| 1 |
<?php
|
| 2 |
// $Id$
|
| 3 |
|
| 4 |
function feedapi_aggregator_refresh($feed) {
|
| 5 |
global $channel, $image;
|
| 6 |
// Generate conditional GET headers.
|
| 7 |
$headers = array();
|
| 8 |
|
| 9 |
if ($feed['etag']) {
|
| 10 |
$headers['If-None-Match'] = $feed['etag'];
|
| 11 |
}
|
| 12 |
if ($feed['modified']) {
|
| 13 |
$headers['If-Modified-Since'] = gmdate('D, d M Y H:i:s', $feed['modified']) .' GMT';
|
| 14 |
}
|
| 15 |
|
| 16 |
// Request feed.
|
| 17 |
$result = drupal_http_request($feed['url'], $headers);
|
| 18 |
|
| 19 |
// Process HTTP response code.
|
| 20 |
switch ($result->code) {
|
| 21 |
case 304:
|
| 22 |
// This is done by feedapi db_query('UPDATE {aggregator_feed} SET checked = %d WHERE fid = %d', time(), $feed['fid']);
|
| 23 |
drupal_set_message(t('There is no new syndicated content from %site.', array('%site' => $feed['title'])));
|
| 24 |
break;
|
| 25 |
case 301:
|
| 26 |
$feed['url'] = $result->redirect_url;
|
| 27 |
watchdog('aggregator', t('Updated URL for feed %title to %url.', array('%title' => $feed['title'], '%url' => $feed['url'])));
|
| 28 |
|
| 29 |
case 200:
|
| 30 |
case 302:
|
| 31 |
case 307:
|
| 32 |
// Filter the input data:
|
| 33 |
if ($parsed_source = feedapi_aggregator_parse_feed($result->data, $feed)) {
|
| 34 |
if ($result->headers['Last-Modified']) {
|
| 35 |
$modified = strtotime($result->headers['Last-Modified']);
|
| 36 |
}
|
| 37 |
|
| 38 |
/*
|
| 39 |
** Prepare the channel data:
|
| 40 |
*/
|
| 41 |
|
| 42 |
foreach ($channel as $key => $value) {
|
| 43 |
$channel[$key] = trim($value);
|
| 44 |
}
|
| 45 |
|
| 46 |
/*
|
| 47 |
** Prepare the image data (if any):
|
| 48 |
*/
|
| 49 |
|
| 50 |
foreach ($image as $key => $value) {
|
| 51 |
$image[$key] = trim($value);
|
| 52 |
}
|
| 53 |
|
| 54 |
if ($image['LINK'] && $image['URL'] && $image['TITLE']) {
|
| 55 |
// Note, we should really use theme_image() here but that only works with local images it won't work with images fetched with a URL unless PHP version > 5
|
| 56 |
$image = '<a href="'. check_url($image['LINK']) .'" class="feed-image"><img src="'. check_url($image['URL']) .'" alt="'. check_plain($image['TITLE']) .'" /></a>';
|
| 57 |
}
|
| 58 |
else {
|
| 59 |
$image = NULL;
|
| 60 |
}
|
| 61 |
|
| 62 |
/*
|
| 63 |
** Update the feed data:
|
| 64 |
*/
|
| 65 |
|
| 66 |
db_query("UPDATE {aggregator_feed} SET url = '%s', checked = %d, link = '%s', description = '%s', image = '%s', etag = '%s', modified = %d WHERE fid = %d", $feed['url'], time(), $channel['LINK'], $channel['DESCRIPTION'], $image, $result->headers['ETag'], $modified, $feed['fid']);
|
| 67 |
|
| 68 |
/*
|
| 69 |
** Clear the cache:
|
| 70 |
*/
|
| 71 |
|
| 72 |
cache_clear_all();
|
| 73 |
|
| 74 |
watchdog('aggregator', t('There is new syndicated content from %site.', array('%site' => $feed['title'])));
|
| 75 |
drupal_set_message(t('There is new syndicated content from %site.', array('%site' => $feed['title'])));
|
| 76 |
|
| 77 |
return $parsed_source;
|
| 78 |
}
|
| 79 |
break;
|
| 80 |
default:
|
| 81 |
watchdog('aggregator', t('The feed from %site seems to be broken, due to "%error".', array('%site' => $feed['title'], '%error' => $result->code .' '. $result->error)), WATCHDOG_WARNING);
|
| 82 |
drupal_set_message(t('The feed from %site seems to be broken, because of error "%error".', array('%site' => $feed['title'], '%error' => $result->code .' '. $result->error)));
|
| 83 |
}
|
| 84 |
}
|
| 85 |
|
| 86 |
|
| 87 |
/**
|
| 88 |
* Call-back function used by the XML parser.
