]> git.wh0rd.org - tt-rss.git/blob - include/rssfuncs.php
Merge branch 'master' of github.com:gothfox/Tiny-Tiny-RSS
[tt-rss.git] / include / rssfuncs.php
1 <?php
2 define_default('DAEMON_UPDATE_LOGIN_LIMIT', 30);
3 define_default('DAEMON_FEED_LIMIT', 500);
4 define_default('DAEMON_SLEEP_INTERVAL', 120);
5
6 function update_feedbrowser_cache() {
7
8 $result = db_query("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");
13
14 db_query("BEGIN");
15
16 db_query("DELETE FROM ttrss_feedbrowser_cache");
17
18 $count = 0;
19
20 while ($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"]);
25
26 $tmp_result = db_query("SELECT subscribers FROM
27 ttrss_feedbrowser_cache WHERE feed_url = '$feed_url'");
28
29 if (db_num_rows($tmp_result) == 0) {
30
31 db_query("INSERT INTO ttrss_feedbrowser_cache
32 (feed_url, site_url, title, subscribers) VALUES ('$feed_url',
33 '$site_url', '$title', '$subscribers')");
34
35 ++$count;
36
37 }
38
39 }
40
41 db_query("COMMIT");
42
43 return $count;
44
45 }
46
47
48 /**
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
52 * by another process.
53 *
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 $debug Set to false to disable debug output. Default to true.
58 * @return void
59 */
60 function update_daemon_common($limit = DAEMON_FEED_LIMIT, $from_http = false, $debug = true) {
61 // Process all other feeds using last_updated and interval parameters
62
63 $schema_version = get_schema_version();
64
65 if ($schema_version != SCHEMA_VERSION) {
66 die("Schema version is wrong, please upgrade the database.\n");
67 }
68
69 define('PREFS_NO_CACHE', true);
70
71 // Test if the user has loggued in recently. If not, it does not update its feeds.
72 if (!SINGLE_USER_MODE && DAEMON_UPDATE_LOGIN_LIMIT > 0) {
73 if (DB_TYPE == "pgsql") {
74 $login_thresh_qpart = "AND ttrss_users.last_login >= NOW() - INTERVAL '".DAEMON_UPDATE_LOGIN_LIMIT." days'";
75 } else {
76 $login_thresh_qpart = "AND ttrss_users.last_login >= DATE_SUB(NOW(), INTERVAL ".DAEMON_UPDATE_LOGIN_LIMIT." DAY)";
77 }
78 } else {
79 $login_thresh_qpart = "";
80 }
81
82 // Test if the feed need a update (update interval exceeded).
83 if (DB_TYPE == "pgsql") {
84 $update_limit_qpart = "AND ((
85 ttrss_feeds.update_interval = 0
86 AND ttrss_user_prefs.value != '-1'
87 AND ttrss_feeds.last_updated < NOW() - CAST((ttrss_user_prefs.value || ' minutes') AS INTERVAL)
88 ) OR (
89 ttrss_feeds.update_interval > 0
90 AND ttrss_feeds.last_updated < NOW() - CAST((ttrss_feeds.update_interval || ' minutes') AS INTERVAL)
91 ) OR (ttrss_feeds.last_updated IS NULL
92 AND ttrss_user_prefs.value != '-1')
93 OR (last_updated = '1970-01-01 00:00:00'
94 AND ttrss_user_prefs.value != '-1'))";
95 } else {
96 $update_limit_qpart = "AND ((
97 ttrss_feeds.update_interval = 0
98 AND ttrss_user_prefs.value != '-1'
99 AND ttrss_feeds.last_updated < DATE_SUB(NOW(), INTERVAL CONVERT(ttrss_user_prefs.value, SIGNED INTEGER) MINUTE)
100 ) OR (
101 ttrss_feeds.update_interval > 0
102 AND ttrss_feeds.last_updated < DATE_SUB(NOW(), INTERVAL ttrss_feeds.update_interval MINUTE)
103 ) OR (ttrss_feeds.last_updated IS NULL
104 AND ttrss_user_prefs.value != '-1')
105 OR (last_updated = '1970-01-01 00:00:00'
106 AND ttrss_user_prefs.value != '-1'))";
107 }
108
109 // Test if feed is currently being updated by another process.
110 if (DB_TYPE == "pgsql") {
111 $updstart_thresh_qpart = "AND (ttrss_feeds.last_update_started IS NULL OR ttrss_feeds.last_update_started < NOW() - INTERVAL '10 minutes')";
112 } else {
113 $updstart_thresh_qpart = "AND (ttrss_feeds.last_update_started IS NULL OR ttrss_feeds.last_update_started < DATE_SUB(NOW(), INTERVAL 10 MINUTE))";
114 }
115
116 // Test if there is a limit to number of updated feeds
117 $query_limit = "";
118 if($limit) $query_limit = sprintf("LIMIT %d", $limit);
119
120 $query = "SELECT DISTINCT ttrss_feeds.feed_url, ttrss_feeds.last_updated
121 FROM
122 ttrss_feeds, ttrss_users, ttrss_user_prefs
123 WHERE
124 ttrss_feeds.owner_uid = ttrss_users.id
125 AND ttrss_user_prefs.profile IS NULL
126 AND ttrss_users.id = ttrss_user_prefs.owner_uid
127 AND ttrss_user_prefs.pref_name = 'DEFAULT_UPDATE_INTERVAL'
128 $login_thresh_qpart $update_limit_qpart
129 $updstart_thresh_qpart
130 ORDER BY last_updated $query_limit";
131
132 // We search for feed needing update.
133 $result = db_query($query);
134
135 if($debug) _debug(sprintf("Scheduled %d feeds to update...", db_num_rows($result)));
136
137 // Here is a little cache magic in order to minimize risk of double feed updates.
138 $feeds_to_update = array();
139 while ($line = db_fetch_assoc($result)) {
140 array_push($feeds_to_update, db_escape_string($line['feed_url']));
141 }
142
143 // We update the feed last update started date before anything else.
144 // There is no lag due to feed contents downloads
145 // It prevent an other process to update the same feed.
146
147 if(count($feeds_to_update) > 0) {
148 $feeds_quoted = array();
149
150 foreach ($feeds_to_update as $feed) {
151 array_push($feeds_quoted, "'" . db_escape_string($feed) . "'");
152 }
153
154 db_query(sprintf("UPDATE ttrss_feeds SET last_update_started = NOW()
155 WHERE feed_url IN (%s)", implode(',', $feeds_quoted)));
156 }
157
158 $nf = 0;
159
160 // For each feed, we call the feed update function.
