feeds_imagegrabber.module in Feeds Image Grabber 6

Grabs image for each feed-item from their respective web pages and stores it in an image field. Requires Feeds module.

File

feeds_imagegrabber.module

<?php

/**
 * @file
 *  Grabs image for each feed-item from their respective web pages and stores
 *  it in an image field. Requires Feeds module.
 *
 */

/**
 * Error code indicating that the request made by feeds_imagegrabber_http_request() exceeded
 * the specified timeout.
 */
define('FIG_HTTP_REQUEST_TIMEOUT', 1);

//=============

//DRUPAL HOOKS.

//=============

/**
 * Implementation of hook_menu().
 */
function feeds_imagegrabber_menu() {
  $items = array();
  $items['admin/settings/feeds_imagegrabber'] = array(
    'title' => 'Feeds Image Grabber',
    'page callback' => 'drupal_get_form',
    'page arguments' => array(
      'feeds_imagegrabber_admin',
    ),
    'access arguments' => array(
      'administer site configuration',
    ),
    'description' => 'Configure default options for Feeds Image Grabber',
  );
  return $items;
}

/**
 * Implementation of hook_form_alter().
 */
function feeds_imagegrabber_form_alter(&$form, $form_state, $form_id) {
  if ($form['#id'] == 'node-form') {
    if ($importer = feeds_get_importer_id($form['type']['#value'])) {

      // Use the values from $form_state if available
      if (isset($form_state['values']['feeds_imagegrabber'])) {
        $settings = $form_state['values']['feeds_imagegrabber'];
      }
      elseif (!isset($form['#node']->nid) || ($settings = feeds_imagegrabber_get_settings($form['#node']->nid)) === FALSE) {
        $settings = feeds_imagegrabber_get_default_settings();
      }
      $form['feeds_imagegrabber'] = array(
        '#type' => 'fieldset',
        '#title' => t('Feeds Image Grabber'),
        '#tree' => TRUE,
        '#collapsible' => TRUE,
        '#collapsed' => TRUE,
      );
      feeds_imagegrabber_form($form, $settings);
      $form['#validate'][] = 'feeds_imagegrabber_form_validate';
    }
  }
}
function feeds_imagegrabber_form_validate($form, &$form_state) {
  if ($form_state['values']['feeds_imagegrabber']['enabled'] == 1) {
    $id_class = $form_state['values']['feeds_imagegrabber']['id_class'];
    $id_class_desc = $form_state['values']['feeds_imagegrabber']['id_class_desc'];
    if ($id_class) {
      if (!isset($id_class_desc) || empty($id_class_desc) || $id_class_desc == '') {
        form_set_error('feeds_imagegrabber][id_class_desc', "Specify the id/class of the desired tag.");
      }
      if ($id_class == 1 && !preg_match('/^[a-zA-Z]+[_a-zA-Z0-9-]*$/', $id_class_desc)) {
        form_set_error('feeds_imagegrabber][id_class_desc', "Only alphabets, digits, hyphens and underscores are allowed in HTML id");
      }
      else {
        if ($id_class == 2 && !preg_match('/^[a-zA-Z]+[_a-zA-Z0-9- ]*$/', $id_class_desc)) {
          form_set_error('feeds_imagegrabber][id_class_desc', "Only alphabets, digits, hyphens, spaces and underscores are allowed in HTML class");
        }
      }
    }
    else {
      form_set_value($form['feeds_imagegrabber']['id_class_desc'], '', $form_state);
    }
    $temp = $form_state['values']['feeds_imagegrabber']['exec_time'];
    if (!is_numeric($temp) || $temp < 10 || $temp != round($temp) || $temp > 75) {
      form_set_error('feeds_imagegrabber][exec_time', t('Select the correct option for FIG execution time.'));
    }
  }
}

/**
 * Implementation of hook_nodeapi().
 */
function feeds_imagegrabber_nodeapi(&$node, $op, $form) {
  switch ($op) {
    case 'insert':
    case 'update':
      if ($importer_id = feeds_get_importer_id($node->type)) {
        if ($node->feeds_imagegrabber['enabled']) {
          $settings = array(
            'feed_nid' => $node->nid,
            'enabled' => $node->feeds_imagegrabber['enabled'],
            'id_class' => $node->feeds_imagegrabber['id_class'],
            'id_class_desc' => $node->feeds_imagegrabber['id_class_desc'],
            'exec_time' => $node->feeds_imagegrabber['exec_time'],
            'feeling_lucky' => $node->feeds_imagegrabber['feeling_lucky'],
          );

