]> 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
pass correct instance of pluginhost to calculate_article_hash
[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 $nf = 0;
172
173 // For each feed, we call the feed update function.
174 foreach ($feeds_to_update as $feed) {
175 if($debug) _debug("Base feed: $feed");
176
177 //update_rss_feed($line["id"], true);
178
179 // since we have the data cached, we can deal with other feeds with the same url
180
181 $tmp_result = db_query("SELECT DISTINCT ttrss_feeds.id,last_updated,ttrss_feeds.owner_uid
182 FROM ttrss_feeds, ttrss_users, ttrss_user_prefs WHERE
183 ttrss_user_prefs.owner_uid = ttrss_feeds.owner_uid AND
184 ttrss_users.id = ttrss_user_prefs.owner_uid AND
185 ttrss_user_prefs.pref_name = 'DEFAULT_UPDATE_INTERVAL' AND
186 ttrss_user_prefs.profile IS NULL AND
187 feed_url = '".db_escape_string($feed)."' AND
188 (ttrss_feeds.update_interval > 0 OR
189 ttrss_user_prefs.value != '-1')
190 $login_thresh_qpart
191 ORDER BY ttrss_feeds.id $query_limit");
192
193 if (db_num_rows($tmp_result) > 0) {
194 $rss = false;
195
196 while ($tline = db_fetch_assoc($tmp_result)) {
197 if($debug) _debug(" => " . $tline["last_updated"] . ", " . $tline["id"] . " " . $tline["owner_uid"]);
198
199 $rss = update_rss_feed($tline["id"], true, false);
200 _debug_suppress(false);
201 ++$nf;
202 }
203 }
204 }
205
206 require_once "digest.php";
207
208 // Send feed digests by email if needed.
209 send_headlines_digests($debug);
210
211 return $nf;
212
213 } // function update_daemon_common
214
215 // ignore_daemon is not used
216 function update_rss_feed($feed, $ignore_daemon = false, $no_cache = false, $rss = false) {
217
218 $debug_enabled = defined('DAEMON_EXTENDED_DEBUG') || $_REQUEST['xdebug'];
219
220 _debug_suppress(!$debug_enabled);
221 _debug("start", $debug_enabled);
222
223 $result = db_query("SELECT id,update_interval,auth_login,
224 feed_url,auth_pass,cache_images,
225 mark_unread_on_update, owner_uid,
226 pubsub_state, auth_pass_encrypted,
227 (SELECT max(date_entered) FROM
228 ttrss_entries, ttrss_user_entries where ref_id = id AND feed_id = '$feed') AS last_article_timestamp
229 FROM ttrss_feeds WHERE id = '$feed'");
230
231 if (db_num_rows($result) == 0) {
232 _debug("feed $feed NOT FOUND/SKIPPED", $debug_enabled);
233 return false;
234 }
235
236 $last_article_timestamp = @strtotime(db_fetch_result($result, 0, "last_article_timestamp"));
237
238 if (defined('_DISABLE_HTTP_304'))
239 $last_article_timestamp = 0;
240
241 $owner_uid = db_fetch_result($result, 0, "owner_uid");
242 $mark_unread_on_update = sql_bool_to_bool(db_fetch_result($result,
243 0, "mark_unread_on_update"));
244 $pubsub_state = db_fetch_result($result, 0, "pubsub_state");
245 $auth_pass_encrypted = sql_bool_to_bool(db_fetch_result($result,
246 0, "auth_pass_encrypted"));
247
248 db_query("UPDATE ttrss_feeds SET last_update_started = NOW()
249 WHERE id = '$feed'");
250
251 $auth_login = db_fetch_result($result, 0, "auth_login");
252 $auth_pass = db_fetch_result($result, 0, "auth_pass");
253
254 if ($auth_pass_encrypted) {
255 require_once "crypt.php";
256 $auth_pass = decrypt_string($auth_pass);
257 }
258
259 $cache_images = sql_bool_to_bool(db_fetch_result($result, 0, "cache_images"));
260 $fetch_url = db_fetch_result($result, 0, "feed_url");
261
262 $feed = db_escape_string($feed);
