Total Complexity | 111 |
Total Lines | 592 |
Duplicated Lines | 0 % |
Changes | 3 | ||
Bugs | 1 | Features | 0 |
Complex classes like Model often do a lot of different things. To break such a class down, we need to identify a cohesive component within that class. A common approach to find such a component is to look for fields/methods that share the same prefixes, or suffixes.
Once you have determined the fields that belong together, you can apply the Extract Class refactoring. If the component makes sense as a sub-class, Extract Subclass is also a candidate, and is often faster.
While breaking up the class, it is a good idea to analyze how other classes use Model, and based on these observations, apply Extract Interface, too.
1 | <?php |
||
8 | class Model |
||
9 | { |
||
10 | /** cache for Vocabulary objects */ |
||
11 | private $allVocabularies = null; |
||
12 | /** cache for Vocabulary objects */ |
||
13 | private $vocabsByGraph = null; |
||
14 | /** cache for Vocabulary objects */ |
||
15 | private $vocabsByUriSpace = null; |
||
16 | /** how long to store retrieved URI information in APC cache */ |
||
17 | const URI_FETCH_TTL = 86400; // 1 day |
||
18 | private $globalConfig; |
||
19 | private $logger; |
||
20 | private $resolver; |
||
21 | |||
22 | /** |
||
23 | * Initializes the Model object |
||
24 | */ |
||
25 | public function __construct($config) |
||
26 | { |
||
27 | $this->globalConfig = $config; |
||
28 | $this->initializeLogging(); |
||
29 | $this->resolver = new Resolver($this); |
||
30 | } |
||
31 | |||
32 | /** |
||
33 | * Returns the GlobalConfig object given to the Model as a constructor parameter. |
||
34 | * @return GlobalConfig |
||
35 | */ |
||
36 | public function getConfig() { |
||
38 | } |
||
39 | |||
40 | /** |
||
41 | * Configures the logging facility |
||
42 | */ |
||
43 | private function initializeLogging() { |
||
61 | } |
||
62 | } |
||
63 | |||
64 | /** |
||
65 | * Return the logging facility |
||
66 | * @return object logger |
||
67 | */ |
||
68 | public function getLogger() { |
||
70 | } |
||
71 | |||
72 | /** |
||
73 | * Return the version of this Skosmos installation, or "unknown" if |
||
74 | * it cannot be determined. The version information is based on Git tags. |
||
75 | * @return string version |
||
76 | */ |
||
77 | public function getVersion() : string |
||
78 | { |
||
79 | $ver = \Composer\InstalledVersions::getRootPackage()['pretty_version']; |
||
80 | if ($ver === null) { |
||
81 | return "unknown"; |
||
82 | } |
||
83 | |||
84 | return $ver; |
||
85 | } |
||
86 | |||
87 | /** |
||
88 | * Return all the vocabularies available. |
||
89 | * @param boolean $categories whether you want everything included in a subarray of |
||
90 | * a category. |
||
91 | * @param boolean $shortname set to true if you want the vocabularies sorted by |
||
92 | * their shortnames instead of their titles. |
||
93 | */ |
||
94 | public function getVocabularyList($categories = true, $shortname = false) |
||
95 | { |
||
96 | $cats = $this->getVocabularyCategories(); |
||
97 | $ret = array(); |
||
98 | foreach ($cats as $cat) { |
||
99 | $catlabel = $cat->getTitle(); |
||
100 | |||
101 | // find all the vocabs in this category |
||
102 | $vocs = array(); |
||
103 | foreach ($cat->getVocabularies() as $voc) { |
||
104 | $vocs[$shortname ? $voc->getConfig()->getShortname() : $voc->getConfig()->getTitle()] = $voc; |
||
105 | } |
||
106 | uksort($vocs, 'strcoll'); |
||
107 | |||
