120 $this->searchTerms[] = $regexp;
121 }
122 wfDebug( __METHOD__ . ": Would search with '$searchon'\n" );
123 wfDebug( __METHOD__ . ': Match with /' . implode( '|', $this->searchTerms ) . "/\n" );
124 } else {
125 wfDebug( __METHOD__ . ": Can't understand search query '{$filteredText}'\n" );
126 }
127
128 $searchon = $this->db->addQuotes( $searchon );
129 $field = $this->getIndexField( $fulltext );
130 return " MATCH($field) AGAINST($searchon IN BOOLEAN MODE) ";
131 }
132
133 function regexTerm( $string, $wildcard ) {
134 global $wgContLang;
135
136 $regex = preg_quote( $string, '/' );
137 if ( $wgContLang->hasWordBreaks() ) {
138 if ( $wildcard ) {
139 // Don't cut off the final bit!
140 $regex = "\b$regex";
141 } else {
142 $regex = "\b$regex\b";
143 }
144 } else {
145 // For Chinese, words may legitimately abut other words in the text literal.
146 // Don't add \b boundary checks... note this could cause false positives
147 // for latin chars.
148 }
149 return $regex;
150 }
151
152 public static function legalSearchChars( $type = self::CHARS_ALL ) {
153 $searchChars = parent::legalSearchChars( $type );
154 if ( $type === self::CHARS_ALL ) {
155 // " for phrase, * for wildcard
156 $searchChars = "\"*" . $searchChars;
157 }
158 return $searchChars;
159 }
160
167 function searchText( $term ) {
168 return $this->searchInternal( $term, true );
169 }
170
177 function searchTitle( $term ) {
178 return $this->searchInternal( $term, false );
179 }
180
181 protected function searchInternal( $term, $fulltext ) {
182 // This seems out of place, why is this called with empty term?
183 if ( trim( $term ) === '' ) {
184 return null;
185 }
186
187 $filteredTerm = $this->filter( $term );
188 $query = $this->getQuery( $filteredTerm, $fulltext );
189 $resultSet = $this->db->select(
190 $query['tables'], $query['fields'], $query['conds'],
191 __METHOD__, $query['options'], $query['joins']
192 );
193
194 $total = null;
195 $query = $this->getCountQuery( $filteredTerm, $fulltext );
196 $totalResult = $this->db->select(
197 $query['tables'], $query['fields'], $query['conds'],
198 __METHOD__, $query['options'], $query['joins']
199 );
200
201 $row = $totalResult->fetchObject();
202 if ( $row ) {
203 $total = intval( $row->c );
204 }
205 $totalResult->free();
206
207 return new SqlSearchResultSet( $resultSet, $this->searchTerms, $total );
208 }
209
210 public function supports( $feature ) {
211 switch ( $feature ) {
212 case 'title-suffix-filter':
213 return true;
214 default:
215 return parent::supports( $feature );
216 }
217 }
218
224 protected function queryFeatures( &$query ) {
225 foreach ( $this->features as $feature => $value ) {
226 if ( $feature === 'title-suffix-filter' && $value ) {
227 $query['conds'][] = 'page_title' . $this->db->buildLike( $this->db->anyString(), $value );
228 }
229 }
230 }
231
237 function queryNamespaces( &$query ) {
238 if ( is_array( $this->namespaces ) ) {
239 if ( count( $this->namespaces ) === 0 ) {
240 $this->namespaces[] = '0';
241 }
242 $query['conds']['page_namespace'] = $this->namespaces;
243 }
244 }
245
251 protected function limitResult( &$query ) {
