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, $line["id"], true);
153 sleep(1); // prevent flood (FIXME make this an option?)
156 require_once "digest.php";
158 // Send feed digests by email if needed.
159 send_headlines_digests($link, $debug);
161 } // function update_daemon_common
163 // ignore_daemon is not used
164 function update_rss_feed($link, $feed, $ignore_daemon = false, $no_cache = false,
165 $override_url = false) {
167 require_once "lib/simplepie/simplepie.inc";
169 $debug_enabled = defined('DAEMON_EXTENDED_DEBUG') ||
$_REQUEST['xdebug'];
171 if ($debug_enabled) {
172 _debug("update_rss_feed: start");
175 $result = db_query($link, "SELECT id,update_interval,auth_login,
176 feed_url,auth_pass,cache_images,last_updated,cache_content,
177 mark_unread_on_update, owner_uid, update_on_checksum_change,
179 FROM ttrss_feeds WHERE id = '$feed'");
181 if (db_num_rows($result) == 0) {
182 if ($debug_enabled) {
183 _debug("update_rss_feed: feed $feed NOT FOUND/SKIPPED");
188 $last_updated = db_fetch_result($result, 0, "last_updated");
189 $owner_uid = db_fetch_result($result, 0, "owner_uid");
190 $mark_unread_on_update = sql_bool_to_bool(db_fetch_result($result,
191 0, "mark_unread_on_update"));
192 $update_on_checksum_change = sql_bool_to_bool(db_fetch_result($result,
193 0, "update_on_checksum_change"));
194 $pubsub_state = db_fetch_result($result, 0, "pubsub_state");
196 db_query($link, "UPDATE ttrss_feeds SET last_update_started = NOW()
197 WHERE id = '$feed'");
199 $auth_login = db_fetch_result($result, 0, "auth_login");
200 $auth_pass = db_fetch_result($result, 0, "auth_pass");
202 $cache_images = sql_bool_to_bool(db_fetch_result($result, 0, "cache_images"));
203 $cache_content = sql_bool_to_bool(db_fetch_result($result, 0, "cache_content"));
204 $fetch_url = db_fetch_result($result, 0, "feed_url");
206 $feed = db_escape_string($feed);
208 if ($auth_login && $auth_pass ){
209 $url_parts = array();
210 preg_match("/(^[^:]*):\/\/(.*)/", $fetch_url, $url_parts);
212 if ($url_parts[1] && $url_parts[2]) {
213 $fetch_url = $url_parts[1] . "://$auth_login:$auth_pass@" . $url_parts[2];
219 $fetch_url = $override_url;
221 if ($debug_enabled) {
222 _debug("update_rss_feed: fetching [$fetch_url]...");
225 // Ignore cache if new feed or manual update.
226 $cache_age = (is_null($last_updated) ||
$last_updated == '1970-01-01 00:00:00') ?
227 -1 : get_feed_update_interval($link, $feed) * 60;
229 $simplepie_cache_dir = CACHE_DIR
. "/simplepie";
231 if (!is_dir($simplepie_cache_dir)) {
232 mkdir($simplepie_cache_dir);
235 $rss = new SimplePie();
236 $rss->set_useragent(SELF_USER_AGENT
);
237 $rss->set_timeout($no_cache ?
15 : 60);
238 $rss->set_feed_url($fetch_url);
239 $rss->set_output_encoding('UTF-8');
240 //$rss->force_feed(true);
242 if ($debug_enabled) {
243 _debug("feed update interval (sec): " .
