SearchMappingService::generateSearchQueryFromQueryContent() - Code Metrics - Inspection of "0.8.1" - nextcloud/fulltextsearch_elasticsearch - Measure and Improve Code Quality continuously with Scrutinizer

Completed

Push — master ( a799ec...765778 )

by Maxence

created 2018-07-28 11:47 UTC

generateSearchQueryFromQueryContent() A

↳ Parent: SearchMappingService

Complexity

Conditions	3
Paths	4

Size

Total Lines

Duplication

Lines	0
Ratio	0 %

Importance

Changes

Metric	Value
dl	0
loc	15
rs	9.7666
c	0
b	0
f	0
cc	3
nc	4
nop	2

<?php
/**
 * 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\IFullTextSearchProvider;
use OCA\FullTextSearch\Model\DocumentAccess;
use OCA\FullTextSearch\Model\SearchRequest;
use OCA\FullTextSearch_ElasticSearch\Exceptions\ConfigurationException;
use OCA\FullTextSearch_ElasticSearch\Exceptions\QueryContentGenerationException;
use OCA\FullTextSearch_ElasticSearch\Exceptions\SearchQueryGenerationException;
use OCA\FullTextSearch_ElasticSearch\Model\QueryContent;


class SearchMappingService {

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

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


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


	/**
	 * @param IFullTextSearchProvider $provider
	 * @param DocumentAccess $access
	 * @param SearchRequest $request
	 *
	 * @return array

	 * @throws ConfigurationException
	 * @throws SearchQueryGenerationException
	 */
	public function generateSearchQuery(
		IFullTextSearchProvider $provider, DocumentAccess $access, SearchRequest $request
	) {
		$query['params'] = $this->generateSearchQueryParams($provider, $access, $request);
foreach ($collection as $item) {
    $myArray['foo'] = $item->getFoo();

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

    // do something with $myArray
}

		return $query;
	}


	/**
	 * @param IFullTextSearchProvider $provider
	 * @param DocumentAccess $access
	 * @param SearchRequest $request
	 *
	 * @return array
	 * @throws ConfigurationException
	 * @throws SearchQueryGenerationException
	 */
	public function generateSearchQueryParams(
		IFullTextSearchProvider $provider, DocumentAccess $access, SearchRequest $request
	) {
		$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' => $provider->getId()]];
		$bool['filter'][]['bool']['should'] = $this->generateSearchQueryAccess($access);
		$bool['filter'][]['bool']['should'] = $this->generateSearchQueryTags($request->getTags());

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

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

		return $params;
	}


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


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

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

	}


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


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

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

	}


	/**
	 * @param SearchRequest $request
	 *
	 * @return array<string,array<string,array>>

	 * @throws SearchQueryGenerationException
	 */
	private function generateSearchQueryContent(SearchRequest $request) {
		$str = strtolower($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 $word
	 *
	 * @return QueryContent
	 * @throws QueryContentGenerationException
	 */
	private function generateQueryContent($word) {

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

		return $searchQueryContent;
	}


	/**
	 * @param SearchRequest $request
	 * @param QueryContent[] $queryContents
	 *
	 * @return array

	 */
	private function generateSearchQueryFromQueryContent(SearchRequest $request, $queryContents) {

		$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 SearchRequest $request
	 * @param QueryContent $content
	 *
	 * @return array

	 */
	private function generateQueryContentFields(SearchRequest $request, QueryContent $content) {
		$parts = array_map(
			function($value) {
				return 'parts.' . $value;
			}, $request->getParts()
		);
		$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<string,array>

	 */
	private function generateSearchQueryAccess(DocumentAccess $access) {

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

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

			return false;
		}

		return true;
	}


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

		$query = [];
		foreach ($tags as $tag) {
			$query[] = ['term' => ['tags' => $tag]];
		}

		return $query;
	}

	private function generateSearchHighlighting() {
		return [
			'fields'    => ['content' => new \stdClass()],
			'pre_tags'  => [''],
			'post_tags' => ['']
		];
	}


