Duplicate code is one of the most pungent code smells. A rule that is often used is to re-structure code once it is duplicated in three or more places.
Common duplication problems, and corresponding solutions are:
Complex classes like DocumentParser 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. You can also have a look at the cohesion graph to spot any un-connected, or weakly-connected components.
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 DocumentParser, and based on these observations, apply Extract Interface, too.
1 | <?php |
||
28 | class DocumentParser |
||
29 | { |
||
30 | const PROPERTY_ANNOTATION = 'ONGR\ElasticsearchBundle\Annotation\Property'; |
||
31 | const EMBEDDED_ANNOTATION = 'ONGR\ElasticsearchBundle\Annotation\Embedded'; |
||
32 | const DOCUMENT_ANNOTATION = 'ONGR\ElasticsearchBundle\Annotation\Document'; |
||
33 | const OBJECT_ANNOTATION = 'ONGR\ElasticsearchBundle\Annotation\Object'; |
||
34 | const NESTED_ANNOTATION = 'ONGR\ElasticsearchBundle\Annotation\Nested'; |
||
35 | |||
36 | // Meta fields |
||
37 | const ID_ANNOTATION = 'ONGR\ElasticsearchBundle\Annotation\Id'; |
||
38 | const PARENT_ANNOTATION = 'ONGR\ElasticsearchBundle\Annotation\ParentDocument'; |
||
39 | const ROUTING_ANNOTATION = 'ONGR\ElasticsearchBundle\Annotation\Routing'; |
||
40 | const VERSION_ANNOTATION = 'ONGR\ElasticsearchBundle\Annotation\Version'; |
||
41 | const HASH_MAP_ANNOTATION = 'ONGR\ElasticsearchBundle\Annotation\HashMap'; |
||
42 | |||
43 | /** |
||
44 | * @var Reader Used to read document annotations. |
||
45 | */ |
||
46 | private $reader; |
||
47 | |||
48 | /** |
||
49 | * @var DocumentFinder Used to find documents. |
||
50 | */ |
||
51 | private $finder; |
||
52 | |||
53 | /** |
||
54 | * @var array Contains gathered objects which later adds to documents. |
||
55 | */ |
||
56 | private $objects = []; |
||
57 | |||
58 | /** |
||
59 | * @var array Document properties aliases. |
||
60 | */ |
||
61 | private $aliases = []; |
||
62 | |||
63 | /** |
||
64 | * @var array Local cache for document properties. |
||
65 | */ |
||
66 | private $properties = []; |
||
67 | |||
68 | /** |
||
69 | * @var array Local cache for documents |
||
70 | */ |
||
71 | private $documents = []; |
||
72 | |||
73 | /** |
||
74 | * @param Reader $reader Used for reading annotations. |
||
75 | * @param DocumentFinder $finder Used for resolving namespaces. |
||
76 | */ |
||
77 | public function __construct(Reader $reader, DocumentFinder $finder) |
||
78 | { |
||
79 | $this->reader = $reader; |
||
80 | $this->finder = $finder; |
||
81 | $this->registerAnnotations(); |
||
82 | } |
||
83 | |||
84 | /** |
||
85 | * Parses documents by used annotations and returns mapping for elasticsearch with some extra metadata. |
||
86 | * |
||
87 | * @param \ReflectionClass $class |
||
88 | * |
||
89 | * @return array|null |
||
90 | * @throws MissingDocumentAnnotationException |
||
91 | */ |
||
92 | public function parse(\ReflectionClass $class) |
||
93 | { |
||
94 | $className = $class->getName(); |
||
95 | |||
96 | if ($class->isTrait()) { |
||
97 | return false; |
||
98 | } |
||
99 | |||
100 | if (!isset($this->documents[$className])) { |
||
101 | /** @var Document $document */ |
||
102 | $document = $this->reader->getClassAnnotation($class, self::DOCUMENT_ANNOTATION); |
||
103 | |||
104 | if ($document === null) { |
||
105 | throw new MissingDocumentAnnotationException( |
||
106 | sprintf( |
||
107 | '"%s" class cannot be parsed as document because @Document annotation is missing.', |
||
108 | $class->getName() |
||
109 | ) |
||
110 | ); |
||
111 | } |
||
112 | |||
