elastic_indexer

<?php

namespace Elastic;
use Elastica;


class Indexer extends Base {
    public $batchSize = 100;
    public $indexSettings = array(
        'analysis' => array(
            'analyzer' => array(
                self::INDEX_ANALYZER => array(
                    'type' => 'custom',
                    'tokenizer' => 'standard',
                    'filter' => array('lowercase', 'russian_morphology', 'english_morphology'),
                    'char_filter' => array('html_strip')
                ),
                self::SEARCH_ANALYZER => array(
                    'type' => 'custom',
                    'tokenizer' => 'standard',
                    'filter' => array('lowercase', 'russian_morphology', 'english_morphology')
                ),
                self::SEARCH_STRICT_ANALYZER => array(
                    'type' => 'custom',
                    'tokenizer' => 'standard',
                    'filter' => array('lowercase')
                )
            )
        )
    );
    public $mappingSettings = array(
        /*'_all'=>array(
            'analyzer'=>self::INDEX_ANALYZER,
            'type'=>'string',
        ),*/
        'title'=>array(
            'type'=>'string',
            'include_in_all'=>true,
            'analyzer'=>self::INDEX_ANALYZER,
            'boost'=>100
        ),
        'content'=>array(
            'type'=>'string',
            'include_in_all'=>true,
            'analyzer'=>self::INDEX_ANALYZER,
            'boost'=>1
        ),
        'url'=>array(
            'type'=>'string',
            "index"=>"not_analyzed",
            'include_in_all'=>false,
        ),
    );


    function initServer($host, $port, $indexName, $typeName)
    {
        parent::initServer($host, $port, $indexName, $typeName);

        $this->index = $this->client->getIndex($indexName);
        $this->index->create($this->indexSettings, true);

        $this->type = $this->index->getType($typeName);
    }

    function initSettings($batchSize) {
        $this->batchSize = $batchSize;

        $this->initMapping();
    }

    /**
     * @param $docsOrClosure - if is array we are just adding it but if is closure we are calling it and add docs from result <br>
     *
     * closure = function($offset, $limit)  return $docs
     */
    function run($docsOrClosure) {
        // array
        if (is_array($docsOrClosure)) {
            $this->addDocsFromArray($docsOrClosure);
        }
        // closure
        else {
            $this->addDocsByPackages($docsOrClosure);
        }

        $this->index->refresh();
    }

    function delete() {
        $this->index->delete();
    }

    protected function addDocsByPackages($closure) {
        $offset = 0;

        while (true) {
            $docs = $closure($offset, $this->batchSize);
            $docsNumber = count($docs);
            $offset += $docsNumber;

            if ($docsNumber < 1) {
                break;
            }

            $this->addDocsFromArray($docs);
        }
    }

    protected function addDocsFromArray($array) {
        $docs = array();
        $fields = $this->getFields();

        foreach ($array as $item) {
            $data = array();

            foreach ($fields as $name=>$field) {
                $data[$name] = $item[$name];
            }

            $docs[] = new Elastica\Document('', $data);
        }

        $this->addDocs($docs);
    }

    protected function addDocs($docs) {
        $this->type->addDocuments($docs);
    }

    protected function initMapping() {
        $mapping = new Elastica\Type\Mapping();
        $mapping->setType( $this->type );
        $mapping->setProperties( $this->mappingSettings );

        $this->type->setMapping($mapping);
    }

    protected function getFields() {
        $fields = $this->mappingSettings;
        unset($fields['_all']);

        return $fields;
    }


}