          // Make sure a source record is present at all time, try to update first,
          // then insert.
          drupal_write_record('feeds_imagegrabber', $settings, array(
            'feed_nid',
          ));
          if (!db_affected_rows()) {
            drupal_write_record('feeds_imagegrabber', $settings);
          }
        }
        else {
          db_query("UPDATE {feeds_imagegrabber} SET enabled = %d WHERE feed_nid = %d", 0, $node->nid);
        }
      }
      break;
    case 'delete':
      @db_query("DELETE FROM {feeds_imagegrabber} where feed_nid = %d", $node->nid);
      break;
  }
}

//==================

// FEEDS HOOKS

//==================

/**
 * Implementation of hook_feeds_node_processor_targets_alter().
 */
function feeds_imagegrabber_feeds_node_processor_targets_alter($targets, $content_type) {
  $info = content_types($content_type);
  $fields = array();
  if (isset($info['fields']) && count($info['fields'])) {
    foreach ($info['fields'] as $field_name => $field) {
      if (isset($field['widget']['type']) && $field['widget']['type'] == 'imagefield_widget') {
        $name = isset($field['widget']['label']) ? $field['widget']['label'] : $field_name;
        $targets[$field_name . ':fig'] = array(
          'name' => $name . ' (FIG)',
          'callback' => 'feeds_imagegrabber_feeds_set_target',
          'description' => t('The Item URL for the CCK @name imagefield of the node.', array(
            '@name' => $name,
          )),
        );
      }
    }
  }
}

/**
 * Implementation of hook_feeds_set_target().
 *
 * @param $node
 *   The target node.
 * @param $target
 *   The name of field on the target node to map to.
 * @param $page_url
 *   This should contain a valid URL for the feed item, from
 *   which the target image is to be grabbed.
 */
function feeds_imagegrabber_feeds_set_target($node, $target, $page_url) {
  $feed_nid = $node->feeds_node_item->feed_nid;
  $settings = feeds_imagegrabber_get_settings($feed_nid);
  if (!$settings || !$settings['enabled']) {
    return;
  }
  if (!feeds_imagegrabber_include_library('url_to_absolute.php', 'feeds_imagegrabber')) {
    drupal_set_message(t('Feeds Image Grabber: The URL conversion script is missing. Go to <a href="!admin-reports-status">Status Report page</a>', array(
      '!admin-reports-status' => url('admin/reports/status'),
    )), 'error');
    return FALSE;
  }
  list($field_name) = split(':', $target);
  $field = content_fields($field_name, $node->type);
  $max_filesize = parse_size(file_upload_max_size());
  if (!empty($field['widget']['max_filesize_per_file']) && parse_size($field['widget']['max_filesize_per_file']) < $max_filesize) {
    $max_filesize = parse_size($field['widget']['max_filesize_per_file']);
  }
  $max_exec_time = ini_get('max_execution_time');
  $timeout = $max_exec_time == 0 ? 10 : $settings['exec_time'] * $max_exec_time / 100;
  $page_time = timer_read('page') / 1000;
  if (function_exists('encode_url')) {
    $page_url = encode_url($page_url);
  }
  if (valid_url($page_url)) {
    $xml = feeds_imagegrabber_webpage_scraper($page_url, $settings['id_class'], $settings['id_class_desc'], $timeout);
    if ($xml == FALSE) {
      return;
    }
    $timeout = $timeout - timer_read('page') / 1000 + $page_time;
    $options = array(
      'max_imagesize' => $max_filesize,
      'timeout' => $timeout,
      'feeling_lucky' => $settings['feeling_lucky'],
    );
    $images = feeds_imagegrabber_scrape_images($xml, $page_url, $options);
    if ($images == FALSE || count($images) <= 0) {
      return;
    }
    asort($images);
    $images = array_reverse($images, TRUE);

    // Map enclosures.
    $items = isset($node->{$field_name}) ? $node->{$field_name} : array();
    foreach ($images as $url => $size) {
      $enclosure = new FeedsEnclosure($url, 'application/octet-stream');
      if (($file = $enclosure
        ->getFile()) && ($file = feeds_imagegrabber_is_image($file))) {
        $target_dir = filefield_widget_file_path($field, user_load($node->uid));
        $file_validators = filefield_widget_upload_validators($field);
        $image_validators = imagefield_widget_upload_validators($field);
        $validators = array_merge($file_validators, $image_validators);
        if (array_key_exists('filefield_validate_is_image', $validators)) {
          unset($validators['filefield_validate_is_image']);
        }
        $info = field_file_save_file($file, $validators, $target_dir, user_load($node->uid));
        if ($info) {
          $info['list'] = array();
          $info['data'] = array(
            'description' => '',
          );
          if ($field['list_field']) {
            $info['list'] = $field['list_default'];
          }
          $items[] = $info;
          break;
        }
      }
    }
    $node->{$field_name} = $items;
  }
}

