2 define('DAEMON_UPDATE_LOGIN_LIMIT', 30);
3 define('DAEMON_FEED_LIMIT', 100);
4 define('DAEMON_SLEEP_INTERVAL', 60);
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 define('PREFS_NO_CACHE', true);
65 // Test if the user has loggued in recently. If not, it does not update its feeds.
66 if (!SINGLE_USER_MODE
&& DAEMON_UPDATE_LOGIN_LIMIT
> 0) {
67 if (DB_TYPE
== "pgsql") {
68 $login_thresh_qpart = "AND ttrss_users.last_login >= NOW() - INTERVAL '".DAEMON_UPDATE_LOGIN_LIMIT
." days'";
70 $login_thresh_qpart = "AND ttrss_users.last_login >= DATE_SUB(NOW(), INTERVAL ".DAEMON_UPDATE_LOGIN_LIMIT
." DAY)";
73 $login_thresh_qpart = "";
76 // Test if the feed need a update (update interval exceded).
77 if (DB_TYPE
== "pgsql") {
78 $update_limit_qpart = "AND ((
79 ttrss_feeds.update_interval = 0
80 AND ttrss_feeds.last_updated < NOW() - CAST((ttrss_user_prefs.value || ' minutes') AS INTERVAL)
82 ttrss_feeds.update_interval > 0
83 AND ttrss_feeds.last_updated < NOW() - CAST((ttrss_feeds.update_interval || ' minutes') AS INTERVAL)
84 ) OR ttrss_feeds.last_updated IS NULL
85 OR last_updated = '1970-01-01 00:00:00')";
87 $update_limit_qpart = "AND ((
88 ttrss_feeds.update_interval = 0
89 AND ttrss_feeds.last_updated < DATE_SUB(NOW(), INTERVAL CONVERT(ttrss_user_prefs.value, SIGNED INTEGER) MINUTE)
91 ttrss_feeds.update_interval > 0
92 AND ttrss_feeds.last_updated < DATE_SUB(NOW(), INTERVAL ttrss_feeds.update_interval MINUTE)
93 ) OR ttrss_feeds.last_updated IS NULL
94 OR last_updated = '1970-01-01 00:00:00')";
97 // Test if feed is currently being updated by another process.
98 if (DB_TYPE
== "pgsql") {
99 $updstart_thresh_qpart = "AND (ttrss_feeds.last_update_started IS NULL OR ttrss_feeds.last_update_started < NOW() - INTERVAL '5 minutes')";
101 $updstart_thresh_qpart = "AND (ttrss_feeds.last_update_started IS NULL OR ttrss_feeds.last_update_started < DATE_SUB(NOW(), INTERVAL 5 MINUTE))";
104 // Test if there is a limit to number of updated feeds
106 if($limit) $query_limit = sprintf("LIMIT %d", $limit);
108 $random_qpart = sql_random_function();
110 // We search for feed needing update.
111 $result = db_query($link, "SELECT ttrss_feeds.feed_url,ttrss_feeds.id, ttrss_feeds.owner_uid,
112 ".SUBSTRING_FOR_DATE
."(ttrss_feeds.last_updated,1,19) AS last_updated,
113 ttrss_feeds.update_interval
115 ttrss_feeds, ttrss_users, ttrss_user_prefs
117 ttrss_feeds.owner_uid = ttrss_users.id
118 AND ttrss_users.id = ttrss_user_prefs.owner_uid
119 AND ttrss_user_prefs.pref_name = 'DEFAULT_UPDATE_INTERVAL'
120 $login_thresh_qpart $update_limit_qpart
121 $updstart_thresh_qpart
122 ORDER BY $random_qpart $query_limit");
124 $user_prefs_cache = array();
126 if($debug) _debug(sprintf("Scheduled %d feeds to update...\n", db_num_rows($result)));
128 // Here is a little cache magic in order to minimize risk of double feed updates.
129 $feeds_to_update = array();
130 while ($line = db_fetch_assoc($result)) {
131 $feeds_to_update[$line['id']] = $line;
134 // We update the feed last update started date before anything else.
135 // There is no lag due to feed contents downloads
136 // It prevent an other process to update the same feed.
137 $feed_ids = array_keys($feeds_to_update);
139 db_query($link, sprintf("UPDATE ttrss_feeds SET last_update_started = NOW()
140 WHERE id IN (%s)", implode(',', $feed_ids)));
143 expire_cached_files($debug);
144 expire_lock_files($debug);
146 // For each feed, we call the feed update function.