113 | $fields = []; |
||
114 | $aliases = $this->getAliases($class, $fields) |
||
115 | |||
116 | $this->documents[$className] = [ |
||
|
|||
117 | 'type' => $document->type ?: Caser::snake($class->getShortName()), |
||
118 | 'properties' => $this->getProperties($class), |
||
119 | 'fields' => array_filter( |
||
120 | array_merge( |
||
121 | $document->dump(), |
||
122 | $fields |
||
123 | ) |
||
124 | ), |
||
125 | 'aliases' => $aliases, |
||
126 | 'analyzers' => $this->getAnalyzers($class), |
||
127 | 'objects' => $this->getObjects(), |
||
128 | 'namespace' => $class->getName(), |
||
129 | 'class' => $class->getShortName(), |
||
130 | ]; |
||
131 | } |
||
132 | return $this->documents[$className]; |
||
133 | } |
||
134 | |||
135 | /** |
||
136 | * Returns document annotation data from reader. |
||
137 | * |
||
138 | * @param \ReflectionClass $document |
||
139 | * |
||
140 | * @return Document|object|null |
||
141 | */ |
||
142 | private function getDocumentAnnotationData($document) |
||
143 | { |
||
144 | return $this->reader->getClassAnnotation($document, self::DOCUMENT_ANNOTATION); |
||
145 | } |
||
146 | |||
147 | /** |
||
148 | * Returns property annotation data from reader. |
||
149 | * |
||
150 | * @param \ReflectionProperty $property |
||
151 | * |
||
152 | * @return Property|object|null |
||
153 | */ |
||
154 | private function getPropertyAnnotationData(\ReflectionProperty $property) |
||
155 | { |
||
156 | $result = $this->reader->getPropertyAnnotation($property, self::PROPERTY_ANNOTATION); |
||
157 | |||
158 | if ($result !== null && $result->name === null) { |
||
159 | $result->name = Caser::snake($property->getName()); |
||
160 | } |
||
161 | |||
162 | return $result; |
||
163 | } |
||
164 | |||
165 | /** |
||
166 | * Returns Embedded annotation data from reader. |
||
167 | * |
||
168 | * @param \ReflectionProperty $property |
||
169 | * |
||
170 | * @return Embedded|object|null |
||
171 | */ |
||
172 | private function getEmbeddedAnnotationData(\ReflectionProperty $property) |
||
173 | { |
||
174 | $result = $this->reader->getPropertyAnnotation($property, self::EMBEDDED_ANNOTATION); |
||
175 | |||
176 | if ($result !== null && $result->name === null) { |
||
177 | $result->name = Caser::snake($property->getName()); |
||
178 | } |
||
179 | |||
180 | return $result; |
||
181 | } |
||
182 | |||
183 | /** |
||
184 | * Returns HashMap annotation data from reader. |
||
185 | * |
||
186 | * @param \ReflectionProperty $property |
||
187 | * |
||
188 | * @return HashMap|object|null |
||
189 | */ |
||
190 | private function getHashMapAnnotationData(\ReflectionProperty $property) |
||
191 | { |
||
192 | $result = $this->reader->getPropertyAnnotation($property, self::HASH_MAP_ANNOTATION); |
||
193 | |||
194 | if ($result !== null && $result->name === null) { |
||
195 | $result->name = Caser::snake($property->getName()); |
||
196 | } |
||
197 | |||
198 | return $result; |
||
199 | } |
||
200 | |||
201 | /** |
||
202 | * Returns meta field annotation data from reader. |
||
203 | * |
||
204 | * @param \ReflectionProperty $property |
||
205 | * @param string $directory The name of the Document directory in the bundle |
||
206 | * |
||
207 | * @return array |
||
208 | */ |
||
209 | private function getMetaFieldAnnotationData($property, $directory) |
||
210 | { |
||
211 | /** @var MetaField $annotation */ |
||
212 | $annotation = $this->reader->getPropertyAnnotation($property, self::ID_ANNOTATION); |
||
213 | $annotation = $annotation ?: $this->reader->getPropertyAnnotation($property, self::PARENT_ANNOTATION); |
||
