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 ttrss_user_prefs.value != '-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 ttrss_user_prefs.value != '-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 DISTINCT ttrss_feeds.id,last_updated
163 FROM ttrss_feeds, ttrss_users, ttrss_user_prefs WHERE
164 ttrss_user_prefs.owner_uid = ttrss_feeds.owner_uid AND
165 ttrss_users.id = ttrss_user_prefs.owner_uid AND
166 ttrss_user_prefs.pref_name = 'DEFAULT_UPDATE_INTERVAL' AND
167 feed_url = '".db_escape_string($link, $feed)."' AND
168 (ttrss_feeds.update_interval > 0 OR
169 ttrss_user_prefs.value != '-1')
171 ORDER BY ttrss_feeds.id $query_limit");
173 if (db_num_rows($tmp_result) > 0) {
174 while ($tline = db_fetch_assoc($tmp_result)) {
175 if($debug) _debug(" => " . $tline["last_updated"] . ", " . $tline["id"]);
176 update_rss_feed($link, $tline["id"], true);
182 require_once "digest.php";
184 // Send feed digests by email if needed.
185 send_headlines_digests($link, $debug);
189 } // function update_daemon_common
191 // ignore_daemon is not used
192 function update_rss_feed($link, $feed, $ignore_daemon = false, $no_cache = false,
193 $override_url = false) {
195 require_once "lib/simplepie/simplepie.inc";
197 $debug_enabled = defined('DAEMON_EXTENDED_DEBUG') ||
$_REQUEST['xdebug'];
199 if ($debug_enabled) {
200 _debug("update_rss_feed: start");
203 $result = db_query($link, "SELECT id,update_interval,auth_login,
204 feed_url,auth_pass,cache_images,last_updated,
205 mark_unread_on_update, owner_uid,
206 pubsub_state, auth_pass_encrypted
207 FROM ttrss_feeds WHERE id = '$feed'");
209 if (db_num_rows($result) == 0) {
210 if ($debug_enabled) {
211 _debug("update_rss_feed: feed $feed NOT FOUND/SKIPPED");
216 $last_updated = db_fetch_result($result, 0, "last_updated");
217 $owner_uid = db_fetch_result($result, 0, "owner_uid");
218 $mark_unread_on_update = sql_bool_to_bool(db_fetch_result($result,
219 0, "mark_unread_on_update"));
220 $pubsub_state = db_fetch_result($result, 0, "pubsub_state");
221 $auth_pass_encrypted = sql_bool_to_bool(db_fetch_result($result,
222 0, "auth_pass_encrypted"));
224 db_query($link, "UPDATE ttrss_feeds SET last_update_started = NOW()
225 WHERE id = '$feed'");
227 $auth_login = db_fetch_result($result, 0, "auth_login");
228 $auth_pass = db_fetch_result($result, 0, "auth_pass");
230 if ($auth_pass_encrypted) {
231 require_once "crypt.php";
232 $auth_pass = decrypt_string($auth_pass);
235 $cache_images = sql_bool_to_bool(db_fetch_result($result, 0, "cache_images"));
236 $fetch_url = db_fetch_result($result, 0, "feed_url");
238 $feed = db_escape_string($link, $feed);
240 if ($override_url) $fetch_url = $override_url;
242 $date_feed_processed = date('Y-m-d H:i');
244 $cache_filename = CACHE_DIR
. "/simplepie/" . sha1($fetch_url) . ".feed";
246 // Ignore cache if new feed or manual update.
247 $cache_age = ($no_cache ||
is_null($last_updated) ||
$last_updated == '1970-01-01 00:00:00') ?
248 30 : get_feed_update_interval($link, $feed) * 60;
250 if ($debug_enabled) {
251 _debug("update_rss_feed: cache filename: $cache_filename exists: " . file_exists($cache_filename));
252 _debug("update_rss_feed: cache age: $cache_age; no cache: $no_cache");
255 $cached_feed_data_hash = false;
259 $cache_timestamp = file_exists($cache_filename) ?
filemtime($cache_filename) : 0;
260 $last_updated_timestamp = strtotime($last_updated);
262 $force_refetch = isset($_REQUEST["force_refetch"]);
264 if (file_exists($cache_filename) &&
265 is_readable($cache_filename) &&
266 !$auth_login && !$auth_pass &&
267 filemtime($cache_filename) > time() - $cache_age) {
269 if ($debug_enabled) {
270 _debug("update_rss_feed: using local cache.");
273 if ($cache_timestamp > $last_updated_timestamp) {
274 @$rss_data = file_get_contents($cache_filename);
277 $rss_hash = sha1($rss_data);
278 @$rss = unserialize($rss_data);
280 } else if (!$force_refetch) {
281 if ($debug_enabled) {
282 _debug("update_rss_feed: local cache valid and older than last_updated, nothing to do.");
291 if ($debug_enabled) {
292 _debug("update_rss_feed: fetching [$fetch_url] (ts: $cache_timestamp/$last_updated_timestamp)");
295 $feed_data = fetch_file_contents($fetch_url, false,
296 $auth_login, $auth_pass, false,
297 $no_cache ? FEED_FETCH_NO_CACHE_TIMEOUT
: FEED_FETCH_TIMEOUT
,
298 $force_refetch ?
