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 12:31 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);

		$this->miscService->log('### ' . json_encode($params['body']['highlight']));
		$this->improveSearchQuerying($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		$this->miscService->log('### ' . json_encode($params['body']['highlight']));
120		$this->improveSearchQuerying($request, $params['body']['query']);
121
122		return $params;
123		}
124
125
126		/**
127		* @param ISearchRequest $request
128		* @param array $arr
129		*/
130		private function improveSearchQuerying(ISearchRequest $request, array &$arr) {
131		// $this->improveSearchWildcardQueries($request, $arr);
132		$this->improveSearchWildcardFilters($request, $arr);
133		$this->improveSearchRegexFilters($request, $arr);
134		}
135
136
137		// /**
138		// * @param SearchRequest $request
139		// * @param array $arr
140		// */
141		// private function improveSearchWildcardQueries(SearchRequest $request, &$arr) {
142		//
143		// $queries = $request->getWildcardQueries();
144		// foreach ($queries as $query) {
145		// $wildcards = [];
146		// foreach ($query as $entry) {
147		// $wildcards[] = ['wildcard' => $entry];
148		// }
149		//
150		// array_push($arr['bool']['must']['bool']['should'], $wildcards);
151		// }
152		//
153		// }
154
155
156		/**
157		* @param ISearchRequest $request
158		* @param array $arr
159		*/
160	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...
161
162		$filters = $request->getWildcardFilters();
163		foreach ($filters as $filter) {
164		$wildcards = [];
165		foreach ($filter as $entry) {
166		$wildcards[] = ['wildcard' => $entry];
167		}
168
169		$arr['bool']['filter'][]['bool']['should'] = $wildcards;
170		}
171
172		}
173
174
175		/**
176		* @param ISearchRequest $request
177		* @param array $arr
178		*/
179	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...
180
181		$filters = $request->getRegexFilters();
182		foreach ($filters as $filter) {
183		$regex = [];
184		foreach ($filter as $entry) {
185		$regex[] = ['regexp' => $entry];
186		}
187
188		$arr['bool']['filter'][]['bool']['should'] = $regex;
189		}
190
191		}
192
193
194		/**
195		* @param ISearchRequest $request
196		*
197		* @return array
198		* @throws SearchQueryGenerationException
199		*/
200		private function generateSearchQueryContent(ISearchRequest $request): array {
201		$str = strtolower($request->getSearch());
202
203		preg_match_all('/[^?]"(?:\\\\.\|[^\\\\"])*"\|\S+/', " $str ", $words);
204		$queryContent = [];
205		foreach ($words[0] as $word) {
206		try {
207		$queryContent[] = $this->generateQueryContent(trim($word));
208		} catch (QueryContentGenerationException $e) {
209		continue;
210		}
211		}
212
213		if (sizeof($queryContent) === 0) {
214		throw new SearchQueryGenerationException();
215		}
216
217		return $this->generateSearchQueryFromQueryContent($request, $queryContent);
218		}
219
220
221		/**
222		* @param string $word
223		*
224		* @return QueryContent
225		* @throws QueryContentGenerationException
226		*/
227		private function generateQueryContent(string $word): QueryContent {
228
229		$searchQueryContent = new QueryContent($word);
230		if (strlen($searchQueryContent->getWord()) === 0) {
231		throw new QueryContentGenerationException();
232		}
233
234		return $searchQueryContent;
235		}
236
237
238		/**
239		* @param ISearchRequest $request
240		* @param QueryContent[] $queryContents
241		*
242		* @return array
243		*/
244		private function generateSearchQueryFromQueryContent(
245		ISearchRequest $request, array $queryContents
246		): array {
247
248		$query = $queryWords = [];
249		foreach ($queryContents as $queryContent) {
250		$queryWords[$queryContent->getShould()][] =
251		$this->generateQueryContentFields($request, $queryContent);