214 | $annotation = $annotation ?: $this->reader->getPropertyAnnotation($property, self::ROUTING_ANNOTATION); |
||
215 | $annotation = $annotation ?: $this->reader->getPropertyAnnotation($property, self::VERSION_ANNOTATION); |
||
216 | |||
217 | if ($annotation === null) { |
||
218 | return null; |
||
219 | } |
||
220 | |||
221 | $data = [ |
||
222 | 'name' => $annotation->getName(), |
||
223 | 'settings' => $annotation->getSettings(), |
||
224 | ]; |
||
225 | |||
226 | if ($annotation instanceof ParentDocument) { |
||
227 | $data['settings']['type'] = $this->getDocumentType($annotation->class, $directory); |
||
228 | } |
||
229 | |||
230 | return $data; |
||
231 | } |
||
232 | |||
233 | /** |
||
234 | * Returns objects used in document. |
||
235 | * |
||
236 | * @return array |
||
237 | */ |
||
238 | private function getObjects() |
||
239 | { |
||
240 | return array_keys($this->objects); |
||
241 | } |
||
242 | |||
243 | /** |
||
244 | * Finds aliases for every property used in document including parent classes. |
||
245 | * |
||
246 | * @param \ReflectionClass $reflectionClass |
||
247 | * @param array $metaFields |
||
248 | * |
||
249 | * @return array |
||
250 | */ |
||
251 | private function getAliases(\ReflectionClass $reflectionClass, array &$metaFields = null) |
||
252 | { |
||
253 | $reflectionName = $reflectionClass->getName(); |
||
254 | $directory = $this->guessDirName($reflectionClass); |
||
255 | |||
256 | // We skip cache in case $metaFields is given. This should not affect performance |
||
257 | // because for each document this method is called only once. For objects it might |
||
258 | // be called few times. |
||
259 | if ($metaFields === null && array_key_exists($reflectionName, $this->aliases)) { |
||
260 | return $this->aliases[$reflectionName]; |
||
261 | } |
||
262 | |||
263 | $alias = []; |
||
264 | |||
265 | /** @var \ReflectionProperty[] $properties */ |
||
266 | $properties = $this->getDocumentPropertiesReflection($reflectionClass); |
||
267 | |||
268 | foreach ($properties as $name => $property) { |
||
269 | $type = $this->getPropertyAnnotationData($property); |
||
270 | $type = $type !== null ? $type : $this->getEmbeddedAnnotationData($property); |
||
271 | $type = $type !== null ? $type : $this->getHashMapAnnotationData($property); |
||
272 | |||
273 | if ($type === null && $metaFields !== null |
||
274 | && ($metaData = $this->getMetaFieldAnnotationData($property, $directory)) !== null) { |
||
275 | $metaFields[$metaData['name']] = $metaData['settings']; |
||
276 | $type = new \stdClass(); |
||
277 | $type->name = $metaData['name']; |
||
278 | } |
||
279 | if ($type !== null) { |
||
280 | $alias[$type->name] = [ |
||
281 | 'propertyName' => $name, |
||
282 | ]; |
||
283 | |||
284 | if ($type instanceof Property) { |
||
285 | $alias[$type->name]['type'] = $type->type; |
||
286 | } |
||
287 | |||
288 | if ($type instanceof HashMap) { |
||
289 | $alias[$type->name]['type'] = HashMap::NAME; |
||
290 | } |
||
291 | |||
292 | $alias[$type->name][HashMap::NAME] = $type instanceof HashMap; |
||
293 | |||
294 | switch (true) { |
||
295 | case $property->isPublic(): |
||
296 | $propertyType = 'public'; |
||
297 | break; |
||
298 | case $property->isProtected(): |
||
299 | case $property->isPrivate(): |
||
300 | $propertyType = 'private'; |
||
301 | $alias[$type->name]['methods'] = $this->getMutatorMethods( |
||
302 | $reflectionClass, |
||
303 | $name, |
||
304 | $type instanceof Property ? $type->type : null |
||
305 | ); |
||
306 | break; |
||
307 | default: |
||
308 | $message = sprintf( |
||
309 | 'Wrong property %s type of %s class types cannot '. |
||
