]> git.wh0rd.org - tt-rss.git/blob - include/rssfuncs.php
git.wh0rd.org / tt-rss.git / blob
? search:
summary | shortlog | log | commit | commitdiff | tree
blame | history | raw | HEAD
rebase translations
[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 define_default('_MIN_CACHE_IMAGE_SIZE', 1024);
6
7 function calculate_article_hash($article, $pluginhost) {
8 $tmp = "";
9
10 foreach ($article as $k => $v) {
11 if ($k != "feed" && isset($v)) {
12 $tmp .= sha1("$k:" . (is_array($v) ? implode(",", $v) : $v));
13 }
14 }
15
16 return sha1(implode(",", $pluginhost->get_plugin_names()) . $tmp);
17 }
18
19 function update_feedbrowser_cache() {
20
21 $result = db_query("SELECT feed_url, site_url, title, COUNT(id) AS subscribers
22 FROM ttrss_feeds WHERE (SELECT COUNT(id) = 0 FROM ttrss_feeds AS tf
23 WHERE tf.feed_url = ttrss_feeds.feed_url
24 AND (private IS true OR auth_login != '' OR auth_pass != '' OR feed_url LIKE '%:%@%/%'))
25 GROUP BY feed_url, site_url, title ORDER BY subscribers DESC LIMIT 1000");
26
27 db_query("BEGIN");
28
29 db_query("DELETE FROM ttrss_feedbrowser_cache");
30
31 $count = 0;
32
33 while ($line = db_fetch_assoc($result)) {
34 $subscribers = db_escape_string($line["subscribers"]);
35 $feed_url = db_escape_string($line["feed_url"]);
36 $title = db_escape_string($line["title"]);
37 $site_url = db_escape_string($line["site_url"]);
38
39 $tmp_result = db_query("SELECT subscribers FROM
40 ttrss_feedbrowser_cache WHERE feed_url = '$feed_url'");
41
42 if (db_num_rows($tmp_result) == 0) {
43
44 db_query("INSERT INTO ttrss_feedbrowser_cache
45 (feed_url, site_url, title, subscribers) VALUES ('$feed_url',
46 '$site_url', '$title', '$subscribers')");
47
48 ++$count;
49
50 }
51
52 }
53
54 db_query("COMMIT");
55
56 return $count;
57
58 }
59
60
61 /**
62 * Update a feed batch.
63 * Used by daemons to update n feeds by run.
64 * Only update feed needing a update, and not being processed
65 * by another process.
66 *
67 * @param mixed $link Database link
68 * @param integer $limit Maximum number of feeds in update batch. Default to DAEMON_FEED_LIMIT.
69 * @param boolean $from_http Set to true if you call this function from http to disable cli specific code.
70 * @param boolean $debug Set to false to disable debug output. Default to true.
71 * @return void
72 */
73 function update_daemon_common($limit = DAEMON_FEED_LIMIT, $from_http = false, $debug = true) {
74 // Process all other feeds using last_updated and interval parameters
75
76 $schema_version = get_schema_version();
77
78 if ($schema_version != SCHEMA_VERSION) {
79 die("Schema version is wrong, please upgrade the database.\n");
80 }
81
82 define('PREFS_NO_CACHE', true);
83
84 // Test if the user has loggued in recently. If not, it does not update its feeds.
85 if (!SINGLE_USER_MODE && DAEMON_UPDATE_LOGIN_LIMIT > 0) {
86 if (DB_TYPE == "pgsql") {
87 $login_thresh_qpart = "AND ttrss_users.last_login >= NOW() - INTERVAL '".DAEMON_UPDATE_LOGIN_LIMIT." days'";
88 } else {
89 $login_thresh_qpart = "AND ttrss_users.last_login >= DATE_SUB(NOW(), INTERVAL ".DAEMON_UPDATE_LOGIN_LIMIT." DAY)";
90 }
91 } else {
92 $login_thresh_qpart = "";
93 }
94
95 // Test if the feed need a update (update interval exceeded).
96 if (DB_TYPE == "pgsql") {
97 $update_limit_qpart = "AND ((
98 ttrss_feeds.update_interval = 0
99 AND ttrss_user_prefs.value != '-1'
100 AND ttrss_feeds.last_updated < NOW() - CAST((ttrss_user_prefs.value || ' minutes') AS INTERVAL)
101 ) OR (
102 ttrss_feeds.update_interval > 0
103 AND ttrss_feeds.last_updated < NOW() - CAST((ttrss_feeds.update_interval || ' minutes') AS INTERVAL)
104 ) OR (ttrss_feeds.last_updated IS NULL
105 AND ttrss_user_prefs.value != '-1')
106 OR (last_updated = '1970-01-01 00:00:00'
107 AND ttrss_user_prefs.value != '-1'))";
108 } else {
109 $update_limit_qpart = "AND ((
110 ttrss_feeds.update_interval = 0
111 AND ttrss_user_prefs.value != '-1'
112 AND ttrss_feeds.last_updated < DATE_SUB(NOW(), INTERVAL CONVERT(ttrss_user_prefs.value, SIGNED INTEGER) MINUTE)
113 ) OR (
114 ttrss_feeds.update_interval > 0
115 AND ttrss_feeds.last_updated < DATE_SUB(NOW(), INTERVAL ttrss_feeds.update_interval MINUTE)
116 ) OR (ttrss_feeds.last_updated IS NULL
117 AND ttrss_user_prefs.value != '-1')
118 OR (last_updated = '1970-01-01 00:00:00'
119 AND ttrss_user_prefs.value != '-1'))";
120 }
121
122 // Test if feed is currently being updated by another process.
123 if (DB_TYPE == "pgsql") {
124 $updstart_thresh_qpart = "AND (ttrss_feeds.last_update_started IS NULL OR ttrss_feeds.last_update_started < NOW() - INTERVAL '10 minutes')";
125 } else {
126 $updstart_thresh_qpart = "AND (ttrss_feeds.last_update_started IS NULL OR ttrss_feeds.last_update_started < DATE_SUB(NOW(), INTERVAL 10 MINUTE))";
127 }
128
129 // Test if there is a limit to number of updated feeds
130 $query_limit = "";
131 if($limit) $query_limit = sprintf("LIMIT %d", $limit);
132
133 $query = "SELECT DISTINCT ttrss_feeds.feed_url, ttrss_feeds.last_updated
134 FROM
135 ttrss_feeds, ttrss_users, ttrss_user_prefs
136 WHERE
137 ttrss_feeds.owner_uid = ttrss_users.id
138 AND ttrss_user_prefs.profile IS NULL
139 AND ttrss_users.id = ttrss_user_prefs.owner_uid
140 AND ttrss_user_prefs.pref_name = 'DEFAULT_UPDATE_INTERVAL'
141 $login_thresh_qpart $update_limit_qpart
142 $updstart_thresh_qpart
143 ORDER BY last_updated $query_limit";
144
145 // We search for feed needing update.
