ElasticScout

A Laravel Scout Driver for Elasticsearch 6

Overview

ElasticScout is a Laravel Scout Elasticsearch 6 compatible engine. It makes critical changes to the old Elasticseach Scout Engine, as well as adds new functionality.

The ElasticScout engine includes an Elasticsearch Query Builder which can be used to create elaborate custom queries and aggregations, allowing full use of Elasticsearch within the Laravel/Scout Paradigm.

License

ElasticScout is released under the MIT Open Source License, https://opensource.org/licenses/MIT

Copyright

ElasticScout © Broker Exchange Network 2018

Installation

• Run composer require command composer require brokerexchange/elasticscout
• Configure Elasticsearch Host (default: localhost:9200)

   ELASTICSEARCH_HOST='elastic1.host.com:9200,elastic2.host.com:9200'

• Add trait to desired model

   use ElasticScout\Searchable;

Usage

   //create search/query object
   $search = $article->search()
       ->boolean()
       ->should(DSL::match('title',$request->input('query')))
       ->should(DSL::match('body',$request->input('query')))
       ->highlight(['body','title'])
       ->filter(DSL::term('published', 1))
       ->aggregate(Agg::terms('categories', 'category.name'));
       
   //fire the search
   $articles = $search->paginate();
       
   //retrieve aggregation results
   $categories = $search->aggregation('categories');
   
   //retrieve highlight results for title field of first result article
   $firstArticleTitleHighlights = $articles->first()->highlight('title');
   

Mappings

You may set a custom mapping by simply defining a "mapping" method on your model.

   public function mappings()
   {
       return [
           $this->searchableType() => [
               'properties' => [
                   'name' => [
                       'type' => 'text',
                       'fields' => [
                           'keyword' => [
                               'type' => 'keyword',
                           ],
                           'autocomplete' => [
                               'type' => 'text',
                               'analyzer' => 'autocomplete',
                               'search_analyzer' => 'autocomplete_search',
                           ],
                       ],
                   ],
               ]
           ]
       ]
   }

Settings

You may create custom settings and analyzers by creating a "settings" method on the model.

   public function settings()
   {
     return [
         'index' => [
             'analysis' => [
                 'analyzer' => [
                     'autocomplete' => [
                         'tokenizer' => 'autocomplete',
                         'filter' => [
                             'lowercase',
                         ],
                     ],
                     'autocomplete_search' => [
                         'tokenizer' => 'lowercase',
                     ],
                 ],
                 'tokenizer' => [
                     'autocomplete' => [
                         'type' => 'edge_ngram',
                         'min_gram' => 1,
                         'max_gram' => 15,
                         'token_chars' => [
                             'letter'
                         ]
                     ]
                 ]
             ],
         ],
     ];
   }