|
| 89 |
*/
|
| 90 |
function feedapi_aggregator_element_start($parser, $name, $attributes) {
|
| 91 |
global $item, $element, $tag, $items, $channel;
|
| 92 |
|
| 93 |
switch ($name) {
|
| 94 |
case 'IMAGE':
|
| 95 |
case 'TEXTINPUT':
|
| 96 |
case 'CONTENT':
|
| 97 |
case 'SUMMARY':
|
| 98 |
case 'TAGLINE':
|
| 99 |
case 'SUBTITLE':
|
| 100 |
case 'LOGO':
|
| 101 |
case 'INFO':
|
| 102 |
$element = $name;
|
| 103 |
break;
|
| 104 |
case 'ID':
|
| 105 |
if ($element != 'ITEM') {
|
| 106 |
$element = $name;
|
| 107 |
}
|
| 108 |
case 'LINK':
|
| 109 |
if ($attributes['REL'] == 'alternate') {
|
| 110 |
if ($element == 'ITEM') {
|
| 111 |
$items[$item]['LINK'] = $attributes['HREF'];
|
| 112 |
}
|
| 113 |
else {
|
| 114 |
$channel['LINK'] = $attributes['HREF'];
|
| 115 |
}
|
| 116 |
}
|
| 117 |
break;
|
| 118 |
case 'ITEM':
|
| 119 |
$element = $name;
|
| 120 |
$item += 1;
|
| 121 |
break;
|
| 122 |
case 'ENTRY':
|
| 123 |
$element = 'ITEM';
|
| 124 |
$item += 1;
|
| 125 |
break;
|
| 126 |
}
|
| 127 |
|
| 128 |
$tag = $name;
|
| 129 |
}
|
| 130 |
|
| 131 |
/**
|
| 132 |
* Call-back function used by the XML parser.
|
| 133 |
*/
|
| 134 |
function feedapi_aggregator_element_end($parser, $name) {
|
| 135 |
global $element;
|
| 136 |
|
| 137 |
switch ($name) {
|
| 138 |
case 'IMAGE':
|
| 139 |
case 'TEXTINPUT':
|
| 140 |
case 'ITEM':
|
| 141 |
case 'ENTRY':
|
| 142 |
case 'CONTENT':
|
| 143 |
case 'INFO':
|
| 144 |
$element = '';
|
| 145 |
break;
|
| 146 |
case 'ID':
|
| 147 |
if ($element == 'ID') {
|
| 148 |
$element = '';
|
| 149 |
}
|
| 150 |
}
|
| 151 |
}
|
| 152 |
|
| 153 |
/**
|
| 154 |
* Call-back function used by the XML parser.
|
| 155 |
*/
|
| 156 |
function feedapi_aggregator_element_data($parser, $data) {
|
| 157 |
global $channel, $element, $items, $item, $image, $tag;
|
| 158 |
switch ($element) {
|
| 159 |
case 'ITEM':
|
| 160 |
$items[$item][$tag] .= $data;
|
| 161 |
break;
|
| 162 |
case 'IMAGE':
|
| 163 |
case 'LOGO':
|
| 164 |
$image[$tag] .= $data;
|
| 165 |
break;
|
| 166 |
case 'LINK':
|
| 167 |
if ($data) {
|
| 168 |
$items[$item][$tag] .= $data;
|
| 169 |
}
|
| 170 |
break;
|
| 171 |
case 'CONTENT':
|
| 172 |
$items[$item]['CONTENT'] .= $data;
|
| 173 |
break;
|
| 174 |
case 'SUMMARY':
|
| 175 |
$items[$item]['SUMMARY'] .= $data;
|
| 176 |
break;
|
| 177 |
case 'TAGLINE':
|
| 178 |
case 'SUBTITLE':
|
| 179 |
$channel['DESCRIPTION'] .= $data;
|
| 180 |
break;
|
| 181 |
case 'INFO':
|
| 182 |
case 'ID':
|
| 183 |
case 'TEXTINPUT':
|
| 184 |
// The sub-element is not supported. However, we must recognize
|
| 185 |
// it or its contents will end up in the item array.
|
| 186 |
break;
|
| 187 |
default:
|
| 188 |
$channel[$tag] .= $data;
|
| 189 |
}
|
| 190 |
}
|
| 191 |
|
| 192 |
/**
|
| 193 |
* Parse the W3C date/time format, a subset of ISO 8601. PHP date parsing
|
| 194 |
* functions do not handle this format.
|
| 195 |
* See http://www.w3.org/TR/NOTE-datetime for more information.