161 foreach ($feeds_to_update as $feed) {
162 if($debug) _debug("Base feed: $feed");
163
164 //update_rss_feed($line["id"], true);
165
166 // since we have the data cached, we can deal with other feeds with the same url
167
168 $tmp_result = db_query("SELECT DISTINCT ttrss_feeds.id,last_updated,ttrss_feeds.owner_uid
169 FROM ttrss_feeds, ttrss_users, ttrss_user_prefs WHERE
170 ttrss_user_prefs.owner_uid = ttrss_feeds.owner_uid AND
171 ttrss_users.id = ttrss_user_prefs.owner_uid AND
172 ttrss_user_prefs.pref_name = 'DEFAULT_UPDATE_INTERVAL' AND
173 ttrss_user_prefs.profile IS NULL AND
174 feed_url = '".db_escape_string($feed)."' AND
175 (ttrss_feeds.update_interval > 0 OR
176 ttrss_user_prefs.value != '-1')
177 $login_thresh_qpart
178 ORDER BY ttrss_feeds.id $query_limit");
179
180 if (db_num_rows($tmp_result) > 0) {
181 $rss = false;
182
183 while ($tline = db_fetch_assoc($tmp_result)) {
184 if($debug) _debug(" => " . $tline["last_updated"] . ", " . $tline["id"] . " " . $tline["owner_uid"]);
185
186 $rss = update_rss_feed($tline["id"], true, false, $rss);
187 _debug_suppress(false);
188 ++$nf;
189 }
190 }
191 }
192
193 require_once "digest.php";
194
195 // Send feed digests by email if needed.
196 send_headlines_digests($debug);
197
198 return $nf;
199
200 } // function update_daemon_common
201
202 // ignore_daemon is not used
203 function update_rss_feed($feed, $ignore_daemon = false, $no_cache = false, $rss = false) {
204
205 $debug_enabled = defined('DAEMON_EXTENDED_DEBUG') || $_REQUEST['xdebug'];
206
207 _debug_suppress(!$debug_enabled);
208 _debug("start", $debug_enabled);
209
210 $result = db_query("SELECT id,update_interval,auth_login,
211 feed_url,auth_pass,cache_images,
212 mark_unread_on_update, owner_uid,
213 pubsub_state, auth_pass_encrypted,
214 (SELECT max(date_entered) FROM
215 ttrss_entries, ttrss_user_entries where ref_id = id AND feed_id = '$feed') AS last_article_timestamp
216 FROM ttrss_feeds WHERE id = '$feed'");
217
218 if (db_num_rows($result) == 0) {
219 _debug("feed $feed NOT FOUND/SKIPPED", $debug_enabled);
220 return false;
221 }
222
223 $last_article_timestamp = @strtotime(db_fetch_result($result, 0, "last_article_timestamp"));
224
225 if (defined('_DISABLE_HTTP_304'))
226 $last_article_timestamp = 0;
227
228 $owner_uid = db_fetch_result($result, 0, "owner_uid");
229 $mark_unread_on_update = sql_bool_to_bool(db_fetch_result($result,
230 0, "mark_unread_on_update"));
231 $pubsub_state = db_fetch_result($result, 0, "pubsub_state");
232 $auth_pass_encrypted = sql_bool_to_bool(db_fetch_result($result,
233 0, "auth_pass_encrypted"));
234
235 db_query("UPDATE ttrss_feeds SET last_update_started = NOW()
236 WHERE id = '$feed'");
237
238 $auth_login = db_fetch_result($result, 0, "auth_login");
239 $auth_pass = db_fetch_result($result, 0, "auth_pass");
240
241 if ($auth_pass_encrypted) {
242 require_once "crypt.php";
243 $auth_pass = decrypt_string($auth_pass);
244 }
245
246 $cache_images = sql_bool_to_bool(db_fetch_result($result, 0, "cache_images"));
247 $fetch_url = db_fetch_result($result, 0, "feed_url");
248
249 $feed = db_escape_string($feed);
250
251 $date_feed_processed = date('Y-m-d H:i');
252
253 $cache_filename = CACHE_DIR . "/simplepie/" . sha1($fetch_url) . ".xml";
254
255 $pluginhost = new PluginHost();
256 $pluginhost->set_debug($debug_enabled);
257 $user_plugins = get_pref("_ENABLED_PLUGINS", $owner_uid);
258
259 $pluginhost->load(PLUGINS, PluginHost::KIND_ALL);
260 $pluginhost->load($user_plugins, PluginHost::KIND_USER, $owner_uid);
261 $pluginhost->load_data();
262
263 if ($rss && is_object($rss) && get_class($rss) == "FeedParser") {
264 _debug("using previously initialized parser object");
265 } else {
266 $rss_hash = false;
267
268 $force_refetch = isset($_REQUEST["force_refetch"]);
269
270 if (file_exists($cache_filename) &&
271 is_readable($cache_filename) &&
272 !$auth_login && !$auth_pass &&
273 filemtime($cache_filename) > time() - 30) {
274
275 _debug("using local cache.", $debug_enabled);
276
277 @$feed_data = file_get_contents($cache_filename);
278
279 if ($feed_data) {
280 $rss_hash = sha1($feed_data);
281 }
282
283 } else {
284 _debug("local cache will not be used for this feed", $debug_enabled);
285 }
286 }
287
288 if (!$rss) {
289
290 foreach ($pluginhost->get_hooks(PluginHost::HOOK_FETCH_FEED) as $plugin) {
291 $feed_data = $plugin->hook_fetch_feed($feed_data, $fetch_url, $owner_uid, $feed);
292 }
293
294 if (!$feed_data) {
295 _debug("fetching [$fetch_url]...", $debug_enabled);
296 _debug("If-Modified-Since: ".gmdate('D, d M Y H:i:s \G\M\T', $last_article_timestamp), $debug_enabled);
297
298 $feed_data = fetch_file_contents($fetch_url, false,
299 $auth_login, $auth_pass, false,
300 $no_cache ? FEED_FETCH_NO_CACHE_TIMEOUT : FEED_FETCH_TIMEOUT,
301 $force_refetch ? 0 : $last_article_timestamp);
302
303 global $fetch_curl_used;
304
305 if (!$fetch_curl_used) {
306 $tmp = @gzdecode($feed_data);
307
308 if ($tmp) $feed_data = $tmp;
309 }
310
311 $feed_data = trim($feed_data);
312
313 _debug("fetch done.", $debug_enabled);
314
315 /* if ($feed_data) {
316 $error = verify_feed_xml($feed_data);
317
318 if ($error) {
319 _debug("error verifying XML, code: " . $error->code, $debug_enabled);
320
321 if ($error->code == 26) {
322 _debug("got error 26, trying to decode entities...", $debug_enabled);
323
324 $feed_data = html_entity_decode($feed_data, ENT_COMPAT, 'UTF-8');
325
326 $error = verify_feed_xml($feed_data);
327
328 if ($error) $feed_data = '';
329 }
330 }
331 } */
332 }
333
334 if (!$feed_data) {
335 global $fetch_last_error;
336 global $fetch_last_error_code;
337
338 _debug("unable to fetch: $fetch_last_error [$fetch_last_error_code]", $debug_enabled);
339
340 $error_escaped = '';
341