244 get_feed_update_interval($link, $feed)*60);
247 $rss->enable_cache(!$no_cache);
250 $rss->set_cache_location($simplepie_cache_dir);
251 $rss->set_cache_duration($cache_age);
258 if ($debug_enabled) {
259 _debug("update_rss_feed: fetch done, parsing...");
262 $feed = db_escape_string($feed);
264 if (!$rss->error()) {
267 $pluginhost->run_hooks($pluginhost::HOOK_FEED_PARSED
, "hook_feed_parsed", $rss);
269 if ($debug_enabled) {
270 _debug("update_rss_feed: processing feed data...");
273 // db_query($link, "BEGIN");
275 if (DB_TYPE
== "pgsql") {
276 $favicon_interval_qpart = "favicon_last_checked < NOW() - INTERVAL '12 hour'";
278 $favicon_interval_qpart = "favicon_last_checked < DATE_SUB(NOW(), INTERVAL 12 HOUR)";
281 $result = db_query($link, "SELECT title,icon_url,site_url,owner_uid,
282 (favicon_last_checked IS NULL OR $favicon_interval_qpart) AS
284 FROM ttrss_feeds WHERE id = '$feed'");
286 $registered_title = db_fetch_result($result, 0, "title");
287 $orig_icon_url = db_fetch_result($result, 0, "icon_url");
288 $orig_site_url = db_fetch_result($result, 0, "site_url");
289 $favicon_needs_check = sql_bool_to_bool(db_fetch_result($result, 0,
290 "favicon_needs_check"));
292 $owner_uid = db_fetch_result($result, 0, "owner_uid");
294 $site_url = db_escape_string(mb_substr(rewrite_relative_url($fetch_url, $rss->get_link()), 0, 250));
296 if ($debug_enabled) {
297 _debug("update_rss_feed: checking favicon...");
300 if ($favicon_needs_check) {
301 check_feed_favicon($site_url, $feed, $link);
303 db_query($link, "UPDATE ttrss_feeds SET favicon_last_checked = NOW()
304 WHERE id = '$feed'");
307 if (!$registered_title ||
$registered_title == "[Unknown]") {
309 $feed_title = db_escape_string($rss->get_title());
311 if ($debug_enabled) {
312 _debug("update_rss_feed: registering title: $feed_title");
315 db_query($link, "UPDATE ttrss_feeds SET
316 title = '$feed_title' WHERE id = '$feed'");
319 if ($site_url && $orig_site_url != $site_url) {
320 db_query($link, "UPDATE ttrss_feeds SET
321 site_url = '$site_url' WHERE id = '$feed'");
324 $icon_url = db_escape_string(mb_substr(
325 rewrite_relative_url($fetch_url, $rss->get_image_url()), 0, 250));
327 if ($icon_url && $orig_icon_url != $icon_url) {
328 db_query($link, "UPDATE ttrss_feeds SET icon_url = '$icon_url' WHERE id = '$feed'");
331 if ($debug_enabled) {
332 _debug("update_rss_feed: loading filters & labels...");
335 $filters = load_filters($link, $feed, $owner_uid);
336 $labels = get_all_labels($link, $owner_uid);
338 if ($debug_enabled) {
340 _debug("update_rss_feed: " . count($filters) . " filters loaded.");
343 $items = $rss->get_items();
345 if (!is_array($items)) {
346 if ($debug_enabled) {
347 _debug("update_rss_feed: no articles found.");
350 db_query($link, "UPDATE ttrss_feeds
351 SET last_updated = NOW(), last_error = '' WHERE id = '$feed'");
353 return; // no articles
356 if ($pubsub_state != 2 && PUBSUBHUBBUB_ENABLED
) {
358 if ($debug_enabled) _debug("update_rss_feed: checking for PUSH hub...");
360 $feed_hub_url = false;
362 $links = $rss->get_links('hub');
364 if ($links && is_array($links)) {
365 foreach ($links as $l) {
371 if ($debug_enabled) _debug("update_rss_feed: feed hub url: $feed_hub_url");
373 if ($feed_hub_url && function_exists('curl_init') &&
374 !ini_get("open_basedir")) {
376 require_once 'lib/pubsubhubbub/subscriber.php';
378 $callback_url = get_self_url_prefix() .
