1 <?php

     2 

     3 /**

     4  * @file

     5  * Parser functions for the aggregator module.

     6  */

     7 

     8 /**

     9  * Implements hook_aggregator_parse_info().

    10  */

    11 function aggregator_aggregator_parse_info() {

    12   return array(

    13     'title' => t('Default parser'),

    14     'description' => t('Parses RSS, Atom and RDF feeds.'),

    15   );

    16 }

    17 

    18 /**

    19  * Implements hook_aggregator_parse().

    20  */

    21 function aggregator_aggregator_parse($feed) {

    22   global $channel, $image;

    23 

    24   // Filter the input data.

    25   if (aggregator_parse_feed($feed->source_string, $feed)) {

    26     $modified = empty($feed->http_headers['last-modified']) ? 0 : strtotime($feed->http_headers['last-modified']);

    27 

    28     // Prepare the channel data.

    29     foreach ($channel as $key => $value) {

    30       $channel[$key] = trim($value);

    31     }

    32 

    33     // Prepare the image data (if any).

    34     foreach ($image as $key => $value) {

    35       $image[$key] = trim($value);

    36     }

    37 

    38     $etag = empty($feed->http_headers['etag']) ? '' : $feed->http_headers['etag'];

    39 

    40     // Add parsed data to the feed object.

    41     $feed->link = !empty($channel['link']) ? $channel['link'] : '';

    42     $feed->description = !empty($channel['description']) ? $channel['description'] : '';

    43     $feed->image = !empty($image['url']) ? $image['url'] : '';

    44     $feed->etag = $etag;

    45     $feed->modified = $modified;

    46 

    47     // Clear the cache.

    48     cache_clear_all();

    49 

    50     return TRUE;

    51   }

    52 

    53   return FALSE;

    54 }

    55 

    56 /**

    57  * Parses a feed and stores its items.

    58  *

    59  * @param $data

    60  *   The feed data.

    61  * @param $feed

    62  *   An object describing the feed to be parsed.

    63  *

    64  * @return

    65  *   FALSE on error, TRUE otherwise.

    66  */

    67 function aggregator_parse_feed(&$data, $feed) {

    68   global $items, $image, $channel;

    69 

    70   // Unset the global variables before we use them.

    71   unset($GLOBALS['element'], $GLOBALS['item'], $GLOBALS['tag']);

    72   $items = array();

    73   $image = array();

    74   $channel = array();

    75 

    76   // Parse the data.

    77   $xml_parser = drupal_xml_parser_create($data);

    78   xml_set_element_handler($xml_parser, 'aggregator_element_start', 'aggregator_element_end');

    79   xml_set_character_data_handler($xml_parser, 'aggregator_element_data');

    80 

    81   if (!xml_parse($xml_parser, $data, 1)) {

    82     watchdog('aggregator', 'The feed from %site seems to be broken, due to an error "%error" on line %line.', array('%site' => $feed->title, '%error' => xml_error_string(xml_get_error_code($xml_parser)), '%line' => xml_get_current_line_number($xml_parser)), WATCHDOG_WARNING);

    83     drupal_set_message(t('The feed from %site seems to be broken, because of error "%error" on line %line.', array('%site' => $feed->title, '%error' => xml_error_string(xml_get_error_code($xml_parser)), '%line' => xml_get_current_line_number($xml_parser))), 'error');

    84     return FALSE;

    85   }

    86   xml_parser_free($xml_parser);

    87 

    88   // We reverse the array such that we store the first item last, and the last

    89   // item first. In the database, the newest item should be at the top.

    90   $items = array_reverse($items);

    91 

    92   // Initialize items array.

    93   $feed->items = array();

    94   foreach ($items as $item) {

    95 

    96     // Prepare the item:

    97     foreach ($item as $key => $value) {

    98       $item[$key] = trim($value);

    99     }

   100 

   101     // Resolve the item's title. If no title is found, we use up to 40

   102     // characters of the description ending at a word boundary, but not

   103     // splitting potential entities.

