2 define('DAEMON_UPDATE_LOGIN_LIMIT', 30);
3 define('DAEMON_FEED_LIMIT', 100);
4 define('DAEMON_SLEEP_INTERVAL', 120);
6 function update_feedbrowser_cache($link) {
8 $result = db_query($link, "SELECT feed_url, site_url, title, COUNT(id) AS subscribers
9 FROM ttrss_feeds WHERE (SELECT COUNT(id) = 0 FROM ttrss_feeds AS tf
10 WHERE tf.feed_url = ttrss_feeds.feed_url
11 AND (private IS true OR auth_login != '' OR auth_pass != '' OR feed_url LIKE '%:%@%/%'))
12 GROUP BY feed_url, site_url, title ORDER BY subscribers DESC LIMIT 1000");
14 db_query($link, "BEGIN");
16 db_query($link, "DELETE FROM ttrss_feedbrowser_cache");
20 while ($line = db_fetch_assoc($result)) {
21 $subscribers = db_escape_string($line["subscribers"]);
22 $feed_url = db_escape_string($line["feed_url"]);
23 $title = db_escape_string($line["title"]);
24 $site_url = db_escape_string($line["site_url"]);
26 $tmp_result = db_query($link, "SELECT subscribers FROM
27 ttrss_feedbrowser_cache WHERE feed_url = '$feed_url'");
29 if (db_num_rows($tmp_result) == 0) {
31 db_query($link, "INSERT INTO ttrss_feedbrowser_cache
32 (feed_url, site_url, title, subscribers) VALUES ('$feed_url',
33 '$site_url', '$title', '$subscribers')");
41 db_query($link, "COMMIT");
49 * Update a feed batch.
50 * Used by daemons to update n feeds by run.
51 * Only update feed needing a update, and not being processed
54 * @param mixed $link Database link
55 * @param integer $limit Maximum number of feeds in update batch. Default to DAEMON_FEED_LIMIT.
56 * @param boolean $from_http Set to true if you call this function from http to disable cli specific code.
57 * @param boolean $debug Set to false to disable debug output. Default to true.
60 function update_daemon_common($link, $limit = DAEMON_FEED_LIMIT
, $from_http = false, $debug = true) {
61 // Process all other feeds using last_updated and interval parameters
63 // Test if the user has loggued in recently. If not, it does not update its feeds.
64 if (!SINGLE_USER_MODE
&& DAEMON_UPDATE_LOGIN_LIMIT
> 0) {
65 if (DB_TYPE
== "pgsql") {
66 $login_thresh_qpart = "AND ttrss_users.last_login >= NOW() - INTERVAL '".DAEMON_UPDATE_LOGIN_LIMIT
." days'";
68 $login_thresh_qpart = "AND ttrss_users.last_login >= DATE_SUB(NOW(), INTERVAL ".DAEMON_UPDATE_LOGIN_LIMIT
." DAY)";
71 $login_thresh_qpart = "";
74 // Test if the feed need a update (update interval exceded).
75 if (DB_TYPE
== "pgsql") {
76 $update_limit_qpart = "AND ((
77 ttrss_feeds.update_interval = 0
78 AND ttrss_feeds.last_updated < NOW() - CAST((ttrss_user_prefs.value || ' minutes') AS INTERVAL)
80 ttrss_feeds.update_interval > 0
81 AND ttrss_feeds.last_updated < NOW() - CAST((ttrss_feeds.update_interval || ' minutes') AS INTERVAL)
82 ) OR ttrss_feeds.last_updated IS NULL)";
84 $update_limit_qpart = "AND ((
85 ttrss_feeds.update_interval = 0
86 AND ttrss_feeds.last_updated < DATE_SUB(NOW(), INTERVAL CONVERT(ttrss_user_prefs.value, SIGNED INTEGER) MINUTE)
88 ttrss_feeds.update_interval > 0
89 AND ttrss_feeds.last_updated < DATE_SUB(NOW(), INTERVAL ttrss_feeds.update_interval MINUTE)
90 ) OR ttrss_feeds.last_updated IS NULL)";
93 // Test if feed is currently being updated by another process.
94 if (DB_TYPE
== "pgsql") {
95 $updstart_thresh_qpart = "AND (ttrss_feeds.last_update_started IS NULL OR ttrss_feeds.last_update_started < NOW() - INTERVAL '5 minutes')";
97 $updstart_thresh_qpart = "AND (ttrss_feeds.last_update_started IS NULL OR ttrss_feeds.last_update_started < DATE_SUB(NOW(), INTERVAL 5 MINUTE))";
100 // Test if there is a limit to number of updated feeds
102 if($limit) $query_limit = sprintf("LIMIT %d", $limit);
104 $random_qpart = sql_random_function();
106 // We search for feed needing update.