263
264 $date_feed_processed = date('Y-m-d H:i');
265
266 $cache_filename = CACHE_DIR . "/simplepie/" . sha1($fetch_url) . ".xml";
267
268 $pluginhost = new PluginHost();
269 $pluginhost->set_debug($debug_enabled);
270 $user_plugins = get_pref("_ENABLED_PLUGINS", $owner_uid);
271
272 $pluginhost->load(PLUGINS, PluginHost::KIND_ALL);
273 $pluginhost->load($user_plugins, PluginHost::KIND_USER, $owner_uid);
274 $pluginhost->load_data();
275
276 if ($rss && is_object($rss) && get_class($rss) == "FeedParser") {
277 _debug("using previously initialized parser object");
278 } else {
279 $rss_hash = false;
280
281 $force_refetch = isset($_REQUEST["force_refetch"]);
282
283 foreach ($pluginhost->get_hooks(PluginHost::HOOK_FETCH_FEED) as $plugin) {
284 $feed_data = $plugin->hook_fetch_feed($feed_data, $fetch_url, $owner_uid, $feed, $last_article_timestamp, $auth_login, $auth_pass);
285 }
286
287 // try cache
288 if (!$feed_data &&
289 file_exists($cache_filename) &&
290 is_readable($cache_filename) &&
291 !$auth_login && !$auth_pass &&
292 filemtime($cache_filename) > time() - 30) {
293
294 _debug("using local cache [$cache_filename].", $debug_enabled);
295
296 @$feed_data = file_get_contents($cache_filename);
297
298 if ($feed_data) {
299 $rss_hash = sha1($feed_data);
300 }
301
302 } else {
303 _debug("local cache will not be used for this feed", $debug_enabled);
304 }
305
306 // fetch feed from source
307 if (!$feed_data) {
308 _debug("fetching [$fetch_url]...", $debug_enabled);
309 _debug("If-Modified-Since: ".gmdate('D, d M Y H:i:s \G\M\T', $last_article_timestamp), $debug_enabled);
310
311 $feed_data = fetch_file_contents($fetch_url, false,
312 $auth_login, $auth_pass, false,
313 $no_cache ? FEED_FETCH_NO_CACHE_TIMEOUT : FEED_FETCH_TIMEOUT,
314 $force_refetch ? 0 : $last_article_timestamp);
315
316 global $fetch_curl_used;
317
318 if (!$fetch_curl_used) {
319 $tmp = @gzdecode($feed_data);
320
321 if ($tmp) $feed_data = $tmp;
322 }
323
324 $feed_data = trim($feed_data);
325
326 _debug("fetch done.", $debug_enabled);
327
328 /* if ($feed_data) {
329 $error = verify_feed_xml($feed_data);
330
331 if ($error) {
332 _debug("error verifying XML, code: " . $error->code, $debug_enabled);
333
334 if ($error->code == 26) {
335 _debug("got error 26, trying to decode entities...", $debug_enabled);
336
337 $feed_data = html_entity_decode($feed_data, ENT_COMPAT, 'UTF-8');
338
339 $error = verify_feed_xml($feed_data);
340
341 if ($error) $feed_data = '';
342 }
343 }
344 } */
345
346 // cache vanilla feed data for re-use
347 if ($feed_data && !$auth_pass && !$auth_login && is_writable(CACHE_DIR . "/simplepie")) {
348 $new_rss_hash = sha1($feed_data);
349
350 if ($new_rss_hash != $rss_hash) {
351 _debug("saving $cache_filename", $debug_enabled);
352 @file_put_contents($cache_filename, $feed_data);
353 }
354 }
355 }
356
357 if (!$feed_data) {
358 global $fetch_last_error;
359 global $fetch_last_error_code;
360
361 _debug("unable to fetch: $fetch_last_error [$fetch_last_error_code]", $debug_enabled);
362
363 $error_escaped = '';
364
365 // If-Modified-Since
366 if ($fetch_last_error_code != 304) {
367 $error_escaped = db_escape_string($fetch_last_error);
368 } else {
369 _debug("source claims data not modified, nothing to do.", $debug_enabled);
370 }
371
372 db_query(
373 "UPDATE ttrss_feeds SET last_error = '$error_escaped',