//==================

//HELPER FUNCTIONS

//==================

/**
 * Retrieve settings for a feed node from the database.
 *
 * @param $feed_nid
 *   The nid of the feed node.
 *
 * @return
 *   An array of settings or FALSE if settings not found.
 *
 */
function feeds_imagegrabber_get_settings($feed_nid) {
  $settings = db_fetch_array(db_query("SELECT enabled, id_class, id_class_desc, feeling_lucky, exec_time FROM {feeds_imagegrabber} WHERE feed_nid = %d", $feed_nid));
  return $settings;
}

/**
 * Retrieve the default settings for a feed node from the database.
 *
 * @return
 *   An array of settings.
 *
 */
function feeds_imagegrabber_get_default_settings() {
  $default = array(
    'enabled' => 0,
    'id_class' => 0,
    'id_class_desc' => '',
    'exec_time' => 10,
    'feeling_lucky' => 0,
  );
  return variable_get('feeds_imagegrabber', $default);
}

/**
 * Implementation of the default settings admin form.
 */
function feeds_imagegrabber_admin(&$form_state) {
  $form = array();
  $settings = feeds_imagegrabber_get_default_settings();
  $form['feeds_imagegrabber'] = array(
    '#type' => 'fieldset',
    '#title' => t('Default Settings'),
    '#tree' => TRUE,
    '#collapsible' => FALSE,
  );
  feeds_imagegrabber_form($form, $settings);
  $form['#validate'][] = 'feeds_imagegrabber_form_validate';
  return system_settings_form($form);
}

/**
 * Appends the form with the Feeds Image Grabber form using the 
 * passed default settings.
 *
 * @param &$form
 *   The form to append under the 'feeds_imagegrabber' fieldset.
 * @param $default_settings
 *   The default values of the form elements.
 */
function feeds_imagegrabber_form(&$form, $default_settings) {
  $form['feeds_imagegrabber']['enabled'] = array(
    '#type' => 'checkbox',
    '#title' => t('Enable Feeds Image Grabber'),
    '#description' => t('Check if you want to download images of the feed items for this feed.'),
    '#default_value' => $default_settings['enabled'],
  );
  $form['feeds_imagegrabber']['id_class'] = array(
    '#type' => 'radios',
    '#title' => t('Search for an image between the tag which is identified by'),
    '#options' => array(
      t('None, search the whole web-page for the image.'),
      t('an Id'),
      t('a Class'),
    ),
    '#default_value' => $default_settings['id_class'],
  );
  $form['feeds_imagegrabber']['id_class_desc'] = array(
    '#type' => 'textfield',
    '#title' => t('<i>Id</i> or <i>Class</i> of the HTML tag (Leave empty if you selected <i>None</i> above.)'),
    '#default_value' => $default_settings['id_class_desc'],
    '#description' => t('Separate multiple classes with spaces (as present in the HTML)'),
    '#maxlength' => 100,
  );
  $form['feeds_imagegrabber']['feeling_lucky'] = array(
    '#type' => 'radios',
    '#title' => t('Feeling lucky, huh?'),
    '#options' => array(
      t('No, select the largest image between the tag.'),
      t('Yes, select the first image between the tag. (Recommended)'),
    ),
    '#default_value' => $default_settings['feeling_lucky'],
  );
  $form['feeds_imagegrabber']['exec_time'] = array(
    '#type' => 'select',
    '#title' => t('Execution time[%]'),
    '#options' => drupal_map_assoc(array(
      10,
      20,
      30,
      50,
      75,
    )),
    '#default_value' => $default_settings['exec_time'],
    '#description' => t('Select the percentage of maximum PHP execution time to take while grabbing image for a feed item.'),
  );
}

