]> 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
define define_default() in installer
[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 exceded).
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 OR last_updated = '1970-01-01 00:00:00')";
93 } else {
94 $update_limit_qpart = "AND ((
95 ttrss_feeds.update_interval = 0
96 AND ttrss_user_prefs.value != '-1'
97 AND ttrss_feeds.last_updated < DATE_SUB(NOW(), INTERVAL CONVERT(ttrss_user_prefs.value, SIGNED INTEGER) MINUTE)
98 ) OR (
99 ttrss_feeds.update_interval > 0
100 AND ttrss_feeds.last_updated < DATE_SUB(NOW(), INTERVAL ttrss_feeds.update_interval MINUTE)
101 ) OR ttrss_feeds.last_updated IS NULL
102 OR last_updated = '1970-01-01 00:00:00')";
103 }
104
105 // Test if feed is currently being updated by another process.
106 if (DB_TYPE == "pgsql") {
107 $updstart_thresh_qpart = "AND (ttrss_feeds.last_update_started IS NULL OR ttrss_feeds.last_update_started < NOW() - INTERVAL '10 minutes')";
108 } else {
109 $updstart_thresh_qpart = "AND (ttrss_feeds.last_update_started IS NULL OR ttrss_feeds.last_update_started < DATE_SUB(NOW(), INTERVAL 10 MINUTE))";
110 }
111
112 // Test if there is a limit to number of updated feeds
113 $query_limit = "";
114 if($limit) $query_limit = sprintf("LIMIT %d", $limit);
115
116 $query = "SELECT DISTINCT ttrss_feeds.feed_url, ttrss_feeds.last_updated
117 FROM
118 ttrss_feeds, ttrss_users, ttrss_user_prefs
119 WHERE
120 ttrss_feeds.owner_uid = ttrss_users.id
121 AND ttrss_users.id = ttrss_user_prefs.owner_uid
122 AND ttrss_user_prefs.pref_name = 'DEFAULT_UPDATE_INTERVAL'
123 $login_thresh_qpart $update_limit_qpart
124 $updstart_thresh_qpart
125 ORDER BY last_updated $query_limit";
126
127 // We search for feed needing update.
128 $result = db_query($query);
129
130 if($debug) _debug(sprintf("Scheduled %d feeds to update...", db_num_rows($result)));
131
132 // Here is a little cache magic in order to minimize risk of double feed updates.
133 $feeds_to_update = array();
134 while ($line = db_fetch_assoc($result)) {
135 array_push($feeds_to_update, db_escape_string($line['feed_url']));
136 }
137
138 // We update the feed last update started date before anything else.
139 // There is no lag due to feed contents downloads
140 // It prevent an other process to update the same feed.
141
142 if(count($feeds_to_update) > 0) {
143 $feeds_quoted = array();
144
145 foreach ($feeds_to_update as $feed) {
146 array_push($feeds_quoted, "'" . db_escape_string($feed) . "'");
147 }
148
149 db_query(sprintf("UPDATE ttrss_feeds SET last_update_started = NOW()
150 WHERE feed_url IN (%s)", implode(',', $feeds_quoted)));
151 }
152
153 $nf = 0;
154
155 // For each feed, we call the feed update function.
156 foreach ($feeds_to_update as $feed) {
157 if($debug) _debug("Base feed: $feed");
158
159 //update_rss_feed($line["id"], true);
160
161 // since we have the data cached, we can deal with other feeds with the same url
162
163 $tmp_result = db_query("SELECT DISTINCT ttrss_feeds.id,last_updated,ttrss_feeds.owner_uid
164 FROM ttrss_feeds, ttrss_users, ttrss_user_prefs WHERE
165 ttrss_user_prefs.owner_uid = ttrss_feeds.owner_uid AND
166 ttrss_users.id = ttrss_user_prefs.owner_uid AND
167 ttrss_user_prefs.pref_name = 'DEFAULT_UPDATE_INTERVAL' AND
168 feed_url = '".db_escape_string($feed)."' AND
169 (ttrss_feeds.update_interval > 0 OR
170 ttrss_user_prefs.value != '-1')
171 $login_thresh_qpart
172 ORDER BY ttrss_feeds.id $query_limit");
173
174 if (db_num_rows($tmp_result) > 0) {
175 $rss = false;
176
177 while ($tline = db_fetch_assoc($tmp_result)) {
178 if($debug) _debug(" => " . $tline["last_updated"] . ", " . $tline["id"] . " " . $tline["owner_uid"]);
179 $rss = update_rss_feed($tline["id"], true, false, $rss);
180 _debug_suppress(false);
181 ++$nf;
182 }
183 }
184 }
185
186 require_once "digest.php";
187
188 // Send feed digests by email if needed.