146 $result = db_query($query);
147
148 if($debug) _debug(sprintf("Scheduled %d feeds to update...", db_num_rows($result)));
149
150 // Here is a little cache magic in order to minimize risk of double feed updates.
151 $feeds_to_update = array();
152 while ($line = db_fetch_assoc($result)) {
153 array_push($feeds_to_update, db_escape_string($line['feed_url']));
154 }
155
156 // We update the feed last update started date before anything else.
157 // There is no lag due to feed contents downloads
158 // It prevent an other process to update the same feed.
159
160 if(count($feeds_to_update) > 0) {
161 $feeds_quoted = array();
162
163 foreach ($feeds_to_update as $feed) {
164 array_push($feeds_quoted, "'" . db_escape_string($feed) . "'");
165 }
166
167 db_query(sprintf("UPDATE ttrss_feeds SET last_update_started = NOW()
168 WHERE feed_url IN (%s)", implode(',', $feeds_quoted)));
169 }
170
171 $nf = 0;
172 $bstarted = microtime(true);
173
174 // For each feed, we call the feed update function.
175 foreach ($feeds_to_update as $feed) {
176 if($debug) _debug("Base feed: $feed");
177
178 //update_rss_feed($line["id"], true);
179
180 // since we have the data cached, we can deal with other feeds with the same url
181
182 $tmp_result = db_query("SELECT DISTINCT ttrss_feeds.id,last_updated,ttrss_feeds.owner_uid
183 FROM ttrss_feeds, ttrss_users, ttrss_user_prefs WHERE
184 ttrss_user_prefs.owner_uid = ttrss_feeds.owner_uid AND
185 ttrss_users.id = ttrss_user_prefs.owner_uid AND
186 ttrss_user_prefs.pref_name = 'DEFAULT_UPDATE_INTERVAL' AND
187 ttrss_user_prefs.profile IS NULL AND
188 feed_url = '".db_escape_string($feed)."' AND
189 (ttrss_feeds.update_interval > 0 OR
190 ttrss_user_prefs.value != '-1')
191 $login_thresh_qpart
192 ORDER BY ttrss_feeds.id $query_limit");
193
194 if (db_num_rows($tmp_result) > 0) {
195 $rss = false;
196
197 while ($tline = db_fetch_assoc($tmp_result)) {
198 if($debug) _debug(" => " . $tline["last_updated"] . ", " . $tline["id"] . " " . $tline["owner_uid"]);
199
200 $fstarted = microtime(true);
201 $rss = update_rss_feed($tline["id"], true, false);
202 _debug_suppress(false);
203
204 _debug(sprintf(" %.4f (sec)", microtime(true) - $fstarted));
205
206 ++$nf;
207 }
208 }
209 }
210
211 if ($nf > 0) {
212 _debug(sprintf("Processed %d feeds in %.4f (sec), %.4f (sec/feed avg)", $nf,
213 microtime(true) - $bstarted, (microtime(true) - $bstarted) / $nf));
214 }
215
216 require_once "digest.php";
217
218 // Send feed digests by email if needed.
219 send_headlines_digests($debug);
220
221 return $nf;
222
223 } // function update_daemon_common
224
225 // ignore_daemon is not used
226 function update_rss_feed($feed, $ignore_daemon = false, $no_cache = false, $rss = false) {
227
228 $debug_enabled = defined('DAEMON_EXTENDED_DEBUG') || $_REQUEST['xdebug'];
229
230 _debug_suppress(!$debug_enabled);
231 _debug("start", $debug_enabled);
232
233 $result = db_query("SELECT id,update_interval,auth_login,
234 feed_url,auth_pass,cache_images,
235 mark_unread_on_update, owner_uid,
236 pubsub_state, auth_pass_encrypted,
237 (SELECT max(date_entered) FROM
238 ttrss_entries, ttrss_user_entries where ref_id = id AND feed_id = '$feed') AS last_article_timestamp
239 FROM ttrss_feeds WHERE id = '$feed'");
240
241 if (db_num_rows($result) == 0) {
242 _debug("feed $feed NOT FOUND/SKIPPED", $debug_enabled);
243 return false;
244 }
245
246 $last_article_timestamp = @strtotime(db_fetch_result($result, 0, "last_article_timestamp"));
247
248 if (defined('_DISABLE_HTTP_304'))
249 $last_article_timestamp = 0;
250
251 $owner_uid = db_fetch_result($result, 0, "owner_uid");
252 $mark_unread_on_update = sql_bool_to_bool(db_fetch_result($result,
253 0, "mark_unread_on_update"));
254 $pubsub_state = db_fetch_result($result, 0, "pubsub_state");
255 $auth_pass_encrypted = sql_bool_to_bool(db_fetch_result($result,
256 0, "auth_pass_encrypted"));
257
258 db_query("UPDATE ttrss_feeds SET last_update_started = NOW()
259 WHERE id = '$feed'");
260
261 $auth_login = db_fetch_result($result, 0, "auth_login");
262 $auth_pass = db_fetch_result($result, 0, "auth_pass");
263
264 if ($auth_pass_encrypted) {
265 require_once "crypt.php";
266 $auth_pass = decrypt_string($auth_pass);
267 }
268
269 $cache_images = sql_bool_to_bool(db_fetch_result($result, 0, "cache_images"));
270 $fetch_url = db_fetch_result($result, 0, "feed_url");
271
272 $feed = db_escape_string($feed);
273
274 $date_feed_processed = date('Y-m-d H:i');
275
276 $cache_filename = CACHE_DIR . "/simplepie/" . sha1($fetch_url) . ".xml";
277
278 $pluginhost = new PluginHost();
279 $pluginhost->set_debug($debug_enabled);
280 $user_plugins = get_pref("_ENABLED_PLUGINS", $owner_uid);
281
282 $pluginhost->load(PLUGINS, PluginHost::KIND_ALL);
283 $pluginhost->load($user_plugins, PluginHost::KIND_USER, $owner_uid);
284 $pluginhost->load_data();
285
286 if ($rss && is_object($rss) && get_class($rss) == "FeedParser") {
287 _debug("using previously initialized parser object");
288 } else {
289 $rss_hash = false;
290
291 $force_refetch = isset($_REQUEST["force_refetch"]);
292
293 foreach ($pluginhost->get_hooks(PluginHost::HOOK_FETCH_FEED) as $plugin) {
294 $feed_data = $plugin->hook_fetch_feed($feed_data, $fetch_url, $owner_uid, $feed, $last_article_timestamp, $auth_login, $auth_pass);
295 }
296
297 // try cache
298 if (!$feed_data &&
299 file_exists($cache_filename) &&
300 is_readable($cache_filename) &&
