SearchMappingService::getPartsFields() - Code Metrics - Inspection of "Get highlight from parts" - nextcloud/fulltextsearch_elasticsearch - Measure and Improve Code Quality continuously with Scrutinizer

Completed

Pull Request — master (#47)

by Maxence

created 2019-02-12 12:32 UTC

SearchMappingService::getPartsFields() A

↳ Parent: SearchMappingService

Complexity

Conditions	1
Paths	1

Size

Total Lines

Duplication

Lines	0
Ratio	0 %

Importance

Changes

Metric	Value
dl	0
loc	7
rs	10
c	0
b	0
f	0
cc	1
nc	1
nop	1

<?php
declare(strict_types=1);


/**
 * FullTextSearch_ElasticSearch - Use Elasticsearch to index the content of your nextcloud
 *
 * This file is licensed under the Affero General Public License version 3 or
 * later. See the COPYING file.
 *
 * @author Maxence Lange <[email protected]>
 * @copyright 2018
 * @license GNU AGPL version 3 or any later version
 *
 * This program is free software: you can redistribute it and/or modify
 * it under the terms of the GNU Affero General Public License as
 * published by the Free Software Foundation, either version 3 of the
 * License, or (at your option) any later version.
 *
 * This program is distributed in the hope that it will be useful,
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 * GNU Affero General Public License for more details.
 *
 * You should have received a copy of the GNU Affero General Public License
 * along with this program.  If not, see <http://www.gnu.org/licenses/>.
 *
 */


namespace OCA\FullTextSearch_ElasticSearch\Service;


use OCA\FullTextSearch_ElasticSearch\Exceptions\ConfigurationException;
use OCA\FullTextSearch_ElasticSearch\Exceptions\QueryContentGenerationException;
use OCA\FullTextSearch_ElasticSearch\Exceptions\SearchQueryGenerationException;
use OCA\FullTextSearch_ElasticSearch\Model\QueryContent;
use OCP\FullTextSearch\Model\DocumentAccess;
use OCP\FullTextSearch\Model\ISearchRequest;


/**
 * Class SearchMappingService
 *
 * @package OCA\FullTextSearch_ElasticSearch\Service
 */
class SearchMappingService {

	/** @var ConfigService */
	private $configService;

	/** @var MiscService */
	private $miscService;


	/**
	 * SearchMappingService constructor.
	 *
	 * @param ConfigService $configService
	 * @param MiscService $miscService
	 */
	public function __construct(ConfigService $configService, MiscService $miscService) {
		$this->configService = $configService;
		$this->miscService = $miscService;
	}


	/**
	 * @param ISearchRequest $request
	 * @param DocumentAccess $access
	 * @param string $providerId
	 *
	 * @return array
	 * @throws ConfigurationException
	 * @throws SearchQueryGenerationException
	 */
	public function generateSearchQuery(
		ISearchRequest $request, DocumentAccess $access, string $providerId
	): array {
		$query['params'] = $this->generateSearchQueryParams($request, $access, $providerId);
foreach ($collection as $item) {
    $myArray['foo'] = $item->getFoo();

    if ($item->hasBar()) {
        $myArray['bar'] = $item->getBar();
    }

    // do something with $myArray
}

		return $query;
	}


	/**
	 * @param ISearchRequest $request
	 * @param DocumentAccess $access
	 * @param string $providerId
	 *
	 * @return array
	 * @throws ConfigurationException
	 * @throws SearchQueryGenerationException
	 */
	public function generateSearchQueryParams(
		ISearchRequest $request, DocumentAccess $access, string $providerId
	): array {
		$params = [
			'index' => $this->configService->getElasticIndex(),
			'type'  => 'standard',
			'size'  => $request->getSize(),
			'from'  => (($request->getPage() - 1) * $request->getSize())
		];

		$bool = [];
		$bool['must']['bool']['should'] = $this->generateSearchQueryContent($request);

		$bool['filter'][]['bool']['must'] = ['term' => ['provider' => $providerId]];
		$bool['filter'][]['bool']['should'] = $this->generateSearchQueryAccess($access);
		$bool['filter'][]['bool']['should'] =
			$this->generateSearchQueryTags('metatags', $request->getMetaTags());
		$bool['filter'][]['bool']['should'] =
			$this->generateSearchQueryTags('subtags', $request->getSubTags(true));
//		$bool['filter'][]['bool']['should'] = $this->generateSearchQueryTags($request->getTags());