379 "/public.php?op=pubsub&id=$feed";
381 $s = new Subscriber($feed_hub_url, $callback_url);
383 $rc = $s->subscribe($fetch_url);
386 _debug("update_rss_feed: feed hub url found, subscribe request sent.");
388 db_query($link, "UPDATE ttrss_feeds SET pubsub_state = 1
389 WHERE id = '$feed'");
393 if ($debug_enabled) {
394 _debug("update_rss_feed: processing articles...");
397 foreach ($items as $item) {
398 if ($_REQUEST['xdebug'] == 2) {
402 $entry_guid = $item->get_id();
403 if (!$entry_guid) $entry_guid = $item->get_link();
404 if (!$entry_guid) $entry_guid = make_guid_from_title($item->get_title());
406 if ($cache_content) {
407 $entry_guid = "ccache:$entry_guid";
410 if ($auth_login ||
$auth_pass) {
411 $entry_guid = "auth,$owner_uid:$entry_guid";
414 if ($debug_enabled) {
415 _debug("update_rss_feed: guid $entry_guid");
418 if (!$entry_guid) continue;
420 $entry_timestamp = "";
422 $entry_timestamp = strtotime($item->get_date());
424 if ($entry_timestamp == -1 ||
!$entry_timestamp) {
425 $entry_timestamp = time();
426 $no_orig_date = 'true';
428 $no_orig_date = 'false';
431 $entry_timestamp_fmt = strftime("%Y/%m/%d %H:%M:%S", $entry_timestamp);
433 if ($debug_enabled) {
434 _debug("update_rss_feed: date $entry_timestamp [$entry_timestamp_fmt]");
437 $entry_title = $item->get_title();
439 $entry_link = rewrite_relative_url($site_url, htmlspecialchars_decode($item->get_link()));
441 if ($debug_enabled) {
442 _debug("update_rss_feed: title $entry_title");
443 _debug("update_rss_feed: link $entry_link");
446 if (!$entry_title) $entry_title = date("Y-m-d H:i:s", $entry_timestamp);;
448 $entry_content = $item->get_content();
449 if (!$entry_content) $entry_content = $item->get_description();
451 if ($cache_images && is_writable(CACHE_DIR
. '/images'))
452 $entry_content = cache_images($entry_content, $site_url, $debug_enabled);
454 if ($_REQUEST["xdebug"] == 2) {
455 print "update_rss_feed: content: ";
456 print $entry_content;
460 $entry_cached_content = "";
462 $entry_comments = $item->data
["comments"];
464 if ($item->get_author()) {
465 $entry_author_item = $item->get_author();
466 $entry_author = $entry_author_item->get_name();
467 if (!$entry_author) $entry_author = $entry_author_item->get_email();
469 $entry_author = db_escape_string($entry_author);
472 $entry_guid = db_escape_string(mb_substr($entry_guid, 0, 250));
474 $result = db_query($link, "SELECT id FROM ttrss_entries
475 WHERE guid = '$entry_guid'");
477 $entry_comments = db_escape_string(mb_substr($entry_comments, 0, 250));
478 $entry_author = db_escape_string(mb_substr($entry_author, 0, 250));
480 $num_comments = $item->get_item_tags('http://purl.org/rss/1.0/modules/slash/', 'comments');
482 if (is_array($num_comments) && is_array($num_comments[0])) {
483 $num_comments = (int) $num_comments[0]["data"];
488 if ($debug_enabled) {
489 _debug("update_rss_feed: num_comments: $num_comments");
490 _debug("update_rss_feed: looking for tags [1]...");
493 // parse <category> entries into tags
495 $additional_tags = array();
497 $additional_tags_src = $item->get_categories();
499 if (is_array($additional_tags_src)) {
500 foreach ($additional_tags_src as $tobj) {
501 array_push($additional_tags, $tobj->get_term());
505 if ($debug_enabled) {
506 _debug("update_rss_feed: category tags:");
507 print_r($additional_tags);
510 if ($debug_enabled) {
511 _debug("update_rss_feed: looking for tags [2]...");