108 | if (sizeof($vocs) > 0 && $categories) { |
||
109 | $ret[$catlabel] = $vocs; |
||
110 | } elseif (sizeof($vocs) > 0) { |
||
111 | $ret = array_merge($ret, $vocs); |
||
112 | } |
||
113 | |||
114 | } |
||
115 | |||
116 | if (!$categories) { |
||
117 | uksort($ret, 'strcoll'); |
||
118 | } |
||
119 | |||
120 | return $ret; |
||
121 | } |
||
122 | |||
123 | /** |
||
124 | * Return all types (RDFS/OWL classes) present in the specified vocabulary or all vocabularies. |
||
125 | * @return array Array with URIs (string) as key and array of (label, superclassURI) as value |
||
126 | */ |
||
127 | public function getTypes($vocid = null, $lang = null) |
||
128 | { |
||
129 | $sparql = (isset($vocid)) ? $this->getVocabulary($vocid)->getSparql() : $this->getDefaultSparql(); |
||
130 | $result = $sparql->queryTypes($lang); |
||
131 | |||
132 | foreach ($result as $uri => $values) { |
||
133 | if (empty($values)) { |
||
134 | $shorteneduri = EasyRdf\RdfNamespace::shorten($uri); |
||
135 | if ($shorteneduri !== null) { |
||
136 | $trans = gettext($shorteneduri); |
||
137 | if ($trans) { |
||
138 | $result[$uri] = array('label' => $trans); |
||
139 | } |
||
140 | } |
||
141 | } |
||
142 | } |
||
143 | |||
144 | return $result; |
||
145 | } |
||
146 | |||
147 | /** |
||
148 | * Return the languages present in the configured vocabularies. |
||
149 | * @return array Array with lang codes (string) |
||
150 | */ |
||
151 | public function getLanguages($lang) |
||
152 | { |
||
153 | $vocabs = $this->getVocabularyList(false); |
||
154 | $ret = array(); |
||
155 | foreach ($vocabs as $vocab) { |
||
156 | foreach ($vocab->getConfig()->getLanguages() as $langcode) { |
||
157 | $langlit = Punic\Language::getName($langcode, $lang); |
||
158 | $ret[$langlit] = $langcode; |
||
159 | } |
||
160 | } |
||
161 | ksort($ret); |
||
162 | return array_unique($ret); |
||
163 | } |
||
164 | |||
165 | /** |
||
166 | * returns a concept's RDF data in downloadable format |
||
167 | * @param string $vocid vocabulary id, or null for global data (from all vocabularies) |
||
168 | * @param string $uri concept URI |
||
169 | * @param string $format the format in which you want to get the result, currently this function supports |
||
170 | * text/turtle, application/rdf+xml and application/json |
||
171 | * @return string RDF data in the requested serialization |
||
172 | */ |
||
173 | public function getRDF($vocid, $uri, $format) |
||
174 | { |
||
175 | |||
176 | if ($format == 'text/turtle') { |
||
177 | $retform = 'turtle'; |
||
178 | $serialiser = new EasyRdf\Serialiser\Turtle(); |
||
179 | } elseif ($format == 'application/ld+json' || $format == 'application/json') { |
||
180 | $retform = 'jsonld'; // serve JSON-LD for both JSON-LD and plain JSON requests |
||
181 | $serialiser = new EasyRdf\Serialiser\JsonLd(); |
||
182 | } else { |
||
183 | $retform = 'rdfxml'; |
||
184 | $serialiser = new EasyRdf\Serialiser\RdfXml(); |
||
185 | } |
||
186 | |||
187 | if ($vocid !== null) { |
||
188 | $vocab = $this->getVocabulary($vocid); |
||
189 | $sparql = $vocab->getSparql(); |
||
190 | $arrayClass = $vocab->getConfig()->getArrayClassURI(); |
||
191 | $vocabs = array($vocab); |
||
192 | } else { |
||
193 | $sparql = $this->getDefaultSparql(); |
||
194 | $arrayClass = null; |
||
195 | $vocabs = null; |
||
196 | } |
||
197 | $result = $sparql->queryConceptInfoGraph($uri, $arrayClass, $vocabs); |
||
198 | |||
199 | if (!$result->isEmpty()) { |
||
