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 $subscribers = db_escape_string($link, $line["subscribers"]);
22 $feed_url = db_escape_string($link, $line["feed_url"]);
23 $title = db_escape_string($link, $line["title"]);
24 $site_url = db_escape_string($link, $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 CAST(ttrss_user_prefs.value AS INTEGER) != -1
81 AND ttrss_feeds.last_updated < NOW() - CAST((ttrss_user_prefs.value || ' minutes') AS INTERVAL)
83 ttrss_feeds.update_interval > 0
84 AND ttrss_feeds.last_updated < NOW() - CAST((ttrss_feeds.update_interval || ' minutes') AS INTERVAL)
85 ) OR ttrss_feeds.last_updated IS NULL
86 OR last_updated = '1970-01-01 00:00:00')";
88 $update_limit_qpart = "AND ((
89 ttrss_feeds.update_interval = 0
90 AND CONVERT(ttrss_user_prefs.value, SIGNED INTEGER) != -1
91 AND ttrss_feeds.last_updated < DATE_SUB(NOW(), INTERVAL CONVERT(ttrss_user_prefs.value, SIGNED INTEGER) MINUTE)
93 ttrss_feeds.update_interval > 0
94 AND ttrss_feeds.last_updated < DATE_SUB(NOW(), INTERVAL ttrss_feeds.update_interval MINUTE)
95 ) OR ttrss_feeds.last_updated IS NULL
96 OR last_updated = '1970-01-01 00:00:00')";
99 // Test if feed is currently being updated by another process.
100 if (DB_TYPE
== "pgsql") {
101 $updstart_thresh_qpart = "AND (ttrss_feeds.last_update_started IS NULL OR ttrss_feeds.last_update_started < NOW() - INTERVAL '5 minutes')";
103 $updstart_thresh_qpart = "AND (ttrss_feeds.last_update_started IS NULL OR ttrss_feeds.last_update_started < DATE_SUB(NOW(), INTERVAL 5 MINUTE))";
106 // Test if there is a limit to number of updated feeds
108 if($limit) $query_limit = sprintf("LIMIT %d", $limit);
110 $random_qpart = sql_random_function();
112 // We search for feed needing update.
113 $result = db_query($link, "SELECT DISTINCT ttrss_feeds.feed_url,$random_qpart
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...", 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($result)) {
131 array_push($feeds_to_update, db_escape_string($link, $line['feed_url']));
134 // We update the feed last update started date before anything else.
135 // There is no lag due to feed contents downloads
136 // It prevent an other process to update the same feed.
138 if(count($feeds_to_update) > 0) {
139 $feeds_quoted = array();
141 foreach ($feeds_to_update as $feed) {
142 array_push($feeds_quoted, "'" . db_escape_string($link, $feed) . "'");
145 db_query($link, sprintf("UPDATE ttrss_feeds SET last_update_started = NOW()
146 WHERE feed_url IN (%s)", implode(',', $feeds_quoted)));
149 expire_cached_files($debug);
150 expire_lock_files($debug);
154 // For each feed, we call the feed update function.
155 foreach ($feeds_to_update as $feed) {
156 if($debug) _debug("Base feed: $feed");
158 //update_rss_feed($link, $line["id"], true);
160 // since we have the data cached, we can deal with other feeds with the same url
162 $tmp_result = db_query($link, "SELECT ttrss_feeds.feed_url,ttrss_feeds.id,last_updated
163 FROM ttrss_feeds, ttrss_users WHERE
164 ttrss_users.id = ttrss_feeds.owner_uid AND
165 feed_url = '".db_escape_string($link, $feed)."' AND
166 ttrss_feeds.update_interval != -1
168 ORDER BY feed_url $query_limit");
170 if (db_num_rows($tmp_result) > 0) {
171 while ($tline = db_fetch_assoc($tmp_result)) {
172 if($debug) _debug(" => " . $tline["last_updated"] . ", " . $tline["id"]);
173 update_rss_feed($link, $tline["id"], true);
179 require_once "digest.php";
181 // Send feed digests by email if needed.