		$params['body']['query']['bool'] = $bool;
		$params['body']['highlight'] = $this->generateSearchHighlighting($request);

		$this->improveSearchQuerying($request, $params['body']['query']);

		return $params;
	}


	/**
	 * @param ISearchRequest $request
	 * @param array $arr
	 */
	private function improveSearchQuerying(ISearchRequest $request, array &$arr) {
//		$this->improveSearchWildcardQueries($request, $arr);
		$this->improveSearchWildcardFilters($request, $arr);
		$this->improveSearchRegexFilters($request, $arr);
	}


//	/**
//	 * @param SearchRequest $request
//	 * @param array $arr
//	 */
//	private function improveSearchWildcardQueries(SearchRequest $request, &$arr) {
//
//		$queries = $request->getWildcardQueries();
//		foreach ($queries as $query) {
//			$wildcards = [];
//			foreach ($query as $entry) {
//				$wildcards[] = ['wildcard' => $entry];
//			}
//
//			array_push($arr['bool']['must']['bool']['should'], $wildcards);
//		}
//
//	}


	/**
	 * @param ISearchRequest $request
	 * @param array $arr
	 */
	private function improveSearchWildcardFilters(ISearchRequest $request, array &$arr) {


		$filters = $request->getWildcardFilters();
		foreach ($filters as $filter) {
			$wildcards = [];
			foreach ($filter as $entry) {
				$wildcards[] = ['wildcard' => $entry];
			}

			$arr['bool']['filter'][]['bool']['should'] = $wildcards;
		}

	}


	/**
	 * @param ISearchRequest $request
	 * @param array $arr
	 */
	private function improveSearchRegexFilters(ISearchRequest $request, array &$arr) {


		$filters = $request->getRegexFilters();
		foreach ($filters as $filter) {
			$regex = [];
			foreach ($filter as $entry) {
				$regex[] = ['regexp' => $entry];
			}

			$arr['bool']['filter'][]['bool']['should'] = $regex;
		}

	}


	/**
	 * @param ISearchRequest $request
	 *
	 * @return array
	 * @throws SearchQueryGenerationException
	 */
	private function generateSearchQueryContent(ISearchRequest $request): array {
		$str = strtolower($request->getSearch());

		preg_match_all('/[^?]"(?:\\\\.|[^\\\\"])*"|\S+/', " $str ", $words);
		$queryContent = [];
		foreach ($words[0] as $word) {
			try {
				$queryContent[] = $this->generateQueryContent(trim($word));
			} catch (QueryContentGenerationException $e) {
				continue;
			}
		}

		if (sizeof($queryContent) === 0) {
			throw new SearchQueryGenerationException();
		}

		return $this->generateSearchQueryFromQueryContent($request, $queryContent);
	}


	/**
	 * @param string $word
	 *
	 * @return QueryContent
	 * @throws QueryContentGenerationException
	 */
	private function generateQueryContent(string $word): QueryContent {

		$searchQueryContent = new QueryContent($word);
		if (strlen($searchQueryContent->getWord()) === 0) {
			throw new QueryContentGenerationException();
		}

		return $searchQueryContent;
	}


	/**
	 * @param ISearchRequest $request
	 * @param QueryContent[] $queryContents
	 *
	 * @return array
	 */
	private function generateSearchQueryFromQueryContent(
		ISearchRequest $request, array $queryContents
	): array {

		$query = $queryWords = [];
		foreach ($queryContents as $queryContent) {
			$queryWords[$queryContent->getShould()][] =
				$this->generateQueryContentFields($request, $queryContent);
		}

		$listShould = array_keys($queryWords);
		foreach ($listShould as $itemShould) {
			$query[$itemShould][] = $queryWords[$itemShould];
		}

		return ['bool' => $query];
	}


	/**
	 * @param ISearchRequest $request
	 * @param QueryContent $content
	 *
	 * @return array
	 */
	private function generateQueryContentFields(ISearchRequest $request, QueryContent $content
	): array {
		$parts = $this->getPartsFields($request);
		$fields = array_merge(['content', 'title'], $request->getFields(), $parts);