|
| 196 |
* Originally from MagpieRSS (http://magpierss.sourceforge.net/).
|
| 197 |
*
|
| 198 |
* @param $date_str A string with a potentially W3C DTF date.
|
| 199 |
* @return A timestamp if parsed successfully or -1 if not.
|
| 200 |
*/
|
| 201 |
function feedapi_aggregator_parse_w3cdtf($date_str) {
|
| 202 |
if (preg_match('/(\d{4})-(\d{2})-(\d{2})T(\d{2}):(\d{2})(:(\d{2}))?(?:([-+])(\d{2}):?(\d{2})|(Z))?/', $date_str, $match)) {
|
| 203 |
list($year, $month, $day, $hours, $minutes, $seconds) = array($match[1], $match[2], $match[3], $match[4], $match[5], $match[6]);
|
| 204 |
// calc epoch for current date assuming GMT
|
| 205 |
$epoch = gmmktime($hours, $minutes, $seconds, $month, $day, $year);
|
| 206 |
if ($match[10] != 'Z') { // Z is zulu time, aka GMT
|
| 207 |
list($tz_mod, $tz_hour, $tz_min) = array($match[8], $match[9], $match[10]);
|
| 208 |
// zero out the variables
|
| 209 |
if (!$tz_hour) {
|
| 210 |
$tz_hour = 0;
|
| 211 |
}
|
| 212 |
if (!$tz_min) {
|
| 213 |
$tz_min = 0;
|
| 214 |
}
|
| 215 |
$offset_secs = (($tz_hour * 60) + $tz_min) * 60;
|
| 216 |
// is timezone ahead of GMT? then subtract offset
|
| 217 |
if ($tz_mod == '+') {
|
| 218 |
$offset_secs *= -1;
|
| 219 |
}
|
| 220 |
$epoch += $offset_secs;
|
| 221 |
}
|
| 222 |
return $epoch;
|
| 223 |
}
|
| 224 |
else {
|
| 225 |
return FALSE;
|
| 226 |
}
|
| 227 |
}
|
| 228 |
|
| 229 |
function feedapi_aggregator_parse_feed(&$data, $feed) {
|
| 230 |
global $items, $image, $channel;
|
| 231 |
|
| 232 |
// Unset the global variables before we use them:
|
| 233 |
unset($GLOBALS['element'], $GLOBALS['item'], $GLOBALS['tag']);
|
| 234 |
$items = array();
|
| 235 |
$image = array();
|
| 236 |
$channel = array();
|
| 237 |
|
| 238 |
// parse the data:
|
| 239 |
$xml_parser = drupal_xml_parser_create($data);
|
| 240 |
xml_set_element_handler($xml_parser, 'feedapi_aggregator_element_start', 'feedapi_aggregator_element_end');
|
| 241 |
xml_set_character_data_handler($xml_parser, 'feedapi_aggregator_element_data');
|
| 242 |
|
| 243 |
if (!xml_parse($xml_parser, $data, 1)) {
|
| 244 |
watchdog('aggregator', t('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);
|
| 245 |
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($xml_parser))), 'error');
|
| 246 |
return 0;
|
| 247 |
}
|
| 248 |
xml_parser_free($xml_parser);
|
| 249 |
|
| 250 |
/*
|
| 251 |
** We reverse the array such that we store the first item last,
|
| 252 |
** and the last item first. In the database, the newest item
|
| 253 |
** should be at the top.
|
| 254 |
*/
|
| 255 |
|
| 256 |
$items = array_reverse($items);
|
| 257 |
|
| 258 |
// Create a feedapi parsed source object
|
| 259 |
$parsed_source = new stdClass();
|
| 260 |
// Detect the title
|
| 261 |
$parsed_source->title = isset($channel['TITLE']) ? $channel['TITLE'] : "";
|
| 262 |
// Detect the description
|
| 263 |
$parsed_source->description = isset($channel['DESCRIPTION']) ? $channel['DESCRIPTION'] : "";
|
| 264 |
$parsed_source->options = new stdClass();
|
| 265 |
// Detect the link
|
| 266 |
$parsed_source->options->link = isset($channel['LINK']) ? $channel['LINK'] : "";
|
| 267 |
$parsed_source->items = array();
|
| 268 |
|
| 269 |
// Initialize variables
|
| 270 |
$title = $link = $author = $description = $guid = NULL;
|
| 271 |
foreach ($items as $item) {
|
| 272 |
unset($title, $link, $author, $description, $guid);
|
| 273 |
|
| 274 |
// Prepare the item:
|
| 275 |
foreach ($item as $key => $value) {
|
| 276 |
$item[$key] = trim($value);