189 send_headlines_digests($debug);
190
191 return $nf;
192
193 } // function update_daemon_common
194
195 // ignore_daemon is not used
196 function update_rss_feed($feed, $ignore_daemon = false, $no_cache = false, $rss = false) {
197
198 $debug_enabled = defined('DAEMON_EXTENDED_DEBUG') || $_REQUEST['xdebug'];
199
200 _debug_suppress(!$debug_enabled);
201 _debug("start", $debug_enabled);
202
203 $result = db_query("SELECT id,update_interval,auth_login,
204 feed_url,auth_pass,cache_images,last_updated,
205 mark_unread_on_update, owner_uid,
206 pubsub_state, auth_pass_encrypted,
207 (SELECT max(date_entered) FROM
208 ttrss_entries, ttrss_user_entries where ref_id = id AND feed_id = '$feed') AS last_article_timestamp
209 FROM ttrss_feeds WHERE id = '$feed'");
210
211 if (db_num_rows($result) == 0) {
212 _debug("feed $feed NOT FOUND/SKIPPED", $debug_enabled);
213 return false;
214 }
215
216 $last_updated = db_fetch_result($result, 0, "last_updated");
217 $last_article_timestamp = @strtotime(db_fetch_result($result, 0, "last_article_timestamp"));
218
219 if (defined('_DISABLE_HTTP_304'))
220 $last_article_timestamp = 0;
221
222 $owner_uid = db_fetch_result($result, 0, "owner_uid");
223 $mark_unread_on_update = sql_bool_to_bool(db_fetch_result($result,
224 0, "mark_unread_on_update"));
225 $pubsub_state = db_fetch_result($result, 0, "pubsub_state");
226 $auth_pass_encrypted = sql_bool_to_bool(db_fetch_result($result,
227 0, "auth_pass_encrypted"));
228
229 db_query("UPDATE ttrss_feeds SET last_update_started = NOW()
230 WHERE id = '$feed'");
231
232 $auth_login = db_fetch_result($result, 0, "auth_login");
233 $auth_pass = db_fetch_result($result, 0, "auth_pass");
234
235 if ($auth_pass_encrypted) {
236 require_once "crypt.php";
237 $auth_pass = decrypt_string($auth_pass);
238 }
239
240 $cache_images = sql_bool_to_bool(db_fetch_result($result, 0, "cache_images"));
241 $fetch_url = db_fetch_result($result, 0, "feed_url");
242
243 $feed = db_escape_string($feed);
244
245 $date_feed_processed = date('Y-m-d H:i');
246
247 $cache_filename = CACHE_DIR . "/simplepie/" . sha1($fetch_url) . ".xml";
248
249 $pluginhost = new PluginHost();
250 $pluginhost->set_debug($debug_enabled);
251 $user_plugins = get_pref("_ENABLED_PLUGINS", $owner_uid);
252
253 $pluginhost->load(PLUGINS, PluginHost::KIND_ALL);
254 $pluginhost->load($user_plugins, PluginHost::KIND_USER, $owner_uid);
255 $pluginhost->load_data();
256
257 if ($rss && is_object($rss) && get_class($rss) == "FeedParser") {
258 _debug("using previously initialized parser object");
259 } else {
260 $rss_hash = false;
261
262 $force_refetch = isset($_REQUEST["force_refetch"]);
263
264 if (file_exists($cache_filename) &&
265 is_readable($cache_filename) &&
266 !$auth_login && !$auth_pass &&
267 filemtime($cache_filename) > time() - 30) {
268
269 _debug("using local cache.", $debug_enabled);
270
271 @$feed_data = file_get_contents($cache_filename);
272
273 if ($feed_data) {
274 $rss_hash = sha1($feed_data);
275 }
276
277 } else {
278 _debug("local cache will not be used for this feed", $debug_enabled);
279 }
280 }
281
282 if (!$rss) {
283
284 foreach ($pluginhost->get_hooks(PluginHost::HOOK_FETCH_FEED) as $plugin) {
285 $feed_data = $plugin->hook_fetch_feed($feed_data, $fetch_url, $owner_uid, $feed);
286 }
287
288 if (!$feed_data) {
289 _debug("fetching [$fetch_url]...", $debug_enabled);
290 _debug("If-Modified-Since: ".gmdate('D, d M Y H:i:s \G\M\T', $last_article_timestamp), $debug_enabled);
291
292 $feed_data = fetch_file_contents($fetch_url, false,
293 $auth_login, $auth_pass, false,
294 $no_cache ? FEED_FETCH_NO_CACHE_TIMEOUT : FEED_FETCH_TIMEOUT,
295 $force_refetch ? 0 : $last_article_timestamp);
296
297 global $fetch_curl_used;
298
299 if (!$fetch_curl_used) {
300 $tmp = @gzdecode($feed_data);
301
302 if ($tmp) $feed_data = $tmp;
303 }
304
305 $feed_data = trim($feed_data);
306
307 _debug("fetch done.", $debug_enabled);
308
309 /* if ($feed_data) {
310 $error = verify_feed_xml($feed_data);
311
312 if ($error) {
313 _debug("error verifying XML, code: " . $error->code, $debug_enabled);
314
315 if ($error->code == 26) {
316 _debug("got error 26, trying to decode entities...", $debug_enabled);
317
318 $feed_data = html_entity_decode($feed_data, ENT_COMPAT, 'UTF-8');
319
320 $error = verify_feed_xml($feed_data);
321
322 if ($error) $feed_data = '';
323 }
324 }
325 } */
326 }
327
328 if (!$feed_data) {
329 global $fetch_last_error;
330 global $fetch_last_error_code;
331
332 _debug("unable to fetch: $fetch_last_error [$fetch_last_error_code]", $debug_enabled);
333
334 $error_escaped = '';
335
336 // If-Modified-Since
337 if ($fetch_last_error_code != 304) {
338 $error_escaped = db_escape_string($fetch_last_error);
339 } else {
340 _debug("source claims data not modified, nothing to do.", $debug_enabled);
341 }
342
343 db_query(
344 "UPDATE ttrss_feeds SET last_error = '$error_escaped',
345 last_updated = NOW() WHERE id = '$feed'");
346
347 return;
348 }
349 }
350
351 foreach ($pluginhost->get_hooks(PluginHost::HOOK_FEED_FETCHED) as $plugin) {
352 $feed_data = $plugin->hook_feed_fetched($feed_data, $fetch_url, $owner_uid, $feed);
353 }
354