182 send_headlines_digests($link, $debug);
186 } // function update_daemon_common
188 // ignore_daemon is not used
189 function update_rss_feed($link, $feed, $ignore_daemon = false, $no_cache = false,
190 $override_url = false) {
192 require_once "lib/simplepie/simplepie.inc";
194 $debug_enabled = defined('DAEMON_EXTENDED_DEBUG') ||
$_REQUEST['xdebug'];
196 if ($debug_enabled) {
197 _debug("update_rss_feed: start");
200 $result = db_query($link, "SELECT id,update_interval,auth_login,
201 feed_url,auth_pass,cache_images,last_updated,
202 mark_unread_on_update, owner_uid,
204 FROM ttrss_feeds WHERE id = '$feed'");
206 if (db_num_rows($result) == 0) {
207 if ($debug_enabled) {
208 _debug("update_rss_feed: feed $feed NOT FOUND/SKIPPED");
213 $last_updated = db_fetch_result($result, 0, "last_updated");
214 $owner_uid = db_fetch_result($result, 0, "owner_uid");
215 $mark_unread_on_update = sql_bool_to_bool(db_fetch_result($result,
216 0, "mark_unread_on_update"));
217 $pubsub_state = db_fetch_result($result, 0, "pubsub_state");
219 db_query($link, "UPDATE ttrss_feeds SET last_update_started = NOW()
220 WHERE id = '$feed'");
222 $auth_login = db_fetch_result($result, 0, "auth_login");
223 $auth_pass = db_fetch_result($result, 0, "auth_pass");
225 $cache_images = sql_bool_to_bool(db_fetch_result($result, 0, "cache_images"));
226 $fetch_url = db_fetch_result($result, 0, "feed_url");
228 $feed = db_escape_string($link, $feed);
230 if ($override_url) $fetch_url = $override_url;
232 $date_feed_processed = date('Y-m-d H:i');
234 $cache_filename = CACHE_DIR
. "/simplepie/" . sha1($fetch_url) . ".feed";
236 // Ignore cache if new feed or manual update.
237 $cache_age = ($no_cache ||
is_null($last_updated) ||
$last_updated == '1970-01-01 00:00:00') ?
238 30 : get_feed_update_interval($link, $feed) * 60;
240 if ($debug_enabled) {
241 _debug("update_rss_feed: cache filename: $cache_filename exists: " . file_exists($cache_filename));
242 _debug("update_rss_feed: cache age: $cache_age; no cache: $no_cache");
245 $cached_feed_data_hash = false;
249 $cache_timestamp = file_exists($cache_filename) ?
filemtime($cache_filename) : 0;
250 $last_updated_timestamp = strtotime($last_updated);
252 if (file_exists($cache_filename) &&
253 is_readable($cache_filename) &&
254 !$auth_login && !$auth_pass &&
255 filemtime($cache_filename) > time() - $cache_age) {
257 if ($debug_enabled) {
258 _debug("update_rss_feed: using local cache.");
261 if ($cache_timestamp > $last_updated_timestamp) {
262 @$rss_data = file_get_contents($cache_filename);
265 $rss_hash = sha1($rss_data);
266 @$rss = unserialize($rss_data);
269 if ($debug_enabled) {
270 _debug("update_rss_feed: local cache valid and older than last_updated, nothing to do.");
279 if ($debug_enabled) {
280 _debug("update_rss_feed: fetching [$fetch_url] (ts: $cache_timestamp/$last_updated_timestamp)");
283 $feed_data = fetch_file_contents($fetch_url, false,
284 $auth_login, $auth_pass, false, $no_cache ?
15 : 45,
285 max($last_updated_timestamp, $cache_timestamp));
287 if ($debug_enabled) {
288 _debug("update_rss_feed: fetch done.");
294 global $fetch_last_error;
295 global $fetch_last_error_code;
297 if ($debug_enabled) {
298 _debug("update_rss_feed: unable to fetch: $fetch_last_error [$fetch_last_error_code]");
304 if ($fetch_last_error_code != 304) {
305 $error_escaped = db_escape_string($link, $fetch_last_error);
307 if ($debug_enabled) {
308 _debug("update_rss_feed: source claims data not modified, nothing to do.");
313 "UPDATE ttrss_feeds SET last_error = '$error_escaped',
314 last_updated = NOW() WHERE id = '$feed'");
320 $pluginhost = new PluginHost($link);
321 $pluginhost->set_debug($debug_enabled);
322 $user_plugins = get_pref($link, "_ENABLED_PLUGINS", $owner_uid);
324 $pluginhost->load(PLUGINS
, $pluginhost::KIND_ALL
);
325 $pluginhost->load($user_plugins, $pluginhost::KIND_USER
, $owner_uid);
326 $pluginhost->load_data();
328 foreach ($pluginhost->get_hooks($pluginhost::HOOK_FEED_FETCHED
) as $plugin) {
329 $feed_data = $plugin->hook_feed_fetched($feed_data);
333 $rss = new SimplePie();
334 $rss->set_sanitize_class("SanitizeDummy");
335 // simplepie ignores the above and creates default sanitizer anyway,
336 // so let's override it...