		$queryFields = [];
		foreach ($fields as $field) {
			if (!$this->fieldIsOutLimit($request, $field)) {
				$queryFields[] = [$content->getMatch() => [$field => $content->getWord()]];
			}
		}

		foreach ($request->getWildcardFields() as $field) {
			if (!$this->fieldIsOutLimit($request, $field)) {
				$queryFields[] = ['wildcard' => [$field => '*' . $content->getWord() . '*']];
			}
		}

		return ['bool' => ['should' => $queryFields]];
	}


	/**
	 * @param DocumentAccess $access
	 *
	 * @return array
	 */
	private function generateSearchQueryAccess(DocumentAccess $access): array {

		$query = [];
		$query[] = ['term' => ['owner' => $access->getViewerId()]];
		$query[] = ['term' => ['users' => $access->getViewerId()]];
		$query[] = ['term' => ['users' => '__all']];

		foreach ($access->getGroups() as $group) {
			$query[] = ['term' => ['groups' => $group]];
		}

		foreach ($access->getCircles() as $circle) {
			$query[] = ['term' => ['circles' => $circle]];
		}

		return $query;
	}


	/**
	 * @param ISearchRequest $request
	 * @param string $field
	 *
	 * @return bool
	 */
	private function fieldIsOutLimit(ISearchRequest $request, string $field): bool {
		$limit = $request->getLimitFields();
		if (sizeof($limit) === 0) {
			return false;
		}

		if (in_array($field, $limit)) {

			return false;
		}

		return true;
	}


	/**
	 * @param string $k
	 * @param array $tags
	 *
	 * @return array
	 */
	private function generateSearchQueryTags(string $k, array $tags): array {

		$query = [];
		foreach ($tags as $t) {
			$query[] = ['term' => [$k => $t]];
		}

		return $query;
	}


	/**
	 * @param ISearchRequest $request
	 *
	 * @return array
	 */
	private function generateSearchHighlighting(ISearchRequest $request): array {

		$parts = $this->getPartsFields($request);
		$fields = ['content' => new \stdClass()];
		foreach ($parts as $part) {
			$fields[$part] = new \stdClass();
		}

		return [
			'fields'    => $fields,
			'pre_tags'  => [''],
			'post_tags' => ['']
		];
	}


	/**
	 * @param string $providerId
	 * @param string $documentId
	 *
	 * @return array
	 * @throws ConfigurationException
	 */
	public function getDocumentQuery(string $providerId, string $documentId): array {
		return [
			'index' => $this->configService->getElasticIndex(),
			'type'  => 'standard',
			'id'    => $providerId . ':' . $documentId
		];
	}


	/**
	 * @param ISearchRequest $request
	 *
	 * @return array
	 */
	private function getPartsFields(ISearchRequest $request) {
		return array_map(
			function($value) {
				return 'parts.' . $value;
			}, $request->getParts()
		);
	}