355 // set last update to now so if anything *simplepie* crashes later we won't be
356 // continuously failing on the same feed
357 //db_query("UPDATE ttrss_feeds SET last_updated = NOW() WHERE id = '$feed'");
358
359 if (!$rss) {
360 $rss = new FeedParser($feed_data);
361 $rss->init();
362 }
363
364 if (DETECT_ARTICLE_LANGUAGE) {
365 require_once "lib/languagedetect/LanguageDetect.php";
366
367 $lang = new Text_LanguageDetect();
368 $lang->setNameMode(2);
369 }
370
371 // print_r($rss);
372
373 $feed = db_escape_string($feed);
374
375 if (!$rss->error()) {
376
377 // cache data for later
378 if (!$auth_pass && !$auth_login && is_writable(CACHE_DIR . "/simplepie")) {
379 $new_rss_hash = sha1($rss_data);
380
381 if ($new_rss_hash != $rss_hash && count($rss->get_items()) > 0 ) {
382 _debug("saving $cache_filename", $debug_enabled);
383 @file_put_contents($cache_filename, $feed_data);
384 }
385 }
386
387 // We use local pluginhost here because we need to load different per-user feed plugins
388 $pluginhost->run_hooks(PluginHost::HOOK_FEED_PARSED, "hook_feed_parsed", $rss);
389
390 _debug("processing feed data...", $debug_enabled);
391
392 // db_query("BEGIN");
393
394 if (DB_TYPE == "pgsql") {
395 $favicon_interval_qpart = "favicon_last_checked < NOW() - INTERVAL '12 hour'";
396 } else {
397 $favicon_interval_qpart = "favicon_last_checked < DATE_SUB(NOW(), INTERVAL 12 HOUR)";
398 }
399
400 $result = db_query("SELECT title,site_url,owner_uid,favicon_avg_color,
401 (favicon_last_checked IS NULL OR $favicon_interval_qpart) AS
402 favicon_needs_check
403 FROM ttrss_feeds WHERE id = '$feed'");
404
405 $registered_title = db_fetch_result($result, 0, "title");
406 $orig_site_url = db_fetch_result($result, 0, "site_url");
407 $favicon_needs_check = sql_bool_to_bool(db_fetch_result($result, 0,
408 "favicon_needs_check"));
409 $favicon_avg_color = db_fetch_result($result, 0, "favicon_avg_color");
410
411 $owner_uid = db_fetch_result($result, 0, "owner_uid");
412
413 $site_url = db_escape_string(mb_substr(rewrite_relative_url($fetch_url, $rss->get_link()), 0, 245));
414
415 _debug("site_url: $site_url", $debug_enabled);
416 _debug("feed_title: " . $rss->get_title(), $debug_enabled);
417
418 if ($favicon_needs_check || $force_refetch) {
419
420 /* terrible hack: if we crash on floicon shit here, we won't check
421 * the icon avgcolor again (unless the icon got updated) */
422
423 $favicon_file = ICONS_DIR . "/$feed.ico";
424 $favicon_modified = @filemtime($favicon_file);
425
426 _debug("checking favicon...", $debug_enabled);
427
428 check_feed_favicon($site_url, $feed);
429 $favicon_modified_new = @filemtime($favicon_file);
430
431 if ($favicon_modified_new > $favicon_modified)
432 $favicon_avg_color = '';
433
434 if (file_exists($favicon_file) && function_exists("imagecreatefromstring") && $favicon_avg_color == '') {
435 require_once "colors.php";
436
437 db_query("UPDATE ttrss_feeds SET favicon_avg_color = 'fail' WHERE
438 id = '$feed'");
439
440 $favicon_color = db_escape_string(
441 calculate_avg_color($favicon_file));
442
443 $favicon_colorstring = ",favicon_avg_color = '".$favicon_color."'";
444 } else if ($favicon_avg_color == 'fail') {
445 _debug("floicon failed on this file, not trying to recalculate avg color", $debug_enabled);
446 }
447
448 db_query("UPDATE ttrss_feeds SET favicon_last_checked = NOW()
449 $favicon_colorstring
450 WHERE id = '$feed'");
451 }
452
453 if (!$registered_title || $registered_title == "[Unknown]") {
454
455 $feed_title = db_escape_string($rss->get_title());
456
457 if ($feed_title) {
458 _debug("registering title: $feed_title", $debug_enabled);
459
460 db_query("UPDATE ttrss_feeds SET
461 title = '$feed_title' WHERE id = '$feed'");
462 }
463 }
464
465 if ($site_url && $orig_site_url != $site_url) {
466 db_query("UPDATE ttrss_feeds SET
467 site_url = '$site_url' WHERE id = '$feed'");
468 }
469
470 _debug("loading filters & labels...", $debug_enabled);
471
472 $filters = load_filters($feed, $owner_uid);
473 $labels = get_all_labels($owner_uid);
474
475 _debug("" . count($filters) . " filters loaded.", $debug_enabled);
476
477 $items = $rss->get_items();
478
479 if (!is_array($items)) {
480 _debug("no articles found.", $debug_enabled);
481
482 db_query("UPDATE ttrss_feeds
483 SET last_updated = NOW(), last_error = '' WHERE id = '$feed'");
484
485 return; // no articles
486 }
487
488 if ($pubsub_state != 2 && PUBSUBHUBBUB_ENABLED) {
489
490 _debug("checking for PUSH hub...", $debug_enabled);
491
492 $feed_hub_url = false;
493
494 $links = $rss->get_links('hub');
495
496 if ($links && is_array($links)) {
497 foreach ($links as $l) {
498 $feed_hub_url = $l;
499 break;
500 }
501 }
502
503 _debug("feed hub url: $feed_hub_url", $debug_enabled);
504
505 $feed_self_url = $fetch_url;
506
507 $links = $rss->get_links('self');
508
509 if ($links && is_array($links)) {
510 foreach ($links as $l) {
511 $feed_self_url = $l;
512 break;
513 }
514 }
515
516 _debug("feed self url = $feed_self_url");
517
518 if ($feed_hub_url && $feed_self_url && function_exists('curl_init') &&
519 !ini_get("open_basedir")) {
520
521 require_once 'lib/pubsubhubbub/subscriber.php';
522
523 $callback_url = get_self_url_prefix() .