107 $result = db_query($link, "SELECT ttrss_feeds.feed_url,ttrss_feeds.id, ttrss_feeds.owner_uid,
108 ".SUBSTRING_FOR_DATE
."(ttrss_feeds.last_updated,1,19) AS last_updated,
109 ttrss_feeds.update_interval
111 ttrss_feeds, ttrss_users, ttrss_user_prefs
113 ttrss_feeds.owner_uid = ttrss_users.id
114 AND ttrss_users.id = ttrss_user_prefs.owner_uid
115 AND ttrss_user_prefs.pref_name = 'DEFAULT_UPDATE_INTERVAL'
116 $login_thresh_qpart $update_limit_qpart
117 $updstart_thresh_qpart
118 ORDER BY $random_qpart $query_limit");
120 $user_prefs_cache = array();
122 if($debug) _debug(sprintf("Scheduled %d feeds to update...\n", db_num_rows($result)));
124 // Here is a little cache magic in order to minimize risk of double feed updates.
125 $feeds_to_update = array();
126 while ($line = db_fetch_assoc($result)) {
127 $feeds_to_update[$line['id']] = $line;
130 // We update the feed last update started date before anything else.
131 // There is no lag due to feed contents downloads
132 // It prevent an other process to update the same feed.
133 $feed_ids = array_keys($feeds_to_update);
135 db_query($link, sprintf("UPDATE ttrss_feeds SET last_update_started = NOW()
136 WHERE id IN (%s)", implode(',', $feed_ids)));
139 // For each feed, we call the feed update function.
140 while ($line = array_pop($feeds_to_update)) {
142 if($debug) _debug("Feed: " . $line["feed_url"] . ", " . $line["last_updated"]);
144 update_rss_feed($link, $line["id"], true);
146 sleep(1); // prevent flood (FIXME make this an option?)
149 // Send feed digests by email if needed.
150 send_headlines_digests($link);
152 } // function update_daemon_common
154 function fetch_twitter_rss($link, $url, $owner_uid) {
156 require_once 'lib/tmhoauth/tmhOAuth.php';
158 $result = db_query($link, "SELECT twitter_oauth FROM ttrss_users
159 WHERE id = $owner_uid");
161 $access_token = json_decode(db_fetch_result($result, 0, 'twitter_oauth'), true);
162 $url_escaped = db_escape_string($url);
166 $tmhOAuth = new tmhOAuth(array(
167 'consumer_key' => CONSUMER_KEY
,
168 'consumer_secret' => CONSUMER_SECRET
,
169 'user_token' => $access_token['oauth_token'],
170 'user_secret' => $access_token['oauth_token_secret'],
173 $code = $tmhOAuth->request('GET', $url);
177 $content = $tmhOAuth->response
['response'];
179 define('MAGPIE_CACHE_ON', false);
181 $rss = new MagpieRSS($content, MAGPIE_OUTPUT_ENCODING
,
182 MAGPIE_INPUT_ENCODING
, MAGPIE_DETECT_ENCODING
);
188 db_query($link, "UPDATE ttrss_feeds
189 SET last_error = 'OAuth authorization failed ($code).'
190 WHERE feed_url = '$url_escaped' AND owner_uid = $owner_uid");
195 db_query($link, "UPDATE ttrss_feeds
196 SET last_error = 'OAuth information not found.'
197 WHERE feed_url = '$url_escaped' AND owner_uid = $owner_uid");
203 function update_rss_feed($link, $feed, $ignore_daemon = false, $no_cache = false) {
207 /* Update all feeds with the same URL to utilize memcache */
210 $result = db_query($link, "SELECT f1.id
211 FROM ttrss_feeds AS f1, ttrss_feeds AS f2
212 WHERE f2.feed_url = f1.feed_url AND f2.id = '$feed'");
214 while ($line = db_fetch_assoc($result)) {
215 update_rss_feed_real($link, $line["id"], $ignore_daemon, $no_cache);
218 update_rss_feed_real($link, $feed, $ignore_daemon, $no_cache);
222 function update_rss_feed_real($link, $feed, $ignore_daemon = false, $no_cache = false,
223 $override_url = false) {
225 require_once "lib/simplepie/simplepie.inc";
226 require_once "lib/magpierss/rss_fetch.inc";
227 require_once 'lib/magpierss/rss_utils.inc';
231 $debug_enabled = defined('DAEMON_EXTENDED_DEBUG') ||
$_REQUEST['xdebug'];
233 if (!$_REQUEST["daemon"] && !$ignore_daemon) {
237 if ($debug_enabled) {
238 _debug("update_rss_feed: start");
241 if (!$ignore_daemon) {
243 if (DB_TYPE
