Elasticsearch compilation target for RulerZ.

elastic/elasticsearch-php is one of the targets supported by RulerZ. It allows the engine to query an Elasticsearch server.

This cookbook will show you how to retrieve objects using the official client for Elasticsearch and RulerZ.

Here is a summary of what you will have to do:

• configure elastic/elasticsearch-php;
• configure RulerZ;
• filter your target.

This subject won't be directly treated here. You can either follow the official documentation or use a bundle/module/whatever the framework you're using promotes.

Once elastic/elasticelasticsearch-php is installed and configured we can the RulerZ engine:

$rulerz = new RulerZ(
    $compiler, [
        new \RulerZ\Elasticsearch\Target\Elasticsearch(), // this line is Elasticsearch-specific
        // other compilation targets...
    ]
);

The only Elasticsearch-related configuration is the Elasticsearch target being added to the list of the known compilation targets.

Now that both elastic/elasticsearch-php and RulerZ are ready, you can use them to retrieve data.

The Elasticsearch instance that we previously injected into the RulerZ engine only knows how to use Elasticsearch\Client objects so the first step is creating one:

$client = new Elasticsearch\Client();

And as usual, we call RulerZ with our target (the Search object) and our rule. RulerZ will build the right executor for the given target and use it to filter the data, or in our case to retrieve data from Elasticsearch.

$rule  = 'gender = :gender and points > :points';
$parameters = [
    'points' => 30,
    'gender' => 'M',
];
$executionContext = [
    'index' => 'index_name',
    'type'  => 'type_name',
];
var_dump(
    iterator_to_array($rulerz->filter($client, $rule, $parameters, $executionContext))
);

N.B: you'll notice an unusual variable named $executionContext. It contains a few parameters needed by the Elasticsearch in order to make the request and are mandatory.

That's it!

This library is under the MIT license.