310 | 'be static or abstract.', |
||
311 | $name, |
||
312 | $reflectionName |
||
313 | ); |
||
314 | throw new \LogicException($message); |
||
315 | } |
||
316 | $alias[$type->name]['propertyType'] = $propertyType; |
||
317 | |||
318 | if ($type instanceof Embedded) { |
||
319 | $child = new \ReflectionClass($this->finder->getNamespace($type->class, $directory)); |
||
320 | $alias[$type->name] = array_merge( |
||
321 | $alias[$type->name], |
||
322 | [ |
||
323 | 'type' => $this->getObjectMapping($type->class, $directory)['type'], |
||
324 | 'multiple' => $type->multiple, |
||
325 | 'aliases' => $this->getAliases($child, $metaFields), |
||
326 | 'namespace' => $child->getName(), |
||
327 | ] |
||
328 | ); |
||
329 | } |
||
330 | } |
||
331 | } |
||
332 | |||
333 | $this->aliases[$reflectionName] = $alias; |
||
334 | |||
335 | return $this->aliases[$reflectionName]; |
||
336 | } |
||
337 | |||
338 | /** |
||
339 | * Checks if class have setter and getter, and returns them in array. |
||
340 | * |
||
341 | * @param \ReflectionClass $reflectionClass |
||
342 | * @param string $property |
||
343 | * |
||
344 | * @return array |
||
345 | */ |
||
346 | private function getMutatorMethods(\ReflectionClass $reflectionClass, $property, $propertyType) |
||
347 | { |
||
348 | $camelCaseName = ucfirst(Caser::camel($property)); |
||
349 | $setterName = 'set'.$camelCaseName; |
||
350 | if (!$reflectionClass->hasMethod($setterName)) { |
||
351 | $message = sprintf( |
||
352 | 'Missing %s() method in %s class. Add it, or change property to public.', |
||
353 | $setterName, |
||
354 | $reflectionClass->getName() |
||
355 | ); |
||
356 | throw new \LogicException($message); |
||
357 | } |
||
358 | |||
359 | if ($reflectionClass->hasMethod('get'.$camelCaseName)) { |
||
360 | return [ |
||
361 | 'getter' => 'get' . $camelCaseName, |
||
362 | 'setter' => $setterName |
||
363 | ]; |
||
364 | } |
||
365 | |||
366 | if ($propertyType === 'boolean') { |
||
367 | if ($reflectionClass->hasMethod('is' . $camelCaseName)) { |
||
368 | return [ |
||
369 | 'getter' => 'is' . $camelCaseName, |
||
370 | 'setter' => $setterName |
||
371 | ]; |
||
372 | } |
||
373 | |||
374 | $message = sprintf( |
||
375 | 'Missing %s() or %s() method in %s class. Add it, or change property to public.', |
||
376 | 'get'.$camelCaseName, |
||
377 | 'is'.$camelCaseName, |
||
378 | $reflectionClass->getName() |
||
379 | ); |
||
380 | throw new \LogicException($message); |
||
381 | } |
||
382 | |||
383 | $message = sprintf( |
||
384 | 'Missing %s() method in %s class. Add it, or change property to public.', |
||
385 | 'get'.$camelCaseName, |
||
386 | $reflectionClass->getName() |
||
387 | ); |
||
388 | throw new \LogicException($message); |
||
389 | } |
||
390 | |||
391 | /** |
||
392 | * Registers annotations to registry so that it could be used by reader. |
||
393 | */ |
||
394 | private function registerAnnotations() |
||
395 | { |
||
396 | $annotations = [ |
||
397 | 'Document', |
||
398 | 'Property', |
||
399 | 'Embedded', |
||
400 | 'Object', |
||
401 | 'Nested', |
||
402 | 'Id', |
||
403 | 'ParentDocument', |
||
404 | 'Routing', |
||
405 | 'Version', |
||
406 | 'HashMap', |
||
407 | ]; |
||
408 | |||
409 | foreach ($annotations as $annotation) { |
||
410 | AnnotationRegistry::registerFile(__DIR__ . "/../Annotation/{$annotation}.php"); |
||
411 | } |
||
412 | } |
||
413 | |||
414 | /** |
||
415 | * Returns document type. |
||
416 | * |
||
417 | * @param string $document Format must be like AcmeBundle:Document. |
||
418 | * @param string $directory The Document directory name of the bundle. |
||
419 | * |
||