337 $rss->sanitize
= new SanitizeDummy();
338 $rss->set_output_encoding('UTF-8');
339 $rss->set_raw_data($feed_data);
340 $rss->enable_cache(false);
347 $feed = db_escape_string($link, $feed);
349 if (!$rss->error()) {
351 // cache data for later
352 if (!$auth_pass && !$auth_login && is_writable(CACHE_DIR
. "/simplepie")) {
353 $rss_data = serialize($rss);
354 $new_rss_hash = sha1($rss_data);
356 if ($new_rss_hash != $rss_hash) {
357 if ($debug_enabled) {
358 _debug("update_rss_feed: saving $cache_filename");
360 @file_put_contents
($cache_filename, serialize($rss));
364 // We use local pluginhost here because we need to load different per-user feed plugins
365 $pluginhost->run_hooks($pluginhost::HOOK_FEED_PARSED
, "hook_feed_parsed", $rss);
367 if ($debug_enabled) {
368 _debug("update_rss_feed: processing feed data...");
371 // db_query($link, "BEGIN");
373 if (DB_TYPE
== "pgsql") {
374 $favicon_interval_qpart = "favicon_last_checked < NOW() - INTERVAL '12 hour'";
376 $favicon_interval_qpart = "favicon_last_checked < DATE_SUB(NOW(), INTERVAL 12 HOUR)";
379 $result = db_query($link, "SELECT title,site_url,owner_uid,
380 (favicon_last_checked IS NULL OR $favicon_interval_qpart) AS
382 FROM ttrss_feeds WHERE id = '$feed'");
384 $registered_title = db_fetch_result($result, 0, "title");
385 $orig_site_url = db_fetch_result($result, 0, "site_url");
386 $favicon_needs_check = sql_bool_to_bool(db_fetch_result($result, 0,
387 "favicon_needs_check"));
389 $owner_uid = db_fetch_result($result, 0, "owner_uid");
391 $site_url = db_escape_string($link, mb_substr(rewrite_relative_url($fetch_url, $rss->get_link()), 0, 245));
393 if ($debug_enabled) {
394 _debug("update_rss_feed: checking favicon...");
397 if ($favicon_needs_check) {
398 check_feed_favicon($site_url, $feed, $link);
400 db_query($link, "UPDATE ttrss_feeds SET favicon_last_checked = NOW()
401 WHERE id = '$feed'");
404 if (!$registered_title ||
$registered_title == "[Unknown]") {
406 $feed_title = db_escape_string($link, $rss->get_title());
408 if ($debug_enabled) {
409 _debug("update_rss_feed: registering title: $feed_title");
412 db_query($link, "UPDATE ttrss_feeds SET
413 title = '$feed_title' WHERE id = '$feed'");
416 if ($site_url && $orig_site_url != $site_url) {
417 db_query($link, "UPDATE ttrss_feeds SET
418 site_url = '$site_url' WHERE id = '$feed'");
421 if ($debug_enabled) {
422 _debug("update_rss_feed: loading filters & labels...");
425 $filters = load_filters($link, $feed, $owner_uid);
426 $labels = get_all_labels($link, $owner_uid);
428 if ($debug_enabled) {
430 _debug("update_rss_feed: " . count($filters) . " filters loaded.");
433 $items = $rss->get_items();
435 if (!is_array($items)) {
436 if ($debug_enabled) {
437 _debug("update_rss_feed: no articles found.");
440 db_query($link, "UPDATE ttrss_feeds
441 SET last_updated = NOW(), last_error = '' WHERE id = '$feed'");
443 return; // no articles
446 if ($pubsub_state != 2 && PUBSUBHUBBUB_ENABLED
) {
448 if ($debug_enabled) _debug("update_rss_feed: checking for PUSH hub...");
450 $feed_hub_url = false;
452 $links = $rss->get_links('hub');
454 if ($links && is_array($links)) {
455 foreach ($links as $l) {
461 if ($debug_enabled) _debug("update_rss_feed: feed hub url: $feed_hub_url");
463 if ($feed_hub_url && function_exists('curl_init') &&
464 !ini_get("open_basedir")) {
466 require_once 'lib/pubsubhubbub/subscriber.php';
468 $callback_url = get_self_url_prefix() .