   104     if (!empty($item['title'])) {

   105       $item['title'] = $item['title'];

   106     }

   107     elseif (!empty($item['description'])) {

   108       $item['title'] = preg_replace('/^(.*)[^\w;&].*?$/', "\\1", truncate_utf8($item['description'], 40));

   109     }

   110     else {

   111       $item['title'] = '';

   112     }

   113 

   114     // Resolve the items link.

   115     if (!empty($item['link'])) {

   116       $item['link'] = $item['link'];

   117     }

   118     else {

   119       $item['link'] = $feed->link;

   120     }

   121 

   122     // Atom feeds have an ID tag instead of a GUID tag.

   123     if (!isset($item['guid'])) {

   124       $item['guid'] = isset($item['id']) ? $item['id'] : '';

   125     }

   126 

   127     // Atom feeds have a content and/or summary tag instead of a description tag.

   128     if (!empty($item['content:encoded'])) {

   129       $item['description'] = $item['content:encoded'];

   130     }

   131     elseif (!empty($item['summary'])) {

   132       $item['description'] = $item['summary'];

   133     }

   134     elseif (!empty($item['content'])) {

   135       $item['description'] = $item['content'];

   136     }

   137 

   138     // Try to resolve and parse the item's publication date.

   139     $date = '';

   140     foreach (array('pubdate', 'dc:date', 'dcterms:issued', 'dcterms:created', 'dcterms:modified', 'issued', 'created', 'modified', 'published', 'updated') as $key) {

   141       if (!empty($item[$key])) {

   142         $date = $item[$key];

   143         break;

   144       }

   145     }

   146 

   147     $item['timestamp'] = strtotime($date);

   148 

   149     if ($item['timestamp'] === FALSE) {

   150       $item['timestamp'] = aggregator_parse_w3cdtf($date); // Aggregator_parse_w3cdtf() returns FALSE on failure.

   151     }

   152 

   153     // Resolve dc:creator tag as the item author if author tag is not set.

   154     if (empty($item['author']) && !empty($item['dc:creator'])) {

   155       $item['author'] = $item['dc:creator'];

   156     }

   157 

   158     $item += array('author' => '', 'description' => '');

   159 

   160     // Store on $feed object. This is where processors will look for parsed items.

   161     $feed->items[] = $item;

   162   }

   163 

   164   return TRUE;

   165 }

   166 

   167 /**

   168  * Performs an action when an opening tag is encountered.

   169  *

   170  * Callback function used by xml_parse() within aggregator_parse_feed().

   171  */

   172 function aggregator_element_start($parser, $name, $attributes) {

   173   global $item, $element, $tag, $items, $channel;

   174 

   175   $name = strtolower($name);

   176   switch ($name) {

   177     case 'image':

   178     case 'textinput':

   179     case 'summary':

   180     case 'tagline':

   181     case 'subtitle':

   182     case 'logo':

   183     case 'info':

   184       $element = $name;

   185       break;

   186     case 'id':

   187     case 'content':

   188       if ($element != 'item') {

   189         $element = $name;

   190       }

   191     case 'link':

   192       // According to RFC 4287, link elements in Atom feeds without a 'rel'

   193       // attribute should be interpreted as though the relation type is

   194       // "alternate".

   195       if (!empty($attributes['HREF']) && (empty($attributes['REL']) || $attributes['REL'] == 'alternate')) {

   196         if ($element == 'item') {

   197           $items[$item]['link'] = $attributes['HREF'];

   198         }

   199         else {

   200           $channel['link'] = $attributes['HREF'];

   201         }

   202       }

   203       break;

   204     case 'item':

   205       $element = $name;

   206       $item += 1;

   207       break;

   208     case 'entry':

   209       $element = 'item';

   210       $item += 1;

   211       break;

   212   }

   213 

   214   $tag = $name;

   215 }

   216 

   217 /**

   218  * Performs an action when a closing tag is encountered.