420 | * @return string |
||
421 | */ |
||
422 | private function getDocumentType($document, $directory) |
||
423 | { |
||
424 | $namespace = $this->finder->getNamespace($document, $directory); |
||
425 | $reflectionClass = new \ReflectionClass($namespace); |
||
426 | $document = $this->getDocumentAnnotationData($reflectionClass); |
||
427 | |||
428 | return empty($document->type) ? Caser::snake($reflectionClass->getShortName()) : $document->type; |
||
429 | } |
||
430 | |||
431 | /** |
||
432 | * Returns all defined properties including private from parents. |
||
433 | * |
||
434 | * @param \ReflectionClass $reflectionClass |
||
435 | * |
||
436 | * @return array |
||
437 | */ |
||
438 | private function getDocumentPropertiesReflection(\ReflectionClass $reflectionClass) |
||
439 | { |
||
440 | if (in_array($reflectionClass->getName(), $this->properties)) { |
||
441 | return $this->properties[$reflectionClass->getName()]; |
||
442 | } |
||
443 | |||
444 | $properties = []; |
||
445 | |||
446 | foreach ($reflectionClass->getProperties() as $property) { |
||
447 | if (!in_array($property->getName(), $properties)) { |
||
448 | $properties[$property->getName()] = $property; |
||
449 | } |
||
450 | } |
||
451 | |||
452 | $parentReflection = $reflectionClass->getParentClass(); |
||
453 | if ($parentReflection !== false) { |
||
454 | $properties = array_merge( |
||
455 | $properties, |
||
456 | array_diff_key($this->getDocumentPropertiesReflection($parentReflection), $properties) |
||
457 | ); |
||
458 | } |
||
459 | |||
460 | $this->properties[$reflectionClass->getName()] = $properties; |
||
461 | |||
462 | return $properties; |
||
463 | } |
||
464 | |||
465 | /** |
||
466 | * Parses analyzers list from document mapping. |
||
467 | * |
||
468 | * @param \ReflectionClass $reflectionClass |
||
469 | * @return array |
||
470 | */ |
||
471 | private function getAnalyzers(\ReflectionClass $reflectionClass) |
||
472 | { |
||
473 | $analyzers = []; |
||
474 | $directory = $this->guessDirName($reflectionClass); |
||
475 | |||
476 | foreach ($this->getDocumentPropertiesReflection($reflectionClass) as $name => $property) { |
||
477 | $type = $this->getPropertyAnnotationData($property); |
||
478 | $type = $type !== null ? $type : $this->getEmbeddedAnnotationData($property); |
||
479 | |||
480 | if ($type instanceof Embedded) { |
||
481 | $analyzers = array_merge( |
||
482 | $analyzers, |
||
483 | $this->getAnalyzers(new \ReflectionClass($this->finder->getNamespace($type->class, $directory))) |
||
484 | ); |
||
485 | } |
||
486 | |||
487 | if ($type instanceof Property) { |
||
488 | if (isset($type->options['analyzer'])) { |
||
489 | $analyzers[] = $type->options['analyzer']; |
||
490 | } |
||
491 | if (isset($type->options['search_analyzer'])) { |
||
492 | $analyzers[] = $type->options['search_analyzer']; |
||
493 | } |
||
494 | |||
495 | if (isset($type->options['fields'])) { |
||
496 | foreach ($type->options['fields'] as $field) { |
||
497 | if (isset($field['analyzer'])) { |
||
498 | $analyzers[] = $field['analyzer']; |
||
499 | } |
||
500 | if (isset($field['search_analyzer'])) { |
||
501 | $analyzers[] = $field['search_analyzer']; |
||
502 | } |
||
503 | } |
||
504 | } |
||
505 | } |
||
506 | } |
||
507 | return array_unique($analyzers); |
||
508 | } |
||
509 | |||
510 | /** |
||
511 | * Returns properties of reflection class. |
||
512 | * |
||
513 | * @param \ReflectionClass $reflectionClass Class to read properties from. |
||
514 | * @param array $properties Properties to skip. |
||
515 | * @param bool $flag If false exludes properties, true only includes properties. |