301 !$auth_login && !$auth_pass &&
302 filemtime($cache_filename) > time() - 30) {
303
304 _debug("using local cache [$cache_filename].", $debug_enabled);
305
306 @$feed_data = file_get_contents($cache_filename);
307
308 if ($feed_data) {
309 $rss_hash = sha1($feed_data);
310 }
311
312 } else {
313 _debug("local cache will not be used for this feed", $debug_enabled);
314 }
315
316 // fetch feed from source
317 if (!$feed_data) {
318 _debug("fetching [$fetch_url]...", $debug_enabled);
319 _debug("If-Modified-Since: ".gmdate('D, d M Y H:i:s \G\M\T', $last_article_timestamp), $debug_enabled);
320
321 $feed_data = fetch_file_contents($fetch_url, false,
322 $auth_login, $auth_pass, false,
323 $no_cache ? FEED_FETCH_NO_CACHE_TIMEOUT : FEED_FETCH_TIMEOUT,
324 $force_refetch ? 0 : $last_article_timestamp);
325
326 global $fetch_curl_used;
327
328 if (!$fetch_curl_used) {
329 $tmp = @gzdecode($feed_data);
330
331 if ($tmp) $feed_data = $tmp;
332 }
333
334 $feed_data = trim($feed_data);
335
336 _debug("fetch done.", $debug_enabled);
337
338 /* if ($feed_data) {
339 $error = verify_feed_xml($feed_data);
340
341 if ($error) {
342 _debug("error verifying XML, code: " . $error->code, $debug_enabled);
343
344 if ($error->code == 26) {
345 _debug("got error 26, trying to decode entities...", $debug_enabled);
346
347 $feed_data = html_entity_decode($feed_data, ENT_COMPAT, 'UTF-8');
348
349 $error = verify_feed_xml($feed_data);
350
351 if ($error) $feed_data = '';
352 }
353 }
354 } */
355
356 // cache vanilla feed data for re-use
357 if ($feed_data && !$auth_pass && !$auth_login && is_writable(CACHE_DIR . "/simplepie")) {
358 $new_rss_hash = sha1($feed_data);
359
360 if ($new_rss_hash != $rss_hash) {
361 _debug("saving $cache_filename", $debug_enabled);
362 @file_put_contents($cache_filename, $feed_data);
363 }
364 }
365 }
366
367 if (!$feed_data) {
368 global $fetch_last_error;
369 global $fetch_last_error_code;
370
371 _debug("unable to fetch: $fetch_last_error [$fetch_last_error_code]", $debug_enabled);
372
373 $error_escaped = '';
374
375 // If-Modified-Since
376 if ($fetch_last_error_code != 304) {
377 $error_escaped = db_escape_string($fetch_last_error);
378 } else {
379 _debug("source claims data not modified, nothing to do.", $debug_enabled);
380 }
381
382 db_query(
383 "UPDATE ttrss_feeds SET last_error = '$error_escaped',
384 last_updated = NOW() WHERE id = '$feed'");
385
386 return;
387 }
388 }
389
390 foreach ($pluginhost->get_hooks(PluginHost::HOOK_FEED_FETCHED) as $plugin) {
391 $feed_data = $plugin->hook_feed_fetched($feed_data, $fetch_url, $owner_uid, $feed);
392 }
393
394 // set last update to now so if anything *simplepie* crashes later we won't be
395 // continuously failing on the same feed
396 //db_query("UPDATE ttrss_feeds SET last_updated = NOW() WHERE id = '$feed'");
397
398 if (!$rss) {
399 $rss = new FeedParser($feed_data);
400 $rss->init();
401 }
402
403 if (DETECT_ARTICLE_LANGUAGE) {
404 require_once "lib/languagedetect/LanguageDetect.php";
405
406 $lang = new Text_LanguageDetect();
407 $lang->setNameMode(2);
408 }
409
410 // print_r($rss);
411
412 $feed = db_escape_string($feed);
413
414 if (!$rss->error()) {
415
416 // We use local pluginhost here because we need to load different per-user feed plugins
417 $pluginhost->run_hooks(PluginHost::HOOK_FEED_PARSED, "hook_feed_parsed", $rss);
418
419 _debug("processing feed data...", $debug_enabled);
420
421 // db_query("BEGIN");
422
423 if (DB_TYPE == "pgsql") {
424 $favicon_interval_qpart = "favicon_last_checked < NOW() - INTERVAL '12 hour'";
425 } else {
426 $favicon_interval_qpart = "favicon_last_checked < DATE_SUB(NOW(), INTERVAL 12 HOUR)";
427 }
428
429 $result = db_query("SELECT title,site_url,owner_uid,favicon_avg_color,
430 (favicon_last_checked IS NULL OR $favicon_interval_qpart) AS
431 favicon_needs_check
432 FROM ttrss_feeds WHERE id = '$feed'");
433
434 $registered_title = db_fetch_result($result, 0, "title");
435 $orig_site_url = db_fetch_result($result, 0, "site_url");
436 $favicon_needs_check = sql_bool_to_bool(db_fetch_result($result, 0,
437 "favicon_needs_check"));
438 $favicon_avg_color = db_fetch_result($result, 0, "favicon_avg_color");
439
440 $owner_uid = db_fetch_result($result, 0, "owner_uid");
441
442 $site_url = db_escape_string(mb_substr(rewrite_relative_url($fetch_url, $rss->get_link()), 0, 245));
443
444 _debug("site_url: $site_url", $debug_enabled);
445 _debug("feed_title: " . $rss->get_title(), $debug_enabled);
446
447 if ($favicon_needs_check || $force_refetch) {
448
449 /* terrible hack: if we crash on floicon shit here, we won't check
450 * the icon avgcolor again (unless the icon got updated) */
451
452 $favicon_file = ICONS_DIR . "/$feed.ico";
453 $favicon_modified = @filemtime($favicon_file);
454
455 _debug("checking favicon...", $debug_enabled);
456
457 check_feed_favicon($site_url, $feed);
458 $favicon_modified_new = @filemtime($favicon_file);
459
460 if ($favicon_modified_new > $favicon_modified)
461 $favicon_avg_color = '';
462
463 if (file_exists($favicon_file) && function_exists("imagecreatefromstring") && $favicon_avg_color == '') {
464 require_once "colors.php";
465
466 db_query("UPDATE ttrss_feeds SET favicon_avg_color = 'fail' WHERE
467 id = '$feed'");
468
469 $favicon_color = db_escape_string(
470 calculate_avg_color($favicon_file));
471
472 $favicon_colorstring = ",favicon_avg_color = '".$favicon_color."'";
473 } else if ($favicon_avg_color == 'fail') {
