search #

Executes a search on the collection.

There is a small delay between the time a document is created and its availability in our search layer (usually a couple of seconds). That means that a document that was just created might not be returned immediately by this function.

Processing large data sets #

When processing a large number of documents (i.e. more than 1000), using search is not always the best option.

Pagination of results can be done by using the from and size but the cost becomes prohibitive when deep pagination is reached. In fact, Elasticsearch, Kuzzle's embedded database, limits results to 10,000 records by default.

Instead, the recommended way to process a large number of documents is to use Collection.scroll or, easier, SearchResult.fetchNext.

See SearchResult.fetchNext for an example of how to process every document of a collection.

search(body, [options], callback) #

Arguments	Type	Description
`body`	JSON object	Search request body, using ElasticSearch Query DSL format. If given an empty object, matches all documents in the collection
`options`	JSON object	Optional parameters
`callback`	function	Callback handling the response

Options #

Option	Type	Description	Default
`from`	number	Provide the starting offset of the request (used to paginate results)	`0`
`queuable`	boolean	Make this request queuable or not	`true`
`scroll`	string	Start a scroll session, with a time to live equals to this parameter's value following the Elastisearch time format	`undefined`
`size`	number	Provide the maximum number of results of the request (used to paginate results)	`10`

To get more information about scroll sessions, please refer to the API reference documentation.

Callback Response #

Returns an instance of SearchResult.

Usage #

<?php
use \Kuzzle\Kuzzle;
use \Kuzzle\Document;
use \Kuzzle\Util\SearchResult;
$body = [
  'query' => [
    'bool' => [
      'must' => [
        [
          'terms' => ['status' => ['idle', 'wantToHire', 'toHire', 'riding']]
        ],
        ['term' => 'cab'],
        [
          'geo_distance' => [
            'distance' => '10km',
            'pos' => [
              'lat' => '48.8566140',
              'lon' => '2.352222'
            ]
          ]
        ]
      ]
    ]
  ],
  'sort' => [
    'status',
    [
      '_geo_distance' => [
        'pos' => ['lat' => '48.8566140', 'lon' => '2.352222'],
        'order' => 'asc'
      ]
    ],
    ['date' => 'desc']
  ],
  'aggregations' => [
    'aggs_name' => [
      'terms' => [
        'field' => 'field_name'
      ]
    ]
  ]
];
$options = [
  'from' => 0,
  'size' => 20
];
$kuzzle = new Kuzzle('localhost');
$dataCollection = $kuzzle->collection('collection', 'index');
try {
  $searchResult = $dataCollection->search($body, $options);
  // $searchResult instanceof SearchResult
  $searchResult->getTotal();
  foreach($searchResult->getDocuments() as $document) {
    // $document instanceof Document
  }
  // return an array representing the aggregations response
  $searchResult->getAggregations();
}
catch (ErrorException $e) {
}

Edit this page on Github