0 : max($last_updated_timestamp, $cache_timestamp));
300 if ($debug_enabled) {
301 _debug("update_rss_feed: fetch done.");
307 global $fetch_last_error;
308 global $fetch_last_error_code;
310 if ($debug_enabled) {
311 _debug("update_rss_feed: unable to fetch: $fetch_last_error [$fetch_last_error_code]");
317 if ($fetch_last_error_code != 304) {
318 $error_escaped = db_escape_string($link, $fetch_last_error);
320 if ($debug_enabled) {
321 _debug("update_rss_feed: source claims data not modified, nothing to do.");
326 "UPDATE ttrss_feeds SET last_error = '$error_escaped',
327 last_updated = NOW() WHERE id = '$feed'");
333 $pluginhost = new PluginHost($link);
334 $pluginhost->set_debug($debug_enabled);
335 $user_plugins = get_pref($link, "_ENABLED_PLUGINS", $owner_uid);
337 $pluginhost->load(PLUGINS
, $pluginhost::KIND_ALL
);
338 $pluginhost->load($user_plugins, $pluginhost::KIND_USER
, $owner_uid);
339 $pluginhost->load_data();
341 foreach ($pluginhost->get_hooks($pluginhost::HOOK_FEED_FETCHED
) as $plugin) {
342 $feed_data = $plugin->hook_feed_fetched($feed_data);
346 $rss = new SimplePie();
347 $rss->set_sanitize_class("SanitizeDummy");
348 // simplepie ignores the above and creates default sanitizer anyway,
349 // so let's override it...
350 $rss->sanitize
= new SanitizeDummy();
351 $rss->set_output_encoding('UTF-8');
352 $rss->set_raw_data($feed_data);
353 $rss->enable_cache(false);
360 $feed = db_escape_string($link, $feed);
362 if (!$rss->error()) {
364 // cache data for later
365 if (!$auth_pass && !$auth_login && is_writable(CACHE_DIR
. "/simplepie")) {
366 $rss_data = serialize($rss);
367 $new_rss_hash = sha1($rss_data);
369 if ($new_rss_hash != $rss_hash) {
370 if ($debug_enabled) {
371 _debug("update_rss_feed: saving $cache_filename");
373 @file_put_contents
($cache_filename, serialize($rss));
377 // We use local pluginhost here because we need to load different per-user feed plugins
378 $pluginhost->run_hooks($pluginhost::HOOK_FEED_PARSED
, "hook_feed_parsed", $rss);
380 if ($debug_enabled) {
381 _debug("update_rss_feed: processing feed data...");
384 // db_query($link, "BEGIN");
386 if (DB_TYPE
== "pgsql") {
387 $favicon_interval_qpart = "favicon_last_checked < NOW() - INTERVAL '12 hour'";
389 $favicon_interval_qpart = "favicon_last_checked < DATE_SUB(NOW(), INTERVAL 12 HOUR)";
392 $result = db_query($link, "SELECT title,site_url,owner_uid,
393 (favicon_last_checked IS NULL OR $favicon_interval_qpart) AS
395 FROM ttrss_feeds WHERE id = '$feed'");
397 $registered_title = db_fetch_result($result, 0, "title");
398 $orig_site_url = db_fetch_result($result, 0, "site_url");
399 $favicon_needs_check = sql_bool_to_bool(db_fetch_result($result, 0,
400 "favicon_needs_check"));
402 $owner_uid = db_fetch_result($result, 0, "owner_uid");
404 $site_url = db_escape_string($link, mb_substr(rewrite_relative_url($fetch_url, $rss->get_link()), 0, 245));
406 if ($favicon_needs_check ||
$force_refetch) {
407 if ($debug_enabled) {
408 _debug("update_rss_feed: checking favicon...");
411 check_feed_favicon($site_url, $feed, $link);
412 $favicon_file = ICONS_DIR
. "/$feed.ico";
414 if (file_exists($favicon_file)) {
415 $favicon_color = calculate_avg_color($favicon_file);
417 require_once "colors.php";
419 if (is_array($favicon_color))
420 $tmp = array($favicon_color['red'],
421 $favicon_color['green'],
422 $favicon_color['blue']);
424 $favicon_colorstring = ",favicon_avg_color = '" .