474 _debug("floicon failed on this file, not trying to recalculate avg color", $debug_enabled);
475 }
476
477 db_query("UPDATE ttrss_feeds SET favicon_last_checked = NOW()
478 $favicon_colorstring
479 WHERE id = '$feed'");
480 }
481
482 if (!$registered_title || $registered_title == "[Unknown]") {
483
484 $feed_title = db_escape_string(mb_substr($rss->get_title(), 0, 199));
485
486 if ($feed_title) {
487 _debug("registering title: $feed_title", $debug_enabled);
488
489 db_query("UPDATE ttrss_feeds SET
490 title = '$feed_title' WHERE id = '$feed'");
491 }
492 }
493
494 if ($site_url && $orig_site_url != $site_url) {
495 db_query("UPDATE ttrss_feeds SET
496 site_url = '$site_url' WHERE id = '$feed'");
497 }
498
499 _debug("loading filters & labels...", $debug_enabled);
500
501 $filters = load_filters($feed, $owner_uid);
502 $labels = get_all_labels($owner_uid);
503
504 _debug("" . count($filters) . " filters loaded.", $debug_enabled);
505
506 $items = $rss->get_items();
507
508 if (!is_array($items)) {
509 _debug("no articles found.", $debug_enabled);
510
511 db_query("UPDATE ttrss_feeds
512 SET last_updated = NOW(), last_error = '' WHERE id = '$feed'");
513
514 return; // no articles
515 }
516
517 if ($pubsub_state != 2 && PUBSUBHUBBUB_ENABLED) {
518
519 _debug("checking for PUSH hub...", $debug_enabled);
520
521 $feed_hub_url = false;
522
523 $links = $rss->get_links('hub');
524
525 if ($links && is_array($links)) {
526 foreach ($links as $l) {
527 $feed_hub_url = $l;
528 break;
529 }
530 }
531
532 _debug("feed hub url: $feed_hub_url", $debug_enabled);
533
534 $feed_self_url = $fetch_url;
535
536 $links = $rss->get_links('self');
537
538 if ($links && is_array($links)) {
539 foreach ($links as $l) {
540 $feed_self_url = $l;
541 break;
542 }
543 }
544
545 _debug("feed self url = $feed_self_url");
546
547 if ($feed_hub_url && $feed_self_url && function_exists('curl_init') &&
548 !ini_get("open_basedir")) {
549
550 require_once 'lib/pubsubhubbub/subscriber.php';
551
552 $callback_url = get_self_url_prefix() .
553 "/public.php?op=pubsub&id=$feed";
554
555 $s = new Subscriber($feed_hub_url, $callback_url);
556
557 $rc = $s->subscribe($feed_self_url);
558
559 _debug("feed hub url found, subscribe request sent. [rc=$rc]", $debug_enabled);
560
561 db_query("UPDATE ttrss_feeds SET pubsub_state = 1
562 WHERE id = '$feed'");
563 }
564 }
565
566 _debug("processing articles...", $debug_enabled);
567
568 foreach ($items as $item) {
569 if ($_REQUEST['xdebug'] == 3) {
570 print_r($item);
571 }
572
573 $entry_guid = $item->get_id();
574 if (!$entry_guid) $entry_guid = $item->get_link();
575 if (!$entry_guid) $entry_guid = make_guid_from_title($item->get_title());
576 if (!$entry_guid) continue;
577
578 $entry_guid = "$owner_uid,$entry_guid";
579
580 $entry_guid_hashed = db_escape_string('SHA1:' . sha1($entry_guid));
581
582 _debug("guid $entry_guid / $entry_guid_hashed", $debug_enabled);
583
584 $entry_timestamp = "";
585
586 $entry_timestamp = $item->get_date();
587
588 _debug("orig date: " . $item->get_date(), $debug_enabled);
589
590 if ($entry_timestamp == -1 || !$entry_timestamp || $entry_timestamp > time()) {
591 $entry_timestamp = time();
592 }
593
594 $entry_timestamp_fmt = strftime("%Y/%m/%d %H:%M:%S", $entry_timestamp);
595
596 _debug("date $entry_timestamp [$entry_timestamp_fmt]", $debug_enabled);
597
598 // $entry_title = html_entity_decode($item->get_title(), ENT_COMPAT, 'UTF-8');
599 // $entry_title = decode_numeric_entities($entry_title);
600 $entry_title = $item->get_title();
601
602 $entry_link = rewrite_relative_url($site_url, $item->get_link());
603
604 _debug("title $entry_title", $debug_enabled);
605 _debug("link $entry_link", $debug_enabled);
606
607 if (!$entry_title) $entry_title = date("Y-m-d H:i:s", $entry_timestamp);;
608
609 $entry_content = $item->get_content();
610 if (!$entry_content) $entry_content = $item->get_description();
611
612 if ($_REQUEST["xdebug"] == 2) {
613 print "content: ";
614 print $entry_content;
615 print "\n";
616 }
617
618 $entry_language = "";
619
620 if (DETECT_ARTICLE_LANGUAGE) {
621 $entry_language = $lang->detect($entry_title . " " . $entry_content, 1);
622
623 if (count($entry_language) > 0) {
624 $possible = array_keys($entry_language);
625 $entry_language = $possible[0];
626
627 _debug("detected language: $entry_language", $debug_enabled);
628 } else {
629 $entry_language = "";
630 }
631 }
632
633 $entry_comments = $item->get_comments_url();
634 $entry_author = $item->get_author();
635
636 $entry_guid = db_escape_string(mb_substr($entry_guid, 0, 245));
637
638 $entry_comments = db_escape_string(mb_substr(trim($entry_comments), 0, 245));
639 $entry_author = db_escape_string(mb_substr(trim($entry_author), 0, 245));
640
641 $num_comments = (int) $item->get_comments_count();
642
643 _debug("author $entry_author", $debug_enabled);
644 _debug("num_comments: $num_comments", $debug_enabled);
645 _debug("looking for tags...", $debug_enabled);
646
647 // parse <category> entries into tags
648
649 $additional_tags = array();
650
651 $additional_tags_src = $item->get_categories();
652
653 if (is_array($additional_tags_src)) {
654 foreach ($additional_tags_src as $tobj) {
655 array_push($additional_tags, $tobj);
656 }
657 }
658
659 $entry_tags = array_unique($additional_tags);
660
661 for ($i = 0; $i < count($entry_tags); $i++)
662 $entry_tags[$i] = mb_strtolower($entry_tags[$i], 'utf-8');