}


1		<?php
2		/**
3		* FullTextSearch_ElasticSearch - Use Elasticsearch to index the content of your nextcloud
4		*
5		* This file is licensed under the Affero General Public License version 3 or
6		* later. See the COPYING file.
7		*
8		* @author Maxence Lange <[email protected]>
9		* @copyright 2018
10		* @license GNU AGPL version 3 or any later version
11		*
12		* This program is free software: you can redistribute it and/or modify
13		* it under the terms of the GNU Affero General Public License as
14		* published by the Free Software Foundation, either version 3 of the
15		* License, or (at your option) any later version.
16		*
17		* This program is distributed in the hope that it will be useful,
18		* but WITHOUT ANY WARRANTY; without even the implied warranty of
19		* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
20		* GNU Affero General Public License for more details.
21		*
22		* You should have received a copy of the GNU Affero General Public License
23		* along with this program. If not, see <http://www.gnu.org/licenses/>.
24		*
25		*/
26
27		namespace OCA\FullTextSearch_ElasticSearch\Service;
28
29		use OCA\FullTextSearch\IFullTextSearchProvider;
30		use OCA\FullTextSearch\Model\DocumentAccess;
31		use OCA\FullTextSearch\Model\SearchRequest;
32		use OCA\FullTextSearch_ElasticSearch\Exceptions\ConfigurationException;
33		use OCA\FullTextSearch_ElasticSearch\Exceptions\QueryContentGenerationException;
34		use OCA\FullTextSearch_ElasticSearch\Exceptions\SearchQueryGenerationException;
35		use OCA\FullTextSearch_ElasticSearch\Model\QueryContent;
36
37
38		class SearchMappingService {
39
40		/** @var ConfigService */
41		private $configService;
42
43		/** @var MiscService */
44		private $miscService;
45
46
47		/**
48		* MappingService constructor.
49		*
50		* @param ConfigService $configService
51		* @param MiscService $miscService
52		*/
53		public function __construct(ConfigService $configService, MiscService $miscService) {
54		$this->configService = $configService;
55		$this->miscService = $miscService;
56		}
57
58
59		/**
60		* @param IFullTextSearchProvider $provider
61		* @param DocumentAccess $access
62		* @param SearchRequest $request
63		*
64		* @return array
		0 ignored issues – show Documentation introduced 2018-01-10 22:16 UTC by Report Bug Copy Issue Report Consider making the return type a bit more specific; maybe use `array<string,array>`. This check looks for the generic type `array` as a return type and suggests a more specific type. This type is inferred from the actual code. Loading history...
65		* @throws ConfigurationException
66		* @throws SearchQueryGenerationException
67		*/
68		public function generateSearchQuery(
69		IFullTextSearchProvider $provider, DocumentAccess $access, SearchRequest $request
70		) {
71		$query['params'] = $this->generateSearchQueryParams($provider, $access, $request);
		0 ignored issues – show Coding Style Comprehensibility introduced 2018-01-10 22:16 UTC by Report Bug Copy Issue Report `$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...
72
73		return $query;
74		}
75
76
77		/**
78		* @param IFullTextSearchProvider $provider
79		* @param DocumentAccess $access
80		* @param SearchRequest $request
81		*
82		* @return array
83		* @throws ConfigurationException
84		* @throws SearchQueryGenerationException
85		*/
86		public function generateSearchQueryParams(
87		IFullTextSearchProvider $provider, DocumentAccess $access, SearchRequest $request
88		) {
89		$params = [
90		'index' => $this->configService->getElasticIndex(),
91		'type' => 'standard',
92		'size' => $request->getSize(),
93		'from' => (($request->getPage() - 1) * $request->getSize())
94		];
95
96		$bool = [];
97		$bool['must']['bool']['should'] = $this->generateSearchQueryContent($request);
98
99		$bool['filter'][]['bool']['must'] = ['term' => ['provider' => $provider->getId()]];
100		$bool['filter'][]['bool']['should'] = $this->generateSearchQueryAccess($access);
101		$bool['filter'][]['bool']['should'] = $this->generateSearchQueryTags($request->getTags());
102
103		$params['body']['query']['bool'] = $bool;
104		$params['body']['highlight'] = $this->generateSearchHighlighting();
105
106		$this->improveSearchQuerying($request, $params['body']['query']);
107
108		return $params;
109		}
110
111
112		/**
113		* @param SearchRequest $request
114		* @param array $arr
115		*/
116		private function improveSearchQuerying(SearchRequest $request, &$arr) {