147 while ($line = array_pop($feeds_to_update)) {
149 if($debug) _debug("Feed: " . $line["feed_url"] . ", " . $line["last_updated"]);
151 update_rss_feed($link, $line["id"], true);
153 sleep(1); // prevent flood (FIXME make this an option?)
156 // Send feed digests by email if needed.
157 send_headlines_digests($link, $debug);
159 } // function update_daemon_common
161 // ignore_daemon is not used
162 function update_rss_feed($link, $feed, $ignore_daemon = false, $no_cache = false,
163 $override_url = false) {
165 require_once "lib/simplepie/simplepie.inc";
167 $debug_enabled = defined('DAEMON_EXTENDED_DEBUG') ||
$_REQUEST['xdebug'];
169 if ($debug_enabled) {
170 _debug("update_rss_feed: start");
173 $result = db_query($link, "SELECT id,update_interval,auth_login,
174 feed_url,auth_pass,cache_images,last_updated,cache_content,
175 mark_unread_on_update, owner_uid, update_on_checksum_change,
177 FROM ttrss_feeds WHERE id = '$feed'");
179 if (db_num_rows($result) == 0) {
180 if ($debug_enabled) {
181 _debug("update_rss_feed: feed $feed NOT FOUND/SKIPPED");
186 $last_updated = db_fetch_result($result, 0, "last_updated");
187 $owner_uid = db_fetch_result($result, 0, "owner_uid");
188 $mark_unread_on_update = sql_bool_to_bool(db_fetch_result($result,
189 0, "mark_unread_on_update"));
190 $update_on_checksum_change = sql_bool_to_bool(db_fetch_result($result,
191 0, "update_on_checksum_change"));
192 $pubsub_state = db_fetch_result($result, 0, "pubsub_state");
194 db_query($link, "UPDATE ttrss_feeds SET last_update_started = NOW()
195 WHERE id = '$feed'");
197 $auth_login = db_fetch_result($result, 0, "auth_login");
198 $auth_pass = db_fetch_result($result, 0, "auth_pass");
200 $cache_images = sql_bool_to_bool(db_fetch_result($result, 0, "cache_images"));
201 $cache_content = sql_bool_to_bool(db_fetch_result($result, 0, "cache_content"));
202 $fetch_url = db_fetch_result($result, 0, "feed_url");
204 $feed = db_escape_string($feed);
206 if ($auth_login && $auth_pass ){
207 $url_parts = array();
208 preg_match("/(^[^:]*):\/\/(.*)/", $fetch_url, $url_parts);
210 if ($url_parts[1] && $url_parts[2]) {
211 $fetch_url = $url_parts[1] . "://$auth_login:$auth_pass@" . $url_parts[2];
217 $fetch_url = $override_url;
219 if ($debug_enabled) {
220 _debug("update_rss_feed: fetching [$fetch_url]...");
223 // Ignore cache if new feed or manual update.
224 $cache_age = (is_null($last_updated) ||
$last_updated == '1970-01-01 00:00:00') ?
225 -1 : get_feed_update_interval($link, $feed) * 60;
227 $simplepie_cache_dir = CACHE_DIR
. "/simplepie";
229 if (!is_dir($simplepie_cache_dir)) {
230 mkdir($simplepie_cache_dir);
233 $rss = new SimplePie();
234 $rss->set_useragent(SELF_USER_AGENT
);
235 $rss->set_timeout($no_cache ?
15 : 60);
236 $rss->set_feed_url($fetch_url);
237 $rss->set_output_encoding('UTF-8');
238 //$rss->force_feed(true);
240 if ($debug_enabled) {
241 _debug("feed update interval (sec): " .
