public function FeedsQueryPathParser::parse in Feeds QueryPath Parser 7

Same name and namespace in other branches

1. 6 FeedsQueryPathParser.inc \FeedsQueryPathParser::parse()

Implements FeedsParser::parse().

File

./FeedsQueryPathParser.inc, line 14

Provides the class for FeedsQueryPathParser.

Class

FeedsQueryPathParser

@file

Code

public function parse(FeedsSource $source, FeedsFetcherResult $fetcher_result) {

  // Setup mappings so they can be used in variable replacement.
  $mappings = $this
    ->getOwnMappings();

  // Set source config, if it's empty get config from importer.
  $this->source_config = $source
    ->getConfigFor($this);

  // Allow config inheritance.
  if (empty($this->source_config)) {
    $this->source_config = $this
      ->getConfig();
  }
  $this->rawXML = array_keys(array_filter($this->source_config['rawXML']));
  $this->debug = array_keys(array_filter($this->source_config['debug']['options']));
  $raw = trim($fetcher_result
    ->getRaw());
  if (empty($raw)) {
    throw new Exception(t('Feeds QueryPath parser: The document is empty.'));
  }
  $opts = array(
    'ignore_parser_warnings' => TRUE,
  );
  $result = new FeedsParserResult();

  // Set link so we can set the result link attribute.
  $fetcher_config = $source
    ->getConfigFor($source->importer->fetcher);
  $result->link = $fetcher_config['source'];
  $this
    ->includeQueryPath();
  $doc = @qp($raw, NULL, $opts);

  // Convert document to UTF-8
  $ContentType = qp($doc, 'meta[http-equiv="content-type"]');
  if ($ContentType
    ->hasAttr('content') && preg_match('/charset=([-\\w]*)/i', $ContentType
    ->attr('content'), $matches)) {
    $ContentType
      ->attr('content', preg_replace('/charset=([-\\w]*)/i', 'charset=utf-8', $ContentType
      ->attr('content')));
    qp($doc, 'meta[http-equiv="content-type"]')
      ->remove();
    qp($doc, 'head')
      ->prepend($ContentType
      ->html());
    $doc = qp(drupal_convert_to_utf8(utf8_decode($doc
      ->html()), $matches[1]), NULL, $opts);
  }
  $result->title = qp($doc, 'title', $opts)
    ->text();
  $context = qp($doc, $this->source_config['context'], $opts);
  $this
    ->debug($context, 'context');
  foreach ($context as $item) {
    $parsed_item = $variables = array();
    foreach ($this->source_config['sources'] as $source => $query) {

      // Variable substitution.
      $query = strtr($query, $variables);
      $parsed = $this
        ->parseSourceElement($item, $query, $source);

      // Avoid null values.
      if (isset($parsed)) {

        // Variable sunstitution can't handle arrays.
        if (!is_array($parsed)) {
          $variables['{' . $mappings[$source] . '}'] = $parsed;
        }
        else {
          $variables['{' . $mappings[$source] . '}'] = '';
        }
        $parsed_item[$source] = $parsed;
      }
    }
    if (!empty($parsed_item)) {
      $result->items[] = $parsed_item;
    }
  }
  return $result;
}