469 "/public.php?op=pubsub&id=$feed";
471 $s = new Subscriber($feed_hub_url, $callback_url);
473 $rc = $s->subscribe($fetch_url);
476 _debug("update_rss_feed: feed hub url found, subscribe request sent.");
478 db_query($link, "UPDATE ttrss_feeds SET pubsub_state = 1
479 WHERE id = '$feed'");
483 if ($debug_enabled) {
484 _debug("update_rss_feed: processing articles...");
487 foreach ($items as $item) {
488 if ($_REQUEST['xdebug'] == 3) {
492 $entry_guid = $item->get_id();
493 if (!$entry_guid) $entry_guid = $item->get_link();
494 if (!$entry_guid) $entry_guid = make_guid_from_title($item->get_title());
496 if ($debug_enabled) {
497 _debug("update_rss_feed: guid $entry_guid");
500 if (!$entry_guid) continue;
502 $entry_guid = "$owner_uid,$entry_guid";
504 $entry_timestamp = "";
506 $entry_timestamp = strtotime($item->get_date());
508 if ($entry_timestamp == -1 ||
!$entry_timestamp ||
$entry_timestamp > time()) {
509 $entry_timestamp = time();
510 $no_orig_date = 'true';
512 $no_orig_date = 'false';
515 $entry_timestamp_fmt = strftime("%Y/%m/%d %H:%M:%S", $entry_timestamp);
517 if ($debug_enabled) {
518 _debug("update_rss_feed: date $entry_timestamp [$entry_timestamp_fmt]");
521 $entry_title = $item->get_title();
523 $entry_link = rewrite_relative_url($site_url, $item->get_link());
525 if ($debug_enabled) {
526 _debug("update_rss_feed: title $entry_title");
527 _debug("update_rss_feed: link $entry_link");
530 if (!$entry_title) $entry_title = date("Y-m-d H:i:s", $entry_timestamp);;
532 $entry_content = $item->get_content();
533 if (!$entry_content) $entry_content = $item->get_description();
535 if ($_REQUEST["xdebug"] == 2) {
536 print "update_rss_feed: content: ";
537 print $entry_content;
541 $entry_comments = $item->data
["comments"];
543 if ($item->get_author()) {
544 $entry_author_item = $item->get_author();
545 $entry_author = $entry_author_item->get_name();
546 if (!$entry_author) $entry_author = $entry_author_item->get_email();
548 $entry_author = db_escape_string($link, $entry_author);
551 $entry_guid = db_escape_string($link, mb_substr($entry_guid, 0, 245));
553 $entry_comments = db_escape_string($link, mb_substr($entry_comments, 0, 245));
554 $entry_author = db_escape_string($link, mb_substr($entry_author, 0, 245));
556 $num_comments = $item->get_item_tags('http://purl.org/rss/1.0/modules/slash/', 'comments');
558 if (is_array($num_comments) && is_array($num_comments[0])) {
559 $num_comments = (int) $num_comments[0]["data"];
564 if ($debug_enabled) {
565 _debug("update_rss_feed: num_comments: $num_comments");
566 _debug("update_rss_feed: looking for tags [1]...");
569 // parse <category> entries into tags
571 $additional_tags = array();
573 $additional_tags_src = $item->get_categories();
575 if (is_array($additional_tags_src)) {
576 foreach ($additional_tags_src as $tobj) {
577 array_push($additional_tags, $tobj->get_term());
581 if ($debug_enabled) {
582 _debug("update_rss_feed: category tags:");
583 print_r($additional_tags);
586 if ($debug_enabled) {
587 _debug("update_rss_feed: looking for tags [2]...");
590 $entry_tags = array_unique($additional_tags);
592 for ($i = 0; $i < count($entry_tags); $i++
)
593 $entry_tags[$i] = mb_strtolower($entry_tags[$i], 'utf-8');
595 if ($debug_enabled) {
596 //_debug("update_rss_feed: unfiltered tags found:");
597 //print_r($entry_tags);
600 if ($debug_enabled) {
601 _debug("update_rss_feed: done collecting data.");
604 // TODO: less memory-hungry implementation
606 if ($debug_enabled) {
607 _debug("update_rss_feed: applying plugin filters..");
610 // FIXME not sure if owner_uid is a good idea here, we may have a base entry without user entry (?)
611 $result = db_query($link, "SELECT plugin_data,title,content,link,tag_cache,author FROM ttrss_entries, ttrss_user_entries
612 WHERE ref_id = id AND guid = '".db_escape_string($link, $entry_guid)."' AND owner_uid = $owner_uid");
614 if (db_num_rows($result) != 0) {
615 $entry_plugin_data = db_fetch_result($result, 0, "plugin_data");
616 $stored_article = array("title" => db_fetch_result($result, 0, "title"),
617 "content" => db_fetch_result($result, 0, "content"),
618 "link" => db_fetch_result($result, 0, "link"),
619 "tags" => explode(",", db_fetch_result($result, 0, "tag_cache")),
620 "author" => db_fetch_result($result, 0, "author"));
622 $entry_plugin_data = "";
623 $stored_article = array();
626 $article = array("owner_uid" => $owner_uid, // read only
627 "guid" => $entry_guid, // read only
628 "title" => $entry_title,
629 "content" => $entry_content,
630 "link" => $entry_link,
631 "tags" => $entry_tags,
632 "plugin_data" => $entry_plugin_data,
633 "author" => $entry_author,
634 "stored" => $stored_article);
636 foreach ($pluginhost->get_hooks($pluginhost::HOOK_ARTICLE_FILTER
) as $plugin) {
637 $article = $plugin->hook_article_filter($article);
640 $entry_tags = $article["tags"];
641 $entry_guid = db_escape_string($link, $entry_guid);
642 $entry_title = db_escape_string($link, $article["title"]);
643 $entry_author = db_escape_string($link, $article["author"]);
644 $entry_link = db_escape_string($link, $article["link"]);
645 $entry_plugin_data = db_escape_string($link, $article["plugin_data"]);
646 $entry_content = $article["content"]; // escaped below
649 if ($debug_enabled) {
650 _debug("update_rss_feed: plugin data: $entry_plugin_data");
653 if ($cache_images && is_writable(CACHE_DIR
. '/images'))
654 cache_images($entry_content, $site_url, $debug_enabled);
656 $entry_content = db_escape_string($link, $entry_content, false);
658 $content_hash = "SHA1:" . sha1($entry_content);
660 db_query($link, "BEGIN");
662 $result = db_query($link, "SELECT id FROM ttrss_entries
663 WHERE guid = '$entry_guid'");
665 if (db_num_rows($result) == 0) {
667 if ($debug_enabled) {
668 _debug("update_rss_feed: base guid [$entry_guid] not found");
671 // base post entry does not exist, create it
673 $result = db_query($link,
674 "INSERT INTO ttrss_entries
693 '$entry_timestamp_fmt',
699 '$date_feed_processed',
702 '$entry_plugin_data',
705 $article_labels = array();
708 // we keep encountering the entry in feeds, so we need to
709 // update date_updated column so that we don't get horrible
710 // dupes when the entry gets purged and reinserted again e.g.