/**
 * Validates the size of an file accessible through a http url.
 *
 * @param $file_url
 *   A string specifying the formatted file url.
 * @param $max_size
 *   Maximum size of the file to be downloaded.
 * @param $timeout
 *   A float representing the maximum number of seconds the function call
 *   may take. The default is 10 seconds. If a timeout occurs, the retuen
 *   code is set to the FIG_HTTP_REQUEST_TIMEOUT constant.
 * @param $max_redirects
 *   An integer representing how many times a redirect may be followed.
 *   Defaults to 3.
 *
 * @return
 *   An integer code containing filesize in case the file exists and conforms to the
 *   size limit, -1 otherwise.
 *
 */
function feeds_imagegrabber_validate_download_size($file_url, $max_size, $timeout = 10, $max_redirects = 3) {
  $options = array(
    'headers' => array(),
    'method' => 'HEAD',
    'data' => NULL,
    'max_redirects' => $max_redirects,
    'timeout' => $timeout,
  );
  $result = feeds_imagegrabber_http_request($file_url, $options);
  if ($result->code == 200 && isset($result->headers) && is_array($result->headers)) {

    //Bug #882992, some servers may return keys with different case.
    $headers = array_change_key_case($result->headers);
    if (isset($headers['content-length']) && $headers['content-length'] <= $max_size) {
      return $headers['content-length'];
    }
  }
  return -1;
}

/**
 * Scrape the webpage using the id or the css class of a tag and returns the
 * HTML between the tag.
 *
 * @param $page_url
 *   A string specifying the page url to scrape. If there is a redirect, it is
 *   changed to the redirect_url.
 * @param $itype
 *   A positive integer value representing the identifier type for the tag:
 *     - 0 : selects content between <body> </body>.
 *     - 1 : selects content between the tag identified by an ID.
 *     - 2 : selects content between the first tag identified by a CSS class.
 * @param $ivalue
 *   A string specifying the ID or the CSS class.
 * @param $timeout
 *   A float representing the maximum number of seconds the function call
 *   may take. The default is 15 seconds. If a timeout occurs, the retuen
 *   code is set to the FIG_HTTP_REQUEST_TIMEOUT constant.
 * @param $max_redirects
 *   An integer representing how many times a redirect may be followed.
 *   Defaults to 3.
 * @param $error_log
 *   An array which contains the error codes and messages in case the functions fails.
 *
 * @return
 *   FALSE on failure, OR content between the tags as XML on success
 *
 */
function feeds_imagegrabber_webpage_scraper(&$page_url, $itype, $ivalue = '', $timeout = 15, $max_redirects = 3, &$error_log = array()) {
  $options = array(
    'headers' => array(),
    'method' => 'GET',
    'data' => NULL,
    'max_redirects' => $max_redirects,
    'timeout' => $timeout,
  );
  $result = feeds_imagegrabber_http_request($page_url, $options);
  if (in_array($result->redirect_code, array(
    301,
    302,
    307,
  ))) {
    $page_url = $result->redirect_url;
  }
  if ($result->code != 200) {
    $error_log['code'] = $result->code;
    $error_log['error'] = "unable to retrieve content from web page";
    return FALSE;
  }
  if (empty($result->data) || drupal_strlen($result->data) <= 0) {
    $error_log['code'] = -1;
    $error_log['error'] = "no data available on url";
    return FALSE;
  }
  $doc = new DOMDocument();
  if (@$doc
    ->loadHTML($result->data) === FALSE) {
    $error_log['code'] = -2;
    $error_log['error'] = "unable to parse the html content";
    return FALSE;
  }
  if ($itype == 0) {
    $items = @$doc
      ->getElementsByTagName("body");
    if ($items != NULL && $items->length > 0) {
      $dist = $items
        ->item(0);
    }
    else {
      $dist = NULL;
    }
  }
  elseif ($itype == 1) {
    $dist = @$doc
      ->getElementById($ivalue);
  }
  elseif ($itype == 2) {
    $xpath = new DOMXPath($doc);

    //Normalize whitespaces.
    $ivalue = preg_replace('/\\s\\s+/', ' ', trim($ivalue));
    $items = $xpath
      ->query("//*[@class and contains(concat(' ',normalize-space(@class),' '), ' {$ivalue} ')]");
    if ($items != NULL && $items->length > 0) {
      $dist = $items
        ->item(0);
    }
    else {
      $dist = NULL;
    }
  }
  else {

    //not supported yet
    $dist = NULL;
  }
  if ($dist == NULL) {
    $error_log['code'] = -3;
    $error_log['error'] = "tag not found";
    return FALSE;
  }
  $content = '';
  if (($content = @$dist->ownerDocument
    ->saveXML($dist)) === FALSE) {
    $error_log['code'] = -4;
    $error_log['error'] = "error converting content to XML";
    return FALSE;
  }
  return $content;
}
function feeds_imagegrabber_scrape_images($content, $base_url, array $options = array(), &$error_log = array()) {