117		// $this->improveSearchWildcardQueries($request, $arr);
		0 ignored issues – show Unused Code Comprehensibility introduced 2018-07-28 11:48 UTC by Report Bug Copy Issue Report `73%` of this comment could be valid code. Did you maybe forget this after debugging? Sometimes obsolete code just ends up commented out instead of removed. In this case it is better to remove the code once you have checked you do not need it. The code might also have been commented out for debugging purposes. In this case it is vital that someone uncomments it again or your project may behave in very unexpected ways in production. This check looks for comments that seem to be mostly valid code and reports them. Loading history...
118		$this->improveSearchWildcardFilters($request, $arr);
119		$this->improveSearchRegexFilters($request, $arr);
120		}
121
122
123		// /**
		0 ignored issues – show Unused Code Comprehensibility introduced 2018-07-28 11:48 UTC by Report Bug Copy Issue Report `49%` of this comment could be valid code. Did you maybe forget this after debugging? Sometimes obsolete code just ends up commented out instead of removed. In this case it is better to remove the code once you have checked you do not need it. The code might also have been commented out for debugging purposes. In this case it is vital that someone uncomments it again or your project may behave in very unexpected ways in production. This check looks for comments that seem to be mostly valid code and reports them. Loading history...
124		// * @param SearchRequest $request
125		// * @param array $arr
126		// */
127		// private function improveSearchWildcardQueries(SearchRequest $request, &$arr) {
128		//
129		// $queries = $request->getWildcardQueries();
130		// foreach ($queries as $query) {
131		// $wildcards = [];
132		// foreach ($query as $entry) {
133		// $wildcards[] = ['wildcard' => $entry];
134		// }
135		//
136		// array_push($arr['bool']['must']['bool']['should'], $wildcards);
137		// }
138		//
139		// }
140
141
142		/**
143		* @param SearchRequest $request
144		* @param array $arr
145		*/
146	View Code Duplication	private function improveSearchWildcardFilters(SearchRequest $request, &$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...
147
148		$filters = $request->getWildcardFilters();
149		foreach ($filters as $filter) {
150		$wildcards = [];
151		foreach ($filter as $entry) {
152		$wildcards[] = ['wildcard' => $entry];
153		}
154
155		$arr['bool']['filter'][]['bool']['should'] = $wildcards;
156		}
157
158		}
159
160
161		/**
162		* @param SearchRequest $request
163		* @param array $arr
164		*/
165	View Code Duplication	private function improveSearchRegexFilters(SearchRequest $request, &$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...
166
167		$filters = $request->getRegexFilters();
168		foreach ($filters as $filter) {
169		$regex = [];
170		foreach ($filter as $entry) {
171		$regex[] = ['regexp' => $entry];
172		}
173
174		$arr['bool']['filter'][]['bool']['should'] = $regex;
175		}
176
177		}
178
179
180		/**
181		* @param SearchRequest $request
182		*
183		* @return array<string,array<string,array>>
		0 ignored issues – show Documentation introduced 2018-05-17 10:42 UTC by Report Bug Copy Issue Report Should the return type not be `array<string,array>`? This check compares the return type specified in the `@return` annotation of a function or method doc comment with the types returned by the function and raises an issue if they mismatch. Loading history...
184		* @throws SearchQueryGenerationException
185		*/
186		private function generateSearchQueryContent(SearchRequest $request) {
187		$str = strtolower($request->getSearch());
188
189		preg_match_all('/[^?]"(?:\\\\.\|[^\\\\"])*"\|\S+/', " $str ", $words);
190		$queryContent = [];
191		foreach ($words[0] as $word) {
192		try {
193		$queryContent[] = $this->generateQueryContent(trim($word));
194		} catch (QueryContentGenerationException $e) {
195		continue;
196		}
197		}
198
199		if (sizeof($queryContent) === 0) {
200		throw new SearchQueryGenerationException();
201		}
202
203		return $this->generateSearchQueryFromQueryContent($request, $queryContent);
204		}
205
206
207		/**
208		* @param string $word
209		*
210		* @return QueryContent
211		* @throws QueryContentGenerationException
212		*/
213		private function generateQueryContent($word) {
214
215		$searchQueryContent = new QueryContent($word);
216		if (strlen($searchQueryContent->getWord()) === 0) {