663
664 _debug("tags found: " . join(",", $entry_tags), $debug_enabled);
665
666 _debug("done collecting data.", $debug_enabled);
667
668 $result = db_query("SELECT id, content_hash FROM ttrss_entries
669 WHERE guid = '".db_escape_string($entry_guid)."' OR guid = '$entry_guid_hashed'");
670
671 if (db_num_rows($result) != 0) {
672 $base_entry_id = db_fetch_result($result, 0, "id");
673 $entry_stored_hash = db_fetch_result($result, 0, "content_hash");
674 $article_labels = get_article_labels($base_entry_id, $owner_uid);
675 } else {
676 $base_entry_id = false;
677 $entry_stored_hash = "";
678 $article_labels = array();
679 }
680
681 $article = array("owner_uid" => $owner_uid, // read only
682 "guid" => $entry_guid, // read only
683 "title" => $entry_title,
684 "content" => $entry_content,
685 "link" => $entry_link,
686 "labels" => $article_labels, // current limitation: can add labels to article, can't remove them
687 "tags" => $entry_tags,
688 "author" => $entry_author,
689 "force_catchup" => false, // ugly hack for the time being
690 "language" => $entry_language, // read only
691 "feed" => array("id" => $feed,
692 "fetch_url" => $fetch_url,
693 "site_url" => $site_url)
694 );
695
696 $entry_plugin_data = "";
697 $entry_current_hash = calculate_article_hash($article, $pluginhost);
698
699 _debug("article hash: $entry_current_hash [stored=$entry_stored_hash]", $debug_enabled);
700
701 if ($entry_current_hash == $entry_stored_hash && !isset($_REQUEST["force_rehash"])) {
702 _debug("stored article seems up to date [IID: $base_entry_id], updating timestamp only", $debug_enabled);
703
704 // we keep encountering the entry in feeds, so we need to
705 // update date_updated column so that we don't get horrible
706 // dupes when the entry gets purged and reinserted again e.g.
707 // in the case of SLOW SLOW OMG SLOW updating feeds
708
709 $base_entry_id = db_fetch_result($result, 0, "id");
710
711 db_query("UPDATE ttrss_entries SET date_updated = NOW()
712 WHERE id = '$base_entry_id'");
713
714 // if we allow duplicate posts, we have to continue to
715 // create the user entries for this feed
716 if (!get_pref("ALLOW_DUPLICATE_POSTS", $owner_uid, false)) {
717 continue;
718 }
719 }
720
721 _debug("hash differs, applying plugin filters:", $debug_enabled);
722
723 foreach ($pluginhost->get_hooks(PluginHost::HOOK_ARTICLE_FILTER) as $plugin) {
724 _debug("... " . get_class($plugin), $debug_enabled);
725
726 $start = microtime(true);
727 $article = $plugin->hook_article_filter($article);
728
729 _debug("=== " . sprintf("%.4f (sec)", microtime(true) - $start), $debug_enabled);
730
731 $entry_plugin_data .= mb_strtolower(get_class($plugin)) . ",";
732 }
733
734 $entry_plugin_data = db_escape_string($entry_plugin_data);
735
736 _debug("plugin data: $entry_plugin_data", $debug_enabled);
737
738 $entry_tags = $article["tags"];
739 $entry_guid = db_escape_string($entry_guid);
740 $entry_title = db_escape_string($article["title"]);
741 $entry_author = db_escape_string($article["author"]);
742 $entry_link = db_escape_string($article["link"]);
743 $entry_content = $article["content"]; // escaped below
744 $entry_force_catchup = $article["force_catchup"];
745 $article_labels = $article["labels"];
746
747 if ($debug_enabled) {
748 _debug("article labels:", $debug_enabled);
749 print_r($article_labels);
750 }
751
752 _debug("force catchup: $entry_force_catchup");
753
754 if ($cache_images && is_writable(CACHE_DIR . '/images'))
755 cache_images($entry_content, $site_url, $debug_enabled);
756
757 $entry_content = db_escape_string($entry_content, false);
758
759 db_query("BEGIN");
760
761 $result = db_query("SELECT id FROM ttrss_entries
762 WHERE (guid = '$entry_guid' OR guid = '$entry_guid_hashed')");
763
764 if (db_num_rows($result) == 0) {
765
766 _debug("base guid [$entry_guid] not found", $debug_enabled);
767
768 // base post entry does not exist, create it
769
770 $result = db_query(
771 "INSERT INTO ttrss_entries
772 (title,
773 guid,
774 link,
775 updated,
776 content,
777 content_hash,
778 no_orig_date,
779 date_updated,
780 date_entered,
781 comments,
782 num_comments,
783 plugin_data,
784 lang,
785 author)
786 VALUES
787 ('$entry_title',
788 '$entry_guid_hashed',
789 '$entry_link',
790 '$entry_timestamp_fmt',
791 '$entry_content',
792 '$entry_current_hash',
793 false,
794 NOW(),
795 '$date_feed_processed',
796 '$entry_comments',
797 '$num_comments',
798 '$entry_plugin_data',
799 '$entry_language',
800 '$entry_author')");
801
802 } else {
803 $base_entry_id = db_fetch_result($result, 0, "id");
804 }
805
806 // now it should exist, if not - bad luck then
807
808 $result = db_query("SELECT id FROM ttrss_entries
809 WHERE guid = '$entry_guid' OR guid = '$entry_guid_hashed'");
810
811 $entry_ref_id = 0;
812 $entry_int_id = 0;
813
814 if (db_num_rows($result) == 1) {
815
816 _debug("base guid found, checking for user record", $debug_enabled);
817
818 $ref_id = db_fetch_result($result, 0, "id");
819 $entry_ref_id = $ref_id;
820
821 /* $stored_guid = db_fetch_result($result, 0, "guid");
822 if ($stored_guid != $entry_guid_hashed) {
823 if ($debug_enabled) _debug("upgrading compat guid to hashed one", $debug_enabled);
824
825 db_query("UPDATE ttrss_entries SET guid = '$entry_guid_hashed' WHERE
826 id = '$ref_id'");
827 } */
828
829 // check for user post link to main table
830
831 // do we allow duplicate posts with same GUID in different feeds?