  // Merge the default options.
  $options += array(
    'expression' => "//img",
    'getsize' => TRUE,
    'max_imagesize' => 512000,
    'timeout' => 10,
    'max_redirects' => 3,
    'feeling_lucky' => 0,
  );
  $doc = new DOMDocument();
  if (@$doc
    ->loadXML($content) === FALSE && @$doc
    ->loadHTML($content) === FALSE) {
    $error_log['code'] = -5;
    $error_log['error'] = "unable to parse the xml//html content";
    return FALSE;
  }
  $xpath = new DOMXPath($doc);
  $hrefs = @$xpath
    ->evaluate($options['expression']);
  if ($options['getsize']) {
    timer_start(__FUNCTION__);
  }
  $images = array();
  $imagesize = 0;
  for ($i = 0; $i < $hrefs->length; $i++) {
    $url = $hrefs
      ->item($i)
      ->getAttribute('src');
    if (!isset($url) || empty($url) || $url == '') {
      continue;
    }
    if (function_exists('encode_url')) {
      $url = encode_url($url);
    }
    $url = url_to_absolute($base_url, $url);
    if ($url == FALSE) {
      continue;
    }
    if ($options['getsize']) {
      if (($imagesize = feeds_imagegrabber_validate_download_size($url, $options['max_imagesize'], $options['timeout'] - timer_read(__FUNCTION__) / 1000)) != -1) {
        $images[$url] = $imagesize;
        if ($settings['feeling_lucky']) {
          break;
        }
      }
      if ($options['timeout'] - timer_read(__FUNCTION__) / 1000 <= 0) {
        $error_log['code'] = FIG_HTTP_REQUEST_TIMEOUT;
        $error_log['error'] = "timeout occured while scraping the content";
        break;
      }
    }
    else {
      $images[$url] = $imagesize;
      if ($settings['feeling_lucky']) {
        break;
      }
    }
  }
  return $images;
}

/**
 * Checks that a file is an image.
 * 
 * This check allows the image formats identified by drupal i.e. jpeg, png 
 * and gif. If the filename is missing an extension but is a valid image, 
 * its actual extension is added to it.
 *
 * @param $filepath
 *   The path to the image file.
 * 
 * @return
 *   The final path to the image file, it may be modified or FALSE on failure.
 */
function feeds_imagegrabber_is_image($filepath) {
  $extensions = 'jpeg jpg png gif';
  if ($filepath) {
    $info = feeds_imagegrabber_get_image_info($filepath);
    if ($info && !empty($info['extension'])) {
      if (!count(feeds_imagegrabber_validate_extensions($filepath, $extensions))) {
        return $filepath;
      }
      else {
        $basename = basename($filepath);
        $directory = dirname($filepath);
        if ($pos = strrpos($basename, '.')) {
          $name = substr($basename, 0, $pos);
          $ext = substr($basename, $pos);
          $regex = '/\\.(' . preg_replace('/ +/', '|', preg_quote($extensions)) . ')/i';
          if (preg_match($regex, $ext, $matches)) {
            $ext = $matches[1];
          }
          else {
            $ext = $info['extension'];
          }
          $basename = $name . '.' . $ext;
        }
        else {
          $basename .= '.' . $info['extension'];
        }
        if ($basename == basename($filepath)) {
          return $filepath;
        }
        $dest = $directory . '/' . $basename;
        if (rename($filepath, $dest)) {
          return $dest;
        }
      }
    }
  }
  return FALSE;
}

/**
 * Check that the filename ends with an allowed extension. This check is also
 * enforced for the user #1 unlike drupal's default file_validate_extensions.
 *
 * @param $filename
 *   A string containing the file name to validate
 * @param $extensions
 *   A string with a space separated
 * @return
 *   An array. If the file extension is not allowed, it will contain an error message.
 */
function feeds_imagegrabber_validate_extensions($filename, $extensions) {
  $errors = array();
  $regex = '/\\.(' . preg_replace('/ +/', '|', preg_quote($extensions)) . ')$/i';
  if (!preg_match($regex, $filename)) {
    $errors[] = t('Only files with the following extensions are allowed: %files-allowed.', array(
      '%files-allowed' => $extensions,
    ));
  }
  return $errors;
}