242 get_feed_update_interval($link, $feed)*60);
245 $rss->enable_cache(!$no_cache);
248 $rss->set_cache_location($simplepie_cache_dir);
249 $rss->set_cache_duration($cache_age);
256 if ($debug_enabled) {
257 _debug("update_rss_feed: fetch done, parsing...");
260 $feed = db_escape_string($feed);
262 if (!$rss->error()) {
264 if ($debug_enabled) {
265 _debug("update_rss_feed: processing feed data...");
268 // db_query($link, "BEGIN");
270 if (DB_TYPE
== "pgsql") {
271 $favicon_interval_qpart = "favicon_last_checked < NOW() - INTERVAL '12 hour'";
273 $favicon_interval_qpart = "favicon_last_checked < DATE_SUB(NOW(), INTERVAL 12 HOUR)";
276 $result = db_query($link, "SELECT title,icon_url,site_url,owner_uid,
277 (favicon_last_checked IS NULL OR $favicon_interval_qpart) AS
279 FROM ttrss_feeds WHERE id = '$feed'");
281 $registered_title = db_fetch_result($result, 0, "title");
282 $orig_icon_url = db_fetch_result($result, 0, "icon_url");
283 $orig_site_url = db_fetch_result($result, 0, "site_url");
284 $favicon_needs_check = sql_bool_to_bool(db_fetch_result($result, 0,
285 "favicon_needs_check"));
287 $owner_uid = db_fetch_result($result, 0, "owner_uid");
289 $site_url = db_escape_string(mb_substr(rewrite_relative_url($fetch_url, $rss->get_link()), 0, 250));
291 if ($debug_enabled) {
292 _debug("update_rss_feed: checking favicon...");
295 if ($favicon_needs_check) {
296 check_feed_favicon($site_url, $feed, $link);
298 db_query($link, "UPDATE ttrss_feeds SET favicon_last_checked = NOW()
299 WHERE id = '$feed'");
302 if (!$registered_title ||
$registered_title == "[Unknown]") {
304 $feed_title = db_escape_string($rss->get_title());
306 if ($debug_enabled) {
307 _debug("update_rss_feed: registering title: $feed_title");
310 db_query($link, "UPDATE ttrss_feeds SET
311 title = '$feed_title' WHERE id = '$feed'");
314 if ($site_url && $orig_site_url != $site_url) {
315 db_query($link, "UPDATE ttrss_feeds SET
316 site_url = '$site_url' WHERE id = '$feed'");
319 $icon_url = db_escape_string(mb_substr(
320 rewrite_relative_url($fetch_url, $rss->get_image_url()), 0, 250));
322 if ($icon_url && $orig_icon_url != $icon_url) {
323 db_query($link, "UPDATE ttrss_feeds SET icon_url = '$icon_url' WHERE id = '$feed'");
326 if ($debug_enabled) {
327 _debug("update_rss_feed: loading filters & labels...");
330 $filters = load_filters($link, $feed, $owner_uid);
331 $labels = get_all_labels($link, $owner_uid);
333 if ($debug_enabled) {
335 _debug("update_rss_feed: " . count($filters) . " filters loaded.");
338 $items = $rss->get_items();
340 if (!is_array($items)) {
341 if ($debug_enabled) {
342 _debug("update_rss_feed: no articles found.");
345 db_query($link, "UPDATE ttrss_feeds
346 SET last_updated = NOW(), last_error = '' WHERE id = '$feed'");
348 return; // no articles
351 if ($pubsub_state != 2 && PUBSUBHUBBUB_ENABLED
) {
353 if ($debug_enabled) _debug("update_rss_feed: checking for PUSH hub...");
355 $feed_hub_url = false;
357 $links = $rss->get_links('hub');
359 if ($links && is_array($links)) {
360 foreach ($links as $l) {
366 if ($debug_enabled) _debug("update_rss_feed: feed hub url: $feed_hub_url");
368 if ($feed_hub_url && function_exists('curl_init') &&
369 !ini_get("open_basedir")) {
371 require_once 'lib/pubsubhubbub/subscriber.php';
373 $callback_url = get_self_url_prefix() .
374 "/public.php?op=pubsub&id=$feed";
376 $s = new Subscriber($feed_hub_url, $callback_url);
378 $rc = $s->subscribe($fetch_url);
381 _debug("update_rss_feed: feed hub url found, subscribe request sent.");
383 db_query($link, "UPDATE ttrss_feeds SET pubsub_state = 1
384 WHERE id = '$feed'");
388 if ($debug_enabled) {
389 _debug("update_rss_feed: processing articles...");
392 foreach ($items as $item) {
393 if ($_REQUEST['xdebug'] == 2) {
397 $entry_guid = $item->get_id();
398 if (!$entry_guid) $entry_guid = $item->get_link();
399 if (!$entry_guid) $entry_guid = make_guid_from_title($item->get_title());
401 if ($cache_content) {
402 $entry_guid = "ccache:$entry_guid";
405 if ($auth_login ||
$auth_pass) {
406 $entry_guid = "auth,$owner_uid:$entry_guid";
409 if ($debug_enabled) {
410 _debug("update_rss_feed: guid $entry_guid");
413 if (!$entry_guid) continue;
415 $entry_timestamp = "";
417 $entry_timestamp = strtotime($item->get_date());
419 if ($entry_timestamp == -1 ||
!$entry_timestamp) {
420 $entry_timestamp = time();
421 $no_orig_date = 'true';
423 $no_orig_date = 'false';