425 _color_pack($tmp) . "'";
428 db_query($link, "UPDATE ttrss_feeds SET favicon_last_checked = NOW() $favicon_colorstring
429 WHERE id = '$feed'");
432 if (!$registered_title ||
$registered_title == "[Unknown]") {
434 $feed_title = db_escape_string($link, $rss->get_title());
436 if ($debug_enabled) {
437 _debug("update_rss_feed: registering title: $feed_title");
440 db_query($link, "UPDATE ttrss_feeds SET
441 title = '$feed_title' WHERE id = '$feed'");
444 if ($site_url && $orig_site_url != $site_url) {
445 db_query($link, "UPDATE ttrss_feeds SET
446 site_url = '$site_url' WHERE id = '$feed'");
449 if ($debug_enabled) {
450 _debug("update_rss_feed: loading filters & labels...");
453 $filters = load_filters($link, $feed, $owner_uid);
454 $labels = get_all_labels($link, $owner_uid);
456 if ($debug_enabled) {
458 _debug("update_rss_feed: " . count($filters) . " filters loaded.");
461 $items = $rss->get_items();
463 if (!is_array($items)) {
464 if ($debug_enabled) {
465 _debug("update_rss_feed: no articles found.");
468 db_query($link, "UPDATE ttrss_feeds
469 SET last_updated = NOW(), last_error = '' WHERE id = '$feed'");
471 return; // no articles
474 if ($pubsub_state != 2 && PUBSUBHUBBUB_ENABLED
) {
476 if ($debug_enabled) _debug("update_rss_feed: checking for PUSH hub...");
478 $feed_hub_url = false;
480 $links = $rss->get_links('hub');
482 if ($links && is_array($links)) {
483 foreach ($links as $l) {
489 if ($debug_enabled) _debug("update_rss_feed: feed hub url: $feed_hub_url");
491 if ($feed_hub_url && function_exists('curl_init') &&
492 !ini_get("open_basedir")) {
494 require_once 'lib/pubsubhubbub/subscriber.php';
496 $callback_url = get_self_url_prefix() .
497 "/public.php?op=pubsub&id=$feed";
499 $s = new Subscriber($feed_hub_url, $callback_url);
501 $rc = $s->subscribe($fetch_url);
504 _debug("update_rss_feed: feed hub url found, subscribe request sent.");
506 db_query($link, "UPDATE ttrss_feeds SET pubsub_state = 1
507 WHERE id = '$feed'");
511 if ($debug_enabled) {
512 _debug("update_rss_feed: processing articles...");
515 foreach ($items as $item) {
516 if ($_REQUEST['xdebug'] == 3) {
520 $entry_guid = $item->get_id();
521 if (!$entry_guid) $entry_guid = $item->get_link();
522 if (!$entry_guid) $entry_guid = make_guid_from_title($item->get_title());
524 if ($debug_enabled) {
525 _debug("update_rss_feed: guid $entry_guid");
528 if (!$entry_guid) continue;
530 $entry_guid = "$owner_uid,$entry_guid";
532 $entry_timestamp = "";
534 $entry_timestamp = strtotime($item->get_date());
536 if ($entry_timestamp == -1 ||
!$entry_timestamp ||
$entry_timestamp > time()) {
537 $entry_timestamp = time();
538 $no_orig_date = 'true';
540 $no_orig_date = 'false';
543 $entry_timestamp_fmt = strftime("%Y/%m/%d %H:%M:%S", $entry_timestamp);
545 if ($debug_enabled) {
546 _debug("update_rss_feed: date $entry_timestamp [$entry_timestamp_fmt]");
549 $entry_title = html_entity_decode($item->get_title());
551 $entry_link = rewrite_relative_url($site_url, $item->get_link());
553 if ($debug_enabled) {
554 _debug("update_rss_feed: title $entry_title");
555 _debug("update_rss_feed: link $entry_link");
558 if (!$entry_title) $entry_title = date("Y-m-d H:i:s", $entry_timestamp);;
560 $entry_content = $item->get_content();
561 if (!$entry_content) $entry_content = $item->get_description();
563 if ($_REQUEST["xdebug"] == 2) {
564 print "update_rss_feed: content: ";
565 print $entry_content;
569 $entry_comments = $item->data
["comments"];
571 if ($item->get_author()) {
572 $entry_author_item = $item->get_author();
573 $entry_author = $entry_author_item->get_name();
574 if (!$entry_author) $entry_author = $entry_author_item->get_email();
576 $entry_author = db_escape_string($link, $entry_author);
579 $entry_guid = db_escape_string($link, mb_substr($entry_guid, 0, 245));
581 $entry_comments = db_escape_string($link, mb_substr($entry_comments, 0, 245));
582 $entry_author = db_escape_string($link, mb_substr($entry_author, 0, 245));