514 $entry_tags = array_unique($additional_tags);
516 for ($i = 0; $i < count($entry_tags); $i++
)
517 $entry_tags[$i] = mb_strtolower($entry_tags[$i], 'utf-8');
519 if ($debug_enabled) {
520 //_debug("update_rss_feed: unfiltered tags found:");
521 //print_r($entry_tags);
524 if ($debug_enabled) {
525 _debug("update_rss_feed: done collecting data [TITLE:$entry_title]");
528 // TODO: less memory-hungry implementation
530 if ($debug_enabled) {
531 _debug("update_rss_feed: applying plugin filters..");
534 $article = array("owner_uid" => $owner_uid,
535 "title" => $entry_title,
536 "content" => $entry_content,
537 "link" => $entry_link,
538 "tags" => $entry_tags,
539 "author" => $entry_author);
542 foreach ($pluginhost->get_hooks($pluginhost::HOOK_ARTICLE_FILTER
) as $plugin) {
543 $article = $plugin->hook_article_filter($article);
546 $entry_tags = $article["tags"];
547 $entry_content = db_escape_string($article["content"], false);
548 $entry_title = db_escape_string($article["title"]);
549 $entry_author = db_escape_string($article["author"]);
550 $entry_link = db_escape_string($article["link"]);
552 $content_hash = "SHA1:" . sha1(strip_tags($entry_content));
554 db_query($link, "BEGIN");
556 if (db_num_rows($result) == 0) {
558 if ($debug_enabled) {
559 _debug("update_rss_feed: base guid not found");
562 if ($cache_content) {
563 if ($debug_enabled) {
564 _debug("update_rss_feed: caching content (initial)...");
567 $entry_cached_content = cache_content($link, $entry_link, $auth_login, $auth_pass);
569 if ($cache_images && is_writable(CACHE_DIR
. '/images'))
570 $entry_cached_content = cache_images($entry_cached_content, $site_url, $debug_enabled);
572 $entry_cached_content = db_escape_string($entry_cached_content, false);
575 // base post entry does not exist, create it
577 $result = db_query($link,
578 "INSERT INTO ttrss_entries
596 '$entry_timestamp_fmt',
599 '$entry_cached_content',
607 $article_labels = array();
610 // we keep encountering the entry in feeds, so we need to
611 // update date_updated column so that we don't get horrible
612 // dupes when the entry gets purged and reinserted again e.g.
613 // in the case of SLOW SLOW OMG SLOW updating feeds
615 $base_entry_id = db_fetch_result($result, 0, "id");
617 db_query($link, "UPDATE ttrss_entries SET date_updated = NOW()
618 WHERE id = '$base_entry_id'");
620 $article_labels = get_article_labels($link, $base_entry_id, $owner_uid);
623 // now it should exist, if not - bad luck then
625 $result = db_query($link, "SELECT
626 id,content_hash,no_orig_date,title,
627 ".SUBSTRING_FOR_DATE
."(date_updated,1,19) as date_updated,
628 ".SUBSTRING_FOR_DATE
."(updated,1,19) as updated,
629 num_comments, cached_content
632 WHERE guid = '$entry_guid'");
637 if (db_num_rows($result) == 1) {
639 if ($debug_enabled) {
640 _debug("update_rss_feed: base guid found, checking for user record");
643 // this will be used below in update handler
644 $orig_content_hash = db_fetch_result($result, 0, "content_hash");
645 $orig_title = db_fetch_result($result, 0, "title");
646 $orig_num_comments = db_fetch_result($result, 0, "num_comments");
647 $orig_cached_content = trim(db_fetch_result($result, 0, "cached_content"));
648 $orig_date_updated = strtotime(db_fetch_result($result,
651 $ref_id = db_fetch_result($result, 0, "id");
652 $entry_ref_id = $ref_id;
654 // check for user post link to main table
656 // do we allow duplicate posts with same GUID in different feeds?