711 // in the case of SLOW SLOW OMG SLOW updating feeds
713 $base_entry_id = db_fetch_result($result, 0, "id");
715 db_query($link, "UPDATE ttrss_entries SET date_updated = NOW()
716 WHERE id = '$base_entry_id'");
718 $article_labels = get_article_labels($link, $base_entry_id, $owner_uid);
721 // now it should exist, if not - bad luck then
723 $result = db_query($link, "SELECT
724 id,content_hash,no_orig_date,title,plugin_data,
725 ".SUBSTRING_FOR_DATE
."(date_updated,1,19) as date_updated,
726 ".SUBSTRING_FOR_DATE
."(updated,1,19) as updated,
730 WHERE guid = '$entry_guid'");
735 if (db_num_rows($result) == 1) {
737 if ($debug_enabled) {
738 _debug("update_rss_feed: base guid [$entry_guid] found, checking for user record");
741 // this will be used below in update handler
742 $orig_content_hash = db_fetch_result($result, 0, "content_hash");
743 $orig_title = db_fetch_result($result, 0, "title");
744 $orig_num_comments = db_fetch_result($result, 0, "num_comments");
745 $orig_date_updated = strtotime(db_fetch_result($result,
747 $orig_plugin_data = db_fetch_result($result, 0, "plugin_data");
749 $ref_id = db_fetch_result($result, 0, "id");
750 $entry_ref_id = $ref_id;
752 // check for user post link to main table
754 // do we allow duplicate posts with same GUID in different feeds?
755 if (get_pref($link, "ALLOW_DUPLICATE_POSTS", $owner_uid, false)) {
756 $dupcheck_qpart = "AND (feed_id = '$feed' OR feed_id IS NULL)";
758 $dupcheck_qpart = "";
761 /* Collect article tags here so we could filter by them: */
763 $article_filters = get_article_filters($filters, $entry_title,
764 $entry_content, $entry_link, $entry_timestamp, $entry_author,
767 if ($debug_enabled) {
768 _debug("update_rss_feed: article filters: ");
769 if (count($article_filters) != 0) {
770 print_r($article_filters);
774 if (find_article_filter($article_filters, "filter")) {
775 db_query($link, "COMMIT"); // close transaction in progress
779 $score = calculate_article_score($article_filters);
781 if ($debug_enabled) {
782 _debug("update_rss_feed: initial score: $score");
785 $query = "SELECT ref_id, int_id FROM ttrss_user_entries WHERE
786 ref_id = '$ref_id' AND owner_uid = '$owner_uid'
789 // if ($_REQUEST["xdebug"]) print "$query\n";
791 $result = db_query($link, $query);
793 // okay it doesn't exist - create user entry
794 if (db_num_rows($result) == 0) {
796 if ($debug_enabled) {
797 _debug("update_rss_feed: user record not found, creating...");
800 if ($score >= -500 && !find_article_filter($article_filters, 'catchup')) {
802 $last_read_qpart = 'NULL';
805 $last_read_qpart = 'NOW()';
808 if (find_article_filter($article_filters, 'mark') ||
$score > 1000) {
814 if (find_article_filter($article_filters, 'publish')) {
817 $published = 'false';
822 if (DB_TYPE
== "pgsql" and defined('_NGRAM_TITLE_DUPLICATE_THRESHOLD')) {
824 $result = db_query($link, "SELECT COUNT(*) AS similar FROM
825 ttrss_entries,ttrss_user_entries
826 WHERE ref_id = id AND updated >= NOW() - INTERVAL '7 day'
827 AND similarity(title, '$entry_title') >= "._NGRAM_TITLE_DUPLICATE_THRESHOLD
."
