You are here

feeds_httpfetcher_append_headers_request.inc in Feeds HTTPFetcher Append Headers 7

Download via HTTP with custom headers appended.

File

libraries/feeds_httpfetcher_append_headers_request.inc
View source
<?php

/**
 * @file
 * Download via HTTP with custom headers appended.
 */

/**
 * Get the content from the given URL.
 *
 * @param string $url
 *   A valid URL (not only web URLs).
 * @param string $username
 *   If the URL uses authentication, supply the username.
 * @param string $password
 *   If the URL uses authentication, supply the password.
 * @param bool $accept_invalid_cert
 *   Whether to accept invalid certificates.
 * @param integer $timeout
 *   Timeout in seconds to wait for an HTTP get request to finish.
 * @param array $append_headers
 *   Custom request headers that should be appended.
 *
 * @return stdClass
 *   An object that describes the data downloaded from $url.
 */
function feeds_httpfetcher_append_headers_request_get($url, $username = NULL, $password = NULL, $accept_invalid_cert = FALSE, $timeout = NULL, $append_headers = NULL) {

  // Intra-pagedownload cache, avoid to download the same content twice within
  // one page download (it's possible, compatible and parse calls).
  static $download_cache = array();
  if (isset($download_cache[$url])) {
    return $download_cache[$url];
  }

  // Determine request timeout.
  $request_timeout = !empty($timeout) ? $timeout : variable_get('http_request_timeout', 30);
  if (!$username && valid_url($url, TRUE)) {

    // Handle password protected feeds.
    $url_parts = parse_url($url);
    if (!empty($url_parts['user'])) {
      $password = $url_parts['pass'];
      $username = $url_parts['user'];
    }
  }
  $curl = http_request_use_curl();

  // Only download and parse data if really needs refresh.
  // Based on "Last-Modified" and "If-Modified-Since".
  $headers = array();
  $headers = array_merge($headers, $append_headers);
  if ($cache = cache_get('feeds_http_download_' . md5($url))) {
    $last_result = $cache->data;
    $last_headers = array_change_key_case($last_result->headers);
    if (!empty($last_headers['etag'])) {
      if ($curl) {
        $headers[] = 'If-None-Match: ' . $last_headers['etag'];
      }
      else {
        $headers['If-None-Match'] = $last_headers['etag'];
      }
    }
    if (!empty($last_headers['last-modified'])) {
      if ($curl) {
        $headers[] = 'If-Modified-Since: ' . $last_headers['last-modified'];
      }
      else {
        $headers['If-Modified-Since'] = $last_headers['last-modified'];
      }
    }
    if (!empty($username) && !$curl) {
      $headers['Authorization'] = 'Basic ' . base64_encode("{$username}:{$password}");
    }
  }

  // Support the 'feed' and 'webcal' schemes by converting them into 'http'.
  $url = strtr($url, array(
    'feed://' => 'http://',
    'webcal://' => 'http://',
  ));
  if ($curl) {
    $headers[] = 'User-Agent: Drupal (+http://drupal.org/)';
    $result = new stdClass();
    $result->headers = array();

    // Parse the URL and make sure we can handle the schema.
    // cURL can only support either http:// or https://.
    // CURLOPT_PROTOCOLS is only supported with cURL 7.19.4
    $uri = parse_url($url);
    if (!isset($uri['scheme'])) {
      $result->error = 'missing schema';
      $result->code = -1002;
    }
    else {
      switch ($uri['scheme']) {
        case 'http':
        case 'https':

          // Valid scheme.
          break;
        default:
          $result->error = 'invalid schema ' . $uri['scheme'];
          $result->code = -1003;
          break;
      }
    }