== "pgsql") {
244 $updstart_thresh_qpart = "(ttrss_feeds.last_update_started IS NULL OR ttrss_feeds.last_update_started < NOW() - INTERVAL '120 seconds')";
246 $updstart_thresh_qpart = "(ttrss_feeds.last_update_started IS NULL OR ttrss_feeds.last_update_started < DATE_SUB(NOW(), INTERVAL 120 SECOND))";
249 $result = db_query($link, "SELECT id,update_interval,auth_login,
250 auth_pass,cache_images,update_method
251 FROM ttrss_feeds WHERE id = '$feed' AND $updstart_thresh_qpart");
255 $result = db_query($link, "SELECT id,update_interval,auth_login,
256 feed_url,auth_pass,cache_images,update_method,last_updated,
257 mark_unread_on_update, owner_uid, update_on_checksum_change,
259 FROM ttrss_feeds WHERE id = '$feed'");
263 if (db_num_rows($result) == 0) {
264 if ($debug_enabled) {
265 _debug("update_rss_feed: feed $feed NOT FOUND/SKIPPED");
270 $update_method = db_fetch_result($result, 0, "update_method");
271 $last_updated = db_fetch_result($result, 0, "last_updated");
272 $owner_uid = db_fetch_result($result, 0, "owner_uid");
273 $mark_unread_on_update = sql_bool_to_bool(db_fetch_result($result,
274 0, "mark_unread_on_update"));
275 $update_on_checksum_change = sql_bool_to_bool(db_fetch_result($result,
276 0, "update_on_checksum_change"));
277 $pubsub_state = db_fetch_result($result, 0, "pubsub_state");
279 db_query($link, "UPDATE ttrss_feeds SET last_update_started = NOW()
280 WHERE id = '$feed'");
282 $auth_login = db_fetch_result($result, 0, "auth_login");
283 $auth_pass = db_fetch_result($result, 0, "auth_pass");
285 if ($update_method == 0)
286 $update_method = DEFAULT_UPDATE_METHOD +
1;
292 if ($update_method == 2)
293 $use_simplepie = true;
295 $use_simplepie = false;
297 if ($debug_enabled) {
298 _debug("update method: $update_method (feed setting: $update_method) (use simplepie: $use_simplepie)\n");
301 if ($update_method == 1) {
302 $auth_login = urlencode($auth_login);
303 $auth_pass = urlencode($auth_pass);
306 $update_interval = db_fetch_result($result, 0, "update_interval");
307 $cache_images = sql_bool_to_bool(db_fetch_result($result, 0, "cache_images"));
308 $fetch_url = db_fetch_result($result, 0, "feed_url");
310 if ($update_interval < 0) { return false; }
312 $feed = db_escape_string($feed);
314 if ($auth_login && $auth_pass ){
315 $url_parts = array();
316 preg_match("/(^[^:]*):\/\/(.*)/", $fetch_url, $url_parts);
318 if ($url_parts[1] && $url_parts[2]) {
319 $fetch_url = $url_parts[1] . "://$auth_login:$auth_pass@" . $url_parts[2];
325 $fetch_url = $override_url;
327 if ($debug_enabled) {
328 _debug("update_rss_feed: fetching [$fetch_url]...");
331 $obj_id = md5("FDATA:$use_simplepie:$fetch_url");
333 if ($memcache && $obj = $memcache->get($obj_id)) {
335 if ($debug_enabled) {
336 _debug("update_rss_feed: data found in memcache.");
343 if ($update_method == 3) {
344 $rss = fetch_twitter_rss($link, $fetch_url, $owner_uid);
345 } else if ($update_method == 1) {
347 define('MAGPIE_CACHE_AGE', get_feed_update_interval($link, $feed) * 60);
348 define('MAGPIE_CACHE_ON', !$no_cache);
349 define('MAGPIE_FETCH_TIME_OUT', 60);
350 define('MAGPIE_CACHE_DIR', CACHE_DIR
. "/magpie");
352 $rss = @fetch_rss
($fetch_url);
354 $simplepie_cache_dir = CACHE_DIR
. "/simplepie";
356 if (!is_dir($simplepie_cache_dir)) {
357 mkdir($simplepie_cache_dir);
360 $rss = new SimplePie();
361 $rss->set_useragent(SELF_USER_AGENT
);
362 # $rss->set_timeout(10);
363 $rss->set_feed_url($fetch_url);
364 $rss->set_output_encoding('UTF-8');
365 $rss->force_feed(true);
367 if (SIMPLEPIE_CACHE_IMAGES
&& $cache_images) {
369 if ($debug_enabled) {
370 _debug("enabling image cache");
373 $rss->set_image_handler("image.php", 'i');
376 if ($debug_enabled) {
377 _debug("feed update interval (sec): " .