584 $num_comments = $item->get_item_tags('http://purl.org/rss/1.0/modules/slash/', 'comments');
586 if (is_array($num_comments) && is_array($num_comments[0])) {
587 $num_comments = (int) $num_comments[0]["data"];
592 if ($debug_enabled) {
593 _debug("update_rss_feed: num_comments: $num_comments");
594 _debug("update_rss_feed: looking for tags [1]...");
597 // parse <category> entries into tags
599 $additional_tags = array();
601 $additional_tags_src = $item->get_categories();
603 if (is_array($additional_tags_src)) {
604 foreach ($additional_tags_src as $tobj) {
605 array_push($additional_tags, $tobj->get_term());
609 if ($debug_enabled) {
610 _debug("update_rss_feed: category tags:");
611 print_r($additional_tags);
614 if ($debug_enabled) {
615 _debug("update_rss_feed: looking for tags [2]...");
618 $entry_tags = array_unique($additional_tags);
620 for ($i = 0; $i < count($entry_tags); $i++
)
621 $entry_tags[$i] = mb_strtolower($entry_tags[$i], 'utf-8');
623 if ($debug_enabled) {
624 //_debug("update_rss_feed: unfiltered tags found:");
625 //print_r($entry_tags);
628 if ($debug_enabled) {
629 _debug("update_rss_feed: done collecting data.");
632 // TODO: less memory-hungry implementation
634 if ($debug_enabled) {
635 _debug("update_rss_feed: applying plugin filters..");
638 // FIXME not sure if owner_uid is a good idea here, we may have a base entry without user entry (?)
639 $result = db_query($link, "SELECT plugin_data,title,content,link,tag_cache,author FROM ttrss_entries, ttrss_user_entries
640 WHERE ref_id = id AND guid = '".db_escape_string($link, $entry_guid)."' AND owner_uid = $owner_uid");
642 if (db_num_rows($result) != 0) {
643 $entry_plugin_data = db_fetch_result($result, 0, "plugin_data");
644 $stored_article = array("title" => db_fetch_result($result, 0, "title"),
645 "content" => db_fetch_result($result, 0, "content"),
646 "link" => db_fetch_result($result, 0, "link"),
647 "tags" => explode(",", db_fetch_result($result, 0, "tag_cache")),
648 "author" => db_fetch_result($result, 0, "author"));
650 $entry_plugin_data = "";
651 $stored_article = array();
654 $article = array("owner_uid" => $owner_uid, // read only
655 "guid" => $entry_guid, // read only
656 "title" => $entry_title,
657 "content" => $entry_content,
658 "link" => $entry_link,
659 "tags" => $entry_tags,
660 "plugin_data" => $entry_plugin_data,
661 "author" => $entry_author,
662 "stored" => $stored_article);
664 foreach ($pluginhost->get_hooks($pluginhost::HOOK_ARTICLE_FILTER
) as $plugin) {
665 $article = $plugin->hook_article_filter($article);
668 $entry_tags = $article["tags"];
669 $entry_guid = db_escape_string($link, $entry_guid);
670 $entry_title = db_escape_string($link, $article["title"]);
671 $entry_author = db_escape_string($link, $article["author"]);
672 $entry_link = db_escape_string($link, $article["link"]);
673 $entry_plugin_data = db_escape_string($link, $article["plugin_data"]);
674 $entry_content = $article["content"]; // escaped below
677 if ($debug_enabled) {
678 _debug("update_rss_feed: plugin data: $entry_plugin_data");
681 if ($cache_images && is_writable(CACHE_DIR
. '/images'))
682 cache_images($entry_content, $site_url, $debug_enabled);
684 $entry_content = db_escape_string($link, $entry_content, false);
686 $content_hash = "SHA1:" . sha1($entry_content);
688 db_query($link, "BEGIN");
690 $result = db_query($link, "SELECT id FROM ttrss_entries
691 WHERE guid = '$entry_guid'");
693 if (db_num_rows($result) == 0) {
695 if ($debug_enabled) {
696 _debug("update_rss_feed: base guid [$entry_guid] not found");
699 // base post entry does not exist, create it
701 $result = db_query($link,
702 "INSERT INTO ttrss_entries
721 '$entry_timestamp_fmt',
727 '$date_feed_processed',
730 '$entry_plugin_data',
733 $article_labels = array();
736 // we keep encountering the entry in feeds, so we need to
737 // update date_updated column so that we don't get horrible
738 // dupes when the entry gets purged and reinserted again e.g.