200 | return $serialiser->serialise($result, $retform); |
||
201 | } |
||
202 | return ""; |
||
203 | } |
||
204 | |||
205 | /** |
||
206 | * Makes a SPARQL-query to the endpoint that retrieves concept |
||
207 | * references as it's search results. |
||
208 | * @param ConceptSearchParameters $params an object that contains all the parameters needed for the search |
||
209 | * @return array search results |
||
210 | */ |
||
211 | public function searchConcepts($params) |
||
212 | { |
||
213 | // don't even try to search for empty prefix if no other search criteria (group or parent concept) has been set |
||
214 | if (($params->getSearchTerm() === "" || !preg_match('/[^*]/', $params->getSearchTerm())) && !$params->getGroupLimit() && !$params->getParentLimit()) { |
||
215 | return array(); |
||
216 | } |
||
217 | |||
218 | $vocabs = $params->getVocabs(); |
||
219 | $showDeprecated=false; |
||
220 | if (sizeof($vocabs) === 1) { // search within vocabulary |
||
221 | $voc = $vocabs[0]; |
||
222 | $sparql = $voc->getSparql(); |
||
223 | $showDeprecated=$voc->getConfig()->getShowDeprecated(); |
||
224 | } else { // multi-vocabulary or global search |
||
225 | $voc = null; |
||
226 | $sparql = $this->getDefaultSparql(); |
||
227 | // @TODO : in a global search showDeprecated will always be false and cannot be set globally |
||
228 | } |
||
229 | |||
230 | $results = $sparql->queryConcepts($vocabs, $params->getAdditionalFields(), $params->getUnique(), $params,$showDeprecated); |
||
231 | if ($params->getRest() && $results && $params->getSearchLimit() !== 0) { |
||
232 | $results = array_slice($results, $params->getOffset(), $params->getSearchLimit()); |
||
233 | } |
||
234 | $ret = array(); |
||
235 | |||
236 | foreach ($results as $hit) { |
||
237 | if (sizeof($vocabs) == 1) { |
||
238 | $hitvoc = $voc; |
||
239 | $hit['vocab'] = $vocabs[0]->getId(); |
||
240 | } else { |
||
241 | try { |
||
242 | $hitvoc = $this->getVocabularyByGraph($hit['graph']); |
||
243 | $hit['vocab'] = $hitvoc->getId(); |
||
244 | } catch (ValueError $e) { |
||
245 | trigger_error($e->getMessage(), E_USER_WARNING); |
||
246 | $hitvoc = null; |
||
247 | $hit['vocab'] = "???"; |
||
248 | } |
||
249 | } |
||
250 | unset($hit['graph']); |
||
251 | |||
252 | $hit['voc'] = $hitvoc; |
||
253 | |||
254 | if ($hitvoc === null || !$hitvoc->containsURI($hit['uri'])) { |
||
255 | // if uri is a external vocab uri that is included in the current vocab |
||
256 | $realvoc = $this->guessVocabularyFromURI($hit['uri'], $voc !== null ? $voc->getId() : null); |
||
257 | if ($realvoc !== $hitvoc) { |
||
258 | unset($hit['localname']); |
||
259 | $hit['exvocab'] = ($realvoc !== null) ? $realvoc->getId() : "???"; |
||
260 | } |
||
261 | } |
||
262 | |||
263 | $ret[] = $hit; |
||
264 | } |
||
265 | |||
266 | return $ret; |
||
267 | } |
||
268 | |||
269 | /** |
||
270 | * Function for performing a search for concepts and their data fields. |
||
271 | * @param ConceptSearchParameters $params an object that contains all the parameters needed for the search |
||
272 | * @return array array with keys 'count' and 'results' |
||
273 | */ |
||
274 | public function searchConceptsAndInfo($params) |
||
275 | { |
||
276 | $params->setUnique(true); |
||
277 | $allhits = $this->searchConcepts($params); |
||
278 | $count = sizeof($allhits); |
||
279 | $hits = array_slice($allhits, intval($params->getOffset()), $params->getSearchLimit()); |
||
280 | |||