}



1		<?php
2		declare(strict_types=1);
3
4
5		/**
6		* FullTextSearch_ElasticSearch - Use Elasticsearch to index the content of your nextcloud
7		*
8		* This file is licensed under the Affero General Public License version 3 or
9		* later. See the COPYING file.
10		*
11		* @author Maxence Lange <[email protected]>
12		* @copyright 2018
13		* @license GNU AGPL version 3 or any later version
14		*
15		* This program is free software: you can redistribute it and/or modify
16		* it under the terms of the GNU Affero General Public License as
17		* published by the Free Software Foundation, either version 3 of the
18		* License, or (at your option) any later version.
19		*
20		* This program is distributed in the hope that it will be useful,
21		* but WITHOUT ANY WARRANTY; without even the implied warranty of
22		* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
23		* GNU Affero General Public License for more details.
24		*
25		* You should have received a copy of the GNU Affero General Public License
26		* along with this program. If not, see <http://www.gnu.org/licenses/>.
27		*
28		*/
29
30
31		namespace OCA\FullTextSearch_ElasticSearch\Service;
32
33
34		use OCA\FullTextSearch_ElasticSearch\Exceptions\ConfigurationException;
35		use OCA\FullTextSearch_ElasticSearch\Exceptions\QueryContentGenerationException;
36		use OCA\FullTextSearch_ElasticSearch\Exceptions\SearchQueryGenerationException;
37		use OCA\FullTextSearch_ElasticSearch\Model\QueryContent;
38		use OCP\FullTextSearch\Model\DocumentAccess;
39		use OCP\FullTextSearch\Model\ISearchRequest;
40
41
42		/**
43		* Class SearchMappingService
44		*
45		* @package OCA\FullTextSearch_ElasticSearch\Service
46		*/
47		class SearchMappingService {
48
49		/** @var ConfigService */
50		private $configService;
51
52		/** @var MiscService */
53		private $miscService;
54
55
56		/**
57		* SearchMappingService constructor.
58		*
59		* @param ConfigService $configService
60		* @param MiscService $miscService
61		*/
62		public function __construct(ConfigService $configService, MiscService $miscService) {
63		$this->configService = $configService;
64		$this->miscService = $miscService;
65		}
66
67
68		/**
69		* @param ISearchRequest $request
70		* @param DocumentAccess $access
71		* @param string $providerId
72		*
73		* @return array
74		* @throws ConfigurationException
75		* @throws SearchQueryGenerationException
76		*/
77		public function generateSearchQuery(
78		ISearchRequest $request, DocumentAccess $access, string $providerId
79		): array {
80		$query['params'] = $this->generateSearchQueryParams($request, $access, $providerId);
		0 ignored issues – show Coding Style Comprehensibility introduced 2018-09-04 13:31 UTC by Report Bug Copy Issue Report `$query` was never initialized. Although not strictly required by PHP, it is generally a good practice to add `$query = array();` before regardless. Adding an explicit array definition is generally preferable to implicit array definition as it guarantees a stable state of the code. Let’s take a look at an example: foreach ($collection as $item) { $myArray['foo'] = $item->getFoo(); if ($item->hasBar()) { $myArray['bar'] = $item->getBar(); } // do something with $myArray } As you can see in this example, the array `$myArray` is initialized the first time when the foreach loop is entered. You can also see that the value of the `bar` key is only written conditionally; thus, its value might result from a previous iteration. This might or might not be intended. To make your intention clear, your code more readible and to avoid accidental bugs, we recommend to add an explicit initialization `$myArray = array()` either outside or inside the foreach loop. Loading history...
81
82		return $query;
83		}
84
85
86		/**
87		* @param ISearchRequest $request
88		* @param DocumentAccess $access
89		* @param string $providerId
90		*
91		* @return array
92		* @throws ConfigurationException
93		* @throws SearchQueryGenerationException
94		*/
95		public function generateSearchQueryParams(
96		ISearchRequest $request, DocumentAccess $access, string $providerId
97		): array {
98		$params = [
99		'index' => $this->configService->getElasticIndex(),
100		'type' => 'standard',
101		'size' => $request->getSize(),
102		'from' => (($request->getPage() - 1) * $request->getSize())
103		];
104
105		$bool = [];
106		$bool['must']['bool']['should'] = $this->generateSearchQueryContent($request);
107
108		$bool['filter'][]['bool']['must'] = ['term' => ['provider' => $providerId]];