252		}
253
254		$listShould = array_keys($queryWords);
255		foreach ($listShould as $itemShould) {
256		$query[$itemShould][] = $queryWords[$itemShould];
257		}
258
259		return ['bool' => $query];
260		}
261
262
263		/**
264		* @param ISearchRequest $request
265		* @param QueryContent $content
266		*
267		* @return array
268		*/
269		private function generateQueryContentFields(ISearchRequest $request, QueryContent $content
270		): array {
271		$parts = $this->getPartsFields($request);
272		$fields = array_merge(['content', 'title'], $request->getFields(), $parts);
273
274		$queryFields = [];
275		foreach ($fields as $field) {
276		if (!$this->fieldIsOutLimit($request, $field)) {
277		$queryFields[] = [$content->getMatch() => [$field => $content->getWord()]];
278		}
279		}
280
281		foreach ($request->getWildcardFields() as $field) {
282		if (!$this->fieldIsOutLimit($request, $field)) {
283		$queryFields[] = ['wildcard' => [$field => '' . $content->getWord() . '']];
284		}
285		}
286
287		return ['bool' => ['should' => $queryFields]];
288		}
289
290
291		/**
292		* @param DocumentAccess $access
293		*
294		* @return array
295		*/
296		private function generateSearchQueryAccess(DocumentAccess $access): array {
297
298		$query = [];
299		$query[] = ['term' => ['owner' => $access->getViewerId()]];
300		$query[] = ['term' => ['users' => $access->getViewerId()]];
301		$query[] = ['term' => ['users' => '__all']];
302
303		foreach ($access->getGroups() as $group) {
304		$query[] = ['term' => ['groups' => $group]];
305		}
306
307		foreach ($access->getCircles() as $circle) {
308		$query[] = ['term' => ['circles' => $circle]];
309		}
310
311		return $query;
312		}
313
314
315		/**
316		* @param ISearchRequest $request
317		* @param string $field
318		*
319		* @return bool
320		*/
321		private function fieldIsOutLimit(ISearchRequest $request, string $field): bool {
322		$limit = $request->getLimitFields();
323		if (sizeof($limit) === 0) {
324		return false;
325		}
326
327		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...
328		return false;
329		}
330
331		return true;
332		}
333
334
335		/**
336		* @param string $k
337		* @param array $tags
338		*
339		* @return array
340		*/
341		private function generateSearchQueryTags(string $k, array $tags): array {
342
343		$query = [];
344		foreach ($tags as $t) {
345		$query[] = ['term' => [$k => $t]];
346		}
347
348		return $query;
349		}
350
351
352		/**
353		* @param ISearchRequest $request
354		*
355		* @return array
356		*/
357		private function generateSearchHighlighting(ISearchRequest $request): array {
358
359		$parts = $this->getPartsFields($request);
360		$fields = ['content' => new \stdClass()];
361		foreach ($parts as $part) {
362		$fields[$part] = new \stdClass();
363		}
364
365		return [
366		'fields' => $fields,
367		'pre_tags' => [''],
368		'post_tags' => ['']
369		];
370		}
371
372
373		/**
374		* @param string $providerId
375		* @param string $documentId
376		*
377		* @return array
378		* @throws ConfigurationException
379		*/
380		public function getDocumentQuery(string $providerId, string $documentId): array {
381		return [
382		'index' => $this->configService->getElasticIndex(),
383		'type' => 'standard',
384		'id' => $providerId . ':' . $documentId
385		];
386		}
387
388
389		/**
390		* @param ISearchRequest $request
391		*
392		* @return array
393		*/
394		private function getPartsFields(ISearchRequest $request) {
395		return array_map(
396		function($value) {
397		return 'parts.' . $value;
398		}, $request->getParts()
399		);
400		}
401
402		}
403
404

nextcloud / fulltextsearch_elasticsearch

Pull Request — master (#47)

SearchMappingService::getPartsFields() A

Complexity

Size

Duplication

Importance

Duplication Side-by-Side

Filter issues like