Indexer.rst

Indexer

The indexer allows collecting data that has to be indexed.

Indexing annotations

The package introduces a new annotation called "Indexable". Use this annotation to define what objects/entities you want to have indexed. If you annotate a class, all supported properties [1] will be mapped. If you annotated single properties, only these will be indexed. However, you have to annotate the class in every case.

Example: Class where every property will be indexed

/**
 * @Flow\Entity
 * @ElasticSearch\Indexable("twitter", typeName="tweet")
 */
class Tweet {

        /**
         * @var string
         */
        protected $username;

        /**
         * @var string
         */
        protected $message;

        /**
         * @var \DateTime
         */
        protected $date;
}

Example: Class where only the ``message`` property will be indexed

/**
 * @Flow\Entity
 * @ElasticSearch\Indexable("twitter", typeName="tweet")
 */
class Tweet {

        /**
         * @var string
         */
        protected $username;

        /**
         * @var string
         * @ElasticSearch\Indexable
         */
        protected $message;

        /**
         * @var \DateTime
         */
        protected $date;
}

Mapping annotations

ElasticSearch allows the mapping configuration done via annotations. See the example how to define mapping annotations:

Example: Annotations to set up mapping directives

/**
 * @var string
 * @ElasticSearch\Mapping(boost=2.0, term_vector="with_offsets")
 */
protected $username;

/**
 * @var string
 */
protected $message;

/**
 * @var \DateTime
 * @ElasticSearch\Mapping(format="YYYY-MM-dd")
 */
protected $date;

Note that for mapping creation, the type will automatically be determined from the PHP type the property is of.

Value transformations

For some properties it'll be necessary to conduct specific conversions in order to meet the requirements of ElasticSearch. Declare custom type converters via their appropriate annotation:

/**
 * @var \DateTime
 * @ElasticSearch\Mapping(format="YYYY-MM-dd")
 * @ElasticSearch\Transform("Date")
 */
protected $date;

This will call the (supplied with the package) Date transformer and hand the converted value over to the ElasticSearch engine.

Setting up the indexes

As soon as you have proper configuration for your entities, you can create your index, with the CLI utility:

flow index:create --index-name twitter

If you need advanced settings you can define them in your Settings.yaml:

Flowpack:
  ElasticSearch:
        indexes:
          default:
                'twitter':
                  analysis:
                        filter:
                          elision:
                                type: 'elision'
                                articles: [ 'l', 'm', 't', 'qu', 'n', 's', 'j', 'd' ]
                  analyzer:
                        custom_french_analyzer:
                          tokenizer: 'letter'
                          filter: [ 'asciifolding', 'lowercase', 'french_stem', 'elision', 'stop' ]
                        tag_analyzer:
                          tokenizer: 'keyword'
                          filter: [ 'asciifolding', 'lowercase' ]

If you use multiple client configurations, please change the default key just below the indexes.

You can update the index configuration with the following CLI:

flow index:updateSettings --index-name twitter

Please check the ElasticSearch configuration to know witch settings are updatable. For any other settings changes, you need to delete your indexes:

flow index:delete --index-name twitter

[1]	supported properties are all scalar types, unless value transformation is applied.