374 last_updated = NOW() WHERE id = '$feed'");
375
376 return;
377 }
378 }
379
380 foreach ($pluginhost->get_hooks(PluginHost::HOOK_FEED_FETCHED) as $plugin) {
381 $feed_data = $plugin->hook_feed_fetched($feed_data, $fetch_url, $owner_uid, $feed);
382 }
383
384 // set last update to now so if anything *simplepie* crashes later we won't be
385 // continuously failing on the same feed
386 //db_query("UPDATE ttrss_feeds SET last_updated = NOW() WHERE id = '$feed'");
387
388 if (!$rss) {
389 $rss = new FeedParser($feed_data);
390 $rss->init();
391 }
392
393 if (DETECT_ARTICLE_LANGUAGE) {
394 require_once "lib/languagedetect/LanguageDetect.php";
395
396 $lang = new Text_LanguageDetect();
397 $lang->setNameMode(2);
398 }
399
400 // print_r($rss);
401
402 $feed = db_escape_string($feed);
403
404 if (!$rss->error()) {
405
406 // We use local pluginhost here because we need to load different per-user feed plugins
407 $pluginhost->run_hooks(PluginHost::HOOK_FEED_PARSED, "hook_feed_parsed", $rss);
408
409 _debug("processing feed data...", $debug_enabled);
410
411 // db_query("BEGIN");
412
413 if (DB_TYPE == "pgsql") {
414 $favicon_interval_qpart = "favicon_last_checked < NOW() - INTERVAL '12 hour'";
415 } else {
416 $favicon_interval_qpart = "favicon_last_checked < DATE_SUB(NOW(), INTERVAL 12 HOUR)";
417 }
418
419 $result = db_query("SELECT title,site_url,owner_uid,favicon_avg_color,
420 (favicon_last_checked IS NULL OR $favicon_interval_qpart) AS
421 favicon_needs_check
422 FROM ttrss_feeds WHERE id = '$feed'");
423
424 $registered_title = db_fetch_result($result, 0, "title");
425 $orig_site_url = db_fetch_result($result, 0, "site_url");
426 $favicon_needs_check = sql_bool_to_bool(db_fetch_result($result, 0,
427 "favicon_needs_check"));
428 $favicon_avg_color = db_fetch_result($result, 0, "favicon_avg_color");
429
430 $owner_uid = db_fetch_result($result, 0, "owner_uid");
431
432 $site_url = db_escape_string(mb_substr(rewrite_relative_url($fetch_url, $rss->get_link()), 0, 245));
433
434 _debug("site_url: $site_url", $debug_enabled);
435 _debug("feed_title: " . $rss->get_title(), $debug_enabled);
436
437 if ($favicon_needs_check || $force_refetch) {
438
439 /* terrible hack: if we crash on floicon shit here, we won't check
440 * the icon avgcolor again (unless the icon got updated) */
441
442 $favicon_file = ICONS_DIR . "/$feed.ico";
443 $favicon_modified = @filemtime($favicon_file);
444
445 _debug("checking favicon...", $debug_enabled);
446
447 check_feed_favicon($site_url, $feed);
448 $favicon_modified_new = @filemtime($favicon_file);
449
450 if ($favicon_modified_new > $favicon_modified)
451 $favicon_avg_color = '';
452
453 if (file_exists($favicon_file) && function_exists("imagecreatefromstring") && $favicon_avg_color == '') {
454 require_once "colors.php";
455
456 db_query("UPDATE ttrss_feeds SET favicon_avg_color = 'fail' WHERE
457 id = '$feed'");
458
459 $favicon_color = db_escape_string(
460 calculate_avg_color($favicon_file));
461
462 $favicon_colorstring = ",favicon_avg_color = '".$favicon_color."'";
463 } else if ($favicon_avg_color == 'fail') {
464 _debug("floicon failed on this file, not trying to recalculate avg color", $debug_enabled);
465 }
466
467 db_query("UPDATE ttrss_feeds SET favicon_last_checked = NOW()
468 $favicon_colorstring
469 WHERE id = '$feed'");