524 "/public.php?op=pubsub&id=$feed";
525
526 $s = new Subscriber($feed_hub_url, $callback_url);
527
528 $rc = $s->subscribe($feed_self_url);
529
530 _debug("feed hub url found, subscribe request sent.", $debug_enabled);
531
532 db_query("UPDATE ttrss_feeds SET pubsub_state = 1
533 WHERE id = '$feed'");
534 }
535 }
536
537 _debug("processing articles...", $debug_enabled);
538
539 foreach ($items as $item) {
540 if ($_REQUEST['xdebug'] == 3) {
541 print_r($item);
542 }
543
544 $entry_guid = $item->get_id();
545 if (!$entry_guid) $entry_guid = $item->get_link();
546 if (!$entry_guid) $entry_guid = make_guid_from_title($item->get_title());
547
548 _debug("f_guid $entry_guid", $debug_enabled);
549
550 if (!$entry_guid) continue;
551
552 $entry_guid = "$owner_uid,$entry_guid";
553
554 $entry_guid_hashed = db_escape_string('SHA1:' . sha1($entry_guid));
555
556 _debug("guid $entry_guid / $entry_guid_hashed", $debug_enabled);
557
558 $entry_timestamp = "";
559
560 $entry_timestamp = $item->get_date();
561
562 _debug("orig date: " . $item->get_date(), $debug_enabled);
563
564 if ($entry_timestamp == -1 || !$entry_timestamp || $entry_timestamp > time()) {
565 $entry_timestamp = time();
566 $no_orig_date = 'true';
567 } else {
568 $no_orig_date = 'false';
569 }
570
571 $entry_timestamp_fmt = strftime("%Y/%m/%d %H:%M:%S", $entry_timestamp);
572
573 _debug("date $entry_timestamp [$entry_timestamp_fmt]", $debug_enabled);
574
575 // $entry_title = html_entity_decode($item->get_title(), ENT_COMPAT, 'UTF-8');
576 // $entry_title = decode_numeric_entities($entry_title);
577 $entry_title = $item->get_title();
578
579 $entry_link = rewrite_relative_url($site_url, $item->get_link());
580
581 _debug("title $entry_title", $debug_enabled);
582 _debug("link $entry_link", $debug_enabled);
583
584 if (!$entry_title) $entry_title = date("Y-m-d H:i:s", $entry_timestamp);;
585
586 $entry_content = $item->get_content();
587 if (!$entry_content) $entry_content = $item->get_description();
588
589 if ($_REQUEST["xdebug"] == 2) {
590 print "content: ";
591 print $entry_content;
592 print "\n";
593 }
594
595 $entry_language = "";
596
597 if (DETECT_ARTICLE_LANGUAGE) {
598 $entry_language = $lang->detect($entry_title . " " . $entry_content, 1);
599
600 if (count($entry_language) > 0) {
601 $entry_language = array_keys($entry_language);
602
603 // the fuck?
604 if (is_array($entry_language))
605 $entry_language = "";
606 else
607 $entry_language = db_escape_string(substr($entry_language[0], 0, 2));
608
609 _debug("detected language: $entry_language", $debug_enabled);
610 }
611 }
612
613 $entry_comments = $item->get_comments_url();
614 $entry_author = $item->get_author();
615
616 $entry_guid = db_escape_string(mb_substr($entry_guid, 0, 245));
617
618 $entry_comments = db_escape_string(mb_substr(trim($entry_comments), 0, 245));
619 $entry_author = db_escape_string(mb_substr(trim($entry_author), 0, 245));
620
621 $num_comments = (int) $item->get_comments_count();
622
623 _debug("author $entry_author", $debug_enabled);
624 _debug("num_comments: $num_comments", $debug_enabled);
625 _debug("looking for tags...", $debug_enabled);
626
627 // parse <category> entries into tags
628
629 $additional_tags = array();
630
631 $additional_tags_src = $item->get_categories();
632
633 if (is_array($additional_tags_src)) {
634 foreach ($additional_tags_src as $tobj) {
635 array_push($additional_tags, $tobj);
636 }
637 }
638
639 $entry_tags = array_unique($additional_tags);
640
641 for ($i = 0; $i < count($entry_tags); $i++)
642 $entry_tags[$i] = mb_strtolower($entry_tags[$i], 'utf-8');
643
644 _debug("tags found: " . join(",", $entry_tags), $debug_enabled);
645
646 _debug("done collecting data.", $debug_enabled);
647
648 // TODO: less memory-hungry implementation
649
650 _debug("applying plugin filters..", $debug_enabled);
651
652 // FIXME not sure if owner_uid is a good idea here, we may have a base entry without user entry (?)
653 $result = db_query("SELECT plugin_data,title,content,link,tag_cache,author FROM ttrss_entries, ttrss_user_entries
654 WHERE ref_id = id AND (guid = '".db_escape_string($entry_guid)."' OR guid = '$entry_guid_hashed') AND owner_uid = $owner_uid");
655
656 if (db_num_rows($result) != 0) {
657 $entry_plugin_data = db_fetch_result($result, 0, "plugin_data");
658 $stored_article = array("title" => db_fetch_result($result, 0, "title"),
659 "content" => db_fetch_result($result, 0, "content"),
660 "link" => db_fetch_result($result, 0, "link"),
661 "tags" => explode(",", db_fetch_result($result, 0, "tag_cache")),
662 "author" => db_fetch_result($result, 0, "author"));
663 } else {
664 $entry_plugin_data = "";
665 $stored_article = array();
666 }
667
668 $article = array("owner_uid" => $owner_uid, // read only
669 "guid" => $entry_guid, // read only
670 "title" => $entry_title,
671 "content" => $entry_content,
672 "link" => $entry_link,
673 "tags" => $entry_tags,
674 "plugin_data" => $entry_plugin_data,
675 "author" => $entry_author,
676 "stored" => $stored_article,
677 "feed" => array("id" => $feed,
678 "fetch_url" => $fetch_url,
679 "site_url" => $site_url)
680 );
681
682 foreach ($pluginhost->get_hooks(PluginHost::HOOK_ARTICLE_FILTER) as $plugin) {
683 $article = $plugin->hook_article_filter($article);
684 }
685
686 $entry_tags = $article["tags"];
687 $entry_guid = db_escape_string($entry_guid);
688 $entry_title = db_escape_string($article["title"]);
689 $entry_author = db_escape_string($article["author"]);
690 $entry_link = db_escape_string($article["link"]);
691 $entry_plugin_data = db_escape_string($article["plugin_data"]);
692 $entry_content = $article["content"]; // escaped below
693
694
695 _debug("plugin data: $entry_plugin_data", $debug_enabled);
696
697 if ($cache_images && is_writable(CACHE_DIR . '/images'))
698 cache_images($entry_content, $site_url, $debug_enabled);
699
700 $entry_content = db_escape_string($entry_content, false);
701
702 $content_hash = "SHA1:" . sha1($entry_content);
703
704 db_query("BEGIN");
705
706 $result = db_query("SELECT id FROM ttrss_entries
707 WHERE (guid = '$entry_guid' OR guid = '$entry_guid_hashed')");
708
709 if (db_num_rows($result) == 0) {
710
711 _debug("base guid [$entry_guid] not found", $debug_enabled);
712
713 // base post entry does not exist, create it
714
715 $result = db_query(
716 "INSERT INTO ttrss_entries
717 (title,
718 guid,
719 link,
720 updated,
721 content,
722 content_hash,
723 no_orig_date,
724 date_updated,
725 date_entered,
726 comments,
727 num_comments,
728 plugin_data,
729 lang,
730 author)
731 VALUES
732 ('$entry_title',
733 '$entry_guid_hashed',
734 '$entry_link',
735 '$entry_timestamp_fmt',
736 '$entry_content',
737 '$content_hash',
738 $no_orig_date,
739 NOW(),
740 '$date_feed_processed',
741 '$entry_comments',
742 '$num_comments',
743 '$entry_plugin_data',
744 '$entry_language',
745 '$entry_author')");
746
747 $article_labels = array();
748
749 } else {
750 // we keep encountering the entry in feeds, so we need to
751 // update date_updated column so that we don't get horrible
752 // dupes when the entry gets purged and reinserted again e.g.