252 $query['options']['LIMIT'] = $this->limit;
253 $query['options']['OFFSET'] = $this->offset;
254 }
255
264 function getQuery( $filteredTerm, $fulltext ) {
265 $query = [
266 'tables' => [],
267 'fields' => [],
268 'conds' => [],
269 'options' => [],
270 'joins' => [],
271 ];
272
273 $this->queryMain( $query, $filteredTerm, $fulltext );
274 $this->queryFeatures( $query );
275 $this->queryNamespaces( $query );
276 $this->limitResult( $query );
277
278 return $query;
279 }
280
286 function getIndexField( $fulltext ) {
287 return $fulltext ? 'si_text' : 'si_title';
288 }
289
298 function queryMain( &$query, $filteredTerm, $fulltext ) {
299 $match = $this->parseQuery( $filteredTerm, $fulltext );
300 $query['tables'][] = 'page';
301 $query['tables'][] = 'searchindex';
302 $query['fields'][] = 'page_id';
303 $query['fields'][] = 'page_namespace';
304 $query['fields'][] = 'page_title';
305 $query['conds'][] = 'page_id=si_page';
306 $query['conds'][] = $match;
307 }
308
315 function getCountQuery( $filteredTerm, $fulltext ) {
316 $match = $this->parseQuery( $filteredTerm, $fulltext );
317
318 $query = [
319 'tables' => [ 'page', 'searchindex' ],
320 'fields' => [ 'COUNT(*) as c' ],
321 'conds' => [ 'page_id=si_page', $match ],
322 'options' => [],
323 'joins' => [],
324 ];
325
326 $this->queryFeatures( $query );
327 $this->queryNamespaces( $query );
328
329 return $query;
330 }
331
340 function update( $id, $title, $text ) {
341 $dbw = wfGetDB( DB_MASTER );
342 $dbw->replace( 'searchindex',
343 [ 'si_page' ],
344 [
345 'si_page' => $id,
346 'si_title' => $this->normalizeText( $title ),
347 'si_text' => $this->normalizeText( $text )
348 ], __METHOD__ );
349 }
350
358 function updateTitle( $id, $title ) {
359 $dbw = wfGetDB( DB_MASTER );
360
361 $dbw->update( 'searchindex',
362 [ 'si_title' => $this->normalizeText( $title ) ],
363 [ 'si_page' => $id ],
364 __METHOD__,
365 [ $dbw->lowPriorityOption() ] );
366 }
367
375 function delete( $id, $title ) {
376 $dbw = wfGetDB( DB_MASTER );
377
378 $dbw->delete( 'searchindex', [ 'si_page' => $id ], __METHOD__ );
379 }
380
387 function normalizeText( $string ) {
388 global $wgContLang;
389
390 $out = parent::normalizeText( $string );
391
392 // MySQL fulltext index doesn't grok utf-8, so we
393 // need to fold cases and convert to hex
394 $out = preg_replace_callback(
395 "/([\\xc0-\\xff][\\x80-\\xbf]*)/",
396 [ $this, 'stripForSearchCallback' ],
397 $wgContLang->lc( $out ) );
398
399 // And to add insult to injury, the default indexing
400 // ignores short words... Pad them so we can pass them
401 // through without reconfiguring the server...
402 $minLength = $this->minSearchLength();
403 if ( $minLength > 1 ) {
404 $n = $minLength - 1;
405 $out = preg_replace(
406 "/\b(\w{1,$n})\b/",
407 "$1u800",
408 $out );
409 }
410
411 // Periods within things like hostnames and IP addresses
412 // are also important -- we want a search for "example.com"
413 // or "192.168.1.1" to work sanely.
414 // MySQL's search seems to ignore them, so you'd match on
415 // "example.wikipedia.com" and "192.168.83.1" as well.
