1 | <?php |
||
2 | |||
3 | namespace Samwilson\SimpleWikidata; |
||
4 | |||
5 | use DateInterval; |
||
6 | use Exception; |
||
7 | use Mediawiki\Api\MediawikiApi; |
||
8 | use Mediawiki\Api\SimpleRequest; |
||
9 | use Psr\Cache\CacheItemPoolInterface; |
||
10 | use Samwilson\SimpleWikidata\Properties\Time; |
||
11 | use Symfony\Component\DomCrawler\Crawler; |
||
12 | |||
13 | class Item { |
||
14 | |||
15 | /** @var string|bool */ |
||
16 | const INSTANCE_OF = false; |
||
17 | |||
18 | const PROP_INSTANCE_OF = 'P31'; |
||
19 | const PROP_TITLE = 'P1476'; |
||
20 | const PROP_IMAGE = 'P18'; |
||
21 | const PROP_AUTHOR = 'P50'; |
||
22 | |||
23 | /** @var string[] List of Q-numbers of registered classes. */ |
||
24 | protected static $registeredClasses = []; |
||
25 | |||
26 | /** @var string */ |
||
27 | protected $id; |
||
28 | |||
29 | /** @var MediawikiApi */ |
||
30 | protected $wdApi; |
||
31 | |||
32 | /** @var string */ |
||
33 | protected $lang; |
||
34 | |||
35 | /** @var CacheItemPoolInterface */ |
||
36 | protected $cache; |
||
37 | |||
38 | /** @var string The base URL of Wikidata, with trailing slash. */ |
||
39 | protected $wikidataUrlBase = 'https://www.wikidata.org/wiki/'; |
||
40 | |||
41 | private function __construct( $id, $lang, CacheItemPoolInterface $cache ) { |
||
42 | if ( !is_string( $id ) || preg_match( '/[QP][0-9]*/i', $id ) !== 1 ) { |
||
43 | throw new Exception( "Not a valid ID: " . var_export( $id, true ) ); |
||
44 | } |
||
45 | $this->id = $id; |
||
46 | $this->wdApi = new MediawikiApi( 'https://www.wikidata.org/w/api.php' ); |
||
47 | $this->entities = []; |
||
0 ignored issues
–
show
Bug
Best Practice
introduced
by
![]() |
|||
48 | $this->lang = $lang; |
||
49 | $this->cache = $cache; |
||
50 | } |
||
51 | |||
52 | /** |
||
53 | * Create a new Item object with class based on the item's 'instance of' statement. |
||
54 | * |
||
55 | * @param string $id The item ID (Q-number). |
||
56 | * @param string $lang The language code. |
||
57 | * @param CacheItemPoolInterface $cache The cache to use. |
||
58 | * @return Item |
||
59 | */ |
||
60 | public static function factory( $id, $lang, CacheItemPoolInterface $cache ) { |
||
61 | $item = new Item( $id, $lang, $cache ); |
||
62 | foreach ( $item->getPropertyOfTypeItem( self::PROP_INSTANCE_OF ) as $instanceOf ) { |
||
63 | // Try to find a class matching the registered 'instance of'. |
||
64 | foreach ( static::$registeredClasses as $classId => $className ) { |
||
65 | // If this 'instance of' is registered, use it. |
||
66 | if ( $classId === $instanceOf->getItem()->getId() ) { |
||
67 | // This won't re-request the metadata, because that's cached. |
||
68 | return new $className( $id, $lang, $cache ); |
||
69 | } |
||
70 | } |
||
71 | } |
||
72 | |||
73 | // If we're here, just leave it as a basic Item. |
||
74 | $item->setCache( $cache ); |
||
75 | return $item; |
||
76 | } |
||
77 | |||
78 | /** |
||
79 | * Register this class as a candidate for being created by Item::factory. |
||
80 | * Should only be called on subclasses of Item. |
||
81 | * @throws Exception if called on Item or the registering class does not have INSTANCE_OF set. |
||
82 | */ |
||
83 | 2 | public static function register() { |
|
84 | 2 | if ( static::class === self::class ) { |