657 if (get_pref($link, "ALLOW_DUPLICATE_POSTS", $owner_uid, false)) {
658 $dupcheck_qpart = "AND (feed_id = '$feed' OR feed_id IS NULL)";
660 $dupcheck_qpart = "";
663 /* Collect article tags here so we could filter by them: */
665 $article_filters = get_article_filters($filters, $entry_title,
666 $entry_content, $entry_link, $entry_timestamp, $entry_author,
669 if ($debug_enabled) {
670 _debug("update_rss_feed: article filters: ");
671 if (count($article_filters) != 0) {
672 print_r($article_filters);
676 if (find_article_filter($article_filters, "filter")) {
677 db_query($link, "COMMIT"); // close transaction in progress
681 $score = calculate_article_score($article_filters);
683 if ($debug_enabled) {
684 _debug("update_rss_feed: initial score: $score");
687 $query = "SELECT ref_id, int_id FROM ttrss_user_entries WHERE
688 ref_id = '$ref_id' AND owner_uid = '$owner_uid'
691 // if ($_REQUEST["xdebug"]) print "$query\n";
693 $result = db_query($link, $query);
695 // okay it doesn't exist - create user entry
696 if (db_num_rows($result) == 0) {
698 if ($debug_enabled) {
699 _debug("update_rss_feed: user record not found, creating...");
702 if ($score >= -500 && !find_article_filter($article_filters, 'catchup')) {
704 $last_read_qpart = 'NULL';
707 $last_read_qpart = 'NOW()';
710 if (find_article_filter($article_filters, 'mark') ||
$score > 1000) {
716 if (find_article_filter($article_filters, 'publish')) {
719 $published = 'false';
724 if (DB_TYPE
== "pgsql" and defined('_NGRAM_TITLE_DUPLICATE_THRESHOLD')) {
726 $result = db_query($link, "SELECT COUNT(*) AS similar FROM
727 ttrss_entries,ttrss_user_entries
728 WHERE ref_id = id AND updated >= NOW() - INTERVAL '7 day'
729 AND similarity(title, '$entry_title') >= "._NGRAM_TITLE_DUPLICATE_THRESHOLD
."
730 AND owner_uid = $owner_uid");
732 $ngram_similar = db_fetch_result($result, 0, "similar");
734 if ($debug_enabled) {
735 _debug("update_rss_feed: N-gram similar results: $ngram_similar");
738 if ($ngram_similar > 0) {
743 $result = db_query($link,
744 "INSERT INTO ttrss_user_entries
745 (ref_id, owner_uid, feed_id, unread, last_read, marked,
746 published, score, tag_cache, label_cache, uuid)
747 VALUES ('$ref_id', '$owner_uid', '$feed', $unread,
748 $last_read_qpart, $marked, $published, '$score', '', '', '')");
750 if (PUBSUBHUBBUB_HUB
&& $published == 'true') {
751 $rss_link = get_self_url_prefix() .
752 "/public.php?op=rss&id=-2&key=" .
753 get_feed_access_key($link, -2, false, $owner_uid);
755 $p = new Publisher(PUBSUBHUBBUB_HUB
);
757 $pubsub_result = $p->publish_update($rss_link);
760 $result = db_query($link,
761 "SELECT int_id FROM ttrss_user_entries WHERE
762 ref_id = '$ref_id' AND owner_uid = '$owner_uid' AND
763 feed_id = '$feed' LIMIT 1");
765 if (db_num_rows($result) == 1) {
766 $entry_int_id = db_fetch_result($result, 0, "int_id");
769 if ($debug_enabled) {
770 _debug("update_rss_feed: user record FOUND");
773 $entry_ref_id = db_fetch_result($result, 0, "ref_id");
774 $entry_int_id = db_fetch_result($result, 0, "int_id");
777 if ($debug_enabled) {
778 _debug("update_rss_feed: RID: $entry_ref_id, IID: $entry_int_id");
781 $post_needs_update = false;
782 $update_insignificant = false;
783 $cached_content_needs_update = false;
785 if ($orig_num_comments != $num_comments) {
786 $post_needs_update = true;
787 $update_insignificant = true;
790 if ($content_hash != $orig_content_hash) {
791 $post_needs_update = true;
792 $update_insignificant = false;
793 $cached_content_needs_update = true;
