SolrCellTextExtractor

Complexity

Total Complexity

11

Size/Duplication

Total Lines	90
Duplicated Lines	0 %

Coupling/Cohesion

Components	1
Dependencies	4

Importance

Changes	5
Bugs	0	Features	1

6 Methods

Rating	Name	Duplication	Size	Complexity
A	getHttpClient()	0	10	3
A	setHttpClient()	0	4	1
A	isAvailable()	0	5	1
A	supportsExtension()	0	11	1
A	supportsMime()	0	5	1
B	getContent()	0	34	4

<?php
use Guzzle\Http\Client;

/**
 * Text extractor that calls an Apache Solr instance
 * and extracts content via the "ExtractingRequestHandler" endpoint.
 * Does not alter the Solr index itself, but uses it purely
 * for its file parsing abilities.
 * 
 * @author ischommer
 * @see  http://wiki.apache.org/solr/ExtractingRequestHandler
 */
class SolrCellTextExtractor extends FileTextExtractor

PSR1 recommends that each class must be in a namespace of at least one level to avoid collisions.

{
    /**
     * Base URL to use for solr text extraction.
     * E.g. http://localhost:8983/solr/update/extract
     *
     * @config
     * @var string
     */
    private static $base_url;

The property $base_url is not used and could be removed.

    private static $priority = 75;

Consider using a different property name as you override a private property of the parent class.

The property $priority is not used and could be removed.

    protected $httpClient;

    public function getHttpClient()
    {
        if (!$this->config()->get('base_url')) {
            throw new InvalidArgumentException('SolrCellTextExtractor.base_url not specified');
        }
        if (!$this->httpClient) {
            $this->httpClient = new Client($this->config()->get('base_url'));
        }
        return $this->httpClient;
    }

    public function setHttpClient($client)
    {
        $this->httpClient = $client;
    }

    public function isAvailable()
    {
        $url = $this->config()->get('base_url');
        return (boolean) $url;
    }

    public function supportsExtension($extension)
    {
        return in_array(
            strtolower($extension),
            array(
                'pdf', 'doc', 'docx', 'xls', 'xlsx',
                'epub', 'rtf', 'odt', 'fodt', 'ods', 'fods',
                'ppt', 'pptx', 'odp', 'fodp', 'csv'
            )
        );
    }

    public function supportsMime($mime)
    {
        // Rely on supportsExtension
        return false;
    }
    
    public function getContent($path)
    {
        if (!$path) {
            return "";
        } // no file

        $fileName = basename($path);
        $client = $this->getHttpClient();
        try {
            $request = $client
                ->post()
                ->addPostFields(array('extractOnly' => 'true', 'extractFormat' => 'text'))
                ->addPostFiles(array('myfile' => $path));
            $response = $request->send();
        } catch (InvalidArgumentException $e) {
            SS_Log::log(
                sprintf(
                    'Error extracting text from "%s" (message: %s)',
                    $path,
                    $e->getMessage()
                ),
                SS_Log::NOTICE
            );
            return null;
        }
        // Use preg match to avoid SimpleXML running out of memory on large text nodes
        preg_match(
            sprintf('/\<str name\="%s"\>(.*?)\<\/str\>/s', preg_quote($fileName)),
            (string)$response->getBody(),
            $matches
        );

        return $matches ? $matches[1] : null;
    }
}