|
0 ignored issues
–
show
|
|||
85 | 1 | throw new Exception( __METHOD__ . ' should only be called on subclasses of Item' ); |
|
86 | } |
||
87 | 1 | if ( !static::INSTANCE_OF ) { |
|
88 | 1 | throw new Exception( 'Please set INSTANCE_OF for ' . static::class ); |
|
89 | } |
||
90 | static::$registeredClasses[ static::INSTANCE_OF ] = static::class; |
||
91 | } |
||
92 | |||
93 | /** |
||
94 | * @param CacheItemPoolInterface $cache The cache to use. |
||
95 | */ |
||
96 | public function setCache( CacheItemPoolInterface $cache ) { |
||
97 | $this->cache = $cache; |
||
98 | } |
||
99 | |||
100 | /** |
||
101 | * Get the ID (Q-number) of this item. |
||
102 | * @return string|bool The ID or false if it couldn't be determined. |
||
103 | */ |
||
104 | 1 | public function getId() { |
|
105 | 1 | $entity = $this->getEntity( $this->id ); |
|
106 | 1 | return isset( $entity['id'] ) ? $entity['id'] : false; |
|
107 | } |
||
108 | |||
109 | /** |
||
110 | * Get this item's label. |
||
111 | * @return string |
||
112 | */ |
||
113 | 1 | public function getLabel() { |
|
114 | 1 | $entity = $this->getEntity( $this->id ); |
|
115 | 1 | if ( !empty( $entity['labels'][ $this->lang ]['value'] ) ) { |
|
116 | // Use the label if there is one. |
||
117 | 1 | return $entity['labels'][ $this->lang ]['value']; |
|
118 | } |
||
119 | // Or just use the ID. |
||
120 | return $entity['id']; |
||
121 | } |
||
122 | |||
123 | /** |
||
124 | * @return string The Wikidata.org URL for this item. |
||
125 | */ |
||
126 | 1 | public function getWikidataUrl() { |
|
127 | 1 | return $this->wikidataUrlBase . $this->id; |
|
128 | } |
||
129 | |||
130 | /** |
||
131 | * Wikiprojects list their properties like this: |
||
132 | * |
||
133 | * {{List of properties/Header}} |
||
134 | * {{List of properties/Row|id=31|example-subject=Q923767|example-object=Q3331189}} |
||
135 | * </table> |
||
136 | * |
||
137 | * @param string $wikiProject The name of the WikiProject (must exist as a Wikidata page e.g. |
||
138 | * [[Wikidata:$wikiProject]]). |
||
139 | * @param string $type |
||
140 | * @return array |
||
141 | */ |
||
142 | public function getStandardProperties( $wikiProject = 'WikiProject_Books', $type = 'work' ) { |
||
143 | if ( $type !== 'work' ) { |
||
144 | $type = 'edition'; |
||
145 | } |
||
146 | $cacheKey = $type . '_item_property_IDs'; |
||
147 | if ( $this->cache->hasItem( $cacheKey ) ) { |
||
148 | $propIds = $this->cache->getItem( $cacheKey )->get(); |
||
149 | } else { |
||
150 | $domCrawler = new Crawler(); |
||
151 | $wikiProjectUrl = 'https://www.wikidata.org/wiki/Wikidata:' . $wikiProject; |
||
152 | $domCrawler->addHtmlContent( file_get_contents( $wikiProjectUrl ) ); |
||
153 | $propAncors = "//h3/span[@id='" . ucfirst( $type ) . "_item_properties']" |
||
154 | . "/../following-sibling::table[1]//td[2]/a"; |
||
155 | $propCells = $domCrawler->filterXPath( $propAncors ); |
||
156 | $propIds = []; |
||
157 | $propCells->each( function ( Crawler $node, $i ) use ( &$propIds ) { |
||
158 | $propId = $node->text(); |
||
159 | $propIds[] = $propId; |
||
160 | } ); |
||
161 | $cacheItem = $this->cache->getItem( $cacheKey ) |
||
162 | ->expiresAfter( new DateInterval( 'PT1H' ) ) |
||
163 | ->set( $propIds ); |
||