828 AND owner_uid = $owner_uid");
830 $ngram_similar = db_fetch_result($result, 0, "similar");
832 if ($debug_enabled) {
833 _debug("update_rss_feed: N-gram similar results: $ngram_similar");
836 if ($ngram_similar > 0) {
841 $last_marked = ($marked == 'true') ?
'NOW()' : 'NULL';
842 $last_published = ($published == 'true') ?
'NOW()' : 'NULL';
844 $result = db_query($link,
845 "INSERT INTO ttrss_user_entries
846 (ref_id, owner_uid, feed_id, unread, last_read, marked,
847 published, score, tag_cache, label_cache, uuid,
848 last_marked, last_published)
849 VALUES ('$ref_id', '$owner_uid', '$feed', $unread,
850 $last_read_qpart, $marked, $published, '$score', '', '',
851 '', $last_marked, $last_published)");
853 if (PUBSUBHUBBUB_HUB
&& $published == 'true') {
854 $rss_link = get_self_url_prefix() .
855 "/public.php?op=rss&id=-2&key=" .
856 get_feed_access_key($link, -2, false, $owner_uid);
858 $p = new Publisher(PUBSUBHUBBUB_HUB
);
860 $pubsub_result = $p->publish_update($rss_link);
863 $result = db_query($link,
864 "SELECT int_id FROM ttrss_user_entries WHERE
865 ref_id = '$ref_id' AND owner_uid = '$owner_uid' AND
866 feed_id = '$feed' LIMIT 1");
868 if (db_num_rows($result) == 1) {
869 $entry_int_id = db_fetch_result($result, 0, "int_id");
872 if ($debug_enabled) {
873 _debug("update_rss_feed: user record FOUND");
876 $entry_ref_id = db_fetch_result($result, 0, "ref_id");
877 $entry_int_id = db_fetch_result($result, 0, "int_id");
880 if ($debug_enabled) {
881 _debug("update_rss_feed: RID: $entry_ref_id, IID: $entry_int_id");
884 $post_needs_update = false;
885 $update_insignificant = false;
887 if ($orig_num_comments != $num_comments) {
888 $post_needs_update = true;
889 $update_insignificant = true;
892 if ($entry_plugin_data != $orig_plugin_data) {
893 $post_needs_update = true;
894 $update_insignificant = true;
897 if ($content_hash != $orig_content_hash) {
898 $post_needs_update = true;
899 $update_insignificant = false;
902 if (db_escape_string($link, $orig_title) != $entry_title) {
903 $post_needs_update = true;
904 $update_insignificant = false;
907 // if post needs update, update it and mark all user entries
908 // linking to this post as updated
909 if ($post_needs_update) {
911 if (defined('DAEMON_EXTENDED_DEBUG')) {
912 _debug("update_rss_feed: post $entry_guid needs update...");
915 // print "<!-- post $orig_title needs update : $post_needs_update -->";
917 db_query($link, "UPDATE ttrss_entries
918 SET title = '$entry_title', content = '$entry_content',
919 content_hash = '$content_hash',
920 updated = '$entry_timestamp_fmt',
921 num_comments = '$num_comments',
922 plugin_data = '$entry_plugin_data'
923 WHERE id = '$ref_id'");
925 if (!$update_insignificant) {
926 if ($mark_unread_on_update) {
927 db_query($link, "UPDATE ttrss_user_entries
928 SET last_read = null, unread = true WHERE ref_id = '$ref_id'");
934 db_query($link, "COMMIT");
936 if ($debug_enabled) {
937 _debug("update_rss_feed: assigning labels...");
940 assign_article_to_label_filters($link, $entry_ref_id, $article_filters,
941 $owner_uid, $article_labels);
943 if ($debug_enabled) {
944 _debug("update_rss_feed: looking for enclosures...");
949 $enclosures = array();
951 $encs = $item->get_enclosures();
953 if (is_array($encs)) {
954 foreach ($encs as $e) {
956 $e->link
, $e->type
, $e->length
);
957 array_push($enclosures, $e_item);
961 if ($debug_enabled) {
962 _debug("update_rss_feed: article enclosures:");
963 print_r($enclosures);
966 db_query($link, "BEGIN");
968 foreach ($enclosures as $enc) {
969 $enc_url = db_escape_string($link, $enc[0]);
970 $enc_type = db_escape_string($link, $enc[1]);
971 $enc_dur = db_escape_string($link, $enc[2]);
973 $result = db_query($link, "SELECT id FROM ttrss_enclosures
974 WHERE content_url = '$enc_url' AND post_id = '$entry_ref_id'");
976 if (db_num_rows($result) == 0) {
977 db_query($link, "INSERT INTO ttrss_enclosures
978 (content_url, content_type, title, duration, post_id) VALUES
979 ('$enc_url', '$enc_type', '', '$enc_dur', '$entry_ref_id')");
983 db_query($link, "COMMIT");
985 // check for manual tags (we have to do it here since they're loaded from filters)