378 get_feed_update_interval($link, $feed)*60);
381 $rss->enable_cache(!$no_cache);
384 $rss->set_cache_location($simplepie_cache_dir);
385 $rss->set_cache_duration(get_feed_update_interval($link, $feed) * 60);
391 if ($memcache && $rss) $memcache->add($obj_id, $rss, 0, 300);
396 if ($debug_enabled) {
397 _debug("update_rss_feed: fetch done, parsing...");
400 $feed = db_escape_string($feed);
402 if ($update_method == 2) {
403 $fetch_ok = !$rss->error();
410 if ($debug_enabled) {
411 _debug("update_rss_feed: processing feed data...");
414 // db_query($link, "BEGIN");
416 $result = db_query($link, "SELECT title,icon_url,site_url,owner_uid
417 FROM ttrss_feeds WHERE id = '$feed'");
419 $registered_title = db_fetch_result($result, 0, "title");
420 $orig_icon_url = db_fetch_result($result, 0, "icon_url");
421 $orig_site_url = db_fetch_result($result, 0, "site_url");
423 $owner_uid = db_fetch_result($result, 0, "owner_uid");
425 if ($use_simplepie) {
426 $site_url = $rss->get_link();
428 $site_url = $rss->channel
["link"];
431 $site_url = rewrite_relative_url($fetch_url, $site_url);
433 if ($debug_enabled) {
434 _debug("update_rss_feed: checking favicon...");
437 check_feed_favicon($site_url, $feed, $link);
439 if (!$registered_title ||
$registered_title == "[Unknown]") {
441 if ($use_simplepie) {
442 $feed_title = db_escape_string($rss->get_title());
444 $feed_title = db_escape_string($rss->channel
["title"]);
447 if ($debug_enabled) {
448 _debug("update_rss_feed: registering title: $feed_title");
451 db_query($link, "UPDATE ttrss_feeds SET
452 title = '$feed_title' WHERE id = '$feed'");
455 // weird, weird Magpie
456 if (!$use_simplepie) {
457 if (!$site_url) $site_url = db_escape_string($rss->channel
["link_"]);
460 if ($site_url && $orig_site_url != db_escape_string($site_url)) {
461 db_query($link, "UPDATE ttrss_feeds SET
462 site_url = '$site_url' WHERE id = '$feed'");
465 // print "I: " . $rss->channel["image"]["url"];
467 if (!$use_simplepie) {
468 $icon_url = db_escape_string($rss->image
["url"]);
470 $icon_url = db_escape_string($rss->get_image_url());
473 $icon_url = substr($icon_url, 0, 250);
475 if ($icon_url && $orig_icon_url != $icon_url) {
476 db_query($link, "UPDATE ttrss_feeds SET icon_url = '$icon_url' WHERE id = '$feed'");
479 if ($debug_enabled) {
480 _debug("update_rss_feed: loading filters...");
483 $filters = load_filters($link, $feed, $owner_uid);
485 // if ($debug_enabled) {
486 // print_r($filters);
489 if ($use_simplepie) {
490 $iterator = $rss->get_items();
492 $iterator = $rss->items
;
493 if (!$iterator ||
!is_array($iterator)) $iterator = $rss->entries
;
494 if (!$iterator ||
!is_array($iterator)) $iterator = $rss;
497 if (!is_array($iterator)) {
498 /* db_query($link, "UPDATE ttrss_feeds
499 SET last_error = 'Parse error: can\'t find any articles.'
500 WHERE id = '$feed'"); */
502 // clear any errors and mark feed as updated if fetched okay
503 // even if it's blank
505 if ($debug_enabled) {
506 _debug("update_rss_feed: entry iterator is not an array, no articles?");
509 db_query($link, "UPDATE ttrss_feeds
510 SET last_updated = NOW(), last_error = '' WHERE id = '$feed'");
512 return; // no articles
515 if ($pubsub_state != 2 && PUBSUBHUBBUB_ENABLED
) {
517 if ($debug_enabled) _debug("update_rss_feed: checking for PUSH hub...");
519 $feed_hub_url = false;
520 if ($use_simplepie) {
521 $links = $rss->get_links('hub');
523 if ($links && is_array($links)) {
524 foreach ($links as $l) {
531 $atom = $rss->channel
['atom'];
534 if ($atom['link@rel'] == 'hub') {
535 $feed_hub_url = $atom['link@href'];
538 if (!$feed_hub_url && $atom['link#'] > 1) {
539 for ($i = 2; $i <= $atom['link#']; $i++
) {
540 if ($atom["link#$i@rel"] == 'hub') {
541 $feed_hub_url = $atom["link#$i@href"];
547 $feed_hub_url = $rss->channel
['link_hub'];
551 if ($debug_enabled) _debug("update_rss_feed: feed hub url: $feed_hub_url");
553 if ($feed_hub_url && function_exists('curl_init') &&
554 !ini_get("open_basedir")) {
556 require_once 'lib/pubsubhubbub/subscriber.php';
558 $callback_url = get_self_url_prefix() .