164 | $this->cache->save( $cacheItem ); |
||
165 | } |
||
166 | $workProperties = []; |
||
167 | foreach ( $propIds as $propId ) { |
||
168 | $workProperties[] = self::factory( $propId, $this->lang, $this->cache ); |
||
169 | } |
||
170 | |||
171 | return $workProperties; |
||
172 | } |
||
173 | |||
174 | /** |
||
175 | * @param string $propertyId |
||
176 | * @return bool|Time[] |
||
177 | */ |
||
178 | public function getPropertyOfTypeTime( $propertyId ) { |
||
179 | $times = []; |
||
180 | $entity = $this->getEntity(); |
||
181 | if ( !isset( $entity['claims'][$propertyId] ) ) { |
||
182 | // No statements for this property. |
||
183 | return $times; |
||
184 | } |
||
185 | // print_r($entity['claims'][$propertyId]);exit(); |
||
186 | foreach ( $entity['claims'][$propertyId] as $claim ) { |
||
187 | // print_r($claim); |
||
188 | $times[] = new Time( $claim, $this->lang, $this->cache ); |
||
189 | // |
||
190 | // $timeValue = $claim['datavalue']['value']['time']; |
||
191 | // // Ugly workaround for imprecise dates. :-( |
||
192 | // if (preg_match('/([0-9]{1,4})-00-00/', $timeValue, $matches) === 1) { |
||
193 | // $timeValue = $matches[1]; |
||
194 | // return $timeValue; |
||
195 | // } |
||
196 | // $time = strtotime($timeValue); |
||
197 | // return date($dateFormat, $time); |
||
198 | // } |
||
199 | } |
||
200 | return $times; |
||
201 | } |
||
202 | |||
203 | /** |
||
204 | * Get the Item that is referred to by the specified item's property. |
||
205 | * |
||
206 | * @param string $propertyId |
||
207 | * |
||
208 | * @return \Samwilson\SimpleWikidata\Properties\Item[] |
||
209 | */ |
||
210 | public function getPropertyOfTypeItem( $propertyId ) { |
||
211 | $entity = $this->getEntity( $this->id ); |
||
212 | if ( !isset( $entity['claims'][$propertyId] ) ) { |
||
213 | return []; |
||
214 | } |
||
215 | $items = []; |
||
216 | foreach ( $entity['claims'][$propertyId] as $claim ) { |
||
217 | $items[] = new Properties\Item( $claim, $this->lang, $this->cache ); |
||
218 | } |
||
219 | |||
220 | return $items; |
||
221 | } |
||
222 | |||
223 | /** |
||
224 | * Set a claim to the given Item. |
||
225 | * @param string $property Property ID, with 'P'. |
||
226 | * @param string $itemId Item ID with 'Q'. |
||
227 | */ |
||
228 | public function setPropertyOfTypeItem( $property, $itemId ) { |
||
229 | $itemIdNumeric = substr( $itemId, 1 ); |
||
230 | |||
231 | // First see if this property already exists, and that it is different from what's being set. |
||
232 | $entity = $this->getEntity( $this->id ); |
||
233 | if ( !empty( $entity['claims'][$property] ) ) { |
||
234 | // Get the first claim, and update it if necessary. |
||
235 | $claim = array_shift( $entity['claims'][$property] ); |
||
236 | if ( $claim['mainsnak']['datavalue']['value']['id'] == $itemId ) { |
||
237 | // Already is the required value, no need to change. |
||
238 | return; |
||
239 | } |
||
240 | $claim['mainsnak']['datavalue']['value']['id'] = $itemId; |
||
241 | $claim['mainsnak']['datavalue']['value']['numeric-id'] = $itemIdNumeric; |
||
242 | $apiParams = [ |
||
243 | 'action' => 'wbsetclaim', |
||
244 | 'claim' => json_encode( $claim ), |
||
245 | ]; |
||
246 | } |
||
247 | |||
248 | // If no claim was found (and modified) above, create a new claim. |