832 if (get_pref("ALLOW_DUPLICATE_POSTS", $owner_uid, false)) {
833 $dupcheck_qpart = "AND (feed_id = '$feed' OR feed_id IS NULL)";
834 } else {
835 $dupcheck_qpart = "";
836 }
837
838 /* Collect article tags here so we could filter by them: */
839
840 $article_filters = get_article_filters($filters, $entry_title,
841 $entry_content, $entry_link, $entry_timestamp, $entry_author,
842 $entry_tags);
843
844 if ($debug_enabled) {
845 _debug("article filters: ", $debug_enabled);
846 if (count($article_filters) != 0) {
847 print_r($article_filters);
848 }
849 }
850
851 if (find_article_filter($article_filters, "filter")) {
852 db_query("COMMIT"); // close transaction in progress
853 continue;
854 }
855
856 $score = calculate_article_score($article_filters);
857
858 _debug("initial score: $score", $debug_enabled);
859
860 $query = "SELECT ref_id, int_id FROM ttrss_user_entries WHERE
861 ref_id = '$ref_id' AND owner_uid = '$owner_uid'
862 $dupcheck_qpart";
863
864 // if ($_REQUEST["xdebug"]) print "$query\n";
865
866 $result = db_query($query);
867
868 // okay it doesn't exist - create user entry
869 if (db_num_rows($result) == 0) {
870
871 _debug("user record not found, creating...", $debug_enabled);
872
873 if ($score >= -500 && !find_article_filter($article_filters, 'catchup') && !$entry_force_catchup) {
874 $unread = 'true';
875 $last_read_qpart = 'NULL';
876 } else {
877 $unread = 'false';
878 $last_read_qpart = 'NOW()';
879 }
880
881 if (find_article_filter($article_filters, 'mark') || $score > 1000) {
882 $marked = 'true';
883 } else {
884 $marked = 'false';
885 }
886
887 if (find_article_filter($article_filters, 'publish')) {
888 $published = 'true';
889 } else {
890 $published = 'false';
891 }
892
893 // N-grams
894
895 /* if (DB_TYPE == "pgsql" and defined('_NGRAM_TITLE_DUPLICATE_THRESHOLD')) {
896
897 $result = db_query("SELECT COUNT(*) AS similar FROM
898 ttrss_entries,ttrss_user_entries
899 WHERE ref_id = id AND updated >= NOW() - INTERVAL '7 day'
900 AND similarity(title, '$entry_title') >= "._NGRAM_TITLE_DUPLICATE_THRESHOLD."
901 AND owner_uid = $owner_uid");
902
903 $ngram_similar = db_fetch_result($result, 0, "similar");
904
905 _debug("N-gram similar results: $ngram_similar", $debug_enabled);
906
907 if ($ngram_similar > 0) {
908 $unread = 'false';
909 }
910 } */
911
912 $last_marked = ($marked == 'true') ? 'NOW()' : 'NULL';
913 $last_published = ($published == 'true') ? 'NOW()' : 'NULL';
914
915 $result = db_query(
916 "INSERT INTO ttrss_user_entries
917 (ref_id, owner_uid, feed_id, unread, last_read, marked,
918 published, score, tag_cache, label_cache, uuid,
919 last_marked, last_published)
920 VALUES ('$ref_id', '$owner_uid', '$feed', $unread,
921 $last_read_qpart, $marked, $published, '$score', '', '',
922 '', $last_marked, $last_published)");
923
924 if (PUBSUBHUBBUB_HUB && $published == 'true') {
925 $rss_link = get_self_url_prefix() .
926 "/public.php?op=rss&id=-2&key=" .
927 get_feed_access_key(-2, false, $owner_uid);
928
929 $p = new Publisher(PUBSUBHUBBUB_HUB);
930
931 /* $pubsub_result = */ $p->publish_update($rss_link);
932 }
933
934 $result = db_query(
935 "SELECT int_id FROM ttrss_user_entries WHERE
936 ref_id = '$ref_id' AND owner_uid = '$owner_uid' AND
937 feed_id = '$feed' LIMIT 1");
938
939 if (db_num_rows($result) == 1) {
940 $entry_int_id = db_fetch_result($result, 0, "int_id");
941 }
942 } else {
943 _debug("user record FOUND", $debug_enabled);
944
945 $entry_ref_id = db_fetch_result($result, 0, "ref_id");
946 $entry_int_id = db_fetch_result($result, 0, "int_id");
947 }
948
949 _debug("RID: $entry_ref_id, IID: $entry_int_id", $debug_enabled);
950
951 db_query("UPDATE ttrss_entries
952 SET title = '$entry_title',
953 content = '$entry_content',
954 content_hash = '$entry_current_hash',
955 updated = '$entry_timestamp_fmt',
956 num_comments = '$num_comments',
957 plugin_data = '$entry_plugin_data',
958 author = '$entry_author',
959 lang = '$entry_language'
960 WHERE id = '$ref_id'");
961
962 if ($mark_unread_on_update) {
963 db_query("UPDATE ttrss_user_entries
964 SET last_read = null, unread = true WHERE ref_id = '$ref_id'");
965 }
966 }
967
968 db_query("COMMIT");
969
970 _debug("assigning labels [other]...", $debug_enabled);
971
972 foreach ($article_labels as $label) {
973 label_add_article($entry_ref_id, $label[1], $owner_uid);
974 }
975
976 _debug("assigning labels [filters]...", $debug_enabled);
977
978 assign_article_to_label_filters($entry_ref_id, $article_filters,
979 $owner_uid, $article_labels);
980
981 _debug("looking for enclosures...", $debug_enabled);
982
983 // enclosures
984
985 $enclosures = array();
986
987 $encs = $item->get_enclosures();
988
989 if (is_array($encs)) {
990 foreach ($encs as $e) {
991 $e_item = array(
992 $e->link, $e->type, $e->length, $e->title, $e->width, $e->height);
993 array_push($enclosures, $e_item);
994 }
995 }
996
997 if ($debug_enabled) {
998 _debug("article enclosures:", $debug_enabled);
999 print_r($enclosures);
1000 }
1001
1002 db_query("BEGIN");
1003
1004 // debugging
1005 // db_query("DELETE FROM ttrss_enclosures WHERE post_id = '$entry_ref_id'");
1006
1007 foreach ($enclosures as $enc) {
1008 $enc_url = db_escape_string($enc[0]);