559 "/public.php?op=pubsub&id=$feed";
561 $s = new Subscriber($feed_hub_url, $callback_url);
563 $rc = $s->subscribe($fetch_url);
566 _debug("update_rss_feed: feed hub url found, subscribe request sent.");
568 db_query($link, "UPDATE ttrss_feeds SET pubsub_state = 1
569 WHERE id = '$feed'");
573 if ($debug_enabled) {
574 _debug("update_rss_feed: processing articles...");
577 foreach ($iterator as $item) {
579 if ($_REQUEST['xdebug'] == 2) {
583 if ($use_simplepie) {
584 $entry_guid = $item->get_id();
585 if (!$entry_guid) $entry_guid = $item->get_link();
586 if (!$entry_guid) $entry_guid = make_guid_from_title($item->get_title());
590 $entry_guid = $item["id"];
592 if (!$entry_guid) $entry_guid = $item["guid"];
593 if (!$entry_guid) $entry_guid = $item["about"];
594 if (!$entry_guid) $entry_guid = $item["link"];
595 if (!$entry_guid) $entry_guid = make_guid_from_title($item["title"]);
598 if ($debug_enabled) {
599 _debug("update_rss_feed: guid $entry_guid");
602 if (!$entry_guid) continue;
604 $entry_timestamp = "";
606 if ($use_simplepie) {
607 $entry_timestamp = strtotime($item->get_date());
609 $rss_2_date = $item['pubdate'];
610 $rss_1_date = $item['dc']['date'];
611 $atom_date = $item['issued'];
612 if (!$atom_date) $atom_date = $item['updated'];
614 if ($atom_date != "") $entry_timestamp = parse_w3cdtf($atom_date);
615 if ($rss_1_date != "") $entry_timestamp = parse_w3cdtf($rss_1_date);
616 if ($rss_2_date != "") $entry_timestamp = strtotime($rss_2_date);
620 if ($entry_timestamp == "" ||
$entry_timestamp == -1 ||
!$entry_timestamp) {
621 $entry_timestamp = time();
622 $no_orig_date = 'true';
624 $no_orig_date = 'false';
627 $entry_timestamp_fmt = strftime("%Y/%m/%d %H:%M:%S", $entry_timestamp);
629 if ($debug_enabled) {
630 _debug("update_rss_feed: date $entry_timestamp [$entry_timestamp_fmt]");
633 if ($use_simplepie) {
634 $entry_title = $item->get_title();
636 $entry_title = trim(strip_tags($item["title"]));
639 if ($use_simplepie) {
640 $entry_link = $item->get_link();
642 // strange Magpie workaround
643 $entry_link = $item["link_"];
644 if (!$entry_link) $entry_link = $item["link"];
647 $entry_link = rewrite_relative_url($site_url, $entry_link);
649 if ($debug_enabled) {
650 _debug("update_rss_feed: title $entry_title");
651 _debug("update_rss_feed: link $entry_link");
654 if (!$entry_title) $entry_title = date("Y-m-d H:i:s", $entry_timestamp);;
656 $entry_link = strip_tags($entry_link);
658 if ($use_simplepie) {
659 $entry_content = $item->get_content();
660 if (!$entry_content) $entry_content = $item->get_description();
662 $entry_content = $item["content:escaped"];
664 if (!$entry_content) $entry_content = $item["content:encoded"];
665 if (!$entry_content) $entry_content = $item["content"]["encoded"];
666 if (!$entry_content) $entry_content = $item["content"];
668 if (is_array($entry_content)) $entry_content = $entry_content[0];
670 // Magpie bugs are getting ridiculous
671 if (trim($entry_content) == "Array") $entry_content = false;
673 if (!$entry_content) $entry_content = $item["atom_content"];
674 if (!$entry_content) $entry_content = $item["summary"];
676 if (!$entry_content ||
677 strlen($entry_content) < strlen($item["description"])) {
678 $entry_content = $item["description"];
682 if (is_array($entry_content)) {
683 $entry_content = $entry_content["encoded"];
684 if (!$entry_content) $entry_content = $entry_content["escaped"];
688 if ($_REQUEST["xdebug"] == 2) {
689 print "update_rss_feed: content: ";
690 print_r(htmlspecialchars($entry_content));
693 $entry_content_unescaped = $entry_content;
695 if ($use_simplepie) {
696 $entry_comments = strip_tags($item->data
["comments"]);
697 if ($item->get_author()) {
698 $entry_author_item = $item->get_author();
699 $entry_author = $entry_author_item->get_name();
700 if (!$entry_author) $entry_author = $entry_author_item->get_email();
702 $entry_author = db_escape_string($entry_author);
705 $entry_comments = strip_tags($item["comments"]);
707 $entry_author = db_escape_string(strip_tags($item['dc']['creator']));
709 if ($item['author']) {
711 if (is_array($item['author'])) {
713 if (!$entry_author) {
714 $entry_author = db_escape_string(strip_tags($item['author']['name']));
717 if (!$entry_author) {
718 $entry_author = db_escape_string(strip_tags($item['author']['email']));
722 if (!$entry_author) {
723 $entry_author = db_escape_string(strip_tags($item['author']));