416 $out = preg_replace(
417 "/(\w)\.(\w|\*)/u",
418 "$1u82e$2",
419 $out );
420
421 return $out;
422 }
423
431 protected function stripForSearchCallback( $matches ) {
432 return 'u8' . bin2hex( $matches[1] );
433 }
434
441 protected function minSearchLength() {
442 if ( is_null( self::$mMinSearchLength ) ) {
443 $sql = "SHOW GLOBAL VARIABLES LIKE 'ft\\_min\\_word\\_len'";
444
446 $result = $dbr->query( $sql, __METHOD__ );
447 $row = $result->fetchObject();
448 $result->free();
449
450 if ( $row && $row->Variable_name == 'ft_min_word_len' ) {
451 self::$mMinSearchLength = intval( $row->Value );
452 } else {
453 self::$mMinSearchLength = 0;
454 }
455 }
457 }
458}
to move a page</td >< td > &*You are moving the page across namespaces
wfDebug( $text, $dest='all', array $context=[])
Sends a line to the debug log if enabled or, optionally, to a comment in output.
wfGetDB( $db, $groups=[], $wiki=false)
Get a Database object.
Base search engine base class for database-backed searches.
filter( $text)
Return a 'cleaned up' search string.
int[] null $namespaces
Search engine hook for MySQL 4+.
parseQuery( $filteredText, $fulltext)
Parse the user's query and transform it into an SQL fragment which will become part of a WHERE clause...
queryMain(&$query, $filteredTerm, $fulltext)
Get the base part of the search query.
static legalSearchChars( $type=self::CHARS_ALL)
Get chars legal for search NOTE: usage as static is deprecated and preserved only as BC measure.
updateTitle( $id, $title)
Update a search index record's title only.
getQuery( $filteredTerm, $fulltext)
Construct the SQL query to do the search.
stripForSearchCallback( $matches)
Armor a case-folded UTF-8 string to get through MySQL's fulltext search without being mucked up by fu...
getIndexField( $fulltext)
Picks which field to index on, depending on what type of query.
queryFeatures(&$query)
Add special conditions.
update( $id, $title, $text)
Create or update the search index record for the given page.
supports( $feature)
searchInternal( $term, $fulltext)
static $mMinSearchLength
limitResult(&$query)
Add limit options.
regexTerm( $string, $wildcard)
queryNamespaces(&$query)
Add namespace conditions.
searchText( $term)
Perform a full text search query and return a result set.
getCountQuery( $filteredTerm, $fulltext)
normalizeText( $string)
Converts some characters for MySQL's indexing to grok it correctly, and pads short words to overcome ...
minSearchLength()
Check MySQL server's ft_min_word_len setting so we know if we need to pad short words....
searchTitle( $term)
Perform a title-only search query and return a result set.
This class is used for different SQL-based search engines shipped with MediaWiki.
if(! $regexes) $dbr
Definition cleanup.php:94
deferred txt A few of the database updates required by various functions here can be deferred until after the result page is displayed to the user For updating the view updating the linked to tables after a etc PHP does not yet have any way to tell the server to actually return and disconnect while still running these but it might have such a feature in the future We handle these by creating a deferred update object and putting those objects on a global list
Definition deferred.txt:11
this class mediates it Skin Encapsulates a look and feel for the wiki All of the functions that render HTML and make choices about how to render it are here and are called from various other places when and is meant to be subclassed with other skins that may override some of its functions The User object contains a reference to a and so rather than having a global skin object we just rely on the global User and get the skin with $wgUser and also has some character encoding functions and other locale stuff The current user interface language is instantiated as and the local content language as $wgContLang
Definition design.txt:57
external whereas SearchGetNearMatch runs after $term
Definition hooks.txt:2814
namespace and then decline to actually register it file or subcat img or subcat $title
Definition hooks.txt:962
this hook is for auditing only or null if authentication failed before getting that far or null if we can t even determine that probably a stub it is not rendered in wiki pages or galleries in category pages allow injecting custom HTML after the section Any uses of the hook need to handle escaping see BaseTemplate::getToolbox and BaseTemplate::makeListItem for details on the format of individual items inside of this array or by returning and letting standard HTTP rendering take place modifiable or by returning false and taking over the output $out
Definition hooks.txt:862
null for the local wiki Added should default to null in handler for backwards compatibility add a value to it if you want to add a cookie that have to vary cache options can modify $query
Definition hooks.txt:1610
const DB_REPLICA
Definition defines.php:25
const DB_MASTER
Definition defines.php:26