/**
 * Get details about an image.
 * Drupal only supports GIF, JPG and PNG file formats.
 * (Mostly from drupal's image.inc.)
 * 
 * @return
 *   FALSE, if the file could not be found or is not an image. Otherwise, a
 *   keyed array containing information about the image:
 *    'width'     - Width in pixels.
 *    'height'    - Height in pixels.
 *    'extension' - Commonly used file extension for the image.
 *    'mime_type' - MIME type ('image/jpeg', 'image/gif', 'image/png').
 */
function feeds_imagegrabber_get_image_info($file) {
  if (!is_file($file)) {
    return FALSE;
  }
  $details = FALSE;
  $data = @getimagesize($file);
  if (isset($data) && is_array($data)) {
    $extensions = array(
      '1' => 'gif',
      '2' => 'jpg',
      '3' => 'png',
    );
    $extension = array_key_exists($data[2], $extensions) ? $extensions[$data[2]] : '';
    $details = array(
      'width' => $data[0],
      'height' => $data[1],
      'extension' => $extension,
      'mime_type' => $data['mime'],
    );
  }
  return $details;
}

/**
 * Perform an HTTP request. Directly ported from D7 for the timeout feature.
 *
 * This is a flexible and powerful HTTP client implementation. Correctly
 * handles GET, POST, PUT or any other HTTP requests. Handles redirects.
 *
 * @param $url
 *   A string containing a fully qualified URI.
 * @param $options
 *   (optional) An array which can have one or more of following keys:
 *   - headers
 *       An array containing request headers to send as name/value pairs.
 *   - method
 *       A string containing the request method. Defaults to 'GET'.
 *   - data
 *       A string containing the request body. Defaults to NULL.
 *   - max_redirects
 *       An integer representing how many times a redirect may be followed.
 *       Defaults to 3.
 *   - timeout
 *       A float representing the maximum number of seconds the function call
 *       may take. The default is 30 seconds. If a timeout occurs, the error
 *       code is set to the FIG_HTTP_REQUEST_TIMEOUT constant.
 *
 * @return
 *   An object which can have one or more of the following parameters:
 *   - request
 *       A string containing the request body that was sent.
 *   - code
 *       An integer containing the response status code, or the error code if
 *       an error occurred.
 *   - protocol
 *       The response protocol (e.g. HTTP/1.1 or HTTP/1.0).
 *   - status_message
 *       The status message from the response, if a response was received.
 *   - redirect_code
 *       If redirected, an integer containing the initial response status code.
 *   - redirect_url
 *       If redirected, a string containing the redirection location.
 *   - error
 *       If an error occurred, the error message. Otherwise not set.
 *   - headers
 *       An array containing the response headers as name/value pairs.
 *   - data
 *       A string containing the response body that was received.
 */
function feeds_imagegrabber_http_request($url, array $options = array()) {
  global $db_prefix;
  $result = new stdClass();

  // Parse the URL and make sure we can handle the schema.
  $uri = @parse_url($url);
  if ($uri == FALSE) {
    $result->error = 'unable to parse URL';
    $result->code = -1001;
    return $result;
  }
  if (!isset($uri['scheme'])) {
    $result->error = 'missing schema';
    $result->code = -1002;
    return $result;
  }
  timer_start(__FUNCTION__);

  // Merge the default options.
  $options += array(
    'headers' => array(),
    'method' => 'GET',
    'data' => NULL,
    'max_redirects' => 3,
    'timeout' => 30,
  );
  switch ($uri['scheme']) {
    case 'http':
      $port = isset($uri['port']) ? $uri['port'] : 80;
      $host = $uri['host'] . ($port != 80 ? ':' . $port : '');
      $fp = @fsockopen($uri['host'], $port, $errno, $errstr, $options['timeout']);
      break;
    case 'https':

      // Note: Only works when PHP is compiled with OpenSSL support.
      $port = isset($uri['port']) ? $uri['port'] : 443;
      $host = $uri['host'] . ($port != 443 ? ':' . $port : '');
      $fp = @fsockopen('ssl://' . $uri['host'], $port, $errno, $errstr, $options['timeout']);
      break;
    default:
      $result->error = 'invalid schema ' . $uri['scheme'];
      $result->code = -1003;
      return $result;
  }

  // Make sure the socket opened properly.
  if (!$fp) {

    // When a network error occurs, we use a negative number so it does not
    // clash with the HTTP status codes.
    $result->code = -$errno;
    $result->error = trim($errstr);

    // Mark that this request failed. This will trigger a check of the web
    // server's ability to make outgoing HTTP requests the next time that
    // requirements checking is performed.
    // @see system_requirements()
    variable_set('drupal_http_request_fails', TRUE);
    return $result;
  }