987 foreach ($article_filters as $f) {
988 if ($f["type"] == "tag") {
990 $manual_tags = trim_array(explode(",", $f["param"]));
992 foreach ($manual_tags as $tag) {
993 if (tag_is_valid($tag)) {
994 array_push($entry_tags, $tag);
1002 $boring_tags = trim_array(explode(",", mb_strtolower(get_pref($link,
1003 'BLACKLISTED_TAGS', $owner_uid, ''), 'utf-8')));
1005 $filtered_tags = array();
1006 $tags_to_cache = array();
1008 if ($entry_tags && is_array($entry_tags)) {
1009 foreach ($entry_tags as $tag) {
1010 if (array_search($tag, $boring_tags) === false) {
1011 array_push($filtered_tags, $tag);
1016 $filtered_tags = array_unique($filtered_tags);
1018 if ($debug_enabled) {
1019 _debug("update_rss_feed: filtered article tags:");
1020 print_r($filtered_tags);
1023 // Save article tags in the database
1025 if (count($filtered_tags) > 0) {
1027 db_query($link, "BEGIN");
1029 foreach ($filtered_tags as $tag) {
1031 $tag = sanitize_tag($tag);
1032 $tag = db_escape_string($link, $tag);
1034 if (!tag_is_valid($tag)) continue;
1036 $result = db_query($link, "SELECT id FROM ttrss_tags
1037 WHERE tag_name = '$tag' AND post_int_id = '$entry_int_id' AND
1038 owner_uid = '$owner_uid' LIMIT 1");
1040 if ($result && db_num_rows($result) == 0) {
1042 db_query($link, "INSERT INTO ttrss_tags
1043 (owner_uid,tag_name,post_int_id)
1044 VALUES ('$owner_uid','$tag', '$entry_int_id')");
1047 array_push($tags_to_cache, $tag);
1050 /* update the cache */
1052 $tags_to_cache = array_unique($tags_to_cache);
1054 $tags_str = db_escape_string($link, join(",", $tags_to_cache));
1056 db_query($link, "UPDATE ttrss_user_entries
1057 SET tag_cache = '$tags_str' WHERE ref_id = '$entry_ref_id'
1058 AND owner_uid = $owner_uid");
1060 db_query($link, "COMMIT");
1063 if (get_pref($link, "AUTO_ASSIGN_LABELS", $owner_uid, false)) {
1064 if ($debug_enabled) {
1065 _debug("update_rss_feed: auto-assigning labels...");
1068 foreach ($labels as $label) {
1069 $caption = preg_quote($label["caption"]);
1071 if ($caption && preg_match("/\b$caption\b/i", "$tags_str " . strip_tags($entry_content) . " $entry_title")) {
1072 if (!labels_contains_caption($article_labels, $caption)) {
1073 label_add_article($link, $entry_ref_id, $caption, $owner_uid);
1079 if ($debug_enabled) {
1080 _debug("update_rss_feed: article processed");
1084 if (!$last_updated) {
1085 if ($debug_enabled) {
1086 _debug("update_rss_feed: new feed, catching it up...");
1088 catchup_feed($link, $feed, false, $owner_uid);
1091 if ($debug_enabled) {
1092 _debug("purging feed...");
1095 purge_feed($link, $feed, 0, $debug_enabled);
1097 db_query($link, "UPDATE ttrss_feeds
1098 SET last_updated = NOW(), last_error = '' WHERE id = '$feed'");
1100 // db_query($link, "COMMIT");
1104 $error_msg = db_escape_string($link, mb_substr($rss->error(), 0, 245));
1106 if ($debug_enabled) {
1107 _debug("update_rss_feed: error fetching feed: $error_msg");
1111 "UPDATE ttrss_feeds SET last_error = '$error_msg',
1112 last_updated = NOW() WHERE id = '$feed'");
1117 if ($debug_enabled) {
1118 _debug("update_rss_feed: done");
1123 function cache_images($html, $site_url, $debug) {
1124 $cache_dir = CACHE_DIR
. "/images";
1126 libxml_use_internal_errors(true);
1128 $charset_hack = '<head>
1129 <meta http-equiv="Content-Type" content="text/html; charset=utf-8"/>
1132 $doc = new DOMDocument();
1133 $doc->loadHTML($charset_hack . $html);
1134 $xpath = new DOMXPath($doc);
1136 $entries = $xpath->query('(//img[@src])');
1138 foreach ($entries as $entry) {
1139 if ($entry->hasAttribute('src')) {
1140 $src = rewrite_relative_url($site_url, $entry->getAttribute('src'));
1142 $local_filename = CACHE_DIR
. "/images/" . sha1($src) . ".png";
1144 if ($debug) _debug("cache_images: downloading: $src to $local_filename");
1146 if (!file_exists($local_filename)) {
1147 $file_content = fetch_file_contents($src);
1149 if ($file_content && strlen($file_content) > 1024) {
1150 file_put_contents($local_filename, $file_content);
1154 if (file_exists($local_filename)) {
1155 $entry->setAttribute('src', SELF_URL_PATH
. '/image.php?url=' .