281 | $ret = array(); |
||
282 | $uris = array(); |
||
283 | $vocabs = array(); |
||
284 | $uniqueVocabs = array(); |
||
285 | foreach ($hits as $hit) { |
||
286 | $uniqueVocabs[$hit['voc']->getId()] = $hit['voc']->getId(); |
||
287 | $vocabs[] = $hit['voc']; |
||
288 | $uris[] = $hit['uri']; |
||
289 | } |
||
290 | if (sizeof($uniqueVocabs) == 1) { |
||
291 | $voc = $vocabs[0]; |
||
292 | $sparql = $voc->getSparql(); |
||
293 | $arrayClass = $voc->getConfig()->getArrayClassURI(); |
||
294 | } else { |
||
295 | $arrayClass = null; |
||
296 | $sparql = $this->getDefaultSparql(); |
||
297 | } |
||
298 | if (sizeof($uris) > 0) { |
||
299 | $ret = $sparql->queryConceptInfo($uris, $arrayClass, $vocabs, $params->getSearchLang()); |
||
300 | } |
||
301 | |||
302 | // For marking that the concept has been found through an alternative label, hidden |
||
303 | // label or a label in another language |
||
304 | foreach ($hits as $idx => $hit) { |
||
305 | if (isset($hit['altLabel']) && isset($ret[$idx])) { |
||
306 | $ret[$idx]->setFoundBy($hit['altLabel'], 'alt'); |
||
307 | } |
||
308 | |||
309 | if (isset($hit['hiddenLabel']) && isset($ret[$idx])) { |
||
310 | $ret[$idx]->setFoundBy($hit['hiddenLabel'], 'hidden'); |
||
311 | } |
||
312 | |||
313 | if (isset($hit['matchedPrefLabel'])) { |
||
314 | $ret[$idx]->setFoundBy($hit['matchedPrefLabel'], 'lang'); |
||
315 | } |
||
316 | |||
317 | if ($ret[$idx] && isset($hit['lang'])) { |
||
318 | $ret[$idx]->setContentLang($hit['lang']); |
||
319 | } |
||
320 | } |
||
321 | |||
322 | return array('count' => $count, 'results' => $ret); |
||
323 | } |
||
324 | |||
325 | /** |
||
326 | * Creates dataobjects from an input array. |
||
327 | * @param string $class the type of class eg. 'Vocabulary'. |
||
328 | * @param array $resarr contains the EasyRdf\Resources. |
||
329 | */ |
||
330 | private function createDataObjects($class, $resarr) |
||
331 | { |
||
332 | $ret = array(); |
||
333 | foreach ($resarr as $res) { |
||
334 | $ret[] = new $class($this, $res); |
||
335 | } |
||
336 | |||
337 | return $ret; |
||
338 | } |
||
339 | |||
340 | /** |
||
341 | * Returns the cached vocabularies. |
||
342 | * @return array of Vocabulary dataobjects |
||
343 | */ |
||
344 | public function getVocabularies() |
||
345 | { |
||
346 | if ($this->allVocabularies === null) { // initialize cache |
||
347 | $vocs = $this->globalConfig->getGraph()->allOfType('skosmos:Vocabulary'); |
||
348 | $this->allVocabularies = $this->createDataObjects("Vocabulary", $vocs); |
||
349 | foreach ($this->allVocabularies as $voc) { |
||
350 | // register vocabulary ids as RDF namespace prefixes |
||
351 | $prefix = preg_replace('/\W+/', '', $voc->getId()); // strip non-word characters |
||
352 | try { |
||
353 | if ($prefix != '' && EasyRdf\RdfNamespace::get($prefix) === null) // if not already defined |
||
354 | { |
||
355 | EasyRdf\RdfNamespace::set($prefix, $voc->getUriSpace()); |
||
356 | } |
||
357 | |||
358 | } catch (Exception $e) { |
||
359 | // not valid as namespace identifier, ignore |
||
360 | } |
||
361 | } |
||
362 | } |
||
363 | |||
364 | return $this->allVocabularies; |
||
365 | } |
||
366 | |||
367 | /** |
||
368 | * Returns the cached vocabularies from a category. |
||
369 | * @param EasyRdf\Resource $cat the category in question |
||
370 | * @return array of vocabulary dataobjects |
||
371 | */ |
||
372 | public function getVocabulariesInCategory($cat) |
||
373 | { |
||