||
249 | if ( !isset( $apiParams ) ) { |
||
250 | $apiParams = [ |
||
0 ignored issues
–
show
|
|||
251 | 'action' => 'wbcreateclaim', |
||
252 | 'entity' => $this->getId(), |
||
253 | 'property' => $property, |
||
254 | 'snaktype' => 'value', |
||
255 | 'value' => json_encode( [ 'entity-type' => 'item', 'numeric-id' => $itemIdNumeric ] ), |
||
256 | ]; |
||
257 | } |
||
258 | |||
259 | // @TODO Save the property. |
||
260 | |||
261 | // Clear the cache. |
||
262 | $this->cache->deleteItem( $this->getEntityCacheKey( $this->id ) ); |
||
263 | } |
||
264 | |||
265 | /** |
||
266 | * @param string $entityId |
||
267 | * @param string $propertyId |
||
268 | * @return array|bool |
||
269 | */ |
||
270 | public function getPropertyOfTypeUrl( $entityId, $propertyId ) { |
||
271 | $entity = $this->getEntity( $entityId ); |
||
272 | if ( !isset( $entity['claims'][$propertyId] ) ) { |
||
273 | return false; |
||
274 | } |
||
275 | $urls = []; |
||
276 | foreach ( $entity['claims'][$propertyId] as $claim ) { |
||
277 | $urls[] = $claim['mainsnak']['datavalue']['value']; |
||
278 | } |
||
279 | |||
280 | return $urls; |
||
281 | } |
||
282 | |||
283 | /** |
||
284 | * @param string $entityId |
||
285 | * @param string $propertyId |
||
286 | * @return array|bool |
||
287 | */ |
||
288 | public function getPropertyOfTypeExternalIdentifier( $entityId, $propertyId ) { |
||
289 | $entity = $this->getEntity( $entityId ); |
||
290 | if ( !isset( $entity['claims'][$propertyId] ) ) { |
||
291 | return false; |
||
292 | } |
||
293 | $idents = []; |
||
294 | foreach ( $entity['claims'][$propertyId] as $claim ) { |
||
295 | $qualifiers = []; |
||
296 | if ( !isset( $claim['qualifiers'] ) ) { |
||
297 | continue; |
||
298 | } |
||
299 | foreach ( $claim['qualifiers'] as $qualsInfo ) { |
||
300 | foreach ( $qualsInfo as $qualInfo ) { |
||
301 | $qualProp = self::factory( $qualInfo['property'], $this->lang, $this->cache ); |
||
302 | $propLabel = $qualProp->getLabel(); |
||
303 | if ( !isset( $qualifiers[$propLabel] ) ) { |
||
304 | $qualifiers[$propLabel] = []; |
||
305 | } |
||
306 | $qualifiers[$propLabel][] = $qualInfo['datavalue']['value']; |
||
307 | } |
||
308 | } |
||
309 | $idents[] = [ |
||
310 | 'qualifiers' => $qualifiers, |
||
311 | 'value' => $claim['mainsnak']['datavalue']['value'], |
||
312 | ]; |
||
313 | } |
||
314 | |||
315 | return $idents; |
||
316 | } |
||
317 | |||
318 | /** |
||
319 | * Get a single-valued text property. |
||
320 | * @param string $property One of the PROP_* constants. |
||
321 | * @return string|bool The value, or false if it can't be found. |
||
322 | */ |
||
323 | public function getPropertyOfTypeText( $property ) { |
||
324 | $entity = $this->getEntity( $this->id ); |
||
325 | if ( isset( $entity['claims'][$property] ) ) { |
||
326 | // Use the first title. |
||
327 | foreach ( $entity['claims'][$property] as $t ) { |
||
328 | if ( !isset( $t['mainsnak']['datavalue']['value']['language'] ) ) { |
||
329 | var_dump( $t['mainsnak']['datavalue']['value'] ); |
||
0 ignored issues
–
show
|
|||
330 | exit(); |
||
0 ignored issues
–
show
|
|||
331 | } |
||
332 | if ( $t['mainsnak']['datavalue']['value']['language'] == $this->lang |
||
333 | && !empty( $t['mainsnak']['datavalue']['value']['text'] ) |