753 // in the case of SLOW SLOW OMG SLOW updating feeds
754
755 $base_entry_id = db_fetch_result($result, 0, "id");
756
757 db_query("UPDATE ttrss_entries SET date_updated = NOW()
758 WHERE id = '$base_entry_id'");
759
760 $article_labels = get_article_labels($base_entry_id, $owner_uid);
761 }
762
763 // now it should exist, if not - bad luck then
764
765 $result = db_query("SELECT
766 id,content_hash,no_orig_date,title,plugin_data,guid,
767 ".SUBSTRING_FOR_DATE."(date_updated,1,19) as date_updated,
768 ".SUBSTRING_FOR_DATE."(updated,1,19) as updated,
769 num_comments
770 FROM
771 ttrss_entries
772 WHERE guid = '$entry_guid' OR guid = '$entry_guid_hashed'");
773
774 $entry_ref_id = 0;
775 $entry_int_id = 0;
776
777 if (db_num_rows($result) == 1) {
778
779 _debug("base guid found, checking for user record", $debug_enabled);
780
781 // this will be used below in update handler
782 $orig_content_hash = db_fetch_result($result, 0, "content_hash");
783 $orig_title = db_fetch_result($result, 0, "title");
784 $orig_num_comments = db_fetch_result($result, 0, "num_comments");
785 $orig_date_updated = strtotime(db_fetch_result($result,
786 0, "date_updated"));
787 $orig_plugin_data = db_fetch_result($result, 0, "plugin_data");
788
789 $ref_id = db_fetch_result($result, 0, "id");
790 $entry_ref_id = $ref_id;
791
792 /* $stored_guid = db_fetch_result($result, 0, "guid");
793 if ($stored_guid != $entry_guid_hashed) {
794 if ($debug_enabled) _debug("upgrading compat guid to hashed one", $debug_enabled);
795
796 db_query("UPDATE ttrss_entries SET guid = '$entry_guid_hashed' WHERE
797 id = '$ref_id'");
798 } */
799
800 // check for user post link to main table
801
802 // do we allow duplicate posts with same GUID in different feeds?
803 if (get_pref("ALLOW_DUPLICATE_POSTS", $owner_uid, false)) {
804 $dupcheck_qpart = "AND (feed_id = '$feed' OR feed_id IS NULL)";
805 } else {
806 $dupcheck_qpart = "";
807 }
808
809 /* Collect article tags here so we could filter by them: */
810
811 $article_filters = get_article_filters($filters, $entry_title,
812 $entry_content, $entry_link, $entry_timestamp, $entry_author,
813 $entry_tags);
814
815 if ($debug_enabled) {
816 _debug("article filters: ", $debug_enabled);
817 if (count($article_filters) != 0) {
818 print_r($article_filters);
819 }
820 }
821
822 if (find_article_filter($article_filters, "filter")) {
823 db_query("COMMIT"); // close transaction in progress
824 continue;
825 }
826
827 $score = calculate_article_score($article_filters);
828
829 _debug("initial score: $score", $debug_enabled);
830
831 $query = "SELECT ref_id, int_id FROM ttrss_user_entries WHERE
832 ref_id = '$ref_id' AND owner_uid = '$owner_uid'
833 $dupcheck_qpart";
834
835 // if ($_REQUEST["xdebug"]) print "$query\n";
836
837 $result = db_query($query);
838
839 // okay it doesn't exist - create user entry
840 if (db_num_rows($result) == 0) {
841
842 _debug("user record not found, creating...", $debug_enabled);
843
844 if ($score >= -500 && !find_article_filter($article_filters, 'catchup')) {
845 $unread = 'true';
846 $last_read_qpart = 'NULL';
847 } else {
848 $unread = 'false';
849 $last_read_qpart = 'NOW()';
850 }
851
852 if (find_article_filter($article_filters, 'mark') || $score > 1000) {
853 $marked = 'true';
854 } else {
855 $marked = 'false';
856 }
857
858 if (find_article_filter($article_filters, 'publish')) {
859 $published = 'true';
860 } else {
861 $published = 'false';
862 }
863
864 // N-grams
865
866 if (DB_TYPE == "pgsql" and defined('_NGRAM_TITLE_DUPLICATE_THRESHOLD')) {
867
868 $result = db_query("SELECT COUNT(*) AS similar FROM
869 ttrss_entries,ttrss_user_entries
870 WHERE ref_id = id AND updated >= NOW() - INTERVAL '7 day'
871 AND similarity(title, '$entry_title') >= "._NGRAM_TITLE_DUPLICATE_THRESHOLD."
872 AND owner_uid = $owner_uid");
873
874 $ngram_similar = db_fetch_result($result, 0, "similar");
875
876 _debug("N-gram similar results: $ngram_similar", $debug_enabled);
877
878 if ($ngram_similar > 0) {
879 $unread = 'false';
880 }
881 }
882
883 $last_marked = ($marked == 'true') ? 'NOW()' : 'NULL';
884 $last_published = ($published == 'true') ? 'NOW()' : 'NULL';
885
886 $result = db_query(
887 "INSERT INTO ttrss_user_entries
888 (ref_id, owner_uid, feed_id, unread, last_read, marked,
889 published, score, tag_cache, label_cache, uuid,
890 last_marked, last_published)
891 VALUES ('$ref_id', '$owner_uid', '$feed', $unread,
892 $last_read_qpart, $marked, $published, '$score', '', '',
893 '', $last_marked, $last_published)");
894
895 if (PUBSUBHUBBUB_HUB && $published == 'true') {
896 $rss_link = get_self_url_prefix() .