1009 $enc_type = db_escape_string($enc[1]);
1010 $enc_dur = db_escape_string($enc[2]);
1011 $enc_title = db_escape_string($enc[3]);
1012 $enc_width = intval($enc[4]);
1013 $enc_height = intval($enc[5]);
1014
1015 $result = db_query("SELECT id FROM ttrss_enclosures
1016 WHERE content_url = '$enc_url' AND post_id = '$entry_ref_id'");
1017
1018 if (db_num_rows($result) == 0) {
1019 db_query("INSERT INTO ttrss_enclosures
1020 (content_url, content_type, title, duration, post_id, width, height) VALUES
1021 ('$enc_url', '$enc_type', '$enc_title', '$enc_dur', '$entry_ref_id', $enc_width, $enc_height)");
1022 }
1023 }
1024
1025 db_query("COMMIT");
1026
1027 // check for manual tags (we have to do it here since they're loaded from filters)
1028
1029 foreach ($article_filters as $f) {
1030 if ($f["type"] == "tag") {
1031
1032 $manual_tags = trim_array(explode(",", $f["param"]));
1033
1034 foreach ($manual_tags as $tag) {
1035 if (tag_is_valid($tag)) {
1036 array_push($entry_tags, $tag);
1037 }
1038 }
1039 }
1040 }
1041
1042 // Skip boring tags
1043
1044 $boring_tags = trim_array(explode(",", mb_strtolower(get_pref(
1045 'BLACKLISTED_TAGS', $owner_uid, ''), 'utf-8')));
1046
1047 $filtered_tags = array();
1048 $tags_to_cache = array();
1049
1050 if ($entry_tags && is_array($entry_tags)) {
1051 foreach ($entry_tags as $tag) {
1052 if (array_search($tag, $boring_tags) === false) {
1053 array_push($filtered_tags, $tag);
1054 }
1055 }
1056 }
1057
1058 $filtered_tags = array_unique($filtered_tags);
1059
1060 if ($debug_enabled) {
1061 _debug("filtered article tags:", $debug_enabled);
1062 print_r($filtered_tags);
1063 }
1064
1065 // Save article tags in the database
1066
1067 if (count($filtered_tags) > 0) {
1068
1069 db_query("BEGIN");
1070
1071 foreach ($filtered_tags as $tag) {
1072
1073 $tag = sanitize_tag($tag);
1074 $tag = db_escape_string($tag);
1075
1076 if (!tag_is_valid($tag)) continue;
1077
1078 $result = db_query("SELECT id FROM ttrss_tags
1079 WHERE tag_name = '$tag' AND post_int_id = '$entry_int_id' AND
1080 owner_uid = '$owner_uid' LIMIT 1");
1081
1082 if ($result && db_num_rows($result) == 0) {
1083
1084 db_query("INSERT INTO ttrss_tags
1085 (owner_uid,tag_name,post_int_id)
1086 VALUES ('$owner_uid','$tag', '$entry_int_id')");
1087 }
1088
1089 array_push($tags_to_cache, $tag);
1090 }
1091
1092 /* update the cache */
1093
1094 $tags_to_cache = array_unique($tags_to_cache);
1095
1096 $tags_str = db_escape_string(join(",", $tags_to_cache));
1097
1098 db_query("UPDATE ttrss_user_entries
1099 SET tag_cache = '$tags_str' WHERE ref_id = '$entry_ref_id'
1100 AND owner_uid = $owner_uid");
1101
1102 db_query("COMMIT");
1103 }
1104
1105 _debug("article processed", $debug_enabled);
1106 }
1107
1108 _debug("purging feed...", $debug_enabled);
1109
1110 purge_feed($feed, 0, $debug_enabled);
1111
1112 db_query("UPDATE ttrss_feeds
1113 SET last_updated = NOW(), last_error = '' WHERE id = '$feed'");
1114
1115 // db_query("COMMIT");
1116
1117 } else {
1118
1119 $error_msg = db_escape_string(mb_substr($rss->error(), 0, 245));
1120
1121 _debug("fetch error: $error_msg", $debug_enabled);
1122
1123 if (count($rss->errors()) > 1) {
1124 foreach ($rss->errors() as $error) {
1125 _debug("+ $error");
1126 }
1127 }
1128
1129 db_query(
1130 "UPDATE ttrss_feeds SET last_error = '$error_msg',
1131 last_updated = NOW() WHERE id = '$feed'");
1132
1133 unset($rss);
1134 }
1135
1136 _debug("done", $debug_enabled);
1137
1138 return $rss;
1139 }
1140
1141 function cache_images($html, $site_url, $debug) {
1142 libxml_use_internal_errors(true);
1143
1144 $charset_hack = '<head>
1145 <meta http-equiv="Content-Type" content="text/html; charset=utf-8"/>
1146 </head>';
1147
1148 $doc = new DOMDocument();
1149 $doc->loadHTML($charset_hack . $html);
1150 $xpath = new DOMXPath($doc);
1151
1152 $entries = $xpath->query('(//img[@src])');
1153
1154 foreach ($entries as $entry) {
1155 if ($entry->hasAttribute('src')) {
1156 $src = rewrite_relative_url($site_url, $entry->getAttribute('src'));
1157
1158 $local_filename = CACHE_DIR . "/images/" . sha1($src) . ".png";
1159
1160 if ($debug) _debug("cache_images: downloading: $src to $local_filename");
1161
1162 if (!file_exists($local_filename)) {
1163 $file_content = fetch_file_contents($src);
1164
1165 if ($file_content && strlen($file_content) > _MIN_CACHE_IMAGE_SIZE) {
1166 file_put_contents($local_filename, $file_content);
1167 }
1168 }
1169 }
1170 }
1171 }
1172
1173 function expire_error_log($debug) {
1174 if ($debug) _debug("Removing old error log entries...");
1175
1176 if (DB_TYPE == "pgsql") {
1177 db_query("DELETE FROM ttrss_error_log
1178 WHERE created_at < NOW() - INTERVAL '7 days'");
1179 } else {
1180 db_query("DELETE FROM ttrss_error_log
1181 WHERE created_at < DATE_SUB(NOW(), INTERVAL 7 DAY)");
1182 }
1183
1184 }
1185
1186 function expire_lock_files($debug) {
1187 //if ($debug) _debug("Removing old lock files...");
1188
1189 $num_deleted = 0;
1190
1191 if (is_writable(LOCK_DIRECTORY)) {
1192 $files = glob(LOCK_DIRECTORY . "/*.lock");
1193
1194 if ($files) {
1195 foreach ($files as $file) {
1196 if (!file_is_locked(basename($file)) && time() - filemtime($file) > 86400*2) {
1197 unlink($file);
1198 ++$num_deleted;
1199 }
1200 }
1201 }
1202 }