217		throw new QueryContentGenerationException();
218		}
219
220		return $searchQueryContent;
221		}
222
223
224		/**
225		* @param SearchRequest $request
226		* @param QueryContent[] $queryContents
227		*
228		* @return array
		0 ignored issues – show Documentation introduced 2018-05-17 10:42 UTC by Report Bug Copy Issue Report Consider making the return type a bit more specific; maybe use `array<string,array>`. This check looks for the generic type `array` as a return type and suggests a more specific type. This type is inferred from the actual code. Loading history...
229		*/
230		private function generateSearchQueryFromQueryContent(SearchRequest $request, $queryContents) {
231
232		$query = $queryWords = [];
233		foreach ($queryContents as $queryContent) {
234		$queryWords[$queryContent->getShould()][] =
235		$this->generateQueryContentFields($request, $queryContent);
236		}
237
238		$listShould = array_keys($queryWords);
239		foreach ($listShould as $itemShould) {
240		$query[$itemShould][] = $queryWords[$itemShould];
241		}
242
243		return ['bool' => $query];
244		}
245
246
247		/**
248		* @param SearchRequest $request
249		* @param QueryContent $content
250		*
251		* @return array
		0 ignored issues – show Documentation introduced 2018-06-04 13:39 UTC by Report Bug Copy Issue Report Consider making the return type a bit more specific; maybe use `array<string,array<string,array>>`. This check looks for the generic type `array` as a return type and suggests a more specific type. This type is inferred from the actual code. Loading history...
252		*/
253		private function generateQueryContentFields(SearchRequest $request, QueryContent $content) {
254		$parts = array_map(
255		function($value) {
256		return 'parts.' . $value;
257		}, $request->getParts()
258		);
259		$fields = array_merge(['content', 'title'], $request->getFields(), $parts);
260
261		$queryFields = [];
262		foreach ($fields as $field) {
263		if (!$this->fieldIsOutLimit($request, $field)) {
264		$queryFields[] = [$content->getMatch() => [$field => $content->getWord()]];
265		}
266		}
267
268		foreach ($request->getWildcardFields() as $field) {
269		if (!$this->fieldIsOutLimit($request, $field)) {
270		$queryFields[] = ['wildcard' => [$field => '' . $content->getWord() . '']];
271		}
272		}
273
274		return ['bool' => ['should' => $queryFields]];
275		}
276
277
278		/**
279		* @param DocumentAccess $access
280		*
281		* @return array<string,array>
		0 ignored issues – show Documentation introduced 2018-01-10 22:16 UTC by Report Bug Copy Issue Report Should the return type not be `array<string,array>[]`? This check compares the return type specified in the `@return` annotation of a function or method doc comment with the types returned by the function and raises an issue if they mismatch. Loading history...
282		*/
283		private function generateSearchQueryAccess(DocumentAccess $access) {
284
285		$query = [];
286		$query[] = ['term' => ['owner' => $access->getViewerId()]];
287		$query[] = ['term' => ['users' => $access->getViewerId()]];
288		$query[] = ['term' => ['users' => '__all']];
289
290		foreach ($access->getGroups() as $group) {
291		$query[] = ['term' => ['groups' => $group]];
292		}
293
294		foreach ($access->getCircles() as $circle) {
295		$query[] = ['term' => ['circles' => $circle]];
296		}
297
298		return $query;
299		}
300
301
302		/**
303		* @param SearchRequest $request
304		* @param string $field
305		*
306		* @return bool
307		*/
308		private function fieldIsOutLimit(SearchRequest $request, $field) {
309		$limit = $request->getLimitFields();
310		if (sizeof($limit) === 0) {
311		return false;
312		}
313
314		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...
315		return false;
316		}
317
318		return true;
319		}
320
321
322		/**
323		* @param array $tags
324		*
325		* @return array<string,array>
326		*/
327		private function generateSearchQueryTags($tags) {
328
329		$query = [];
330		foreach ($tags as $tag) {
331		$query[] = ['term' => ['tags' => $tag]];
332		}
333
334		return $query;
335		}
336
337		private function generateSearchHighlighting() {
338		return [
339		'fields' => ['content' => new \stdClass()],
340		'pre_tags' => [''],
341		'post_tags' => ['']
342		];
343		}
344
345
346		}
347

nextcloud / fulltextsearch_elasticsearch

Push — master ( a799ec...765778 )

generateSearchQueryFromQueryContent() A

Complexity

Size

Duplication

Importance

Duplication Side-by-Side

Filter issues like