426 $entry_timestamp_fmt = strftime("%Y/%m/%d %H:%M:%S", $entry_timestamp);
428 if ($debug_enabled) {
429 _debug("update_rss_feed: date $entry_timestamp [$entry_timestamp_fmt]");
432 $entry_title = $item->get_title();
434 $entry_link = rewrite_relative_url($site_url, htmlspecialchars_decode($item->get_link()));
436 if ($debug_enabled) {
437 _debug("update_rss_feed: title $entry_title");
438 _debug("update_rss_feed: link $entry_link");
441 if (!$entry_title) $entry_title = date("Y-m-d H:i:s", $entry_timestamp);;
443 $entry_content = $item->get_content();
444 if (!$entry_content) $entry_content = $item->get_description();
446 if ($cache_images && is_writable(CACHE_DIR
. '/images'))
447 $entry_content = cache_images($entry_content, $site_url, $debug_enabled);
449 if ($_REQUEST["xdebug"] == 2) {
450 print "update_rss_feed: content: ";
451 print $entry_content;
455 $entry_cached_content = "";
457 $entry_comments = $item->data
["comments"];
459 if ($item->get_author()) {
460 $entry_author_item = $item->get_author();
461 $entry_author = $entry_author_item->get_name();
462 if (!$entry_author) $entry_author = $entry_author_item->get_email();
464 $entry_author = db_escape_string($entry_author);
467 $entry_guid = db_escape_string(mb_substr($entry_guid, 0, 250));
469 $result = db_query($link, "SELECT id FROM ttrss_entries
470 WHERE guid = '$entry_guid'");
472 $entry_comments = db_escape_string(mb_substr($entry_comments, 0, 250));
473 $entry_author = db_escape_string(mb_substr($entry_author, 0, 250));
475 $num_comments = $item->get_item_tags('http://purl.org/rss/1.0/modules/slash/', 'comments');
477 if (is_array($num_comments) && is_array($num_comments[0])) {
478 $num_comments = (int) $num_comments[0]["data"];
483 if ($debug_enabled) {
484 _debug("update_rss_feed: num_comments: $num_comments");
485 _debug("update_rss_feed: looking for tags [1]...");
488 // parse <category> entries into tags
490 $additional_tags = array();
492 $additional_tags_src = $item->get_categories();
494 if (is_array($additional_tags_src)) {
495 foreach ($additional_tags_src as $tobj) {
496 array_push($additional_tags, $tobj->get_term());
500 if ($debug_enabled) {
501 _debug("update_rss_feed: category tags:");
502 print_r($additional_tags);
505 if ($debug_enabled) {
506 _debug("update_rss_feed: looking for tags [2]...");
509 $entry_tags = array_unique($additional_tags);
511 for ($i = 0; $i < count($entry_tags); $i++
)
512 $entry_tags[$i] = mb_strtolower($entry_tags[$i], 'utf-8');
514 if ($debug_enabled) {
515 //_debug("update_rss_feed: unfiltered tags found:");
516 //print_r($entry_tags);
519 if ($debug_enabled) {
520 _debug("update_rss_feed: done collecting data [TITLE:$entry_title]");
523 // TODO: less memory-hungry implementation
526 if ($debug_enabled) {
527 _debug("update_rss_feed: applying plugin filters..");
530 $article = array("owner_uid" => $owner_uid,
531 "title" => $entry_title,
532 "content" => $entry_content,
533 "link" => $entry_link,
534 "tags" => $entry_tags,
535 "author" => $entry_author);
537 foreach ($pluginhost->get_hooks($pluginhost::HOOK_ARTICLE_FILTER
) as $plugin) {
538 $article = $plugin->hook_article_filter($article);
541 $entry_tags = $article["tags"];
542 $entry_content = db_escape_string($article["content"], false);
543 $entry_title = db_escape_string($article["title"]);
544 $entry_author = db_escape_string($article["author"]);
545 $entry_link = db_escape_string($article["link"]);
547 $content_hash = "SHA1:" . sha1(strip_tags($entry_content));
549 db_query($link, "BEGIN");
551 if (db_num_rows($result) == 0) {
553 if ($debug_enabled) {
554 _debug("update_rss_feed: base guid not found");
557 if ($cache_content) {
558 if ($debug_enabled) {
559 _debug("update_rss_feed: caching content (initial)...");
562 $entry_cached_content = cache_content($link, $entry_link, $auth_login, $auth_pass);
564 if ($cache_images && is_writable(CACHE_DIR
. '/images'))
565 $entry_cached_content = cache_images($entry_cached_content, $site_url, $debug_enabled);
567 $entry_cached_content = db_escape_string($entry_cached_content, false);
570 // base post entry does not exist, create it
572 $result = db_query($link,
573 "INSERT INTO ttrss_entries
591 '$entry_timestamp_fmt',
594 '$entry_cached_content',
602 $article_labels = array();
605 // we keep encountering the entry in feeds, so we need to
606 // update date_updated column so that we don't get horrible
607 // dupes when the entry gets purged and reinserted again e.g.