1203
1204 if ($debug) _debug("Removed $num_deleted old lock files.");
1205 }
1206
1207 function expire_cached_files($debug) {
1208 foreach (array("simplepie", "images", "export", "upload") as $dir) {
1209 $cache_dir = CACHE_DIR . "/$dir";
1210
1211 // if ($debug) _debug("Expiring $cache_dir");
1212
1213 $num_deleted = 0;
1214
1215 if (is_writable($cache_dir)) {
1216 $files = glob("$cache_dir/*");
1217
1218 if ($files) {
1219 foreach ($files as $file) {
1220 if (time() - filemtime($file) > 86400*7) {
1221 unlink($file);
1222
1223 ++$num_deleted;
1224 }
1225 }
1226 }
1227 }
1228
1229 if ($debug) _debug("$cache_dir: removed $num_deleted files.");
1230 }
1231 }
1232
1233 /**
1234 * Source: http://www.php.net/manual/en/function.parse-url.php#104527
1235 * Returns the url query as associative array
1236 *
1237 * @param string query
1238 * @return array params
1239 */
1240 function convertUrlQuery($query) {
1241 $queryParts = explode('&', $query);
1242
1243 $params = array();
1244
1245 foreach ($queryParts as $param) {
1246 $item = explode('=', $param);
1247 $params[$item[0]] = $item[1];
1248 }
1249
1250 return $params;
1251 }
1252
1253 function get_article_filters($filters, $title, $content, $link, $timestamp, $author, $tags) {
1254 $matches = array();
1255
1256 foreach ($filters as $filter) {
1257 $match_any_rule = $filter["match_any_rule"];
1258 $inverse = $filter["inverse"];
1259 $filter_match = false;
1260
1261 foreach ($filter["rules"] as $rule) {
1262 $match = false;
1263 $reg_exp = str_replace('/', '\/', $rule["reg_exp"]);
1264 $rule_inverse = $rule["inverse"];
1265
1266 if (!$reg_exp)
1267 continue;
1268
1269 switch ($rule["type"]) {
1270 case "title":
1271 $match = @preg_match("/$reg_exp/i", $title);
1272 break;
1273 case "content":
1274 // we don't need to deal with multiline regexps
1275 $content = preg_replace("/[\r\n\t]/", "", $content);
1276
1277 $match = @preg_match("/$reg_exp/i", $content);
1278 break;
1279 case "both":
1280 // we don't need to deal with multiline regexps
1281 $content = preg_replace("/[\r\n\t]/", "", $content);
1282
1283 $match = (@preg_match("/$reg_exp/i", $title) || @preg_match("/$reg_exp/i", $content));
1284 break;
1285 case "link":
1286 $match = @preg_match("/$reg_exp/i", $link);
1287 break;
1288 case "author":
1289 $match = @preg_match("/$reg_exp/i", $author);
1290 break;
1291 case "tag":
1292 foreach ($tags as $tag) {
1293 if (@preg_match("/$reg_exp/i", $tag)) {
1294 $match = true;
1295 break;
1296 }
1297 }
1298 break;
1299 }
1300
1301 if ($rule_inverse) $match = !$match;
1302
1303 if ($match_any_rule) {
1304 if ($match) {
1305 $filter_match = true;
1306 break;
1307 }
1308 } else {
1309 $filter_match = $match;
1310 if (!$match) {
1311 break;
1312 }
1313 }
1314 }
1315
1316 if ($inverse) $filter_match = !$filter_match;
1317
1318 if ($filter_match) {
1319 foreach ($filter["actions"] AS $action) {
1320 array_push($matches, $action);
1321
1322 // if Stop action encountered, perform no further processing
1323 if ($action["type"] == "stop") return $matches;
1324 }
1325 }
1326 }
1327
1328 return $matches;
1329 }
1330
1331 function find_article_filter($filters, $filter_name) {
1332 foreach ($filters as $f) {
1333 if ($f["type"] == $filter_name) {
1334 return $f;
1335 };
1336 }
1337 return false;
1338 }
1339
1340 function find_article_filters($filters, $filter_name) {
1341 $results = array();
1342
1343 foreach ($filters as $f) {
1344 if ($f["type"] == $filter_name) {
1345 array_push($results, $f);
1346 };
1347 }
1348 return $results;
1349 }
1350
1351 function calculate_article_score($filters) {
1352 $score = 0;
1353
1354 foreach ($filters as $f) {
1355 if ($f["type"] == "score") {
1356 $score += $f["param"];
1357 };
1358 }
1359 return $score;
1360 }
1361
1362 function labels_contains_caption($labels, $caption) {
1363 foreach ($labels as $label) {
1364 if ($label[1] == $caption) {
1365 return true;
1366 }
1367 }
1368
1369 return false;
1370 }
1371
1372 function assign_article_to_label_filters($id, $filters, $owner_uid, $article_labels) {
1373 foreach ($filters as $f) {
1374 if ($f["type"] == "label") {
1375 if (!labels_contains_caption($article_labels, $f["param"])) {
1376 label_add_article($id, $f["param"], $owner_uid);
1377 }
1378 }
1379 }
1380 }
1381
1382 function make_guid_from_title($title) {
1383 return preg_replace("/[ \"\',.:;]/", "-",
1384 mb_strtolower(strip_tags($title), 'utf-8'));
1385 }
1386
1387 /* function verify_feed_xml($feed_data) {
1388 libxml_use_internal_errors(true);
1389 $doc = new DOMDocument();
1390 $doc->loadXML($feed_data);
1391 $error = libxml_get_last_error();
1392 libxml_clear_errors();
1393 return $error;
1394 } */
1395
1396 function housekeeping_common($debug) {
1397 expire_cached_files($debug);
1398 expire_lock_files($debug);
1399 expire_error_log($debug);
1400
1401 $count = update_feedbrowser_cache();
1402 _debug("Feedbrowser updated, $count feeds processed.");
1403
1404 purge_orphans( true);
1405 $rc = cleanup_tags( 14, 50000);
1406
1407 _debug("Cleaned $rc cached tags.");
1408
1409 PluginHost::getInstance()->run_hooks(PluginHost::HOOK_HOUSE_KEEPING, "hook_house_keeping", "");
1410
1411 }
1412 ?>
web-based news feed (RSS/Atom) reader and aggregator; see https://tt-rss.org/