897 "/public.php?op=rss&id=-2&key=" .
898 get_feed_access_key(-2, false, $owner_uid);
899
900 $p = new Publisher(PUBSUBHUBBUB_HUB);
901
902 $pubsub_result = $p->publish_update($rss_link);
903 }
904
905 $result = db_query(
906 "SELECT int_id FROM ttrss_user_entries WHERE
907 ref_id = '$ref_id' AND owner_uid = '$owner_uid' AND
908 feed_id = '$feed' LIMIT 1");
909
910 if (db_num_rows($result) == 1) {
911 $entry_int_id = db_fetch_result($result, 0, "int_id");
912 }
913 } else {
914 _debug("user record FOUND", $debug_enabled);
915
916 $entry_ref_id = db_fetch_result($result, 0, "ref_id");
917 $entry_int_id = db_fetch_result($result, 0, "int_id");
918 }
919
920 _debug("RID: $entry_ref_id, IID: $entry_int_id", $debug_enabled);
921
922 $post_needs_update = false;
923 $update_insignificant = false;
924
925 if ($orig_num_comments != $num_comments) {
926 $post_needs_update = true;
927 $update_insignificant = true;
928 }
929
930 if ($entry_plugin_data != $orig_plugin_data) {
931 $post_needs_update = true;
932 $update_insignificant = true;
933 }
934
935 if ($content_hash != $orig_content_hash) {
936 $post_needs_update = true;
937 $update_insignificant = false;
938 }
939
940 if (db_escape_string($orig_title) != $entry_title) {
941 $post_needs_update = true;
942 $update_insignificant = false;
943 }
944
945 // if post needs update, update it and mark all user entries
946 // linking to this post as updated
947 if ($post_needs_update) {
948
949 if (defined('DAEMON_EXTENDED_DEBUG')) {
950 _debug("post $entry_guid_hashed needs update...", $debug_enabled);
951 }
952
953 // print "<!-- post $orig_title needs update : $post_needs_update -->";
954
955 db_query("UPDATE ttrss_entries
956 SET title = '$entry_title', content = '$entry_content',
957 content_hash = '$content_hash',
958 updated = '$entry_timestamp_fmt',
959 num_comments = '$num_comments',
960 plugin_data = '$entry_plugin_data'
961 WHERE id = '$ref_id'");
962
963 if (!$update_insignificant) {
964 if ($mark_unread_on_update) {
965 db_query("UPDATE ttrss_user_entries
966 SET last_read = null, unread = true WHERE ref_id = '$ref_id'");
967 }
968 }
969 }
970 }
971
972 db_query("COMMIT");
973
974 _debug("assigning labels...", $debug_enabled);
975
976 assign_article_to_label_filters($entry_ref_id, $article_filters,
977 $owner_uid, $article_labels);
978
979 _debug("looking for enclosures...", $debug_enabled);
980
981 // enclosures
982
983 $enclosures = array();
984
985 $encs = $item->get_enclosures();
986
987 if (is_array($encs)) {
988 foreach ($encs as $e) {
989 $e_item = array(
990 $e->link, $e->type, $e->length, $e->title);
991 array_push($enclosures, $e_item);
992 }
993 }
994
995 if ($debug_enabled) {
996 _debug("article enclosures:", $debug_enabled);
997 print_r($enclosures);
998 }
999
1000 db_query("BEGIN");
1001
1002 // debugging
1003 // db_query("DELETE FROM ttrss_enclosures WHERE post_id = '$entry_ref_id'");
1004
1005 foreach ($enclosures as $enc) {
1006 $enc_url = db_escape_string($enc[0]);
1007 $enc_type = db_escape_string($enc[1]);
1008 $enc_dur = db_escape_string($enc[2]);
1009 $enc_title = db_escape_string($enc[3]);
1010
1011 $result = db_query("SELECT id FROM ttrss_enclosures
1012 WHERE content_url = '$enc_url' AND post_id = '$entry_ref_id'");
1013
1014 if (db_num_rows($result) == 0) {
1015 db_query("INSERT INTO ttrss_enclosures
1016 (content_url, content_type, title, duration, post_id) VALUES
1017 ('$enc_url', '$enc_type', '$enc_title', '$enc_dur', '$entry_ref_id')");
1018 }
1019 }
1020
1021 db_query("COMMIT");
1022
1023 // check for manual tags (we have to do it here since they're loaded from filters)
1024
1025 foreach ($article_filters as $f) {
1026 if ($f["type"] == "tag") {
1027
1028 $manual_tags = trim_array(explode(",", $f["param"]));
1029
1030 foreach ($manual_tags as $tag) {
1031 if (tag_is_valid($tag)) {
1032 array_push($entry_tags, $tag);
1033 }
1034 }
1035 }
1036 }
1037
1038 // Skip boring tags
1039
1040 $boring_tags = trim_array(explode(",", mb_strtolower(get_pref(