796 if ($cache_content) {
797 if ($debug_enabled) {
798 _debug("update_rss_feed: caching content because original checksum changed...");
801 $entry_cached_content = cache_content($link, $entry_link, $auth_login, $auth_pass);
803 if ($entry_cached_content) {
804 if ($cache_images && is_writable(CACHE_DIR
. '/images'))
805 $entry_cached_content = cache_images($entry_cached_content, $site_url, $debug_enabled);
807 $entry_cached_content = db_escape_string($entry_cached_content, false);
808 $post_needs_update = true;
810 $entry_cached_content = db_escape_string($orig_cached_content);
813 $entry_cached_content = db_escape_string($orig_cached_content);
816 if (db_escape_string($orig_title) != $entry_title) {
817 $post_needs_update = true;
818 $update_insignificant = false;
821 // if post needs update, update it and mark all user entries
822 // linking to this post as updated
823 if ($post_needs_update) {
825 if (defined('DAEMON_EXTENDED_DEBUG')) {
826 _debug("update_rss_feed: post $entry_guid needs update...");
829 // print "<!-- post $orig_title needs update : $post_needs_update -->";
831 db_query($link, "UPDATE ttrss_entries
832 SET title = '$entry_title', content = '$entry_content',
833 content_hash = '$content_hash',
834 cached_content = '$entry_cached_content',
835 updated = '$entry_timestamp_fmt',
836 num_comments = '$num_comments'
837 WHERE id = '$ref_id'");
839 if (!$update_insignificant) {
840 if ($mark_unread_on_update) {
841 db_query($link, "UPDATE ttrss_user_entries
842 SET last_read = null, unread = true WHERE ref_id = '$ref_id'");
843 } else if ($update_on_checksum_change) {
844 db_query($link, "UPDATE ttrss_user_entries
845 SET last_read = null WHERE ref_id = '$ref_id'
846 AND unread = false");
852 db_query($link, "COMMIT");
854 if ($debug_enabled) {
855 _debug("update_rss_feed: assigning labels...");
858 assign_article_to_label_filters($link, $entry_ref_id, $article_filters,
859 $owner_uid, $article_labels);
861 if ($debug_enabled) {
862 _debug("update_rss_feed: looking for enclosures...");
867 $enclosures = array();
869 $encs = $item->get_enclosures();
871 if (is_array($encs)) {
872 foreach ($encs as $e) {
874 $e->link
, $e->type
, $e->length
);
875 array_push($enclosures, $e_item);
879 if ($debug_enabled) {
880 _debug("update_rss_feed: article enclosures:");
881 print_r($enclosures);
884 db_query($link, "BEGIN");
886 foreach ($enclosures as $enc) {
887 $enc_url = db_escape_string($enc[0]);
888 $enc_type = db_escape_string($enc[1]);
889 $enc_dur = db_escape_string($enc[2]);
891 $result = db_query($link, "SELECT id FROM ttrss_enclosures
892 WHERE content_url = '$enc_url' AND post_id = '$entry_ref_id'");
894 if (db_num_rows($result) == 0) {
895 db_query($link, "INSERT INTO ttrss_enclosures
896 (content_url, content_type, title, duration, post_id) VALUES
897 ('$enc_url', '$enc_type', '', '$enc_dur', '$entry_ref_id')");
901 db_query($link, "COMMIT");
903 // check for manual tags (we have to do it here since they're loaded from filters)
905 foreach ($article_filters as $f) {
906 if ($f["type"] == "tag") {
908 $manual_tags = trim_array(explode(",", $f["param"]));
910 foreach ($manual_tags as $tag) {
911 if (tag_is_valid($tag)) {
912 array_push($entry_tags, $tag);
920 $boring_tags = trim_array(explode(",", mb_strtolower(get_pref($link,
921 'BLACKLISTED_TAGS', $owner_uid, ''), 'utf-8')));
923 $filtered_tags = array();
924 $tags_to_cache = array();
926 if ($entry_tags && is_array($entry_tags)) {
927 foreach ($entry_tags as $tag) {
