Code Coverage |
||||||||||
Lines |
Functions and Methods |
Classes and Traits |
||||||||
| Total | |
88.89% |
32 / 36 |
|
66.67% |
2 / 3 |
CRAP | |
0.00% |
0 / 1 |
| LabelsField | |
88.89% |
32 / 36 |
|
66.67% |
2 / 3 |
12.20 | |
0.00% |
0 / 1 |
| getMapping | |
82.61% |
19 / 23 |
|
0.00% |
0 / 1 |
4.08 | |||
| getFieldData | |
100.00% |
3 / 3 |
|
100.00% |
1 / 1 |
2 | |||
| getLabelsIndexedData | |
100.00% |
10 / 10 |
|
100.00% |
1 / 1 |
6 | |||
| 1 | <?php |
| 2 | namespace Wikibase\Search\Elastic\Fields; |
| 3 | |
| 4 | use CirrusSearch\CirrusSearch; |
| 5 | use SearchEngine; |
| 6 | use Wikibase\DataModel\Entity\EntityDocument; |
| 7 | use Wikibase\DataModel\Term\AliasesProvider; |
| 8 | use Wikibase\DataModel\Term\LabelsProvider; |
| 9 | |
| 10 | /** |
| 11 | * Field which contains per-language specific labels. |
| 12 | * |
| 13 | * @license GPL-2.0-or-later |
| 14 | * @author Stas Malyshev |
| 15 | */ |
| 16 | class LabelsField extends TermIndexField implements WikibaseLabelsIndexField { |
| 17 | |
| 18 | use LabelsDescriptionsFieldTrait; |
| 19 | |
| 20 | /** |
| 21 | * Field name |
| 22 | */ |
| 23 | public const NAME = "labels"; |
| 24 | |
| 25 | /** |
| 26 | * @param SearchEngine $engine |
| 27 | * @return array |
| 28 | */ |
| 29 | public function getMapping( SearchEngine $engine ) { |
| 30 | // Since we need a specially tuned field, we can not use |
| 31 | // standard search engine types. |
| 32 | if ( !( $engine instanceof CirrusSearch ) ) { |
| 33 | // For now only Cirrus/Elastic is supported |
| 34 | return []; |
| 35 | } |
| 36 | |
| 37 | $config = [ |
| 38 | 'type' => 'object', |
| 39 | 'properties' => [] |
| 40 | ]; |
| 41 | foreach ( $this->languages as $language ) { |
| 42 | if ( empty( $this->stemmingSettings[$language]['index'] ) ) { |
| 43 | $langConfig = $this->getUnindexedField(); |
| 44 | } else { |
| 45 | $langConfig = $this->getTokenizedSubfield( $engine->getConfig(), |
| 46 | $language . '_text', |
| 47 | $language . '_text_search' |
| 48 | ); |
| 49 | } |
| 50 | |
| 51 | $langConfig['fields']['prefix'] = |
| 52 | $this->getSubfield( 'prefix_asciifolding', 'near_match_asciifolding' ); |
| 53 | $langConfig['fields']['near_match_folded'] = |
| 54 | $this->getSubfield( 'near_match_asciifolding' ); |
| 55 | $langConfig['fields']['near_match'] = $this->getSubfield( 'near_match' ); |
| 56 | // This one is for full-text search, will tokenize |
| 57 | // TODO: here we probably will need better language-specific analyzers |
| 58 | $langConfig['fields']['plain'] = $this->getTokenizedSubfield( $engine->getConfig(), |
| 59 | $language . '_plain', $language . '_plain_search' ); |
| 60 | // All labels are copies to labels_all |
| 61 | $langConfig['copy_to'] = 'labels_all'; |
| 62 | |
| 63 | $config['properties'][$language] = $langConfig; |
| 64 | } |
| 65 | |
| 66 | return $config; |
| 67 | } |
| 68 | |
| 69 | /** |
| 70 | * @param EntityDocument $entity |
| 71 | * |
| 72 | * @return mixed Get the value of the field to be indexed when a page/document |
| 73 | * is indexed. This might be an array with nested data, if the field |
| 74 | * is defined with nested type or an int or string for simple field types. |
| 75 | */ |
| 76 | public function getFieldData( EntityDocument $entity ) { |
| 77 | if ( !( $entity instanceof LabelsProvider ) ) { |
| 78 | return null; |
| 79 | } |
| 80 | return $this->getLabelsIndexedData( $entity ); |
| 81 | } |
| 82 | |
| 83 | /** |
| 84 | * @inheritDoc |
| 85 | */ |
| 86 | public function getLabelsIndexedData( LabelsProvider $entity ) { |
| 87 | $data = []; |
| 88 | foreach ( $entity->getLabels() as $language => $label ) { |
| 89 | $data[$language][] = $label->getText(); |
| 90 | } |
| 91 | if ( $entity instanceof AliasesProvider ) { |
| 92 | foreach ( $entity->getAliasGroups() as $aliases ) { |
| 93 | $language = $aliases->getLanguageCode(); |
| 94 | if ( !isset( $data[$language] ) ) { |
| 95 | $data[$language][] = ''; |
| 96 | } |
| 97 | $data[$language] = array_merge( $data[$language], $aliases->getAliases() ); |
| 98 | } |
| 99 | } |
| 100 | // Shouldn't return empty arrays, that will be encoded to json as an |
| 101 | // empty list instead of an empty map. Elastic doesn't mind, but this |
| 102 | // allows more consistency working with the resulting cirrus docs |
| 103 | return $data ?: null; |
| 104 | } |
| 105 | } |