739 // in the case of SLOW SLOW OMG SLOW updating feeds
741 $base_entry_id = db_fetch_result($result, 0, "id");
743 db_query($link, "UPDATE ttrss_entries SET date_updated = NOW()
744 WHERE id = '$base_entry_id'");
746 $article_labels = get_article_labels($link, $base_entry_id, $owner_uid);
749 // now it should exist, if not - bad luck then
751 $result = db_query($link, "SELECT
752 id,content_hash,no_orig_date,title,plugin_data,
753 ".SUBSTRING_FOR_DATE
."(date_updated,1,19) as date_updated,
754 ".SUBSTRING_FOR_DATE
."(updated,1,19) as updated,
758 WHERE guid = '$entry_guid'");
763 if (db_num_rows($result) == 1) {
765 if ($debug_enabled) {
766 _debug("update_rss_feed: base guid [$entry_guid] found, checking for user record");
769 // this will be used below in update handler
770 $orig_content_hash = db_fetch_result($result, 0, "content_hash");
771 $orig_title = db_fetch_result($result, 0, "title");
772 $orig_num_comments = db_fetch_result($result, 0, "num_comments");
773 $orig_date_updated = strtotime(db_fetch_result($result,
775 $orig_plugin_data = db_fetch_result($result, 0, "plugin_data");
777 $ref_id = db_fetch_result($result, 0, "id");
778 $entry_ref_id = $ref_id;
780 // check for user post link to main table
782 // do we allow duplicate posts with same GUID in different feeds?
783 if (get_pref($link, "ALLOW_DUPLICATE_POSTS", $owner_uid, false)) {
784 $dupcheck_qpart = "AND (feed_id = '$feed' OR feed_id IS NULL)";
786 $dupcheck_qpart = "";
789 /* Collect article tags here so we could filter by them: */
791 $article_filters = get_article_filters($filters, $entry_title,
792 $entry_content, $entry_link, $entry_timestamp, $entry_author,
795 if ($debug_enabled) {
796 _debug("update_rss_feed: article filters: ");
797 if (count($article_filters) != 0) {
798 print_r($article_filters);
802 if (find_article_filter($article_filters, "filter")) {
803 db_query($link, "COMMIT"); // close transaction in progress
807 $score = calculate_article_score($article_filters);
809 if ($debug_enabled) {
810 _debug("update_rss_feed: initial score: $score");
813 $query = "SELECT ref_id, int_id FROM ttrss_user_entries WHERE
814 ref_id = '$ref_id' AND owner_uid = '$owner_uid'
817 // if ($_REQUEST["xdebug"]) print "$query\n";
819 $result = db_query($link, $query);
821 // okay it doesn't exist - create user entry
822 if (db_num_rows($result) == 0) {
824 if ($debug_enabled) {
825 _debug("update_rss_feed: user record not found, creating...");
828 if ($score >= -500 && !find_article_filter($article_filters, 'catchup')) {
830 $last_read_qpart = 'NULL';
833 $last_read_qpart = 'NOW()';
836 if (find_article_filter($article_filters, 'mark') ||
$score > 1000) {
842 if (find_article_filter($article_filters, 'publish')) {
845 $published = 'false';
850 if (DB_TYPE
== "pgsql" and defined('_NGRAM_TITLE_DUPLICATE_THRESHOLD')) {
852 $result = db_query($link, "SELECT COUNT(*) AS similar FROM
853 ttrss_entries,ttrss_user_entries
854 WHERE ref_id = id AND updated >= NOW() - INTERVAL '7 day'
855 AND similarity(title, '$entry_title') >= "._NGRAM_TITLE_DUPLICATE_THRESHOLD
."
856 AND owner_uid = $owner_uid");
858 $ngram_similar = db_fetch_result($result, 0, "similar");
860 if ($debug_enabled) {
861 _debug("update_rss_feed: N-gram similar results: $ngram_similar");
864 if ($ngram_similar > 0) {
869 $last_marked = ($marked == 'true') ?
'NOW()' : 'NULL';
870 $last_published = ($published == 'true') ?