  // Construct the path to act on.
  $path = isset($uri['path']) ? $uri['path'] : '/';
  if (isset($uri['query'])) {
    $path .= '?' . $uri['query'];
  }

  // Merge the default headers.
  $options['headers'] += array(
    'User-Agent' => 'Drupal (+http://drupal.org/)',
  );

  // RFC 2616: "non-standard ports MUST, default ports MAY be included".
  // We don't add the standard port to prevent from breaking rewrite rules
  // checking the host that do not take into account the port number.
  $options['headers']['Host'] = $host;

  // Only add Content-Length if we actually have any content or if it is a POST
  // or PUT request. Some non-standard servers get confused by Content-Length in
  // at least HEAD/GET requests, and Squid always requires Content-Length in
  // POST/PUT requests.
  $content_length = strlen($options['data']);
  if ($content_length > 0 || $options['method'] == 'POST' || $options['method'] == 'PUT') {
    $options['headers']['Content-Length'] = $content_length;
  }

  // If the server URL has a user then attempt to use basic authentication.
  if (isset($uri['user'])) {
    $options['headers']['Authorization'] = 'Basic ' . base64_encode($uri['user'] . (!empty($uri['pass']) ? ":" . $uri['pass'] : ''));
  }

  // If the database prefix is being used by SimpleTest to run the tests in a copied
  // database then set the user-agent header to the database prefix so that any
  // calls to other Drupal pages will run the SimpleTest prefixed database. The
  // user-agent is used to ensure that multiple testing sessions running at the
  // same time won't interfere with each other as they would if the database
  // prefix were stored statically in a file or database variable.
  if (is_string($db_prefix) && preg_match("/simpletest\\d+/", $db_prefix, $matches)) {
    $options['headers']['User-Agent'] = drupal_generate_test_ua($matches[0]);
  }
  $request = $options['method'] . ' ' . $path . " HTTP/1.0\r\n";
  foreach ($options['headers'] as $name => $value) {
    $request .= $name . ': ' . trim($value) . "\r\n";
  }
  $request .= "\r\n" . $options['data'];
  $result->request = $request;
  fwrite($fp, $request);

  // Fetch response.
  $response = '';
  while (!feof($fp)) {

    // Calculate how much time is left of the original timeout value.
    $timeout = $options['timeout'] - timer_read(__FUNCTION__) / 1000;
    if ($timeout <= 0) {
      $result->code = FIG_HTTP_REQUEST_TIMEOUT;
      $result->error = 'request timed out';
      return $result;
    }
    stream_set_timeout($fp, floor($timeout), floor(1000000 * fmod($timeout, 1)));
    $response .= fread($fp, 1024);
  }
  fclose($fp);

  // Parse response headers from the response body.
  list($response, $result->data) = explode("\r\n\r\n", $response, 2);
  $response = preg_split("/\r\n|\n|\r/", $response);

  // Parse the response status line.
  list($protocol, $code, $status_message) = explode(' ', trim(array_shift($response)), 3);
  $result->protocol = $protocol;
  $result->status_message = $status_message;
  $result->headers = array();

  // Parse the response headers.
  while ($line = trim(array_shift($response))) {
    list($header, $value) = explode(':', $line, 2);
    if (isset($result->headers[$header]) && $header == 'Set-Cookie') {

      // RFC 2109: the Set-Cookie response header comprises the token Set-
      // Cookie:, followed by a comma-separated list of one or more cookies.
      $result->headers[$header] .= ',' . trim($value);
    }
    else {
      $result->headers[$header] = trim($value);
    }
  }
  $responses = array(
    100 => 'Continue',
    101 => 'Switching Protocols',
    200 => 'OK',
    201 => 'Created',
    202 => 'Accepted',
    203 => 'Non-Authoritative Information',
    204 => 'No Content',
    205 => 'Reset Content',
    206 => 'Partial Content',
    300 => 'Multiple Choices',
    301 => 'Moved Permanently',
    302 => 'Found',
    303 => 'See Other',
    304 => 'Not Modified',
    305 => 'Use Proxy',
    307 => 'Temporary Redirect',
    400 => 'Bad Request',
    401 => 'Unauthorized',
    402 => 'Payment Required',
    403 => 'Forbidden',
    404 => 'Not Found',
    405 => 'Method Not Allowed',
    406 => 'Not Acceptable',
    407 => 'Proxy Authentication Required',
    408 => 'Request Time-out',
    409 => 'Conflict',
    410 => 'Gone',
    411 => 'Length Required',
    412 => 'Precondition Failed',
    413 => 'Request Entity Too Large',
    414 => 'Request-URI Too Large',
    415 => 'Unsupported Media Type',
    416 => 'Requested range not satisfiable',
    417 => 'Expectation Failed',
    500 => 'Internal Server Error',
    501 => 'Not Implemented',
    502 => 'Bad Gateway',
    503 => 'Service Unavailable',
    504 => 'Gateway Time-out',
    505 => 'HTTP Version not supported',
  );