   219  *

   220  * Callback function used by xml_parse() within aggregator_parse_feed().

   221  */

   222 function aggregator_element_end($parser, $name) {

   223   global $element;

   224 

   225   switch ($name) {

   226     case 'image':

   227     case 'textinput':

   228     case 'item':

   229     case 'entry':

   230     case 'info':

   231       $element = '';

   232       break;

   233     case 'id':

   234     case 'content':

   235       if ($element == $name) {

   236         $element = '';

   237       }

   238   }

   239 }

   240 

   241 /**

   242  * Performs an action when data is encountered.

   243  *

   244  * Callback function used by xml_parse() within aggregator_parse_feed().

   245  */

   246 function aggregator_element_data($parser, $data) {

   247   global $channel, $element, $items, $item, $image, $tag;

   248   $items += array($item => array());

   249   switch ($element) {

   250     case 'item':

   251       $items[$item] += array($tag => '');

   252       $items[$item][$tag] .= $data;

   253       break;

   254     case 'image':

   255     case 'logo':

   256       $image += array($tag => '');

   257       $image[$tag] .= $data;

   258       break;

   259     case 'link':

   260       if ($data) {

   261         $items[$item] += array($tag => '');

   262         $items[$item][$tag] .= $data;

   263       }

   264       break;

   265     case 'content':

   266       $items[$item] += array('content' => '');

   267       $items[$item]['content'] .= $data;

   268       break;

   269     case 'summary':

   270       $items[$item] += array('summary' => '');

   271       $items[$item]['summary'] .= $data;

   272       break;

   273     case 'tagline':

   274     case 'subtitle':

   275       $channel += array('description' => '');

   276       $channel['description'] .= $data;

   277       break;

   278     case 'info':

   279     case 'id':

   280     case 'textinput':

   281       // The sub-element is not supported. However, we must recognize

   282       // it or its contents will end up in the item array.

   283       break;

   284     default:

   285       $channel += array($tag => '');

   286       $channel[$tag] .= $data;

   287   }

   288 }

   289 

   290 /**

   291  * Parses the W3C date/time format, a subset of ISO 8601.

   292  *

   293  * PHP date parsing functions do not handle this format. See

   294  * http://www.w3.org/TR/NOTE-datetime for more information. Originally from

   295  * MagpieRSS (http://magpierss.sourceforge.net/).

   296  *

   297  * @param $date_str

   298  *   A string with a potentially W3C DTF date.

   299  *

   300  * @return

   301  *   A timestamp if parsed successfully or FALSE if not.

   302  */

   303 function aggregator_parse_w3cdtf($date_str) {

   304   if (preg_match('/(\d{4})-(\d{2})-(\d{2})T(\d{2}):(\d{2})(:(\d{2}))?(?:([-+])(\d{2}):?(\d{2})|(Z))?/', $date_str, $match)) {

   305     list($year, $month, $day, $hours, $minutes, $seconds) = array($match[1], $match[2], $match[3], $match[4], $match[5], $match[6]);

   306     // Calculate the epoch for current date assuming GMT.

   307     $epoch = gmmktime($hours, $minutes, $seconds, $month, $day, $year);

   308     if ($match[10] != 'Z') { // Z is zulu time, aka GMT

   309       list($tz_mod, $tz_hour, $tz_min) = array($match[8], $match[9], $match[10]);

   310       // Zero out the variables.

   311       if (!$tz_hour) {

   312         $tz_hour = 0;

   313       }

   314       if (!$tz_min) {

   315         $tz_min = 0;

   316       }

   317       $offset_secs = (($tz_hour * 60) + $tz_min) * 60;

   318       // Is timezone ahead of GMT?  If yes, subtract offset.

   319       if ($tz_mod == '+') {

   320         $offset_secs *= -1;

   321       }

   322       $epoch += $offset_secs;

   323     }

   324     return $epoch;

   325   }

   326   else {

   327     return FALSE;

   328   }

   329 }