'NOW()' : 'NULL';
872 $result = db_query($link,
873 "INSERT INTO ttrss_user_entries
874 (ref_id, owner_uid, feed_id, unread, last_read, marked,
875 published, score, tag_cache, label_cache, uuid,
876 last_marked, last_published)
877 VALUES ('$ref_id', '$owner_uid', '$feed', $unread,
878 $last_read_qpart, $marked, $published, '$score', '', '',
879 '', $last_marked, $last_published)");
881 if (PUBSUBHUBBUB_HUB
&& $published == 'true') {
882 $rss_link = get_self_url_prefix() .
883 "/public.php?op=rss&id=-2&key=" .
884 get_feed_access_key($link, -2, false, $owner_uid);
886 $p = new Publisher(PUBSUBHUBBUB_HUB
);
888 $pubsub_result = $p->publish_update($rss_link);
891 $result = db_query($link,
892 "SELECT int_id FROM ttrss_user_entries WHERE
893 ref_id = '$ref_id' AND owner_uid = '$owner_uid' AND
894 feed_id = '$feed' LIMIT 1");
896 if (db_num_rows($result) == 1) {
897 $entry_int_id = db_fetch_result($result, 0, "int_id");
900 if ($debug_enabled) {
901 _debug("update_rss_feed: user record FOUND");
904 $entry_ref_id = db_fetch_result($result, 0, "ref_id");
905 $entry_int_id = db_fetch_result($result, 0, "int_id");
908 if ($debug_enabled) {
909 _debug("update_rss_feed: RID: $entry_ref_id, IID: $entry_int_id");
912 $post_needs_update = false;
913 $update_insignificant = false;
915 if ($orig_num_comments != $num_comments) {
916 $post_needs_update = true;
917 $update_insignificant = true;
920 if ($entry_plugin_data != $orig_plugin_data) {
921 $post_needs_update = true;
922 $update_insignificant = true;
925 if ($content_hash != $orig_content_hash) {
926 $post_needs_update = true;
927 $update_insignificant = false;
930 if (db_escape_string($link, $orig_title) != $entry_title) {
931 $post_needs_update = true;
932 $update_insignificant = false;
935 // if post needs update, update it and mark all user entries
936 // linking to this post as updated
937 if ($post_needs_update) {
939 if (defined('DAEMON_EXTENDED_DEBUG')) {
940 _debug("update_rss_feed: post $entry_guid needs update...");
943 // print "<!-- post $orig_title needs update : $post_needs_update -->";
945 db_query($link, "UPDATE ttrss_entries
946 SET title = '$entry_title', content = '$entry_content',
947 content_hash = '$content_hash',
948 updated = '$entry_timestamp_fmt',
949 num_comments = '$num_comments',
950 plugin_data = '$entry_plugin_data'
951 WHERE id = '$ref_id'");
953 if (!$update_insignificant) {
954 if ($mark_unread_on_update) {
955 db_query($link, "UPDATE ttrss_user_entries
956 SET last_read = null, unread = true WHERE ref_id = '$ref_id'");
962 db_query($link, "COMMIT");
964 if ($debug_enabled) {
965 _debug("update_rss_feed: assigning labels...");
968 assign_article_to_label_filters($link, $entry_ref_id, $article_filters,
969 $owner_uid, $article_labels);
971 if ($debug_enabled) {
972 _debug("update_rss_feed: looking for enclosures...");
977 $enclosures = array();
979 $encs = $item->get_enclosures();
981 if (is_array($encs)) {
982 foreach ($encs as $e) {
984 $e->link
, $e->type
, $e->length
);
985 array_push($enclosures, $e_item);
989 if ($debug_enabled) {
990 _debug("update_rss_feed: article enclosures:");
991 print_r($enclosures);
994 db_query($link, "BEGIN");
996 foreach ($enclosures as $enc) {
997 $enc_url = db_escape_string($link, $enc[0]);
998 $enc_type = db_escape_string($link, $enc[1]);
999 $enc_dur = db_escape_string($link, $enc[2]);
1001 $result = db_query($link, "SELECT id FROM ttrss_enclosures
1002 WHERE content_url = '$enc_url' AND post_id = '$entry_ref_id'");
1004 if (db_num_rows($result) == 0) {
1005 db_query($link, "INSERT INTO ttrss_enclosures
1006 (content_url, content_type, title, duration, post_id) VALUES
1007 ('$enc_url', '$enc_type', '', '$enc_dur', '$entry_ref_id')");
1011 db_query($link, "COMMIT");
1013 // check for manual tags (we have to do it here since they're loaded from filters)
1015 foreach ($article_filters as $f) {
1016 if ($f["type"] == "tag") {
1018 $manual_tags = trim_array(explode(",", $f["param"]));