374 | $vocs = $this->globalConfig->getGraph()->resourcesMatching('dc:subject', $cat); |
||
375 | return $this->createDataObjects("Vocabulary", $vocs); |
||
376 | } |
||
377 | |||
378 | /** |
||
379 | * Creates dataobjects of all the different vocabulary categories (Health etc.). |
||
380 | * @return array of Dataobjects of the type VocabularyCategory. |
||
381 | */ |
||
382 | public function getVocabularyCategories() |
||
383 | { |
||
384 | $cats = $this->globalConfig->getGraph()->allOfType('skos:Concept'); |
||
385 | if(empty($cats)) { |
||
386 | return array(new VocabularyCategory($this, null)); |
||
387 | } |
||
388 | |||
389 | return $this->createDataObjects("VocabularyCategory", $cats); |
||
390 | } |
||
391 | |||
392 | /** |
||
393 | * Returns the label defined in config.ttl with the appropriate language. |
||
394 | * @param string $lang language code of returned labels, eg. 'fi' |
||
395 | * @return string the label for vocabulary categories. |
||
396 | */ |
||
397 | public function getClassificationLabel($lang) |
||
398 | { |
||
399 | $cats = $this->globalConfig->getGraph()->allOfType('skos:ConceptScheme'); |
||
400 | return $cats ? $cats[0]->label($lang) : null; |
||
401 | } |
||
402 | |||
403 | /** |
||
404 | * Returns a single cached vocabulary. |
||
405 | * @param string $vocid the vocabulary id eg. 'mesh'. |
||
406 | * @return Vocabulary dataobject |
||
407 | */ |
||
408 | public function getVocabulary($vocid): Vocabulary |
||
409 | { |
||
410 | $vocs = $this->getVocabularies(); |
||
411 | foreach ($vocs as $voc) { |
||
412 | if ($voc->getId() == $vocid) { |
||
413 | return $voc; |
||
414 | } |
||
415 | } |
||
416 | throw new ValueError("Vocabulary id '$vocid' not found in configuration."); |
||
417 | } |
||
418 | |||
419 | /** |
||
420 | * Return the vocabulary that is stored in the given graph on the given endpoint. |
||
421 | * |
||
422 | * @param $graph string graph URI |
||
423 | * @param $endpoint string endpoint URL (default SPARQL endpoint if omitted) |
||
424 | * @return Vocabulary vocabulary of this URI, or null if not found |
||
425 | */ |
||
426 | public function getVocabularyByGraph($graph, $endpoint = null) |
||
444 | } |
||
445 | |||
446 | } |
||
447 | |||
448 | /** |
||
449 | * When multiple vocabularies share same URI namespace, return the |
||
450 | * vocabulary in which the URI is actually defined (has a label). |
||
451 | * |
||
452 | * @param Vocabulary[] $vocabs vocabularies to search |
||
453 | * @param string $uri URI to look for |
||
454 | * @param $preferredVocabId string ID of the preferred vocabulary to return if more than one is found |
||
455 | * @return Vocabulary the vocabulary with the URI |
||
456 | */ |
||
457 | |||
458 | private function disambiguateVocabulary($vocabs, $uri, $preferredVocabId = null) |
||
503 | } |
||
504 | |||
505 | /** |
||
506 | * Guess which vocabulary a URI originates from, based on the declared |
||
507 | * vocabulary URI spaces. |
||
508 | * |
||
509 | * @param $uri string URI to search |
||
510 | * @param $preferredVocabId string ID of the preferred vocabulary to return if more than one is found |
||
511 | * @return Vocabulary vocabulary of this URI, or null if not found |
||
512 | */ |
||
513 | public function guessVocabularyFromURI($uri, $preferredVocabId = null) |
||
514 | { |
||
515 | if ($this->vocabsByUriSpace === null) { // initialize cache |
||
516 | $this->vocabsByUriSpace = array(); |
||
517 | foreach ($this->getVocabularies() as $voc) { |