608 // in the case of SLOW SLOW OMG SLOW updating feeds
610 $base_entry_id = db_fetch_result($result, 0, "id");
612 db_query($link, "UPDATE ttrss_entries SET date_updated = NOW()
613 WHERE id = '$base_entry_id'");
615 $article_labels = get_article_labels($link, $base_entry_id, $owner_uid);
618 // now it should exist, if not - bad luck then
620 $result = db_query($link, "SELECT
621 id,content_hash,no_orig_date,title,
622 ".SUBSTRING_FOR_DATE
."(date_updated,1,19) as date_updated,
623 ".SUBSTRING_FOR_DATE
."(updated,1,19) as updated,
624 num_comments, cached_content
627 WHERE guid = '$entry_guid'");
632 if (db_num_rows($result) == 1) {
634 if ($debug_enabled) {
635 _debug("update_rss_feed: base guid found, checking for user record");
638 // this will be used below in update handler
639 $orig_content_hash = db_fetch_result($result, 0, "content_hash");
640 $orig_title = db_fetch_result($result, 0, "title");
641 $orig_num_comments = db_fetch_result($result, 0, "num_comments");
642 $orig_cached_content = trim(db_fetch_result($result, 0, "cached_content"));
643 $orig_date_updated = strtotime(db_fetch_result($result,
646 $ref_id = db_fetch_result($result, 0, "id");
647 $entry_ref_id = $ref_id;
649 // check for user post link to main table
651 // do we allow duplicate posts with same GUID in different feeds?
652 if (get_pref($link, "ALLOW_DUPLICATE_POSTS", $owner_uid, false)) {
653 $dupcheck_qpart = "AND (feed_id = '$feed' OR feed_id IS NULL)";
655 $dupcheck_qpart = "";
658 /* Collect article tags here so we could filter by them: */
660 $article_filters = get_article_filters($filters, $entry_title,
661 $entry_content, $entry_link, $entry_timestamp, $entry_author,
664 if ($debug_enabled) {
665 _debug("update_rss_feed: article filters: ");
666 if (count($article_filters) != 0) {
667 print_r($article_filters);
671 if (find_article_filter($article_filters, "filter")) {
672 db_query($link, "COMMIT"); // close transaction in progress
676 $score = calculate_article_score($article_filters);
678 if ($debug_enabled) {
679 _debug("update_rss_feed: initial score: $score");
682 $query = "SELECT ref_id, int_id FROM ttrss_user_entries WHERE
683 ref_id = '$ref_id' AND owner_uid = '$owner_uid'
686 // if ($_REQUEST["xdebug"]) print "$query\n";
688 $result = db_query($link, $query);
690 // okay it doesn't exist - create user entry
691 if (db_num_rows($result) == 0) {
693 if ($debug_enabled) {
694 _debug("update_rss_feed: user record not found, creating...");
697 if ($score >= -500 && !find_article_filter($article_filters, 'catchup')) {
699 $last_read_qpart = 'NULL';
702 $last_read_qpart = 'NOW()';
705 if (find_article_filter($article_filters, 'mark') ||
$score > 1000) {
711 if (find_article_filter($article_filters, 'publish')) {
714 $published = 'false';
719 if (DB_TYPE
== "pgsql" and defined('_NGRAM_TITLE_DUPLICATE_THRESHOLD')) {
721 $result = db_query($link, "SELECT COUNT(*) AS similar FROM
722 ttrss_entries,ttrss_user_entries
723 WHERE ref_id = id AND updated >= NOW() - INTERVAL '7 day'
724 AND similarity(title, '$entry_title') >= "._NGRAM_TITLE_DUPLICATE_THRESHOLD
."
725 AND owner_uid = $owner_uid");
727 $ngram_similar = db_fetch_result($result, 0, "similar");
729 if ($debug_enabled) {
730 _debug("update_rss_feed: N-gram similar results: $ngram_similar");
733 if ($ngram_similar > 0) {
738 $result = db_query($link,
739 "INSERT INTO ttrss_user_entries
740 (ref_id, owner_uid, feed_id, unread, last_read, marked,
741 published, score, tag_cache, label_cache, uuid)
742 VALUES ('$ref_id', '$owner_uid', '$feed', $unread,
743 $last_read_qpart, $marked, $published, '$score', '', '', '')");
745 if (PUBSUBHUBBUB_HUB
&& $published == 'true') {
746 $rss_link = get_self_url_prefix() .
747 "/public.php?op=rss&id=-2&key=" .