728 if (preg_match('/^[\t\n\r ]*$/', $entry_author)) $entry_author = '';
730 $entry_guid = db_escape_string(strip_tags($entry_guid));
731 $entry_guid = mb_substr($entry_guid, 0, 250);
733 $result = db_query($link, "SELECT id FROM ttrss_entries
734 WHERE guid = '$entry_guid'");
736 $entry_content = db_escape_string($entry_content, false);
738 $content_hash = "SHA1:" . sha1(strip_tags($entry_content));
740 $entry_title = db_escape_string($entry_title);
741 $entry_link = db_escape_string($entry_link);
742 $entry_comments = mb_substr(db_escape_string($entry_comments), 0, 250);
743 $entry_author = mb_substr($entry_author, 0, 250);
745 if ($use_simplepie) {
746 $num_comments = 0; #FIXME#
748 $num_comments = db_escape_string($item["slash"]["comments"]);
751 if (!$num_comments) $num_comments = 0;
753 if ($debug_enabled) {
754 _debug("update_rss_feed: looking for tags [1]...");
757 // parse <category> entries into tags
759 $additional_tags = array();
761 if ($use_simplepie) {
763 $additional_tags_src = $item->get_categories();
765 if (is_array($additional_tags_src)) {
766 foreach ($additional_tags_src as $tobj) {
767 array_push($additional_tags, $tobj->get_term());
771 if ($debug_enabled) {
772 _debug("update_rss_feed: category tags:");
773 print_r($additional_tags);
778 $t_ctr = $item['category#'];
781 $additional_tags = array();
782 } else if ($t_ctr > 0) {
783 $additional_tags = array($item['category']);
785 if ($item['category@term']) {
786 array_push($additional_tags, $item['category@term']);
789 for ($i = 0; $i <= $t_ctr; $i++
) {
790 if ($item["category#$i"]) {
791 array_push($additional_tags, $item["category#$i"]);
794 if ($item["category#$i@term"]) {
795 array_push($additional_tags, $item["category#$i@term"]);
800 // parse <dc:subject> elements
802 $t_ctr = $item['dc']['subject#'];
805 array_push($additional_tags, $item['dc']['subject']);
807 for ($i = 0; $i <= $t_ctr; $i++
) {
808 if ($item['dc']["subject#$i"]) {
809 array_push($additional_tags, $item['dc']["subject#$i"]);
815 if ($debug_enabled) {
816 _debug("update_rss_feed: looking for tags [2]...");
820 // <a href="..." rel="tag">Xorg</a>, //
824 preg_match_all("/<a.*?rel=['\"]tag['\"].*?\>([^<]+)<\/a>/i",
825 $entry_content_unescaped, $entry_tags);
827 $entry_tags = $entry_tags[1];
829 $entry_tags = array_merge($entry_tags, $additional_tags);
830 $entry_tags = array_unique($entry_tags);
832 for ($i = 0; $i < count($entry_tags); $i++
)
833 $entry_tags[$i] = mb_strtolower($entry_tags[$i], 'utf-8');
835 if ($debug_enabled) {
836 _debug("update_rss_feed: unfiltered tags found:");
837 print_r($entry_tags);
842 $entry_content = sanitize_article_content($entry_content);
843 $entry_title = sanitize_article_content($entry_title);
845 if ($debug_enabled) {
846 _debug("update_rss_feed: done collecting data [TITLE:$entry_title]");
849 db_query($link, "BEGIN");
851 if (db_num_rows($result) == 0) {
853 if ($debug_enabled) {
854 _debug("update_rss_feed: base guid not found");
857 // base post entry does not exist, create it
859 $result = db_query($link,
860 "INSERT INTO ttrss_entries
877 '$entry_timestamp_fmt',
887 // we keep encountering the entry in feeds, so we need to
888 // update date_updated column so that we don't get horrible
889 // dupes when the entry gets purged and reinserted again e.g.
890 // in the case of SLOW SLOW OMG SLOW updating feeds
892 $base_entry_id = db_fetch_result($result, 0, "id");
894 db_query($link, "UPDATE ttrss_entries SET date_updated = NOW()
895 WHERE id = '$base_entry_id'");
898 // now it should exist, if not - bad luck then
900 $result = db_query($link, "SELECT
901 id,content_hash,no_orig_date,title,
902 ".SUBSTRING_FOR_DATE
."(date_updated,1,19) as date_updated,
903 ".SUBSTRING_FOR_DATE
."(updated,1,19) as updated,
907 WHERE guid = '$entry_guid'");
912 if (db_num_rows($result) == 1) {
914 if ($debug_enabled) {
915 _debug("update_rss_feed: base guid found, checking for user record");
918 // this will be used below in update handler
919 $orig_content_hash = db_fetch_result($result, 0, "content_hash");
920 $orig_title = db_fetch_result($result, 0, "title");
921 $orig_num_comments = db_fetch_result($result, 0, "num_comments");
922 $orig_date_updated = strtotime(db_fetch_result($result,
925 $ref_id = db_fetch_result($result, 0, "id");
926 $entry_ref_id = $ref_id;
928 // check for user post link to main table
930 // do we allow duplicate posts with same GUID in different feeds?