342 // If-Modified-Since
343 if ($fetch_last_error_code != 304) {
344 $error_escaped = db_escape_string($fetch_last_error);
345 } else {
346 _debug("source claims data not modified, nothing to do.", $debug_enabled);
347 }
348
349 db_query(
350 "UPDATE ttrss_feeds SET last_error = '$error_escaped',
351 last_updated = NOW() WHERE id = '$feed'");
352
353 return;
354 }
355 }
356
357 foreach ($pluginhost->get_hooks(PluginHost::HOOK_FEED_FETCHED) as $plugin) {
358 $feed_data = $plugin->hook_feed_fetched($feed_data, $fetch_url, $owner_uid, $feed);
359 }
360
361 // set last update to now so if anything *simplepie* crashes later we won't be
362 // continuously failing on the same feed
363 //db_query("UPDATE ttrss_feeds SET last_updated = NOW() WHERE id = '$feed'");
364
365 if (!$rss) {
366 $rss = new FeedParser($feed_data);
367 $rss->init();
368 }
369
370 if (DETECT_ARTICLE_LANGUAGE) {
371 require_once "lib/languagedetect/LanguageDetect.php";
372
373 $lang = new Text_LanguageDetect();
374 $lang->setNameMode(2);
375 }
376
377 // print_r($rss);
378
379 $feed = db_escape_string($feed);
380
381 if (!$rss->error()) {
382
383 // cache data for later
384 if (!$auth_pass && !$auth_login && is_writable(CACHE_DIR . "/simplepie")) {
385 $new_rss_hash = sha1($feed_data);
386
387 if ($new_rss_hash != $rss_hash && count($rss->get_items()) > 0 ) {
388 _debug("saving $cache_filename", $debug_enabled);
389 @file_put_contents($cache_filename, $feed_data);
390 }
391 }
392
393 // We use local pluginhost here because we need to load different per-user feed plugins
394 $pluginhost->run_hooks(PluginHost::HOOK_FEED_PARSED, "hook_feed_parsed", $rss);
395
396 _debug("processing feed data...", $debug_enabled);
397
398 // db_query("BEGIN");
399
400 if (DB_TYPE == "pgsql") {
401 $favicon_interval_qpart = "favicon_last_checked < NOW() - INTERVAL '12 hour'";
402 } else {
403 $favicon_interval_qpart = "favicon_last_checked < DATE_SUB(NOW(), INTERVAL 12 HOUR)";
404 }
405
406 $result = db_query("SELECT title,site_url,owner_uid,favicon_avg_color,
407 (favicon_last_checked IS NULL OR $favicon_interval_qpart) AS
408 favicon_needs_check
409 FROM ttrss_feeds WHERE id = '$feed'");
410
411 $registered_title = db_fetch_result($result, 0, "title");
412 $orig_site_url = db_fetch_result($result, 0, "site_url");
413 $favicon_needs_check = sql_bool_to_bool(db_fetch_result($result, 0,
414 "favicon_needs_check"));
415 $favicon_avg_color = db_fetch_result($result, 0, "favicon_avg_color");
416
417 $owner_uid = db_fetch_result($result, 0, "owner_uid");
418
419 $site_url = db_escape_string(mb_substr(rewrite_relative_url($fetch_url, $rss->get_link()), 0, 245));
420
421 _debug("site_url: $site_url", $debug_enabled);
422 _debug("feed_title: " . $rss->get_title(), $debug_enabled);
423
424 if ($favicon_needs_check || $force_refetch) {
425
426 /* terrible hack: if we crash on floicon shit here, we won't check
427 * the icon avgcolor again (unless the icon got updated) */
428
429 $favicon_file = ICONS_DIR . "/$feed.ico";
430 $favicon_modified = @filemtime($favicon_file);
431
432 _debug("checking favicon...", $debug_enabled);
433
434 check_feed_favicon($site_url, $feed);
435 $favicon_modified_new = @filemtime($favicon_file);
436
437 if ($favicon_modified_new > $favicon_modified)
438 $favicon_avg_color = '';
439
440 if (file_exists($favicon_file) && function_exists("imagecreatefromstring") && $favicon_avg_color == '') {
441 require_once "colors.php";
442
443 db_query("UPDATE ttrss_feeds SET favicon_avg_color = 'fail' WHERE
444 id = '$feed'");
445
446 $favicon_color = db_escape_string(
447 calculate_avg_color($favicon_file));
448
449 $favicon_colorstring = ",favicon_avg_color = '".$favicon_color."'";
450 } else if ($favicon_avg_color == 'fail') {
451 _debug("floicon failed on this file, not trying to recalculate avg color", $debug_enabled);
452 }
453
454 db_query("UPDATE ttrss_feeds SET favicon_last_checked = NOW()
455 $favicon_colorstring
456 WHERE id = '$feed'");
457 }
458
459 if (!$registered_title || $registered_title == "[Unknown]") {
460
461 $feed_title = db_escape_string($rss->get_title());
462
463 if ($feed_title) {
464 _debug("registering title: $feed_title", $debug_enabled);
465
466 db_query("UPDATE ttrss_feeds SET
467 title = '$feed_title' WHERE id = '$feed'");
468 }
469 }
470
471 if ($site_url && $orig_site_url != $site_url) {
472 db_query("UPDATE ttrss_feeds SET
473 site_url = '$site_url' WHERE id = '$feed'");
474 }
475
476 _debug("loading filters & labels...", $debug_enabled);
477
478 $filters = load_filters($feed, $owner_uid);
479 $labels = get_all_labels($owner_uid);
480
481 _debug("" . count($filters) . " filters loaded.", $debug_enabled);
482
483 $items = $rss->get_items();
484
485 if (!is_array($items)) {
486 _debug("no articles found.", $debug_enabled);
487
488 db_query("UPDATE ttrss_feeds
489 SET last_updated = NOW(), last_error = '' WHERE id = '$feed'");
490
491 return; // no articles
492 }
493
494 if ($pubsub_state != 2 && PUBSUBHUBBUB_ENABLED) {
495
496 _debug("checking for PUSH hub...", $debug_enabled);
497
498 $feed_hub_url = false;
499
500 $links = $rss->get_links('hub');
501
502 if ($links && is_array($links)) {
503 foreach ($links as $l) {
504 $feed_hub_url = $l;
505 break;
506 }
507 }
508
509 _debug("feed hub url: $feed_hub_url", $debug_enabled);
510
511 $feed_self_url = $fetch_url;
512
513 $links = $rss->get_links('self');
514
515 if ($links && is_array($links)) {
516 foreach ($links as $l) {
517 $feed_self_url = $l;
518 break;
519 }
520 }
521
522 _debug("feed self url = $feed_self_url");
523
524 if ($feed_hub_url && $feed_self_url && function_exists('curl_init') &&
525 !ini_get("open_basedir")) {
526
527 require_once 'lib/pubsubhubbub/subscriber.php';
528
529 $callback_url = get_self_url_prefix() .