  // RFC 2616 states that all unknown HTTP codes must be treated the same as the
  // base code in their class.
  if (!isset($responses[$code])) {
    $code = floor($code / 100) * 100;
  }
  $result->code = $code;
  switch ($code) {
    case 200:

    // OK
    case 304:

      // Not modified
      break;
    case 301:

    // Moved permanently
    case 302:

    // Moved temporarily
    case 307:

      // Moved temporarily
      $location = $result->headers['Location'];
      $options['timeout'] -= timer_read(__FUNCTION__) / 1000;
      if ($options['timeout'] <= 0) {
        $result->code = FIG_HTTP_REQUEST_TIMEOUT;
        $result->error = 'request timed out';
      }
      elseif ($options['max_redirects']) {

        // Redirect to the new location.
        $options['max_redirects']--;
        $result = feeds_imagegrabber_http_request($location, $options);
        $result->redirect_code = $code;
      }
      $result->redirect_url = $location;
      break;
    default:
      $result->error = $status_message;
  }
  return $result;
}

/**
 * Includes a required library file.
 *
 * @param $file
 *   The filename to load from.
 * @param $library
 *   The name of the library. If libraries module is installed,
 *   feeds_imagegrabber_include_library() will look for libraries 
 *   with this name managed by libraries module.
 */
function feeds_imagegrabber_include_library($file, $library) {
  if (module_exists('libraries') && file_exists(libraries_get_path($library) . "/{$file}")) {
    require_once libraries_get_path($library) . "/{$file}";
    return TRUE;
  }
  else {
    $paths = array(
      drupal_get_path('module', 'feeds_imagegrabber'),
      drupal_get_path('module', 'feeds_imagegrabber') . "/libraries",
      'sites/all/libraries',
      'sites/all/libraries/feeds_imagegrabber',
      'sites/all/libraries/absoluteurl',
      'sites/all/libraries/AbsoluteUrl',
    );
    foreach ($paths as $library_path) {
      $path = $library_path . "/{$file}";
      if (file_exists($path)) {
        require_once $path;
        return TRUE;
      }
    }
  }
  return FALSE;
}

Functions

Name	Description
feeds_imagegrabber_admin	Implementation of the default settings admin form.
feeds_imagegrabber_feeds_node_processor_targets_alter	Implementation of hook_feeds_node_processor_targets_alter().
feeds_imagegrabber_feeds_set_target	Implementation of hook_feeds_set_target().
feeds_imagegrabber_form	Appends the form with the Feeds Image Grabber form using the passed default settings.
feeds_imagegrabber_form_alter	Implementation of hook_form_alter().
feeds_imagegrabber_form_validate
feeds_imagegrabber_get_default_settings	Retrieve the default settings for a feed node from the database.
feeds_imagegrabber_get_image_info	Get details about an image. Drupal only supports GIF, JPG and PNG file formats. (Mostly from drupal's image.inc.)
feeds_imagegrabber_get_settings	Retrieve settings for a feed node from the database.
feeds_imagegrabber_http_request	Perform an HTTP request. Directly ported from D7 for the timeout feature.
feeds_imagegrabber_include_library	Includes a required library file.
feeds_imagegrabber_is_image	Checks that a file is an image.
feeds_imagegrabber_menu	Implementation of hook_menu().
feeds_imagegrabber_nodeapi	Implementation of hook_nodeapi().
feeds_imagegrabber_scrape_images
feeds_imagegrabber_validate_download_size	Validates the size of an file accessible through a http url.
feeds_imagegrabber_validate_extensions	Check that the filename ends with an allowed extension. This check is also enforced for the user #1 unlike drupal's default file_validate_extensions.
feeds_imagegrabber_webpage_scraper	Scrape the webpage using the id or the css class of a tag and returns the HTML between the tag.

Constants

Name	Description
FIG_HTTP_REQUEST_TIMEOUT	Error code indicating that the request made by feeds_imagegrabber_http_request() exceeded the specified timeout.