928 if (array_search($tag, $boring_tags) === false) {
929 array_push($filtered_tags, $tag);
934 $filtered_tags = array_unique($filtered_tags);
936 if ($debug_enabled) {
937 _debug("update_rss_feed: filtered article tags:");
938 print_r($filtered_tags);
941 // Save article tags in the database
943 if (count($filtered_tags) > 0) {
945 db_query($link, "BEGIN");
947 foreach ($filtered_tags as $tag) {
949 $tag = sanitize_tag($tag);
950 $tag = db_escape_string($tag);
952 if (!tag_is_valid($tag)) continue;
954 $result = db_query($link, "SELECT id FROM ttrss_tags
955 WHERE tag_name = '$tag' AND post_int_id = '$entry_int_id' AND
956 owner_uid = '$owner_uid' LIMIT 1");
958 if ($result && db_num_rows($result) == 0) {
960 db_query($link, "INSERT INTO ttrss_tags
961 (owner_uid,tag_name,post_int_id)
962 VALUES ('$owner_uid','$tag', '$entry_int_id')");
965 array_push($tags_to_cache, $tag);
968 /* update the cache */
970 $tags_to_cache = array_unique($tags_to_cache);
972 $tags_str = db_escape_string(join(",", $tags_to_cache));
974 db_query($link, "UPDATE ttrss_user_entries
975 SET tag_cache = '$tags_str' WHERE ref_id = '$entry_ref_id'
976 AND owner_uid = $owner_uid");
978 db_query($link, "COMMIT");
981 if (get_pref($link, "AUTO_ASSIGN_LABELS", $owner_uid, false)) {
982 if ($debug_enabled) {
983 _debug("update_rss_feed: auto-assigning labels...");
986 foreach ($labels as $label) {
987 $caption = $label["caption"];
989 if (preg_match("/\b$caption\b/i", "$tags_str " . strip_tags($entry_content) . " $entry_title")) {
990 if (!labels_contains_caption($article_labels, $caption)) {
991 label_add_article($link, $entry_ref_id, $caption, $owner_uid);
997 if ($debug_enabled) {
998 _debug("update_rss_feed: article processed");
1002 if (!$last_updated) {
1003 if ($debug_enabled) {
1004 _debug("update_rss_feed: new feed, catching it up...");
1006 catchup_feed($link, $feed, false, $owner_uid);
1009 if ($debug_enabled) {
1010 _debug("purging feed...");
1013 purge_feed($link, $feed, 0, $debug_enabled);
1015 db_query($link, "UPDATE ttrss_feeds
1016 SET last_updated = NOW(), last_error = '' WHERE id = '$feed'");
1018 // db_query($link, "COMMIT");
1022 $error_msg = db_escape_string(mb_substr($rss->error(), 0, 250));
1024 if ($debug_enabled) {
1025 _debug("update_rss_feed: error fetching feed: $error_msg");
1029 "UPDATE ttrss_feeds SET last_error = '$error_msg',
1030 last_updated = NOW() WHERE id = '$feed'");
1035 if ($debug_enabled) {
1036 _debug("update_rss_feed: done");
1041 function cache_images($html, $site_url, $debug) {
1042 $cache_dir = CACHE_DIR
. "/images";
1044 libxml_use_internal_errors(true);
1046 $charset_hack = '<head>
1047 <meta http-equiv="Content-Type" content="text/html; charset=utf-8"/>
1050 $doc = new DOMDocument();
1051 $doc->loadHTML($charset_hack . $html);
1052 $xpath = new DOMXPath($doc);
1054 $entries = $xpath->query('(//img[@src])');
1056 foreach ($entries as $entry) {
1057 if ($entry->hasAttribute('src')) {
1058 $src = rewrite_relative_url($site_url, $entry->getAttribute('src'));
1060 $local_filename = CACHE_DIR
. "/images/" . sha1($src) . ".png";
1062 if ($debug) _debug("cache_images: downloading: $src to $local_filename");
1064 if (!file_exists($local_filename)) {
1065 $file_content = fetch_file_contents($src);
1067 if ($file_content && strlen($file_content) > 1024) {
1068 file_put_contents($local_filename, $file_content);
1072 if (file_exists($local_filename)) {
1073 $entry->setAttribute('src', SELF_URL_PATH
. '/image.php?url=' .