||
334 | ) { |
||
335 | return $t['mainsnak']['datavalue']['value']['text']; |
||
336 | } |
||
337 | } |
||
338 | } |
||
339 | return false; |
||
340 | } |
||
341 | |||
342 | /** |
||
343 | * Literal data field for a quantity that relates to some kind of well-defined unit. |
||
344 | * The actual unit goes in the data values that is entered. |
||
345 | * - amount – implicit part of the string (mapping of unit prefix is unclear) |
||
346 | * - unit – implicit part of the string that defaults to "1" (mapping to standardizing |
||
347 | * body is unclear) |
||
348 | * - upperbound - quantity's upper bound |
||
349 | * - lowerbound - quantity's lower bound |
||
350 | * @param string $property |
||
351 | * @return mixed[]|bool If it's not false it's an array with 'amount', 'unit', etc. |
||
352 | */ |
||
353 | public function getPropertyOfTypeQuantity( $property ) { |
||
354 | $quantities = []; |
||
355 | $entity = $this->getEntity( $this->id ); |
||
356 | if ( !isset( $entity['claims'][$property] ) ) { |
||
357 | return false; |
||
358 | } |
||
359 | foreach ( $entity['claims'][$property] as $t ) { |
||
360 | $quantity = $t['mainsnak']['datavalue']['value']; |
||
361 | $unitId = substr( $quantity['unit'], strlen( $this->wikidataUrlBase ) + 1 ); |
||
362 | $quantity['unit'] = self::factory( $unitId, $this->lang, $this->cache ); |
||
363 | $quantities[] = $quantity; |
||
364 | } |
||
365 | return $quantities; |
||
366 | } |
||
367 | |||
368 | /** |
||
369 | * Set a single-valued text property. |
||
370 | * @param string $property One of the PROP_* constants. |
||
371 | * @param string $value The value. |
||
372 | */ |
||
373 | public function setPropertyOfTypeText( $property, $value ) { |
||
374 | // First see if this property already exists, and that it is different from what's being set. |
||
375 | $entity = $this->getEntity( $this->id ); |
||
376 | if ( !empty( $entity['claims'][$property] ) ) { |
||
377 | // Find this language's claim (if there is one). |
||
378 | foreach ( $entity['claims'][$property] as $claim ) { |
||
379 | if ( $claim['mainsnak']['datavalue']['value']['language'] == $this->lang ) { |
||
380 | // Modify this claim's text value. |
||
381 | $titleClaim = $claim; |
||
382 | $titleClaim['mainsnak']['datavalue']['value']['text'] = $value; |
||
383 | $setTitleParams = [ |
||
384 | 'action' => 'wbsetclaim', |
||
385 | 'claim' => \GuzzleHttp\json_encode( $titleClaim ), |
||
386 | ]; |
||
387 | continue; |
||
388 | } |
||
389 | } |
||
390 | } |
||
391 | |||
392 | // If no claim was found (and modified) above, create a new claim. |
||
393 | if ( !isset( $setTitleParams ) ) { |
||
394 | $setTitleParams = [ |
||
395 | 'action' => 'wbcreateclaim', |
||
396 | 'entity' => $this->getId(), |
||
397 | 'property' => $property, |
||
398 | 'snaktype' => 'value', |
||
399 | 'value' => \GuzzleHttp\json_encode( [ 'text' => $value, 'language' => $this->lang ] ), |
||
400 | ]; |
||
401 | } |
||
402 | |||
403 | // Save the property. |
||
404 | $wdWpOauth = new WdWpOauth(); |
||
0 ignored issues
–
show
The type
Samwilson\SimpleWikidata\WdWpOauth was not found. Maybe you did not declare it correctly or list all dependencies?
The issue could also be caused by a filter entry in the build configuration.