530 "/public.php?op=pubsub&id=$feed";
531
532 $s = new Subscriber($feed_hub_url, $callback_url);
533
534 $rc = $s->subscribe($feed_self_url);
535
536 _debug("feed hub url found, subscribe request sent. [rc=$rc]", $debug_enabled);
537
538 db_query("UPDATE ttrss_feeds SET pubsub_state = 1
539 WHERE id = '$feed'");
540 }
541 }
542
543 _debug("processing articles...", $debug_enabled);
544
545 foreach ($items as $item) {
546 if ($_REQUEST['xdebug'] == 3) {
547 print_r($item);
548 }
549
550 $entry_guid = $item->get_id();
551 if (!$entry_guid) $entry_guid = $item->get_link();
552 if (!$entry_guid) $entry_guid = make_guid_from_title($item->get_title());
553
554 _debug("f_guid $entry_guid", $debug_enabled);
555
556 if (!$entry_guid) continue;
557
558 $entry_guid = "$owner_uid,$entry_guid";
559
560 $entry_guid_hashed = db_escape_string('SHA1:' . sha1($entry_guid));
561
562 _debug("guid $entry_guid / $entry_guid_hashed", $debug_enabled);
563
564 $entry_timestamp = "";
565
566 $entry_timestamp = $item->get_date();
567
568 _debug("orig date: " . $item->get_date(), $debug_enabled);
569
570 if ($entry_timestamp == -1 || !$entry_timestamp || $entry_timestamp > time()) {
571 $entry_timestamp = time();
572 }
573
574 $entry_timestamp_fmt = strftime("%Y/%m/%d %H:%M:%S", $entry_timestamp);
575
576 _debug("date $entry_timestamp [$entry_timestamp_fmt]", $debug_enabled);
577
578 // $entry_title = html_entity_decode($item->get_title(), ENT_COMPAT, 'UTF-8');
579 // $entry_title = decode_numeric_entities($entry_title);
580 $entry_title = $item->get_title();
581
582 $entry_link = rewrite_relative_url($site_url, $item->get_link());
583
584 _debug("title $entry_title", $debug_enabled);
585 _debug("link $entry_link", $debug_enabled);
586
587 if (!$entry_title) $entry_title = date("Y-m-d H:i:s", $entry_timestamp);;
588
589 $entry_content = $item->get_content();
590 if (!$entry_content) $entry_content = $item->get_description();
591
592 if ($_REQUEST["xdebug"] == 2) {
593 print "content: ";
594 print $entry_content;
595 print "\n";
596 }
597
598 $entry_language = "";
599
600 if (DETECT_ARTICLE_LANGUAGE) {
601 $entry_language = $lang->detect($entry_title . " " . $entry_content, 1);
602
603 if (count($entry_language) > 0) {
604 $entry_language = array_keys($entry_language);
605
606 // the fuck?
607 if (is_array($entry_language))
608 $entry_language = "";
609 else
610 $entry_language = db_escape_string(substr($entry_language[0], 0, 2));
611
612 _debug("detected language: $entry_language", $debug_enabled);
613 }
614 }
615
616 $entry_comments = $item->get_comments_url();
617 $entry_author = $item->get_author();
618
619 $entry_guid = db_escape_string(mb_substr($entry_guid, 0, 245));
620
621 $entry_comments = db_escape_string(mb_substr(trim($entry_comments), 0, 245));
622 $entry_author = db_escape_string(mb_substr(trim($entry_author), 0, 245));
623
624 $num_comments = (int) $item->get_comments_count();
625
626 _debug("author $entry_author", $debug_enabled);
627 _debug("num_comments: $num_comments", $debug_enabled);
628 _debug("looking for tags...", $debug_enabled);
629
630 // parse <category> entries into tags
631
632 $additional_tags = array();
633
634 $additional_tags_src = $item->get_categories();
635
636 if (is_array($additional_tags_src)) {
637 foreach ($additional_tags_src as $tobj) {
638 array_push($additional_tags, $tobj);
639 }
640 }
641
642 $entry_tags = array_unique($additional_tags);
643
644 for ($i = 0; $i < count($entry_tags); $i++)
645 $entry_tags[$i] = mb_strtolower($entry_tags[$i], 'utf-8');
646
647 _debug("tags found: " . join(",", $entry_tags), $debug_enabled);
648
649 _debug("done collecting data.", $debug_enabled);
650
651 // TODO: less memory-hungry implementation
652
653 _debug("applying plugin filters..", $debug_enabled);
654
655 // FIXME not sure if owner_uid is a good idea here, we may have a base entry without user entry (?)