||
518 | $uriSpace = $voc->getUriSpace(); |
||
519 | if ($uriSpace) { |
||
520 | $this->vocabsByUriSpace[$uriSpace][] = $voc; |
||
521 | } |
||
522 | } |
||
523 | } |
||
524 | |||
525 | // try to guess the URI space and look it up in the cache |
||
526 | $res = new EasyRdf\Resource($uri); |
||
527 | $namespace = substr(strval($uri), 0, -strlen(strval($res->localName()))); |
||
528 | if ($namespace && array_key_exists($namespace, $this->vocabsByUriSpace)) { |
||
529 | $vocabs = $this->vocabsByUriSpace[$namespace]; |
||
530 | return $this->disambiguateVocabulary($vocabs, $uri, $preferredVocabId); |
||
531 | } |
||
532 | |||
533 | // didn't work, try to match with each URI space separately |
||
534 | foreach ($this->vocabsByUriSpace as $urispace => $vocabs) { |
||
535 | if (strpos($uri, $urispace) === 0) { |
||
536 | return $this->disambiguateVocabulary($vocabs, $uri, $preferredVocabId); |
||
537 | } |
||
538 | } |
||
539 | |||
540 | // not found |
||
541 | return null; |
||
542 | } |
||
543 | |||
544 | /** |
||
545 | * Get the label for a resource, preferring 1. the given language 2. configured languages 3. any language. |
||
546 | * @param EasyRdf\Resource $res resource whose label to return |
||
547 | * @param string $lang preferred language |
||
548 | * @return EasyRdf\Literal label as an EasyRdf\Literal object, or null if not found |
||
549 | */ |
||
550 | public function getResourceLabel($res, $lang) |
||
551 | { |
||
552 | $langs = array_merge(array($lang), array_keys($this->getConfig()->getLanguages())); |
||
553 | foreach ($langs as $l) { |
||
554 | $label = $res->label($l); |
||
555 | if ($label !== null) { |
||
556 | return $label; |
||
557 | } |
||
558 | |||
559 | } |
||
560 | return $res->label(); // desperate check for label in any language; will return null if even this fails |
||
561 | } |
||
562 | |||
563 | public function getResourceFromUri($uri) |
||
564 | { |
||
565 | // using apc cache for the resource if available |
||
566 | if ($this->globalConfig->getCache()->isAvailable()) { |
||
567 | // @codeCoverageIgnoreStart |
||
568 | $key = 'fetch: ' . $uri; |
||
569 | $resource = $this->globalConfig->getCache()->fetch($key); |
||
570 | if ($resource === null || $resource === false) { // was not found in cache, or previous request failed |
||
571 | $resource = $this->resolver->resolve($uri, $this->getConfig()->getHttpTimeout()); |
||
572 | $this->globalConfig->getCache()->store($key, $resource, self::URI_FETCH_TTL); |
||
573 | } |
||
574 | // @codeCoverageIgnoreEnd |
||
575 | } else { // APC not available, parse on every request |
||
576 | $resource = $this->resolver->resolve($uri, $this->getConfig()->getHttpTimeout()); |
||
577 | } |
||
578 | return $resource; |
||
579 | } |
||
580 | |||
581 | /** |
||
582 | * Returns a SPARQL endpoint object. |
||
583 | * @param string $dialect eg. 'JenaText'. |
||
584 | * @param string $endpoint url address of endpoint |
||
585 | * @param string|null $graph uri for the target graph. |
||
586 | */ |
||
587 | public function getSparqlImplementation($dialect, $endpoint, $graph) |
||
588 | { |
||
589 | $classname = $dialect . "Sparql"; |
||
590 | |||
591 | return new $classname($endpoint, $graph, $this); |
||
592 | } |
||
593 | |||
594 | /** |
||
595 | * Returns a SPARQL endpoint object using the default implementation set in the config.ttl. |
||
596 | */ |
||
597 | public function getDefaultSparql() |
||
600 | } |
||
601 | |||
602 | } |
||
603 |