109		$bool['filter'][]['bool']['should'] = $this->generateSearchQueryAccess($access);
110		$bool['filter'][]['bool']['should'] =
111		$this->generateSearchQueryTags('metatags', $request->getMetaTags());
112		$bool['filter'][]['bool']['should'] =
113		$this->generateSearchQueryTags('subtags', $request->getSubTags(true));
114		// $bool['filter'][]['bool']['should'] = $this->generateSearchQueryTags($request->getTags());
115
116		$params['body']['query']['bool'] = $bool;
117		$params['body']['highlight'] = $this->generateSearchHighlighting($request);
118
119		$this->improveSearchQuerying($request, $params['body']['query']);
120
121		return $params;
122		}
123
124
125		/**
126		* @param ISearchRequest $request
127		* @param array $arr
128		*/
129		private function improveSearchQuerying(ISearchRequest $request, array &$arr) {
130		// $this->improveSearchWildcardQueries($request, $arr);
131		$this->improveSearchWildcardFilters($request, $arr);
132		$this->improveSearchRegexFilters($request, $arr);
133		}
134
135
136		// /**
137		// * @param SearchRequest $request
138		// * @param array $arr
139		// */
140		// private function improveSearchWildcardQueries(SearchRequest $request, &$arr) {
141		//
142		// $queries = $request->getWildcardQueries();
143		// foreach ($queries as $query) {
144		// $wildcards = [];
145		// foreach ($query as $entry) {
146		// $wildcards[] = ['wildcard' => $entry];
147		// }
148		//
149		// array_push($arr['bool']['must']['bool']['should'], $wildcards);
150		// }
151		//
152		// }
153
154
155		/**
156		* @param ISearchRequest $request
157		* @param array $arr
158		*/
159	View Code Duplication	private function improveSearchWildcardFilters(ISearchRequest $request, array &$arr) {
		0 ignored issues – show Duplication introduced 2018-03-15 11:51 UTC by Report Bug Copy Issue Report This method seems to be duplicated in your project. Duplicated code is one of the most pungent code smells. If you need to duplicate the same code in three or more different places, we strongly encourage you to look into extracting the code into a single class or operation. You can also find more detailed suggestions in the “Code” section of your repository. Loading history...
160
161		$filters = $request->getWildcardFilters();
162		foreach ($filters as $filter) {
163		$wildcards = [];
164		foreach ($filter as $entry) {
165		$wildcards[] = ['wildcard' => $entry];
166		}
167
168		$arr['bool']['filter'][]['bool']['should'] = $wildcards;
169		}
170
171		}
172
173
174		/**
175		* @param ISearchRequest $request
176		* @param array $arr
177		*/
178	View Code Duplication	private function improveSearchRegexFilters(ISearchRequest $request, array &$arr) {
		0 ignored issues – show Duplication introduced 2018-05-30 12:57 UTC by Report Bug Copy Issue Report This method seems to be duplicated in your project. Duplicated code is one of the most pungent code smells. If you need to duplicate the same code in three or more different places, we strongly encourage you to look into extracting the code into a single class or operation. You can also find more detailed suggestions in the “Code” section of your repository. Loading history...
179
180		$filters = $request->getRegexFilters();
181		foreach ($filters as $filter) {
182		$regex = [];
183		foreach ($filter as $entry) {
184		$regex[] = ['regexp' => $entry];
185		}
186
187		$arr['bool']['filter'][]['bool']['should'] = $regex;
188		}
189
190		}
191
192
193		/**
194		* @param ISearchRequest $request
195		*
196		* @return array
197		* @throws SearchQueryGenerationException
198		*/
199		private function generateSearchQueryContent(ISearchRequest $request): array {
200		$str = strtolower($request->getSearch());
201
202		preg_match_all('/[^?]"(?:\\\\.\|[^\\\\"])*"\|\S+/', " $str ", $words);
203		$queryContent = [];
204		foreach ($words[0] as $word) {
205		try {
206		$queryContent[] = $this->generateQueryContent(trim($word));
207		} catch (QueryContentGenerationException $e) {
208		continue;
209		}
210		}
211
212		if (sizeof($queryContent) === 0) {
213		throw new SearchQueryGenerationException();
214		}
215
216		return $this->generateSearchQueryFromQueryContent($request, $queryContent);
217		}
218
219
220		/**
221		* @param string $word
222		*
223		* @return QueryContent
224		* @throws QueryContentGenerationException
225		*/
226		private function generateQueryContent(string $word): QueryContent {
227
228		$searchQueryContent = new QueryContent($word);
229		if (strlen($searchQueryContent->getWord()) === 0) {
230		throw new QueryContentGenerationException();