748 get_feed_access_key($link, -2, false, $owner_uid);
750 $p = new Publisher(PUBSUBHUBBUB_HUB
);
752 $pubsub_result = $p->publish_update($rss_link);
755 $result = db_query($link,
756 "SELECT int_id FROM ttrss_user_entries WHERE
757 ref_id = '$ref_id' AND owner_uid = '$owner_uid' AND
758 feed_id = '$feed' LIMIT 1");
760 if (db_num_rows($result) == 1) {
761 $entry_int_id = db_fetch_result($result, 0, "int_id");
764 if ($debug_enabled) {
765 _debug("update_rss_feed: user record FOUND");
768 $entry_ref_id = db_fetch_result($result, 0, "ref_id");
769 $entry_int_id = db_fetch_result($result, 0, "int_id");
772 if ($debug_enabled) {
773 _debug("update_rss_feed: RID: $entry_ref_id, IID: $entry_int_id");
776 $post_needs_update = false;
777 $update_insignificant = false;
778 $cached_content_needs_update = false;
780 if ($orig_num_comments != $num_comments) {
781 $post_needs_update = true;
782 $update_insignificant = true;
785 if ($content_hash != $orig_content_hash) {
786 $post_needs_update = true;
787 $update_insignificant = false;
788 $cached_content_needs_update = true;
791 if ($cache_content) {
792 if ($debug_enabled) {
793 _debug("update_rss_feed: caching content because original checksum changed...");
796 $entry_cached_content = cache_content($link, $entry_link, $auth_login, $auth_pass);
798 if ($entry_cached_content) {
799 if ($cache_images && is_writable(CACHE_DIR
. '/images'))
800 $entry_cached_content = cache_images($entry_cached_content, $site_url, $debug_enabled);
802 $entry_cached_content = db_escape_string($entry_cached_content, false);
803 $post_needs_update = true;
805 $entry_cached_content = db_escape_string($orig_cached_content);
808 $entry_cached_content = db_escape_string($orig_cached_content);
811 if (db_escape_string($orig_title) != $entry_title) {
812 $post_needs_update = true;
813 $update_insignificant = false;
816 // if post needs update, update it and mark all user entries
817 // linking to this post as updated
818 if ($post_needs_update) {
820 if (defined('DAEMON_EXTENDED_DEBUG')) {
821 _debug("update_rss_feed: post $entry_guid needs update...");
824 // print "<!-- post $orig_title needs update : $post_needs_update -->";
826 db_query($link, "UPDATE ttrss_entries
827 SET title = '$entry_title', content = '$entry_content',
828 content_hash = '$content_hash',
829 cached_content = '$entry_cached_content',
830 updated = '$entry_timestamp_fmt',
831 num_comments = '$num_comments'
832 WHERE id = '$ref_id'");
834 if (!$update_insignificant) {
835 if ($mark_unread_on_update) {
836 db_query($link, "UPDATE ttrss_user_entries
837 SET last_read = null, unread = true WHERE ref_id = '$ref_id'");
838 } else if ($update_on_checksum_change) {
839 db_query($link, "UPDATE ttrss_user_entries
840 SET last_read = null WHERE ref_id = '$ref_id'
841 AND unread = false");
847 db_query($link, "COMMIT");
849 if ($debug_enabled) {
850 _debug("update_rss_feed: assigning labels...");
853 assign_article_to_label_filters($link, $entry_ref_id, $article_filters,
854 $owner_uid, $article_labels);
856 if ($debug_enabled) {
857 _debug("update_rss_feed: looking for enclosures...");
862 $enclosures = array();
864 $encs = $item->get_enclosures();
866 if (is_array($encs)) {
867 foreach ($encs as $e) {
869 $e->link
, $e->type
, $e->length
);
870 array_push($enclosures, $e_item);
874 if ($debug_enabled) {
875 _debug("update_rss_feed: article enclosures:");
876 print_r($enclosures);
879 db_query($link, "BEGIN");
881 foreach ($enclosures as $enc) {
882 $enc_url = db_escape_string($enc[0]);
883 $enc_type = db_escape_string($enc[1]);
884 $enc_dur = db_escape_string($enc[2]);
886 $result = db_query($link, "SELECT id FROM ttrss_enclosures
887 WHERE content_url = '$enc_url' AND post_id = '$entry_ref_id'");
889 if (db_num_rows($result) == 0) {
890 db_query($link, "INSERT INTO ttrss_enclosures
891 (content_url, content_type, title, duration, post_id) VALUES
892 ('$enc_url', '$enc_type', '', '$enc_dur', '$entry_ref_id')");
896 db_query($link, "COMMIT");
898 // check for manual tags (we have to do it here since they're loaded from filters)