1041 'BLACKLISTED_TAGS', $owner_uid, ''), 'utf-8')));
1042
1043 $filtered_tags = array();
1044 $tags_to_cache = array();
1045
1046 if ($entry_tags && is_array($entry_tags)) {
1047 foreach ($entry_tags as $tag) {
1048 if (array_search($tag, $boring_tags) === false) {
1049 array_push($filtered_tags, $tag);
1050 }
1051 }
1052 }
1053
1054 $filtered_tags = array_unique($filtered_tags);
1055
1056 if ($debug_enabled) {
1057 _debug("filtered article tags:", $debug_enabled);
1058 print_r($filtered_tags);
1059 }
1060
1061 // Save article tags in the database
1062
1063 if (count($filtered_tags) > 0) {
1064
1065 db_query("BEGIN");
1066
1067 foreach ($filtered_tags as $tag) {
1068
1069 $tag = sanitize_tag($tag);
1070 $tag = db_escape_string($tag);
1071
1072 if (!tag_is_valid($tag)) continue;
1073
1074 $result = db_query("SELECT id FROM ttrss_tags
1075 WHERE tag_name = '$tag' AND post_int_id = '$entry_int_id' AND
1076 owner_uid = '$owner_uid' LIMIT 1");
1077
1078 if ($result && db_num_rows($result) == 0) {
1079
1080 db_query("INSERT INTO ttrss_tags
1081 (owner_uid,tag_name,post_int_id)
1082 VALUES ('$owner_uid','$tag', '$entry_int_id')");
1083 }
1084
1085 array_push($tags_to_cache, $tag);
1086 }
1087
1088 /* update the cache */
1089
1090 $tags_to_cache = array_unique($tags_to_cache);
1091
1092 $tags_str = db_escape_string(join(",", $tags_to_cache));
1093
1094 db_query("UPDATE ttrss_user_entries
1095 SET tag_cache = '$tags_str' WHERE ref_id = '$entry_ref_id'
1096 AND owner_uid = $owner_uid");
1097
1098 db_query("COMMIT");
1099 }
1100
1101 if (get_pref("AUTO_ASSIGN_LABELS", $owner_uid, false)) {
1102 _debug("auto-assigning labels...", $debug_enabled);
1103
1104 foreach ($labels as $label) {
1105 $caption = preg_quote($label["caption"]);
1106
1107 if ($caption && preg_match("/\b$caption\b/i", "$tags_str " . strip_tags($entry_content) . " $entry_title")) {
1108 if (!labels_contains_caption($article_labels, $caption)) {
1109 label_add_article($entry_ref_id, $caption, $owner_uid);
1110 }
1111 }
1112 }
1113 }
1114
1115 _debug("article processed", $debug_enabled);
1116 }
1117
1118 _debug("purging feed...", $debug_enabled);
1119
1120 purge_feed($feed, 0, $debug_enabled);
1121
1122 db_query("UPDATE ttrss_feeds
1123 SET last_updated = NOW(), last_error = '' WHERE id = '$feed'");
1124
1125 // db_query("COMMIT");
1126
1127 } else {
1128
1129 $error_msg = db_escape_string(mb_substr($rss->error(), 0, 245));
1130
1131 _debug("fetch error: $error_msg", $debug_enabled);
1132
1133 if (count($rss->errors()) > 1) {
1134 foreach ($rss->errors() as $error) {
1135 _debug("+ $error");
1136 }
1137 }
1138
1139 db_query(
1140 "UPDATE ttrss_feeds SET last_error = '$error_msg',
1141 last_updated = NOW() WHERE id = '$feed'");
1142
1143 unset($rss);
1144 }
1145
1146 _debug("done", $debug_enabled);
1147
1148 return $rss;
1149 }
1150
1151 function cache_images($html, $site_url, $debug) {
1152 $cache_dir = CACHE_DIR . "/images";
1153
1154 libxml_use_internal_errors(true);
1155
1156 $charset_hack = '<head>
1157 <meta http-equiv="Content-Type" content="text/html; charset=utf-8"/>
1158 </head>';
1159
1160 $doc = new DOMDocument();
1161 $doc->loadHTML($charset_hack . $html);
1162 $xpath = new DOMXPath($doc);
1163
1164 $entries = $xpath->query('(//img[@src])');
1165
1166 foreach ($entries as $entry) {
1167 if ($entry->hasAttribute('src')) {
1168 $src = rewrite_relative_url($site_url, $entry->getAttribute('src'));
1169
1170 $local_filename = CACHE_DIR . "/images/" . sha1($src) . ".png";
1171
1172 if ($debug) _debug("cache_images: downloading: $src to $local_filename");
1173
1174 if (!file_exists($local_filename)) {
1175 $file_content = fetch_file_contents($src);
1176
1177 if ($file_content && strlen($file_content) > 1024) {
1178 file_put_contents($local_filename, $file_content);
1179 }
1180 }
1181
1182 /* if (file_exists($local_filename)) {
1183 $entry->setAttribute('src', SELF_URL_PATH . '/image.php?url=' .