|
| 277 |
}
|
| 278 |
|
| 279 |
/*
|
| 280 |
** Resolve the item's title. If no title is found, we use
|
| 281 |
** up to 40 characters of the description ending at a word
|
| 282 |
** boundary but not splitting potential entities.
|
| 283 |
*/
|
| 284 |
|
| 285 |
if ($item['TITLE']) {
|
| 286 |
$title = $item['TITLE'];
|
| 287 |
}
|
| 288 |
else {
|
| 289 |
$title = preg_replace('/^(.*)[^\w;&].*?$/', "\\1", truncate_utf8($item['DESCRIPTION'], 40));
|
| 290 |
}
|
| 291 |
|
| 292 |
/*
|
| 293 |
** Resolve the items link.
|
| 294 |
*/
|
| 295 |
|
| 296 |
if ($item['LINK']) {
|
| 297 |
$link = $item['LINK'];
|
| 298 |
}
|
| 299 |
else {
|
| 300 |
$link = $feed['link'];
|
| 301 |
}
|
| 302 |
if ($item['GUID']) {
|
| 303 |
$guid = $item['GUID'];
|
| 304 |
}
|
| 305 |
|
| 306 |
/**
|
| 307 |
* Atom feeds have a CONTENT and/or SUMMARY tag instead of a DESCRIPTION tag
|
| 308 |
*/
|
| 309 |
if ($item['CONTENT:ENCODED']) {
|
| 310 |
$item['DESCRIPTION'] = $item['CONTENT:ENCODED'];
|
| 311 |
}
|
| 312 |
else if ($item['SUMMARY']) {
|
| 313 |
$item['DESCRIPTION'] = $item['SUMMARY'];
|
| 314 |
}
|
| 315 |
else if ($item['CONTENT']) {
|
| 316 |
$item['DESCRIPTION'] = $item['CONTENT'];
|
| 317 |
}
|
| 318 |
|
| 319 |
/*
|
| 320 |
** Try to resolve and parse the item's publication date. If no
|
| 321 |
** date is found, we use the current date instead.
|
| 322 |
*/
|
| 323 |
|
| 324 |
if ($item['PUBDATE']) $date = $item['PUBDATE']; // RSS 2.0
|
| 325 |
else if ($item['DC:DATE']) $date = $item['DC:DATE']; // Dublin core
|
| 326 |
else if ($item['DCTERMS:ISSUED']) $date = $item['DCTERMS:ISSUED']; // Dublin core
|
| 327 |
else if ($item['DCTERMS:CREATED']) $date = $item['DCTERMS:CREATED']; // Dublin core
|
| 328 |
else if ($item['DCTERMS:MODIFIED']) $date = $item['DCTERMS:MODIFIED']; // Dublin core
|
| 329 |
else if ($item['ISSUED']) $date = $item['ISSUED']; // Atom XML
|
| 330 |
else if ($item['CREATED']) $date = $item['CREATED']; // Atom XML
|
| 331 |
else if ($item['MODIFIED']) $date = $item['MODIFIED']; // Atom XML
|
| 332 |
else if ($item['PUBLISHED']) $date = $item['PUBLISHED']; // Atom XML
|
| 333 |
else if ($item['UPDATED']) $date = $item['UPDATED']; // Atom XML
|
| 334 |
else $date = 'now';
|
| 335 |
|
| 336 |
$timestamp = strtotime($date); // As of PHP 5.1.0, strtotime returns FALSE on failure instead of -1.
|
| 337 |
if ($timestamp <= 0) {
|
| 338 |
$timestamp = feedapi_aggregator_parse_w3cdtf($date); // Returns FALSE on failure
|
| 339 |
if (!$timestamp) {
|
| 340 |
$timestamp = time(); // better than nothing
|
| 341 |
}
|
| 342 |
}
|
| 343 |
|
| 344 |
// Add item to feedapi items array.
|
| 345 |
$parsed_item = new stdClass();
|
| 346 |
$parsed_item->title = $title;
|
| 347 |
$parsed_item->description = $item['DESCRIPTION'];
|
| 348 |
$parsed_item->options = new stdClass();
|
| 349 |
$parsed_item->options->original_author = $item['AUTHOR'];
|
| 350 |
$parsed_item->options->timestamp = $timestamp;
|
| 351 |
$parsed_item->options->original_url = $link;
|
| 352 |
$parsed_item->options->guid = $guid;
|
| 353 |
$parsed_source->items[] = $parsed_item;
|
| 354 |
|
| 355 |
}
|
| 356 |
|
| 357 |
return $parsed_source;
|
| 358 |
}
|
Parent Directory
| 
Revision Log
| 
Revision Graph