||
516 | * |
||
517 | * @return array |
||
518 | */ |
||
519 | private function getProperties(\ReflectionClass $reflectionClass, $properties = [], $flag = false) |
||
520 | { |
||
521 | $mapping = []; |
||
522 | $directory = $this->guessDirName($reflectionClass); |
||
523 | |||
524 | /** @var \ReflectionProperty $property */ |
||
525 | foreach ($this->getDocumentPropertiesReflection($reflectionClass) as $name => $property) { |
||
526 | $type = $this->getPropertyAnnotationData($property); |
||
527 | $type = $type !== null ? $type : $this->getEmbeddedAnnotationData($property); |
||
528 | $type = $type !== null ? $type : $this->getHashMapAnnotationData($property); |
||
529 | |||
530 | if ((in_array($name, $properties) && !$flag) |
||
531 | || (!in_array($name, $properties) && $flag) |
||
532 | || empty($type) |
||
533 | ) { |
||
534 | continue; |
||
535 | } |
||
536 | |||
537 | $map = $type->dump(); |
||
538 | |||
539 | // Inner object |
||
540 | if ($type instanceof Embedded) { |
||
541 | $map = array_replace_recursive($map, $this->getObjectMapping($type->class, $directory)); |
||
542 | } |
||
543 | |||
544 | // HashMap object |
||
545 | if ($type instanceof HashMap) { |
||
546 | $map = array_replace_recursive($map, [ |
||
547 | 'type' => Nested::NAME, |
||
548 | 'dynamic' => true, |
||
549 | ]); |
||
550 | } |
||
551 | |||
552 | // If there is set some Raw options, it will override current ones. |
||
553 | if (isset($map['options'])) { |
||
554 | $options = $map['options']; |
||
555 | unset($map['options']); |
||
556 | $map = array_merge($map, $options); |
||
557 | } |
||
558 | |||
559 | $mapping[$type->name] = $map; |
||
560 | } |
||
561 | |||
562 | return $mapping; |
||
563 | } |
||
564 | |||
565 | /** |
||
566 | * Returns object mapping. |
||
567 | * |
||
568 | * Loads from cache if it's already loaded. |
||
569 | * |
||
570 | * @param string $className |
||
571 | * @param string $directory Name of the directory where the Document is |
||
572 | * |
||
573 | * @return array |
||
574 | */ |
||
575 | private function getObjectMapping($className, $directory) |
||
576 | { |
||
577 | $namespace = $this->finder->getNamespace($className, $directory); |
||
578 | |||
579 | if (array_key_exists($namespace, $this->objects)) { |
||
580 | return $this->objects[$namespace]; |
||
581 | } |
||
582 | |||
583 | $reflectionClass = new \ReflectionClass($namespace); |
||
584 | |||
585 | switch (true) { |
||
586 | case $this->reader->getClassAnnotation($reflectionClass, self::OBJECT_ANNOTATION): |
||
587 | $type = 'object'; |
||
588 | break; |
||
589 | case $this->reader->getClassAnnotation($reflectionClass, self::NESTED_ANNOTATION): |
||
590 | $type = 'nested'; |
||
591 | break; |
||
592 | default: |
||
593 | throw new \LogicException( |
||
594 | sprintf( |
||
595 | '%s should have @Object or @Nested annotation to be used as embeddable object.', |
||
596 | $className |
||
597 | ) |
||
598 | ); |
||
599 | } |
||
600 | |||
601 | $this->objects[$namespace] = [ |
||
602 | 'type' => $type, |
||
603 | 'properties' => $this->getProperties($reflectionClass), |
||
604 | ]; |
||
605 | |||
606 | return $this->objects[$namespace]; |
||
607 | } |
||
608 | |||
609 | /** |
||
610 | * @param \ReflectionClass $reflection |
||
611 | * |
||
612 | * @return string |
||
613 | */ |
||
614 | private function guessDirName(\ReflectionClass $reflection) |
||
615 | { |
||
616 | return substr( |
||
617 | $directory = $reflection->getName(), |
||
618 | $start = strpos($directory, '\\') + 1, |
||
619 | strrpos($directory, '\\') - $start |
||
620 | ); |
||
621 | } |
||
622 | } |
||
623 |