1020 foreach ($manual_tags as $tag) {
1021 if (tag_is_valid($tag)) {
1022 array_push($entry_tags, $tag);
1030 $boring_tags = trim_array(explode(",", mb_strtolower(get_pref($link,
1031 'BLACKLISTED_TAGS', $owner_uid, ''), 'utf-8')));
1033 $filtered_tags = array();
1034 $tags_to_cache = array();
1036 if ($entry_tags && is_array($entry_tags)) {
1037 foreach ($entry_tags as $tag) {
1038 if (array_search($tag, $boring_tags) === false) {
1039 array_push($filtered_tags, $tag);
1044 $filtered_tags = array_unique($filtered_tags);
1046 if ($debug_enabled) {
1047 _debug("update_rss_feed: filtered article tags:");
1048 print_r($filtered_tags);
1051 // Save article tags in the database
1053 if (count($filtered_tags) > 0) {
1055 db_query($link, "BEGIN");
1057 foreach ($filtered_tags as $tag) {
1059 $tag = sanitize_tag($tag);
1060 $tag = db_escape_string($link, $tag);
1062 if (!tag_is_valid($tag)) continue;
1064 $result = db_query($link, "SELECT id FROM ttrss_tags
1065 WHERE tag_name = '$tag' AND post_int_id = '$entry_int_id' AND
1066 owner_uid = '$owner_uid' LIMIT 1");
1068 if ($result && db_num_rows($result) == 0) {
1070 db_query($link, "INSERT INTO ttrss_tags
1071 (owner_uid,tag_name,post_int_id)
1072 VALUES ('$owner_uid','$tag', '$entry_int_id')");
1075 array_push($tags_to_cache, $tag);
1078 /* update the cache */
1080 $tags_to_cache = array_unique($tags_to_cache);
1082 $tags_str = db_escape_string($link, join(",", $tags_to_cache));
1084 db_query($link, "UPDATE ttrss_user_entries
1085 SET tag_cache = '$tags_str' WHERE ref_id = '$entry_ref_id'
1086 AND owner_uid = $owner_uid");
1088 db_query($link, "COMMIT");
1091 if (get_pref($link, "AUTO_ASSIGN_LABELS", $owner_uid, false)) {
1092 if ($debug_enabled) {
1093 _debug("update_rss_feed: auto-assigning labels...");
1096 foreach ($labels as $label) {
1097 $caption = preg_quote($label["caption"]);
1099 if ($caption && preg_match("/\b$caption\b/i", "$tags_str " . strip_tags($entry_content) . " $entry_title")) {
1100 if (!labels_contains_caption($article_labels, $caption)) {
1101 label_add_article($link, $entry_ref_id, $caption, $owner_uid);
1107 if ($debug_enabled) {
1108 _debug("update_rss_feed: article processed");
1112 if (!$last_updated) {
1113 if ($debug_enabled) {
1114 _debug("update_rss_feed: new feed, catching it up...");
1116 catchup_feed($link, $feed, false, $owner_uid);
1119 if ($debug_enabled) {
1120 _debug("purging feed...");
1123 purge_feed($link, $feed, 0, $debug_enabled);
1125 db_query($link, "UPDATE ttrss_feeds
1126 SET last_updated = NOW(), last_error = '' WHERE id = '$feed'");
1128 // db_query($link, "COMMIT");
1132 $error_msg = db_escape_string($link, mb_substr($rss->error(), 0, 245));
1134 if ($debug_enabled) {
1135 _debug("update_rss_feed: error fetching feed: $error_msg");
1139 "UPDATE ttrss_feeds SET last_error = '$error_msg',
1140 last_updated = NOW() WHERE id = '$feed'");
1145 if ($debug_enabled) {
1146 _debug("update_rss_feed: done");
1151 function cache_images($html, $site_url, $debug) {
1152 $cache_dir = CACHE_DIR
. "/images";
1154 libxml_use_internal_errors(true);
1156 $charset_hack = '<head>
1157 <meta http-equiv="Content-Type" content="text/html; charset=utf-8"/>
1160 $doc = new DOMDocument();
1161 $doc->loadHTML($charset_hack . $html);
1162 $xpath = new DOMXPath($doc);
1164 $entries = $xpath->query('(//img[@src])');
1166 foreach ($entries as $entry) {
1167 if ($entry->hasAttribute('src')) {
1168 $src = rewrite_relative_url($site_url, $entry->getAttribute('src'));
1170 $local_filename = CACHE_DIR
. "/images/" . sha1($src) . ".png";
1172 if ($debug) _debug("cache_images: downloading: $src to $local_filename");
1174 if (!file_exists($local_filename)) {
1175 $file_content = fetch_file_contents($src);
1177 if ($file_content && strlen($file_content) > 1024) {
1178 file_put_contents($local_filename, $file_content);
1182 if (file_exists($local_filename)) {
1183 $entry->setAttribute('src', SELF_URL_PATH
. '/image.php?url=' .