    // If the scheme was valid, continue to request the feed using cURL.
    if (empty($result->error)) {
      $download = curl_init($url);
      curl_setopt($download, CURLOPT_FOLLOWLOCATION, TRUE);
      if (!empty($username)) {
        curl_setopt($download, CURLOPT_USERPWD, "{$username}:{$password}");
        curl_setopt($download, CURLOPT_HTTPAUTH, CURLAUTH_ANY);
      }
      curl_setopt($download, CURLOPT_HTTPHEADER, $headers);
      curl_setopt($download, CURLOPT_HEADER, TRUE);
      curl_setopt($download, CURLOPT_RETURNTRANSFER, TRUE);
      curl_setopt($download, CURLOPT_ENCODING, '');
      curl_setopt($download, CURLOPT_TIMEOUT, $request_timeout);
      $proxy_server = variable_get('proxy_server');
      if ($proxy_server && _drupal_http_use_proxy($uri['host'])) {
        curl_setopt($download, CURLOPT_PROXY, $proxy_server);
        curl_setopt($download, CURLOPT_PROXYPORT, variable_get('proxy_port', 8080));

        // Proxy user/password.
        if ($proxy_username = variable_get('proxy_username')) {
          $username_password = $proxy_username . ':' . variable_get('proxy_password', '');
          curl_setopt($download, CURLOPT_PROXYUSERPWD, $username_password);
          curl_setopt($download, CURLOPT_PROXYAUTH, variable_get('proxy_auth_method', CURLAUTH_BASIC));
        }
      }
      if ($accept_invalid_cert) {
        curl_setopt($download, CURLOPT_SSL_VERIFYPEER, 0);
      }
      $header = '';
      $data = curl_exec($download);
      if (curl_error($download)) {
        throw new HRCurlException(t('cURL error (@code) @error for @url', array(
          '@code' => curl_errno($download),
          '@error' => curl_error($download),
          '@url' => $url,
        )), curl_errno($download));
      }

      // When using a proxy, remove extra data from the header which is not
      // considered by CURLINFO_HEADER_SIZE (possibly cURL bug).
      // This data is only added when to HTTP header when working with a proxy.
      // Example string added: <HTTP/1.0 200 Connection established\r\n\r\n>
      if ($proxy_server && _drupal_http_use_proxy($uri['host'])) {
        $http_header_break = "\r\n\r\n";
        $response = explode($http_header_break, $data);
        if (count($response) > 2) {
          $data = substr($data, strlen($response[0] . $http_header_break), strlen($data));
        }
      }
      $header_size = curl_getinfo($download, CURLINFO_HEADER_SIZE);
      $header = substr($data, 0, $header_size - 1);
      $result->data = substr($data, $header_size);
      $headers = preg_split("/(\r\n){2}/", $header);
      $header_lines = preg_split("/\r\n|\n|\r/", end($headers));

      // Skip HTTP response status.
      array_shift($header_lines);
      while ($line = trim(array_shift($header_lines))) {
        list($header, $value) = explode(':', $line, 2);

        // Normalize the headers.
        $header = strtolower($header);
        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);
        }
      }
      $result->code = curl_getinfo($download, CURLINFO_HTTP_CODE);
      curl_close($download);
    }
  }
  else {
    $result = drupal_http_request($url, array(
      'headers' => $headers,
      'timeout' => $request_timeout,
    ));
    $result->headers = isset($result->headers) ? $result->headers : array();
  }
  $result->code = isset($result->code) ? $result->code : 200;

  // In case of 304 Not Modified try to return cached data.
  if ($result->code == 304) {
    if (isset($last_result)) {
      $last_result->from_cache = TRUE;
      return $last_result;
    }
    else {

      // It's a tragedy, this file must exist and contain good data.
      // In this case, clear cache and repeat.
      cache_clear_all('feeds_http_download_' . md5($url), 'cache');
      return feeds_httpfetcher_append_headers_request_get($url, $username, $password, $accept_invalid_cert, $request_timeout);
    }
  }

  // Set caches.
  cache_set('feeds_http_download_' . md5($url), $result);
  $download_cache[$url] = $result;
  return $result;
}

Functions

Namesort descending Description
feeds_httpfetcher_append_headers_request_get Get the content from the given URL.