900 foreach ($article_filters as $f) {
901 if ($f["type"] == "tag") {
903 $manual_tags = trim_array(explode(",", $f["param"]));
905 foreach ($manual_tags as $tag) {
906 if (tag_is_valid($tag)) {
907 array_push($entry_tags, $tag);
915 $boring_tags = trim_array(explode(",", mb_strtolower(get_pref($link,
916 'BLACKLISTED_TAGS', $owner_uid, ''), 'utf-8')));
918 $filtered_tags = array();
919 $tags_to_cache = array();
921 if ($entry_tags && is_array($entry_tags)) {
922 foreach ($entry_tags as $tag) {
923 if (array_search($tag, $boring_tags) === false) {
924 array_push($filtered_tags, $tag);
929 $filtered_tags = array_unique($filtered_tags);
931 if ($debug_enabled) {
932 _debug("update_rss_feed: filtered article tags:");
933 print_r($filtered_tags);
936 // Save article tags in the database
938 if (count($filtered_tags) > 0) {
940 db_query($link, "BEGIN");
942 foreach ($filtered_tags as $tag) {
944 $tag = sanitize_tag($tag);
945 $tag = db_escape_string($tag);
947 if (!tag_is_valid($tag)) continue;
949 $result = db_query($link, "SELECT id FROM ttrss_tags
950 WHERE tag_name = '$tag' AND post_int_id = '$entry_int_id' AND
951 owner_uid = '$owner_uid' LIMIT 1");
953 if ($result && db_num_rows($result) == 0) {
955 db_query($link, "INSERT INTO ttrss_tags
956 (owner_uid,tag_name,post_int_id)
957 VALUES ('$owner_uid','$tag', '$entry_int_id')");
960 array_push($tags_to_cache, $tag);
963 /* update the cache */
965 $tags_to_cache = array_unique($tags_to_cache);
967 $tags_str = db_escape_string(join(",", $tags_to_cache));
969 db_query($link, "UPDATE ttrss_user_entries
970 SET tag_cache = '$tags_str' WHERE ref_id = '$entry_ref_id'
971 AND owner_uid = $owner_uid");
973 db_query($link, "COMMIT");
976 if (get_pref($link, "AUTO_ASSIGN_LABELS", $owner_uid, false)) {
977 if ($debug_enabled) {
978 _debug("update_rss_feed: auto-assigning labels...");
981 foreach ($labels as $label) {
982 $caption = $label["caption"];
984 if (preg_match("/\b$caption\b/i", "$tags_str " . strip_tags($entry_content) . " $entry_title")) {
985 if (!labels_contains_caption($article_labels, $caption)) {
986 label_add_article($link, $entry_ref_id, $caption, $owner_uid);
992 if ($debug_enabled) {
993 _debug("update_rss_feed: article processed");
997 if (!$last_updated) {
998 if ($debug_enabled) {
999 _debug("update_rss_feed: new feed, catching it up...");
1001 catchup_feed($link, $feed, false, $owner_uid);
1004 if ($debug_enabled) {
1005 _debug("purging feed...");
1008 purge_feed($link, $feed, 0, $debug_enabled);
1010 db_query($link, "UPDATE ttrss_feeds
1011 SET last_updated = NOW(), last_error = '' WHERE id = '$feed'");
1013 // db_query($link, "COMMIT");
1017 $error_msg = mb_substr($rss->error(), 0, 250);
1019 if ($debug_enabled) {
1020 _debug("update_rss_feed: error fetching feed: $error_msg");
1023 $error_msg = db_escape_string($error_msg);
1026 "UPDATE ttrss_feeds SET last_error = '$error_msg',
1027 last_updated = NOW() WHERE id = '$feed'");
1032 if ($debug_enabled) {
1033 _debug("update_rss_feed: done");
1038 function cache_images($html, $site_url, $debug) {
1039 $cache_dir = CACHE_DIR
. "/images";
1041 libxml_use_internal_errors(true);
1043 $charset_hack = '<head>
1044 <meta http-equiv="Content-Type" content="text/html; charset=utf-8"/>
1047 $doc = new DOMDocument();
1048 $doc->loadHTML($charset_hack . $html);
1049 $xpath = new DOMXPath($doc);
1051 $entries = $xpath->query('(//img[@src])');
1053 foreach ($entries as $entry) {
1054 if ($entry->hasAttribute('src')) {
1055 $src = rewrite_relative_url($site_url, $entry->getAttribute('src'));
1057 $local_filename = CACHE_DIR
. "/images/" . sha1($src) . ".png";
1059 if ($debug) _debug("cache_images: downloading: $src to $local_filename");
1061 if (!file_exists($local_filename)) {
1062 $file_content = fetch_file_contents($src);
1064 if ($file_content && strlen($file_content) > 1024) {
1065 file_put_contents($local_filename, $file_content);
1069 if (file_exists($local_filename)) {
1070 $entry->setAttribute('src', SELF_URL_PATH
. '/image.php?url=' .