931 if (get_pref($link, "ALLOW_DUPLICATE_POSTS", $owner_uid, false)) {
932 $dupcheck_qpart = "AND (feed_id = '$feed' OR feed_id IS NULL)";
934 $dupcheck_qpart = "";
937 /* Collect article tags here so we could filter by them: */
939 $article_filters = get_article_filters($filters, $entry_title,
940 $entry_content, $entry_link, $entry_timestamp, $entry_author,
943 if ($debug_enabled) {
944 _debug("update_rss_feed: article filters: ");
945 if (count($article_filters) != 0) {
946 print_r($article_filters);
950 if (find_article_filter($article_filters, "filter")) {
951 db_query($link, "COMMIT"); // close transaction in progress
955 $score = calculate_article_score($article_filters);
957 if ($debug_enabled) {
958 _debug("update_rss_feed: initial score: $score");
961 $query = "SELECT ref_id, int_id FROM ttrss_user_entries WHERE
962 ref_id = '$ref_id' AND owner_uid = '$owner_uid'
965 // if ($_REQUEST["xdebug"]) print "$query\n";
967 $result = db_query($link, $query);
969 // okay it doesn't exist - create user entry
970 if (db_num_rows($result) == 0) {
972 if ($debug_enabled) {
973 _debug("update_rss_feed: user record not found, creating...");
976 if ($score >= -500 && !find_article_filter($article_filters, 'catchup')) {
978 $last_read_qpart = 'NULL';
981 $last_read_qpart = 'NOW()';
984 if (find_article_filter($article_filters, 'mark') ||
$score > 1000) {
990 if (find_article_filter($article_filters, 'publish')) {
993 $published = 'false';
996 $result = db_query($link,
997 "INSERT INTO ttrss_user_entries
998 (ref_id, owner_uid, feed_id, unread, last_read, marked,
999 published, score, tag_cache, label_cache, uuid)
1000 VALUES ('$ref_id', '$owner_uid', '$feed', $unread,
1001 $last_read_qpart, $marked, $published, '$score', '', '', '')");
1003 if (PUBSUBHUBBUB_HUB
&& $published == 'true') {
1004 $rss_link = get_self_url_prefix() .
1005 "/public.php?op=rss&id=-2&key=" .
1006 get_feed_access_key($link, -2, false, $owner_uid);
1008 $p = new Publisher(PUBSUBHUBBUB_HUB
);
1010 $pubsub_result = $p->publish_update($rss_link);
1013 $result = db_query($link,
1014 "SELECT int_id FROM ttrss_user_entries WHERE
1015 ref_id = '$ref_id' AND owner_uid = '$owner_uid' AND
1016 feed_id = '$feed' LIMIT 1");
1018 if (db_num_rows($result) == 1) {
1019 $entry_int_id = db_fetch_result($result, 0, "int_id");
1022 if ($debug_enabled) {
1023 _debug("update_rss_feed: user record FOUND");
1026 $entry_ref_id = db_fetch_result($result, 0, "ref_id");
1027 $entry_int_id = db_fetch_result($result, 0, "int_id");
1030 if ($debug_enabled) {
1031 _debug("update_rss_feed: RID: $entry_ref_id, IID: $entry_int_id");
1034 $post_needs_update = false;
1035 $update_insignificant = false;
1037 if ($orig_num_comments != $num_comments) {
1038 $post_needs_update = true;
1039 $update_insignificant = true;
1042 if ($content_hash != $orig_content_hash) {
1043 $post_needs_update = true;
1044 $update_insignificant = false;
1047 if (db_escape_string($orig_title) != $entry_title) {
1048 $post_needs_update = true;
1049 $update_insignificant = false;
1052 // if post needs update, update it and mark all user entries
1053 // linking to this post as updated
1054 if ($post_needs_update) {
1056 if (defined('DAEMON_EXTENDED_DEBUG')) {
1057 _debug("update_rss_feed: post $entry_guid needs update...");
1060 // print "<!-- post $orig_title needs update : $post_needs_update -->";
1062 db_query($link, "UPDATE ttrss_entries
1063 SET title = '$entry_title', content = '$entry_content',
1064 content_hash = '$content_hash',
1065 updated = '$entry_timestamp_fmt',
1066 num_comments = '$num_comments'
1067 WHERE id = '$ref_id'");
1069 if (!$update_insignificant) {
1070 if ($mark_unread_on_update) {
1071 db_query($link, "UPDATE ttrss_user_entries
1072 SET last_read = null, unread = true WHERE ref_id = '$ref_id'");
1073 } else if ($update_on_checksum_change) {
1074 db_query($link, "UPDATE ttrss_user_entries
1075 SET last_read = null WHERE ref_id = '$ref_id'
1076 AND unread = false");
1082 db_query($link, "COMMIT");
1084 if ($debug_enabled) {
1085 _debug("update_rss_feed: assigning labels...");
1088 assign_article_to_labels($link, $entry_ref_id, $article_filters,
1091 if ($debug_enabled) {
1092 _debug("update_rss_feed: looking for enclosures...");
1097 $enclosures = array();
1099 if ($use_simplepie) {
1100 $encs = $item->get_enclosures();
1102 if (is_array($encs)) {
1103 foreach ($encs as $e) {
1105 $e->link
, $e->type
, $e->length
);