1156 base64_encode($src));
1161 $node = $doc->getElementsByTagName('body')->item(0);
1163 return $doc->saveXML($node);
1166 function expire_lock_files($debug) {
1167 if ($debug) _debug("Removing old lock files...");
1171 if (is_writable(LOCK_DIRECTORY
)) {
1172 $files = glob(LOCK_DIRECTORY
. "/*.lock");
1175 foreach ($files as $file) {
1176 if (!file_is_locked($file) && time() - filemtime($file) > 86400*2) {
1184 if ($debug) _debug("Removed $num_deleted files.");
1187 function expire_cached_files($debug) {
1188 foreach (array("simplepie", "images", "export") as $dir) {
1189 $cache_dir = CACHE_DIR
. "/$dir";
1191 if ($debug) _debug("Expiring $cache_dir");
1195 if (is_writable($cache_dir)) {
1196 $files = glob("$cache_dir/*");
1199 foreach ($files as $file) {
1200 if (time() - filemtime($file) > 86400*7) {
1209 if ($debug) _debug("Removed $num_deleted files.");
1214 * Source: http://www.php.net/manual/en/function.parse-url.php#104527
1215 * Returns the url query as associative array
1217 * @param string query
1218 * @return array params
1220 function convertUrlQuery($query) {
1221 $queryParts = explode('&', $query);
1225 foreach ($queryParts as $param) {
1226 $item = explode('=', $param);
1227 $params[$item[0]] = $item[1];
1233 function get_article_filters($filters, $title, $content, $link, $timestamp, $author, $tags) {
1236 foreach ($filters as $filter) {
1237 $match_any_rule = $filter["match_any_rule"];
1238 $inverse = $filter["inverse"];
1239 $filter_match = false;
1241 foreach ($filter["rules"] as $rule) {
1243 $reg_exp = $rule["reg_exp"];
1244 $rule_inverse = $rule["inverse"];
1249 switch ($rule["type"]) {
1251 $match = @preg_match
("/$reg_exp/i", $title);
1254 // we don't need to deal with multiline regexps
1255 $content = preg_replace("/[\r\n\t]/", "", $content);
1257 $match = @preg_match
("/$reg_exp/i", $content);
1260 // we don't need to deal with multiline regexps
1261 $content = preg_replace("/[\r\n\t]/", "", $content);
1263 $match = (@preg_match
("/$reg_exp/i", $title) || @preg_match
("/$reg_exp/i", $content));
1266 $match = @preg_match
("/$reg_exp/i", $link);
1269 $match = @preg_match
("/$reg_exp/i", $author);
1272 $tag_string = join(",", $tags);
1273 $match = @preg_match
("/$reg_exp/i", $tag_string);
1277 if ($rule_inverse) $match = !$match;
1279 if ($match_any_rule) {
1281 $filter_match = true;
1285 $filter_match = $match;
1292 if ($inverse) $filter_match = !$filter_match;
1294 if ($filter_match) {
1295 foreach ($filter["actions"] AS $action) {
1296 array_push($matches, $action);
1304 function find_article_filter($filters, $filter_name) {
1305 foreach ($filters as $f) {
1306 if ($f["type"] == $filter_name) {
1313 function find_article_filters($filters, $filter_name) {
1316 foreach ($filters as $f) {
1317 if ($f["type"] == $filter_name) {
1318 array_push($results, $f);
1324 function calculate_article_score($filters) {
1327 foreach ($filters as $f) {
1328 if ($f["type"] == "score") {
1329 $score +
= $f["param"];
1335 function labels_contains_caption($labels, $caption) {
1336 foreach ($labels as $label) {
1337 if ($label[1] == $caption) {
1345 function assign_article_to_label_filters($link, $id, $filters, $owner_uid, $article_labels) {
1346 foreach ($filters as $f) {
1347 if ($f["type"] == "label") {
1348 if (!labels_contains_caption($article_labels, $f["param"])) {
1349 label_add_article($link, $id, $f["param"], $owner_uid);
1355 function make_guid_from_title($title) {
1356 return preg_replace("/[ \"\',.:;]/", "-",
1357 mb_strtolower(strip_tags($title), 'utf-8'));