1074 base64_encode($src));
1079 $node = $doc->getElementsByTagName('body')->item(0);
1081 return $doc->saveXML($node, LIBXML_NOEMPTYTAG
);
1084 function expire_lock_files($debug) {
1085 if ($debug) _debug("Removing old lock files...");
1089 if (is_writable(LOCK_DIRECTORY
)) {
1090 $files = glob(LOCK_DIRECTORY
. "/*.lock");
1093 foreach ($files as $file) {
1094 if (!file_is_locked($file) && time() - filemtime($file) > 86400*2) {
1102 if ($debug) _debug("Removed $num_deleted files.");
1105 function expire_cached_files($debug) {
1106 foreach (array("simplepie", "images", "export") as $dir) {
1107 $cache_dir = CACHE_DIR
. "/$dir";
1109 if ($debug) _debug("Expiring $cache_dir");
1113 if (is_writable($cache_dir)) {
1114 $files = glob("$cache_dir/*");
1117 foreach ($files as $file) {
1118 if (time() - filemtime($file) > 86400*7) {
1127 if ($debug) _debug("Removed $num_deleted files.");
1132 * Source: http://www.php.net/manual/en/function.parse-url.php#104527
1133 * Returns the url query as associative array
1135 * @param string query
1136 * @return array params
1138 function convertUrlQuery($query) {
1139 $queryParts = explode('&', $query);
1143 foreach ($queryParts as $param) {
1144 $item = explode('=', $param);
1145 $params[$item[0]] = $item[1];
1151 function get_article_filters($filters, $title, $content, $link, $timestamp, $author, $tags) {
1154 foreach ($filters as $filter) {
1155 $match_any_rule = $filter["match_any_rule"];
1156 $filter_match = false;
1158 foreach ($filter["rules"] as $rule) {
1160 $reg_exp = $rule["reg_exp"];
1165 switch ($rule["type"]) {
1167 $match = @preg_match
("/$reg_exp/i", $title);
1170 // we don't need to deal with multiline regexps
1171 $content = preg_replace("/[\r\n\t]/", "", $content);
1173 $match = @preg_match
("/$reg_exp/i", $content);
1176 // we don't need to deal with multiline regexps
1177 $content = preg_replace("/[\r\n\t]/", "", $content);
1179 $match = (@preg_match
("/$reg_exp/i", $title) || @preg_match
("/$reg_exp/i", $content));
1182 $match = @preg_match
("/$reg_exp/i", $link);
1185 $match = @preg_match
("/$reg_exp/i", $author);
1188 $tag_string = join(",", $tags);
1189 $match = @preg_match
("/$reg_exp/i", $tag_string);
1193 if ($match_any_rule) {
1195 $filter_match = true;
1199 $filter_match = $match;
1206 if ($filter_match) {
1207 foreach ($filter["actions"] AS $action) {
1208 array_push($matches, $action);
1216 function find_article_filter($filters, $filter_name) {
1217 foreach ($filters as $f) {
1218 if ($f["type"] == $filter_name) {
1225 function find_article_filters($filters, $filter_name) {
1228 foreach ($filters as $f) {
1229 if ($f["type"] == $filter_name) {
1230 array_push($results, $f);
1236 function calculate_article_score($filters) {
1239 foreach ($filters as $f) {
1240 if ($f["type"] == "score") {
1241 $score +
= $f["param"];
1247 function labels_contains_caption($labels, $caption) {
1248 foreach ($labels as $label) {
1249 if ($label[1] == $caption) {
1257 function assign_article_to_label_filters($link, $id, $filters, $owner_uid, $article_labels) {
1258 foreach ($filters as $f) {
1259 if ($f["type"] == "label") {
1260 if (!labels_contains_caption($article_labels, $f["param"])) {
1261 label_add_article($link, $id, $f["param"], $owner_uid);
1267 function cache_content($link, $url, $login, $pass) {
1269 $content = fetch_file_contents($url, $login, $pass);
1272 $doc = new DOMDocument();
1273 @$doc->loadHTML($content);
1274 $xpath = new DOMXPath($doc);
1276 $node = $doc->getElementsByTagName('body')->item(0);
1279 $content = $doc->saveXML($node, LIBXML_NOEMPTYTAG
);