1107 array_push($enclosures, $e_item);
1114 $e_ctr = $item['enclosure#'];
1117 $e_item = array($item['enclosure@url'],
1118 $item['enclosure@type'],
1119 $item['enclosure@length']);
1121 array_push($enclosures, $e_item);
1123 for ($i = 0; $i <= $e_ctr; $i++
) {
1125 if ($item["enclosure#$i@url"]) {
1126 $e_item = array($item["enclosure#$i@url"],
1127 $item["enclosure#$i@type"],
1128 $item["enclosure#$i@length"]);
1129 array_push($enclosures, $e_item);
1135 // can there be many of those? yes -fox
1137 $m_ctr = $item['media']['content#'];
1140 $e_item = array($item['media']['content@url'],
1141 $item['media']['content@medium'],
1142 $item['media']['content@length']);
1144 array_push($enclosures, $e_item);
1146 for ($i = 0; $i <= $m_ctr; $i++
) {
1148 if ($item["media"]["content#$i@url"]) {
1149 $e_item = array($item["media"]["content#$i@url"],
1150 $item["media"]["content#$i@medium"],
1151 $item["media"]["content#$i@length"]);
1152 array_push($enclosures, $e_item);
1160 if ($debug_enabled) {
1161 _debug("update_rss_feed: article enclosures:");
1162 print_r($enclosures);
1165 db_query($link, "BEGIN");
1167 foreach ($enclosures as $enc) {
1168 $enc_url = db_escape_string($enc[0]);
1169 $enc_type = db_escape_string($enc[1]);
1170 $enc_dur = db_escape_string($enc[2]);
1172 $result = db_query($link, "SELECT id FROM ttrss_enclosures
1173 WHERE content_url = '$enc_url' AND post_id = '$entry_ref_id'");
1175 if (db_num_rows($result) == 0) {
1176 db_query($link, "INSERT INTO ttrss_enclosures
1177 (content_url, content_type, title, duration, post_id) VALUES
1178 ('$enc_url', '$enc_type', '', '$enc_dur', '$entry_ref_id')");
1182 db_query($link, "COMMIT");
1184 // check for manual tags (we have to do it here since they're loaded from filters)
1186 foreach ($article_filters as $f) {
1187 if ($f[0] == "tag") {
1189 $manual_tags = trim_array(explode(",", $f[1]));
1191 foreach ($manual_tags as $tag) {
1192 if (tag_is_valid($tag)) {
1193 array_push($entry_tags, $tag);
1201 $boring_tags = trim_array(explode(",", mb_strtolower(get_pref($link,
1202 'BLACKLISTED_TAGS', $owner_uid, ''), 'utf-8')));
1204 $filtered_tags = array();
1205 $tags_to_cache = array();
1207 if ($entry_tags && is_array($entry_tags)) {
1208 foreach ($entry_tags as $tag) {
1209 if (array_search($tag, $boring_tags) === false) {
1210 array_push($filtered_tags, $tag);
1215 $filtered_tags = array_unique($filtered_tags);
1217 if ($debug_enabled) {
1218 _debug("update_rss_feed: filtered article tags:");
1219 print_r($filtered_tags);
1222 // Save article tags in the database
1224 if (count($filtered_tags) > 0) {
1226 db_query($link, "BEGIN");
1228 foreach ($filtered_tags as $tag) {
1230 $tag = sanitize_tag($tag);
1231 $tag = db_escape_string($tag);
1233 if (!tag_is_valid($tag)) continue;
1235 $result = db_query($link, "SELECT id FROM ttrss_tags
1236 WHERE tag_name = '$tag' AND post_int_id = '$entry_int_id' AND
1237 owner_uid = '$owner_uid' LIMIT 1");
1239 if ($result && db_num_rows($result) == 0) {
1241 db_query($link, "INSERT INTO ttrss_tags
1242 (owner_uid,tag_name,post_int_id)
1243 VALUES ('$owner_uid','$tag', '$entry_int_id')");
1246 array_push($tags_to_cache, $tag);
1249 /* update the cache */
1251 $tags_to_cache = array_unique($tags_to_cache);
1253 $tags_str = db_escape_string(join(",", $tags_to_cache));
1255 db_query($link, "UPDATE ttrss_user_entries
1256 SET tag_cache = '$tags_str' WHERE ref_id = '$entry_ref_id'
1257 AND owner_uid = $owner_uid");
1259 db_query($link, "COMMIT");
1262 if ($debug_enabled) {
1263 _debug("update_rss_feed: article processed");
1267 if (!$last_updated) {
1268 if ($debug_enabled) {
1269 _debug("update_rss_feed: new feed, catching it up...");
1271 catchup_feed($link, $feed, false, $owner_uid);
1274 if ($debug_enabled) {
1275 _debug("purging feed...");
1278 purge_feed($link, $feed, 0, $debug_enabled);
1280 db_query($link, "UPDATE ttrss_feeds
1281 SET last_updated = NOW(), last_error = '' WHERE id = '$feed'");
1283 // db_query($link, "COMMIT");
1287 if ($use_simplepie) {
1288 $error_msg = mb_substr($rss->error(), 0, 250);
1290 $error_msg = mb_substr(magpie_error(), 0, 250);
1293 if ($debug_enabled) {
1294 _debug("update_rss_feed: error fetching feed: $error_msg");
1297 $error_msg = db_escape_string($error_msg);
1300 "UPDATE ttrss_feeds SET last_error = '$error_msg',
1301 last_updated = NOW() WHERE id = '$feed'");
1304 if ($use_simplepie) {
1308 if ($debug_enabled) {
1309 _debug("update_rss_feed: done");