1184 base64_encode($src));
1185 } */
1186 }
1187 }
1188
1189 //$node = $doc->getElementsByTagName('body')->item(0);
1190 //return $doc->saveXML($node);
1191 }
1192
1193 function expire_error_log($debug) {
1194 if ($debug) _debug("Removing old error log entries...");
1195
1196 if (DB_TYPE == "pgsql") {
1197 db_query("DELETE FROM ttrss_error_log
1198 WHERE created_at < NOW() - INTERVAL '7 days'");
1199 } else {
1200 db_query("DELETE FROM ttrss_error_log
1201 WHERE created_at < DATE_SUB(NOW(), INTERVAL 7 DAY)");
1202 }
1203
1204 }
1205
1206 function expire_lock_files($debug) {
1207 //if ($debug) _debug("Removing old lock files...");
1208
1209 $num_deleted = 0;
1210
1211 if (is_writable(LOCK_DIRECTORY)) {
1212 $files = glob(LOCK_DIRECTORY . "/*.lock");
1213
1214 if ($files) {
1215 foreach ($files as $file) {
1216 if (!file_is_locked(basename($file)) && time() - filemtime($file) > 86400*2) {
1217 unlink($file);
1218 ++$num_deleted;
1219 }
1220 }
1221 }
1222 }
1223
1224 if ($debug) _debug("Removed $num_deleted old lock files.");
1225 }
1226
1227 function expire_cached_files($debug) {
1228 foreach (array("simplepie", "images", "export", "upload") as $dir) {
1229 $cache_dir = CACHE_DIR . "/$dir";
1230
1231 // if ($debug) _debug("Expiring $cache_dir");
1232
1233 $num_deleted = 0;
1234
1235 if (is_writable($cache_dir)) {
1236 $files = glob("$cache_dir/*");
1237
1238 if ($files) {
1239 foreach ($files as $file) {
1240 if (time() - filemtime($file) > 86400*7) {
1241 unlink($file);
1242
1243 ++$num_deleted;
1244 }
1245 }
1246 }
1247 }
1248
1249 if ($debug) _debug("$cache_dir: removed $num_deleted files.");
1250 }
1251 }
1252
1253 /**
1254 * Source: http://www.php.net/manual/en/function.parse-url.php#104527
1255 * Returns the url query as associative array
1256 *
1257 * @param string query
1258 * @return array params
1259 */
1260 function convertUrlQuery($query) {
1261 $queryParts = explode('&', $query);
1262
1263 $params = array();
1264
1265 foreach ($queryParts as $param) {
1266 $item = explode('=', $param);
1267 $params[$item[0]] = $item[1];
1268 }
1269
1270 return $params;
1271 }
1272
1273 function get_article_filters($filters, $title, $content, $link, $timestamp, $author, $tags) {
1274 $matches = array();
1275
1276 foreach ($filters as $filter) {
1277 $match_any_rule = $filter["match_any_rule"];
1278 $inverse = $filter["inverse"];
1279 $filter_match = false;
1280
1281 foreach ($filter["rules"] as $rule) {
1282 $match = false;
1283 $reg_exp = str_replace('/', '\/', $rule["reg_exp"]);
1284 $rule_inverse = $rule["inverse"];
1285
1286 if (!$reg_exp)
1287 continue;
1288
1289 switch ($rule["type"]) {
1290 case "title":
1291 $match = @preg_match("/$reg_exp/i", $title);
1292 break;
1293 case "content":
1294 // we don't need to deal with multiline regexps
1295 $content = preg_replace("/[\r\n\t]/", "", $content);
1296
1297 $match = @preg_match("/$reg_exp/i", $content);
1298 break;
1299 case "both":
1300 // we don't need to deal with multiline regexps
1301 $content = preg_replace("/[\r\n\t]/", "", $content);
1302
1303 $match = (@preg_match("/$reg_exp/i", $title) || @preg_match("/$reg_exp/i", $content));
1304 break;
1305 case "link":
1306 $match = @preg_match("/$reg_exp/i", $link);
1307 break;
1308 case "author":
1309 $match = @preg_match("/$reg_exp/i", $author);
1310 break;
1311 case "tag":
1312 foreach ($tags as $tag) {
1313 if (@preg_match("/$reg_exp/i", $tag)) {
1314 $match = true;
1315 break;
1316 }
1317 }
1318 break;
1319 }
1320
1321 if ($rule_inverse) $match = !$match;
1322
1323 if ($match_any_rule) {
1324 if ($match) {
1325 $filter_match = true;
1326 break;
1327 }
1328 } else {
1329 $filter_match = $match;
1330 if (!$match) {
1331 break;
1332 }
1333 }
1334 }
1335
1336 if ($inverse) $filter_match = !$filter_match;
1337
1338 if ($filter_match) {
1339 foreach ($filter["actions"] AS $action) {
1340 array_push($matches, $action);
1341
1342 // if Stop action encountered, perform no further processing
1343 if ($action["type"] == "stop") return $matches;
1344 }
1345 }
1346 }
1347
1348 return $matches;
1349 }
1350
1351 function find_article_filter($filters, $filter_name) {
1352 foreach ($filters as $f) {
1353 if ($f["type"] == $filter_name) {
1354 return $f;
1355 };
1356 }
1357 return false;
1358 }
1359
1360 function find_article_filters($filters, $filter_name) {
1361 $results = array();
1362
1363 foreach ($filters as $f) {
1364 if ($f["type"] == $filter_name) {
1365 array_push($results, $f);
1366 };
1367 }
1368 return $results;
1369 }
1370
1371 function calculate_article_score($filters) {
1372 $score = 0;
1373
1374 foreach ($filters as $f) {
1375 if ($f["type"] == "score") {
1376 $score += $f["param"];
1377 };
1378 }
1379 return $score;
1380 }
1381
1382 function labels_contains_caption($labels, $caption) {
1383 foreach ($labels as $label) {
1384 if ($label[1] == $caption) {
1385 return true;
1386 }
1387 }
1388
1389 return false;
1390 }
1391
1392 function assign_article_to_label_filters($id, $filters, $owner_uid, $article_labels) {
1393 foreach ($filters as $f) {
1394 if ($f["type"] == "label") {
1395 if (!labels_contains_caption($article_labels, $f["param"])) {
1396 label_add_article($id, $f["param"], $owner_uid);
1397 }
1398 }
1399 }
1400 }
1401
1402 function make_guid_from_title($title) {
1403 return preg_replace("/[ \"\',.:;]/", "-",
1404 mb_strtolower(strip_tags($title), 'utf-8'));
1405 }
1406
1407 /* function verify_feed_xml($feed_data) {
1408 libxml_use_internal_errors(true);
1409 $doc = new DOMDocument();
1410 $doc->loadXML($feed_data);
1411 $error = libxml_get_last_error();
1412 libxml_clear_errors();
1413 return $error;
1414 } */
1415
1416 function housekeeping_common($debug) {
1417 expire_cached_files($debug);
1418 expire_lock_files($debug);
1419 expire_error_log($debug);
1420
1421 $count = update_feedbrowser_cache();
1422 _debug("Feedbrowser updated, $count feeds processed.");
1423
1424 purge_orphans( true);
1425 $rc = cleanup_tags( 14, 50000);
1426
1427 _debug("Cleaned $rc cached tags.");
1428
1429 PluginHost::getInstance()->run_hooks(PluginHost::HOOK_HOUSE_KEEPING, "hook_house_keeping", "");
1430
1431 }
1432 ?>
web-based news feed (RSS/Atom) reader and aggregator; see https://tt-rss.org/