1184 base64_encode($src));
1189 $node = $doc->getElementsByTagName('body')->item(0);
1191 return $doc->saveXML($node);
1194 function expire_lock_files($debug) {
1195 if ($debug) _debug("Removing old lock files...");
1199 if (is_writable(LOCK_DIRECTORY
)) {
1200 $files = glob(LOCK_DIRECTORY
. "/*.lock");
1203 foreach ($files as $file) {
1204 if (!file_is_locked($file) && time() - filemtime($file) > 86400*2) {
1212 if ($debug) _debug("Removed $num_deleted files.");
1215 function expire_cached_files($debug) {
1216 foreach (array("simplepie", "images", "export", "upload") as $dir) {
1217 $cache_dir = CACHE_DIR
. "/$dir";
1219 if ($debug) _debug("Expiring $cache_dir");
1223 if (is_writable($cache_dir)) {
1224 $files = glob("$cache_dir/*");
1227 foreach ($files as $file) {
1228 if (time() - filemtime($file) > 86400*7) {
1237 if ($debug) _debug("Removed $num_deleted files.");
1242 * Source: http://www.php.net/manual/en/function.parse-url.php#104527
1243 * Returns the url query as associative array
1245 * @param string query
1246 * @return array params
1248 function convertUrlQuery($query) {
1249 $queryParts = explode('&', $query);
1253 foreach ($queryParts as $param) {
1254 $item = explode('=', $param);
1255 $params[$item[0]] = $item[1];
1261 function get_article_filters($filters, $title, $content, $link, $timestamp, $author, $tags) {
1264 foreach ($filters as $filter) {
1265 $match_any_rule = $filter["match_any_rule"];
1266 $inverse = $filter["inverse"];
1267 $filter_match = false;
1269 foreach ($filter["rules"] as $rule) {
1271 $reg_exp = $rule["reg_exp"];
1272 $rule_inverse = $rule["inverse"];
1277 switch ($rule["type"]) {
1279 $match = @preg_match
("/$reg_exp/i", $title);
1282 // we don't need to deal with multiline regexps
1283 $content = preg_replace("/[\r\n\t]/", "", $content);
1285 $match = @preg_match
("/$reg_exp/i", $content);
1288 // we don't need to deal with multiline regexps
1289 $content = preg_replace("/[\r\n\t]/", "", $content);
1291 $match = (@preg_match
("/$reg_exp/i", $title) || @preg_match
("/$reg_exp/i", $content));
1294 $match = @preg_match
("/$reg_exp/i", $link);
1297 $match = @preg_match
("/$reg_exp/i", $author);
1300 $tag_string = join(",", $tags);
1301 $match = @preg_match
("/$reg_exp/i", $tag_string);
1305 if ($rule_inverse) $match = !$match;
1307 if ($match_any_rule) {
1309 $filter_match = true;
1313 $filter_match = $match;
1320 if ($inverse) $filter_match = !$filter_match;
1322 if ($filter_match) {
1323 foreach ($filter["actions"] AS $action) {
1324 array_push($matches, $action);
1326 // if Stop action encountered, perform no further processing
1327 if ($action["type"] == "stop") return $matches;
1335 function find_article_filter($filters, $filter_name) {
1336 foreach ($filters as $f) {
1337 if ($f["type"] == $filter_name) {
1344 function find_article_filters($filters, $filter_name) {
1347 foreach ($filters as $f) {
1348 if ($f["type"] == $filter_name) {
1349 array_push($results, $f);
1355 function calculate_article_score($filters) {
1358 foreach ($filters as $f) {
1359 if ($f["type"] == "score") {
1360 $score +
= $f["param"];
1366 function labels_contains_caption($labels, $caption) {
1367 foreach ($labels as $label) {
1368 if ($label[1] == $caption) {
1376 function assign_article_to_label_filters($link, $id, $filters, $owner_uid, $article_labels) {
1377 foreach ($filters as $f) {
1378 if ($f["type"] == "label") {
1379 if (!labels_contains_caption($article_labels, $f["param"])) {
1380 label_add_article($link, $id, $f["param"], $owner_uid);
1386 function make_guid_from_title($title) {
1387 return preg_replace("/[ \"\',.:;]/", "-",
1388 mb_strtolower(strip_tags($title), 'utf-8'));