656 $result = db_query("SELECT plugin_data,title,content,link,tag_cache,author FROM ttrss_entries, ttrss_user_entries
657 WHERE ref_id = id AND (guid = '".db_escape_string($entry_guid)."' OR guid = '$entry_guid_hashed') AND owner_uid = $owner_uid");
658
659 if (db_num_rows($result) != 0) {
660 $entry_plugin_data = db_fetch_result($result, 0, "plugin_data");
661 $stored_article = array("title" => db_fetch_result($result, 0, "title"),
662 "content" => db_fetch_result($result, 0, "content"),
663 "link" => db_fetch_result($result, 0, "link"),
664 "tags" => explode(",", db_fetch_result($result, 0, "tag_cache")),
665 "author" => db_fetch_result($result, 0, "author"));
666 } else {
667 $entry_plugin_data = "";
668 $stored_article = array();
669 }
670
671 $article = array("owner_uid" => $owner_uid, // read only
672 "guid" => $entry_guid, // read only
673 "title" => $entry_title,
674 "content" => $entry_content,
675 "link" => $entry_link,
676 "tags" => $entry_tags,
677 "plugin_data" => $entry_plugin_data,
678 "author" => $entry_author,
679 "stored" => $stored_article,
680 "feed" => array("id" => $feed,
681 "fetch_url" => $fetch_url,
682 "site_url" => $site_url)
683 );
684
685 foreach ($pluginhost->get_hooks(PluginHost::HOOK_ARTICLE_FILTER) as $plugin) {
686 $article = $plugin->hook_article_filter($article);
687 }
688
689 $entry_tags = $article["tags"];
690 $entry_guid = db_escape_string($entry_guid);
691 $entry_title = db_escape_string($article["title"]);
692 $entry_author = db_escape_string($article["author"]);
693 $entry_link = db_escape_string($article["link"]);
694 $entry_plugin_data = db_escape_string($article["plugin_data"]);
695 $entry_content = $article["content"]; // escaped below
696
697
698 _debug("plugin data: $entry_plugin_data", $debug_enabled);
699
700 if ($cache_images && is_writable(CACHE_DIR . '/images'))
701 cache_images($entry_content, $site_url, $debug_enabled);
702
703 $entry_content = db_escape_string($entry_content, false);
704
705 $content_hash = "SHA1:" . sha1($entry_content);
706
707 db_query("BEGIN");
708
709 $result = db_query("SELECT id FROM ttrss_entries
710 WHERE (guid = '$entry_guid' OR guid = '$entry_guid_hashed')");
711
712 if (db_num_rows($result) == 0) {
713
714 _debug("base guid [$entry_guid] not found", $debug_enabled);
715
716 // base post entry does not exist, create it
717
718 $result = db_query(
719 "INSERT INTO ttrss_entries
720 (title,
721 guid,
722 link,
723 updated,
724 content,
725 content_hash,
726 no_orig_date,
727 date_updated,
728 date_entered,
729 comments,
730 num_comments,
731 plugin_data,
732 lang,
733 author)
734 VALUES
735 ('$entry_title',
736 '$entry_guid_hashed',
737 '$entry_link',
738 '$entry_timestamp_fmt',
739 '$entry_content',
740 '$content_hash',
741 false,
742 NOW(),
743 '$date_feed_processed',
744 '$entry_comments',
745 '$num_comments',
746 '$entry_plugin_data',
747 '$entry_language',
748 '$entry_author')");
749
750 $article_labels = array();
751
752 } else {
753 // we keep encountering the entry in feeds, so we need to
754 // update date_updated column so that we don't get horrible
755 // dupes when the entry gets purged and reinserted again e.g.
756 // in the case of SLOW SLOW OMG SLOW updating feeds
757
758 $base_entry_id = db_fetch_result($result, 0, "id");
759
760 db_query("UPDATE ttrss_entries SET date_updated = NOW()
761 WHERE id = '$base_entry_id'");
762
763 $article_labels = get_article_labels($base_entry_id, $owner_uid);
764 }
765
766 // now it should exist, if not - bad luck then
767
768 $result = db_query("SELECT
769 id,content_hash,title,plugin_data,guid,
770 num_comments
771 FROM
772 ttrss_entries
773 WHERE guid = '$entry_guid' OR guid = '$entry_guid_hashed'");
774
775 $entry_ref_id = 0;
776 $entry_int_id = 0;
777
778 if (db_num_rows($result) == 1) {
779
780 _debug("base guid found, checking for user record", $debug_enabled);
781
782 // this will be used below in update handler
783 $orig_content_hash = db_fetch_result($result, 0, "content_hash");
784 $orig_title = db_fetch_result($result, 0, "title");
785 $orig_num_comments = db_fetch_result($result, 0, "num_comments");
786 $orig_plugin_data = db_fetch_result($result, 0, "plugin_data");
787
788 $ref_id = db_fetch_result($result, 0, "id");
789 $entry_ref_id = $ref_id;
790
791 /* $stored_guid = db_fetch_result($result, 0, "guid");
792 if ($stored_guid != $entry_guid_hashed) {
793 if ($debug_enabled) _debug("upgrading compat guid to hashed one", $debug_enabled);
794
795 db_query("UPDATE ttrss_entries SET guid = '$entry_guid_hashed' WHERE
796 id = '$ref_id'");
797 } */
798
799 // check for user post link to main table
800
801 // do we allow duplicate posts with same GUID in different feeds?
802 if (get_pref("ALLOW_DUPLICATE_POSTS", $owner_uid, false)) {
803 $dupcheck_qpart = "AND (feed_id = '$feed' OR feed_id IS NULL)";
804 } else {
805 $dupcheck_qpart = "";
806 }
807
808 /* Collect article tags here so we could filter by them: */
809
810 $article_filters = get_article_filters($filters, $entry_title,
811 $entry_content, $entry_link, $entry_timestamp, $entry_author,
812 $entry_tags);
813
814 if ($debug_enabled) {
815 _debug("article filters: ", $debug_enabled);
816 if (count($article_filters) != 0) {
817 print_r($article_filters);
818 }
819 }
820
821 if (find_article_filter($article_filters, "filter")) {
822 db_query("COMMIT"); // close transaction in progress
823 continue;
824 }
825
826 $score = calculate_article_score($article_filters);
827
828 _debug("initial score: $score", $debug_enabled);
829
830 $query = "SELECT ref_id, int_id FROM ttrss_user_entries WHERE
831 ref_id = '$ref_id' AND owner_uid = '$owner_uid'
832 $dupcheck_qpart";
833
834 // if ($_REQUEST["xdebug"]) print "$query\n";
835
836 $result = db_query($query);
837
838 // okay it doesn't exist - create user entry
839 if (db_num_rows($result) == 0) {
840
841 _debug("user record not found, creating...", $debug_enabled);
842
843 if ($score >= -500 && !find_article_filter($article_filters, 'catchup')) {
844 $unread = 'true';
845 $last_read_qpart = 'NULL';
846 } else {
847 $unread = 'false';
848 $last_read_qpart = 'NOW()';
849 }
850
851 if (find_article_filter($article_filters, 'mark') || $score > 1000) {
852 $marked = 'true';
853 } else {
854 $marked = 'false';
855 }
856
857 if (find_article_filter($article_filters, 'publish')) {
858 $published = 'true';
859 } else {
860 $published = 'false';