231		}
232
233		return $searchQueryContent;
234		}
235
236
237		/**
238		* @param ISearchRequest $request
239		* @param QueryContent[] $queryContents
240		*
241		* @return array
242		*/
243		private function generateSearchQueryFromQueryContent(
244		ISearchRequest $request, array $queryContents
245		): array {
246
247		$query = $queryWords = [];
248		foreach ($queryContents as $queryContent) {
249		$queryWords[$queryContent->getShould()][] =
250		$this->generateQueryContentFields($request, $queryContent);
251		}
252
253		$listShould = array_keys($queryWords);
254		foreach ($listShould as $itemShould) {
255		$query[$itemShould][] = $queryWords[$itemShould];
256		}
257
258		return ['bool' => $query];
259		}
260
261
262		/**
263		* @param ISearchRequest $request
264		* @param QueryContent $content
265		*
266		* @return array
267		*/
268		private function generateQueryContentFields(ISearchRequest $request, QueryContent $content
269		): array {
270		$parts = $this->getPartsFields($request);
271		$fields = array_merge(['content', 'title'], $request->getFields(), $parts);
272
273		$queryFields = [];
274		foreach ($fields as $field) {
275		if (!$this->fieldIsOutLimit($request, $field)) {
276		$queryFields[] = [$content->getMatch() => [$field => $content->getWord()]];
277		}
278		}
279
280		foreach ($request->getWildcardFields() as $field) {
281		if (!$this->fieldIsOutLimit($request, $field)) {
282		$queryFields[] = ['wildcard' => [$field => '' . $content->getWord() . '']];
283		}
284		}
285
286		return ['bool' => ['should' => $queryFields]];
287		}
288
289
290		/**
291		* @param DocumentAccess $access
292		*
293		* @return array
294		*/
295		private function generateSearchQueryAccess(DocumentAccess $access): array {
296
297		$query = [];
298		$query[] = ['term' => ['owner' => $access->getViewerId()]];
299		$query[] = ['term' => ['users' => $access->getViewerId()]];
300		$query[] = ['term' => ['users' => '__all']];
301
302		foreach ($access->getGroups() as $group) {
303		$query[] = ['term' => ['groups' => $group]];
304		}
305
306		foreach ($access->getCircles() as $circle) {
307		$query[] = ['term' => ['circles' => $circle]];
308		}
309
310		return $query;
311		}
312
313
314		/**
315		* @param ISearchRequest $request
316		* @param string $field
317		*
318		* @return bool
319		*/
320		private function fieldIsOutLimit(ISearchRequest $request, string $field): bool {
321		$limit = $request->getLimitFields();
322		if (sizeof($limit) === 0) {
323		return false;
324		}
325
326		if (in_array($field, $limit)) {
		0 ignored issues – show Unused Code introduced 2018-06-13 18:14 UTC by Report Bug Copy Issue Report This `if` statement, and the following `return` statement can be replaced with `return !in_array($field, $limit);`. Loading history...
327		return false;
328		}
329
330		return true;
331		}
332
333
334		/**
335		* @param string $k
336		* @param array $tags
337		*
338		* @return array
339		*/
340		private function generateSearchQueryTags(string $k, array $tags): array {
341
342		$query = [];
343		foreach ($tags as $t) {
344		$query[] = ['term' => [$k => $t]];
345		}
346
347		return $query;
348		}
349
350
351		/**
352		* @param ISearchRequest $request
353		*
354		* @return array
355		*/
356		private function generateSearchHighlighting(ISearchRequest $request): array {
357
358		$parts = $this->getPartsFields($request);
359		$fields = ['content' => new \stdClass()];
360		foreach ($parts as $part) {
361		$fields[$part] = new \stdClass();
362		}
363
364		return [
365		'fields' => $fields,
366		'pre_tags' => [''],
367		'post_tags' => ['']
368		];
369		}
370
371
372		/**
373		* @param string $providerId
374		* @param string $documentId
375		*
376		* @return array
377		* @throws ConfigurationException
378		*/
379		public function getDocumentQuery(string $providerId, string $documentId): array {
380		return [
381		'index' => $this->configService->getElasticIndex(),
382		'type' => 'standard',
383		'id' => $providerId . ':' . $documentId
384		];
385		}
386
387
388		/**
389		* @param ISearchRequest $request
390		*
391		* @return array
392		*/
393		private function getPartsFields(ISearchRequest $request) {
394		return array_map(
395		function($value) {
396		return 'parts.' . $value;
397		}, $request->getParts()
398		);
399		}
400
401		}
402
403

nextcloud / fulltextsearch_elasticsearch

Pull Request — master (#47)

SearchMappingService::getPartsFields() A

Complexity

Size

Duplication

Importance

Duplication Side-by-Side

Filter issues like