1071 base64_encode($src));
1076 $node = $doc->getElementsByTagName('body')->item(0);
1078 return $doc->saveXML($node, LIBXML_NOEMPTYTAG
);
1081 function expire_lock_files($debug) {
1082 if ($debug) _debug("Removing old lock files...");
1086 if (is_writable(LOCK_DIRECTORY
)) {
1087 $files = glob(LOCK_DIRECTORY
. "/*.lock");
1090 foreach ($files as $file) {
1091 if (!file_is_locked($file) && time() - filemtime($file) > 86400*2) {
1099 if ($debug) _debug("Removed $num_deleted files.");
1102 function expire_cached_files($debug) {
1103 foreach (array("simplepie", "images", "export") as $dir) {
1104 $cache_dir = CACHE_DIR
. "/$dir";
1106 if ($debug) _debug("Expiring $cache_dir");
1110 if (is_writable($cache_dir)) {
1111 $files = glob("$cache_dir/*");
1114 foreach ($files as $file) {
1115 if (time() - filemtime($file) > 86400*7) {
1124 if ($debug) _debug("Removed $num_deleted files.");
1129 * Source: http://www.php.net/manual/en/function.parse-url.php#104527
1130 * Returns the url query as associative array
1132 * @param string query
1133 * @return array params
1135 function convertUrlQuery($query) {
1136 $queryParts = explode('&', $query);
1140 foreach ($queryParts as $param) {
1141 $item = explode('=', $param);
1142 $params[$item[0]] = $item[1];
1148 function get_article_filters($filters, $title, $content, $link, $timestamp, $author, $tags) {
1151 foreach ($filters as $filter) {
1152 $match_any_rule = $filter["match_any_rule"];
1153 $filter_match = false;
1155 foreach ($filter["rules"] as $rule) {
1157 $reg_exp = $rule["reg_exp"];
1162 switch ($rule["type"]) {
1164 $match = @preg_match
("/$reg_exp/i", $title);
1167 // we don't need to deal with multiline regexps
1168 $content = preg_replace("/[\r\n\t]/", "", $content);
1170 $match = @preg_match
("/$reg_exp/i", $content);
1173 // we don't need to deal with multiline regexps
1174 $content = preg_replace("/[\r\n\t]/", "", $content);
1176 $match = (@preg_match
("/$reg_exp/i", $title) || @preg_match
("/$reg_exp/i", $content));
1179 $match = @preg_match
("/$reg_exp/i", $link);
1182 $match = @preg_match
("/$reg_exp/i", $author);
1185 $tag_string = join(",", $tags);
1186 $match = @preg_match
("/$reg_exp/i", $tag_string);
1190 if ($match_any_rule) {
1192 $filter_match = true;
1196 $filter_match = $match;
1203 if ($filter_match) {
1204 foreach ($filter["actions"] AS $action) {
1205 array_push($matches, $action);
1213 function find_article_filter($filters, $filter_name) {
1214 foreach ($filters as $f) {
1215 if ($f["type"] == $filter_name) {
1222 function find_article_filters($filters, $filter_name) {
1225 foreach ($filters as $f) {
1226 if ($f["type"] == $filter_name) {
1227 array_push($results, $f);
1233 function calculate_article_score($filters) {
1236 foreach ($filters as $f) {
1237 if ($f["type"] == "score") {
1238 $score +
= $f["param"];
1244 function labels_contains_caption($labels, $caption) {
1245 foreach ($labels as $label) {
1246 if ($label[1] == $caption) {
1254 function assign_article_to_label_filters($link, $id, $filters, $owner_uid, $article_labels) {
1255 foreach ($filters as $f) {
1256 if ($f["type"] == "label") {
1257 if (!labels_contains_caption($article_labels, $f["param"])) {
1258 label_add_article($link, $id, $f["param"], $owner_uid);
1264 function cache_content($link, $url, $login, $pass) {
1266 $content = fetch_file_contents($url, $login, $pass);
1269 $doc = new DOMDocument();
1270 @$doc->loadHTML($content);
1271 $xpath = new DOMXPath($doc);
1273 $node = $doc->getElementsByTagName('body')->item(0);
1276 $content = $doc->saveXML($node, LIBXML_NOEMPTYTAG
);