861 }
862
863 // N-grams
864
865 if (DB_TYPE == "pgsql" and defined('_NGRAM_TITLE_DUPLICATE_THRESHOLD')) {
866
867 $result = db_query("SELECT COUNT(*) AS similar FROM
868 ttrss_entries,ttrss_user_entries
869 WHERE ref_id = id AND updated >= NOW() - INTERVAL '7 day'
870 AND similarity(title, '$entry_title') >= "._NGRAM_TITLE_DUPLICATE_THRESHOLD."
871 AND owner_uid = $owner_uid");
872
873 $ngram_similar = db_fetch_result($result, 0, "similar");
874
875 _debug("N-gram similar results: $ngram_similar", $debug_enabled);
876
877 if ($ngram_similar > 0) {
878 $unread = 'false';
879 }
880 }
881
882 $last_marked = ($marked == 'true') ? 'NOW()' : 'NULL';
883 $last_published = ($published == 'true') ? 'NOW()' : 'NULL';
884
885 $result = db_query(
886 "INSERT INTO ttrss_user_entries
887 (ref_id, owner_uid, feed_id, unread, last_read, marked,
888 published, score, tag_cache, label_cache, uuid,
889 last_marked, last_published)
890 VALUES ('$ref_id', '$owner_uid', '$feed', $unread,
891 $last_read_qpart, $marked, $published, '$score', '', '',
892 '', $last_marked, $last_published)");
893
894 if (PUBSUBHUBBUB_HUB && $published == 'true') {
895 $rss_link = get_self_url_prefix() .
896 "/public.php?op=rss&id=-2&key=" .
897 get_feed_access_key(-2, false, $owner_uid);
898
899 $p = new Publisher(PUBSUBHUBBUB_HUB);
900
901 /* $pubsub_result = */ $p->publish_update($rss_link);
902 }
903
904 $result = db_query(
905 "SELECT int_id FROM ttrss_user_entries WHERE
906 ref_id = '$ref_id' AND owner_uid = '$owner_uid' AND
907 feed_id = '$feed' LIMIT 1");
908
909 if (db_num_rows($result) == 1) {
910 $entry_int_id = db_fetch_result($result, 0, "int_id");
911 }
912 } else {
913 _debug("user record FOUND", $debug_enabled);
914
915 $entry_ref_id = db_fetch_result($result, 0, "ref_id");
916 $entry_int_id = db_fetch_result($result, 0, "int_id");
917 }
918
919 _debug("RID: $entry_ref_id, IID: $entry_int_id", $debug_enabled);
920
921 $post_needs_update = false;
922 $update_insignificant = false;
923
924 if ($orig_num_comments != $num_comments) {
925 $post_needs_update = true;
926 $update_insignificant = true;
927 }
928
929 if ($entry_plugin_data != $orig_plugin_data) {
930 $post_needs_update = true;
931 $update_insignificant = true;
932 }
933
934 if ($content_hash != $orig_content_hash) {
935 $post_needs_update = true;
936 $update_insignificant = false;
937 }
938
939 if (db_escape_string($orig_title) != $entry_title) {
940 $post_needs_update = true;
941 $update_insignificant = false;
942 }
943
944 // if post needs update, update it and mark all user entries
945 // linking to this post as updated
946 if ($post_needs_update) {
947
948 if (defined('DAEMON_EXTENDED_DEBUG')) {
949 _debug("post $entry_guid_hashed needs update...", $debug_enabled);
950 }
951
952 // print "<!-- post $orig_title needs update : $post_needs_update -->";
953
954 db_query("UPDATE ttrss_entries
955 SET title = '$entry_title', content = '$entry_content',
956 content_hash = '$content_hash',
957 updated = '$entry_timestamp_fmt',
958 num_comments = '$num_comments',
959 plugin_data = '$entry_plugin_data'
960 WHERE id = '$ref_id'");
961
962 if (!$update_insignificant) {
963 if ($mark_unread_on_update) {
964 db_query("UPDATE ttrss_user_entries
965 SET last_read = null, unread = true WHERE ref_id = '$ref_id'");
966 }
967 }
968 }
969 }
970
971 db_query("COMMIT");
972
973 _debug("assigning labels...", $debug_enabled);
974
975 assign_article_to_label_filters($entry_ref_id, $article_filters,
976 $owner_uid, $article_labels);
977
978 _debug("looking for enclosures...", $debug_enabled);
979
980 // enclosures
981
982 $enclosures = array();
983
984 $encs = $item->get_enclosures();
985
986 if (is_array($encs)) {
987 foreach ($encs as $e) {
988 $e_item = array(
989 $e->link, $e->type, $e->length, $e->title);
990 array_push($enclosures, $e_item);
991 }
992 }
993
994 if ($debug_enabled) {
995 _debug("article enclosures:", $debug_enabled);
996 print_r($enclosures);
997 }
998
999 db_query("BEGIN");
1000
1001 // debugging
1002 // db_query("DELETE FROM ttrss_enclosures WHERE post_id = '$entry_ref_id'");
1003
1004 foreach ($enclosures as $enc) {
1005 $enc_url = db_escape_string($enc[0]);
1006 $enc_type = db_escape_string($enc[1]);
1007 $enc_dur = db_escape_string($enc[2]);
1008 $enc_title = db_escape_string($enc[3]);
1009
1010 $result = db_query("SELECT id FROM ttrss_enclosures
1011 WHERE content_url = '$enc_url' AND post_id = '$entry_ref_id'");
1012
1013 if (db_num_rows($result) == 0) {
1014 db_query("INSERT INTO ttrss_enclosures