If the path has been excluded in your configuration, e.g. filter:
dependency_paths: ["lib/*"]
For further information see https://scrutinizer-ci.com/docs/tools/php/php-scrutinizer/#list-dependency-paths ![]() |
|||
405 | $wdWpOauth->makeCall( $setTitleParams, true ); |
||
406 | |||
407 | // Clear the cache. |
||
408 | $this->cache->deleteItem( $this->getEntityCacheKey( $this->id ) ); |
||
409 | } |
||
410 | |||
411 | /** |
||
412 | * Does this item exist? |
||
413 | * @return bool |
||
414 | */ |
||
415 | public function exists() { |
||
416 | return $this->getId() !== false; |
||
417 | } |
||
418 | |||
419 | /** |
||
420 | * @return array With 'html' and 'title' keys. |
||
421 | */ |
||
422 | public function getWikipediaIntro() { |
||
423 | $cacheKey = 'wikipedia-intro-' . $this->id . $this->lang; |
||
424 | if ( $this->cache->hasItem( $cacheKey ) ) { |
||
425 | return $this->cache->getItem( $cacheKey )->get(); |
||
426 | } |
||
427 | $entity = $this->getEntity( $this->id ); |
||
428 | if ( !isset( $entity['sitelinks'] ) ) { |
||
429 | return []; |
||
430 | } |
||
431 | foreach ( $entity['sitelinks'] as $sitelink ) { |
||
432 | if ( $sitelink['site'] == $this->lang . 'wiki' ) { |
||
433 | $api = new MediawikiApi( 'https://' . $this->lang . '.wikipedia.org/w/api.php' ); |
||
434 | $req = new SimpleRequest( 'query', [ |
||
435 | 'prop' => 'extracts', |
||
436 | 'exintro' => true, |
||
437 | 'titles' => $sitelink['title'], |
||
438 | ] ); |
||
439 | $response = $api->getRequest( $req ); |
||
440 | $page = array_shift( $response['query']['pages'] ); |
||
441 | $out = [ |
||
442 | 'title' => $page['title'], |
||
443 | 'html' => $page['extract'], |
||
444 | ]; |
||
445 | $cacheItem = $this->cache->getItem( $cacheKey ) |
||
446 | ->expiresAfter( new DateInterval( 'P1D' ) ) |
||
447 | ->set( $out ); |
||
448 | $this->cache->save( $cacheItem ); |
||
449 | |||
450 | return $out; |
||
451 | } |
||
452 | } |
||
453 | |||
454 | return []; |
||
455 | } |
||
456 | |||
457 | /** |
||
458 | * Get the raw entity data from the 'wbgetentities' API call. |
||
459 | * @param string|null $id The Q-number. |
||
460 | * @param bool $ignoreCache |
||
461 | * @return array|bool |
||
462 | */ |
||
463 | public function getEntity( $id = null, $ignoreCache = false ) { |
||
464 | $idActual = $id ?: $this->id; |
||
465 | $cacheKey = $this->getEntityCacheKey( $idActual ); |
||
466 | if ( !$ignoreCache && $this->cache->hasItem( $cacheKey ) ) { |
||
467 | return $this->cache->getItem( $cacheKey )->get(); |
||
468 | } |
||
469 | $metadataRequest = new SimpleRequest( 'wbgetentities', [ 'ids' => $idActual ] ); |
||
470 | $itemResult = $this->wdApi->getRequest( $metadataRequest ); |
||
471 | if ( !isset( $itemResult['success'] ) || !isset( $itemResult['entities'][$id] ) ) { |
||
472 | return false; |
||
473 | } |
||
474 | $metadata = $itemResult['entities'][$idActual]; |
||
475 | $cacheItem = $this->cache->getItem( $cacheKey ) |
||
476 | ->expiresAfter( new DateInterval( 'PT10M' ) ) |
||
477 | ->set( $metadata ); |
||
478 | $this->cache->save( $cacheItem ); |
||
479 | return $metadata; |
||
480 | } |
||
481 | |||
482 | /** |
||
483 | * @param string $id |
||
484 | * |
||
485 | * @return string |
||
486 | */ |
||
487 | protected function getEntityCacheKey( $id ) { |
||
488 | return 'entities' . $id; |
||
489 | } |
||
490 | } |
||
491 |