1015 (content_url, content_type, title, duration, post_id) VALUES
1016 ('$enc_url', '$enc_type', '$enc_title', '$enc_dur', '$entry_ref_id')");
1017 }
1018 }
1019
1020 db_query("COMMIT");
1021
1022 // check for manual tags (we have to do it here since they're loaded from filters)
1023
1024 foreach ($article_filters as $f) {
1025 if ($f["type"] == "tag") {
1026
1027 $manual_tags = trim_array(explode(",", $f["param"]));
1028
1029 foreach ($manual_tags as $tag) {
1030 if (tag_is_valid($tag)) {
1031 array_push($entry_tags, $tag);
1032 }
1033 }
1034 }
1035 }
1036
1037 // Skip boring tags
1038
1039 $boring_tags = trim_array(explode(",", mb_strtolower(get_pref(
1040 'BLACKLISTED_TAGS', $owner_uid, ''), 'utf-8')));
1041
1042 $filtered_tags = array();
1043 $tags_to_cache = array();
1044
1045 if ($entry_tags && is_array($entry_tags)) {
1046 foreach ($entry_tags as $tag) {
1047 if (array_search($tag, $boring_tags) === false) {
1048 array_push($filtered_tags, $tag);
1049 }
1050 }
1051 }
1052
1053 $filtered_tags = array_unique($filtered_tags);
1054
1055 if ($debug_enabled) {
1056 _debug("filtered article tags:", $debug_enabled);
1057 print_r($filtered_tags);
1058 }
1059
1060 // Save article tags in the database
1061
1062 if (count($filtered_tags) > 0) {
1063
1064 db_query("BEGIN");
1065
1066 foreach ($filtered_tags as $tag) {
1067
1068 $tag = sanitize_tag($tag);
1069 $tag = db_escape_string($tag);
1070
1071 if (!tag_is_valid($tag)) continue;
1072
1073 $result = db_query("SELECT id FROM ttrss_tags
1074 WHERE tag_name = '$tag' AND post_int_id = '$entry_int_id' AND
1075 owner_uid = '$owner_uid' LIMIT 1");
1076
1077 if ($result && db_num_rows($result) == 0) {
1078
1079 db_query("INSERT INTO ttrss_tags
1080 (owner_uid,tag_name,post_int_id)
1081 VALUES ('$owner_uid','$tag', '$entry_int_id')");
1082 }
1083
1084 array_push($tags_to_cache, $tag);
1085 }
1086
1087 /* update the cache */
1088
1089 $tags_to_cache = array_unique($tags_to_cache);
1090
1091 $tags_str = db_escape_string(join(",", $tags_to_cache));
1092
1093 db_query("UPDATE ttrss_user_entries
1094 SET tag_cache = '$tags_str' WHERE ref_id = '$entry_ref_id'
1095 AND owner_uid = $owner_uid");
1096
1097 db_query("COMMIT");
1098 }
1099
1100 if (get_pref("AUTO_ASSIGN_LABELS", $owner_uid, false)) {
1101 _debug("auto-assigning labels...", $debug_enabled);
1102
1103 foreach ($labels as $label) {
1104 $caption = preg_quote($label["caption"]);
1105
1106 if ($caption && preg_match("/\b$caption\b/i", "$tags_str " . strip_tags($entry_content) . " $entry_title")) {
1107 if (!labels_contains_caption($article_labels, $caption)) {
1108 label_add_article($entry_ref_id, $caption, $owner_uid);
1109 }
1110 }
1111 }
1112 }
1113
1114 _debug("article processed", $debug_enabled);
1115 }
1116
1117 _debug("purging feed...", $debug_enabled);
1118
1119 purge_feed($feed, 0, $debug_enabled);
1120
1121 db_query("UPDATE ttrss_feeds
1122 SET last_updated = NOW(), last_error = '' WHERE id = '$feed'");
1123
1124 // db_query("COMMIT");
1125
1126 } else {
1127
1128 $error_msg = db_escape_string(mb_substr($rss->error(), 0, 245));
1129
1130 _debug("fetch error: $error_msg", $debug_enabled);
1131
1132 if (count($rss->errors()) > 1) {
1133 foreach ($rss->errors() as $error) {
1134 _debug("+ $error");
1135 }
1136 }
1137
1138 db_query(
1139 "UPDATE ttrss_feeds SET last_error = '$error_msg',
1140 last_updated = NOW() WHERE id = '$feed'");
1141
1142 unset($rss);
1143 }
1144
1145 _debug("done", $debug_enabled);
1146
1147 return $rss;
1148 }
1149
1150 function cache_images($html, $site_url, $debug) {
1151 libxml_use_internal_errors(true);
1152
1153 $charset_hack = '<head>
1154 <meta http-equiv="Content-Type" content="text/html; charset=utf-8"/>
1155 </head>';
1156
1157 $doc = new DOMDocument();
1158 $doc->loadHTML($charset_hack . $html);
1159 $xpath = new DOMXPath($doc);
1160
1161 $entries = $xpath->query('(//img[@src])');
1162
1163 foreach ($entries as $entry) {
1164 if ($entry->hasAttribute('src')) {
1165 $src = rewrite_relative_url($site_url, $entry->getAttribute('src'));
1166
1167 $local_filename = CACHE_DIR . "/images/" . sha1($src) . ".png";
1168
1169 if ($debug) _debug("cache_images: downloading: $src to $local_filename");
1170
1171 if (!file_exists($local_filename)) {
1172 $file_content = fetch_file_contents($src);
1173
1174 if ($file_content && strlen($file_content) > 1024) {
1175 file_put_contents($local_filename, $file_content);
1176 }
1177 }
1178
1179 /* if (file_exists($local_filename)) {
1180 $entry->setAttribute('src', SELF_URL_PATH . '/image.php?url=' .
1181 base64_encode($src));
1182 } */
1183 }
1184 }
1185
1186 //$node = $doc->getElementsByTagName('body')->item(0);
1187 //return $doc->saveXML($node);
1188 }
1189
1190 function expire_error_log($debug) {
1191 if ($debug) _debug("Removing old error log entries...");
1192
1193 if (DB_TYPE == "pgsql") {
1194 db_query("DELETE FROM ttrss_error_log
1195 WHERE created_at < NOW() - INTERVAL '7 days'");
1196 } else {
1197 db_query("DELETE FROM ttrss_error_log
1198 WHERE created_at < DATE_SUB(NOW(), INTERVAL 7 DAY)");
1199 }
1200
1201 }
1202
1203 function expire_lock_files($debug) {
1204 //if ($debug) _debug("Removing old lock files...");
1205
1206 $num_deleted = 0;
1207
1208 if (is_writable(LOCK_DIRECTORY)) {
1209 $files = glob(LOCK_DIRECTORY . "/*.lock");
1210
1211 if ($files) {
1212 foreach ($files as $file) {
1213 if (!file_is_locked(basename($file)) && time() - filemtime($file) > 86400*2) {
1214 unlink($file);
1215 ++$num_deleted;
1216 }
1217 }
1218 }
1219 }
1220
1221 if ($debug) _debug("Removed $num_deleted old lock files.");
1222 }
1223
1224 function expire_cached_files($debug) {
1225 foreach (array("simplepie", "images", "export", "upload") as $dir) {
1226 $cache_dir = CACHE_DIR . "/$dir";
1227
1228 // if ($debug) _debug("Expiring $cache_dir");
1229
1230 $num_deleted = 0;
1231
1232 if (is_writable($cache_dir)) {
1233 $files = glob("$cache_dir/*");
1234
1235 if ($files) {
1236 foreach ($files as $file) {
1237 if (time() - filemtime($file) > 86400*7) {
1238 unlink($file);
1239
1240 ++$num_deleted;
1241 }
1242 }
1243 }
1244 }
1245
1246 if ($debug) _debug("$cache_dir: removed $num_deleted files.");
1247 }
1248 }
1249
1250 /**
1251 * Source: http://www.php.net/manual/en/function.parse-url.php#104527
1252 * Returns the url query as associative array
1253 *
1254 * @param string query
1255 * @return array params
1256 */
1257 function convertUrlQuery($query) {
1258 $queryParts = explode('&', $query);
1259
1260 $params = array();
1261
1262 foreach ($queryParts as $param) {
1263 $item = explode('=', $param);
1264 $params[$item[0]] = $item[1];
1265 }
1266
1267 return $params;
1268 }
1269
1270 function get_article_filters($filters, $title, $content, $link, $timestamp, $author, $tags) {
1271 $matches = array();
1272
1273 foreach ($filters as $filter) {
1274 $match_any_rule = $filter["match_any_rule"];
1275 $inverse = $filter["inverse"];
1276 $filter_match = false;
1277
1278 foreach ($filter["rules"] as $rule) {
1279 $match = false;
1280 $reg_exp = str_replace('/', '\/', $rule["reg_exp"]);
1281 $rule_inverse = $rule["inverse"];
1282
1283 if (!$reg_exp)
1284 continue;
1285
1286 switch ($rule["type"]) {
1287 case "title":
1288 $match = @preg_match("/$reg_exp/i", $title);
1289 break;
1290 case "content":
1291 // we don't need to deal with multiline regexps
1292 $content = preg_replace("/[\r\n\t]/", "", $content);
1293
1294 $match = @preg_match("/$reg_exp/i", $content);
1295 break;
1296 case "both":
1297 // we don't need to deal with multiline regexps
1298 $content = preg_replace("/[\r\n\t]/", "", $content);
1299
1300 $match = (@preg_match("/$reg_exp/i", $title) || @preg_match("/$reg_exp/i", $content));
1301 break;
1302 case "link":
1303 $match = @preg_match("/$reg_exp/i", $link);
1304 break;
1305 case "author":
1306 $match = @preg_match("/$reg_exp/i", $author);
1307 break;
1308 case "tag":
1309 foreach ($tags as $tag) {
1310 if (@preg_match("/$reg_exp/i", $tag)) {
1311 $match = true;
1312 break;
1313 }
1314 }
1315 break;
1316 }
1317
1318 if ($rule_inverse) $match = !$match;
1319
1320 if ($match_any_rule) {
1321 if ($match) {
1322 $filter_match = true;
1323 break;
1324 }
1325 } else {
1326 $filter_match = $match;
1327 if (!$match) {
1328 break;
1329 }
1330 }
1331 }
1332
1333 if ($inverse) $filter_match = !$filter_match;
1334
1335 if ($filter_match) {
1336 foreach ($filter["actions"] AS $action) {
1337 array_push($matches, $action);
1338
1339 // if Stop action encountered, perform no further processing
1340 if ($action["type"] == "stop") return $matches;
1341 }
1342 }
1343 }
1344
1345 return $matches;
1346 }
1347
1348 function find_article_filter($filters, $filter_name) {
1349 foreach ($filters as $f) {
1350 if ($f["type"] == $filter_name) {
1351 return $f;
1352 };
1353 }
1354 return false;
1355 }
1356
1357 function find_article_filters($filters, $filter_name) {
1358 $results = array();
1359
1360 foreach ($filters as $f) {
1361 if ($f["type"] == $filter_name) {
1362 array_push($results, $f);
1363 };
1364 }
1365 return $results;
1366 }
1367
1368 function calculate_article_score($filters) {
1369 $score = 0;
1370
1371 foreach ($filters as $f) {
1372 if ($f["type"] == "score") {
1373 $score += $f["param"];
1374 };
1375 }
1376 return $score;
1377 }
1378
1379 function labels_contains_caption($labels, $caption) {
1380 foreach ($labels as $label) {
1381 if ($label[1] == $caption) {
1382 return true;
1383 }
1384 }
1385
1386 return false;
1387 }
1388
1389 function assign_article_to_label_filters($id, $filters, $owner_uid, $article_labels) {
1390 foreach ($filters as $f) {
1391 if ($f["type"] == "label") {
1392 if (!labels_contains_caption($article_labels, $f["param"])) {
1393 label_add_article($id, $f["param"], $owner_uid);
1394 }
1395 }
1396 }
1397 }
1398
1399 function make_guid_from_title($title) {
1400 return preg_replace("/[ \"\',.:;]/", "-",
1401 mb_strtolower(strip_tags($title), 'utf-8'));
1402 }
1403
1404 /* function verify_feed_xml($feed_data) {
1405 libxml_use_internal_errors(true);
1406 $doc = new DOMDocument();
1407 $doc->loadXML($feed_data);
1408 $error = libxml_get_last_error();
1409 libxml_clear_errors();
1410 return $error;
1411 } */
1412
1413 function housekeeping_common($debug) {
1414 expire_cached_files($debug);
1415 expire_lock_files($debug);
1416 expire_error_log($debug);
1417
1418 $count = update_feedbrowser_cache();
1419 _debug("Feedbrowser updated, $count feeds processed.");
1420
1421 purge_orphans( true);
1422 $rc = cleanup_tags( 14, 50000);
1423
1424 _debug("Cleaned $rc cached tags.");
1425
1426 PluginHost::getInstance()->run_hooks(PluginHost::HOOK_HOUSE_KEEPING, "hook_house_keeping", "");
1427
1428 }
1429 ?>