470 }
471
472 if (!$registered_title || $registered_title == "[Unknown]") {
473
474 $feed_title = db_escape_string($rss->get_title());
475
476 if ($feed_title) {
477 _debug("registering title: $feed_title", $debug_enabled);
478
479 db_query("UPDATE ttrss_feeds SET
480 title = '$feed_title' WHERE id = '$feed'");
481 }
482 }
483
484 if ($site_url && $orig_site_url != $site_url) {
485 db_query("UPDATE ttrss_feeds SET
486 site_url = '$site_url' WHERE id = '$feed'");
487 }
488
489 _debug("loading filters & labels...", $debug_enabled);
490
491 $filters = load_filters($feed, $owner_uid);
492 $labels = get_all_labels($owner_uid);
493
494 _debug("" . count($filters) . " filters loaded.", $debug_enabled);
495
496 $items = $rss->get_items();
497
498 if (!is_array($items)) {
499 _debug("no articles found.", $debug_enabled);
500
501 db_query("UPDATE ttrss_feeds
502 SET last_updated = NOW(), last_error = '' WHERE id = '$feed'");
503
504 return; // no articles
505 }
506
507 if ($pubsub_state != 2 && PUBSUBHUBBUB_ENABLED) {
508
509 _debug("checking for PUSH hub...", $debug_enabled);
510
511 $feed_hub_url = false;
512
513 $links = $rss->get_links('hub');
514
515 if ($links && is_array($links)) {
516 foreach ($links as $l) {
517 $feed_hub_url = $l;
518 break;
519 }
520 }
521
522 _debug("feed hub url: $feed_hub_url", $debug_enabled);
523
524 $feed_self_url = $fetch_url;
525
526 $links = $rss->get_links('self');
527
528 if ($links && is_array($links)) {
529 foreach ($links as $l) {
530 $feed_self_url = $l;
531 break;
532 }
533 }
534
535 _debug("feed self url = $feed_self_url");
536
537 if ($feed_hub_url && $feed_self_url && function_exists('curl_init') &&
538 !ini_get("open_basedir")) {
539
540 require_once 'lib/pubsubhubbub/subscriber.php';
541
542 $callback_url = get_self_url_prefix() .
543 "/public.php?op=pubsub&id=$feed";
544
545 $s = new Subscriber($feed_hub_url, $callback_url);
546
547 $rc = $s->subscribe($feed_self_url);
548
549 _debug("feed hub url found, subscribe request sent. [rc=$rc]", $debug_enabled);
550
551 db_query("UPDATE ttrss_feeds SET pubsub_state = 1
552 WHERE id = '$feed'");
553 }
554 }
555
556 _debug("processing articles...", $debug_enabled);
557
558 foreach ($items as $item) {
559 if ($_REQUEST['xdebug'] == 3) {
560 print_r($item);
561 }
562
563 $entry_guid = $item->get_id();
564 if (!$entry_guid) $entry_guid = $item->get_link();
565 if (!$entry_guid) $entry_guid = make_guid_from_title($item->get_title());
566
567 _debug("f_guid $entry_guid", $debug_enabled);
568
569 if (!$entry_guid) continue;
570
571 $entry_guid = "$owner_uid,$entry_guid";
572
573 $entry_guid_hashed = db_escape_string('SHA1:' . sha1($entry_guid));
574
575 _debug("guid $entry_guid / $entry_guid_hashed", $debug_enabled);
576
577 $entry_timestamp = "";
578
579 $entry_timestamp = $item->get_date();
580
581 _debug("orig date: " . $item->get_date(), $debug_enabled);
582
583 if ($entry_timestamp == -1 || !$entry_timestamp || $entry_timestamp > time()) {
584 $entry_timestamp = time();
585 }
586
587 $entry_timestamp_fmt = strftime("%Y/%m/%d %H:%M:%S", $entry_timestamp);
588
589 _debug("date $entry_timestamp [$entry_timestamp_fmt]", $debug_enabled);
590
591 // $entry_title = html_entity_decode($item->get_title(), ENT_COMPAT, 'UTF-8');
592 // $entry_title = decode_numeric_entities($entry_title);
593 $entry_title = $item->get_title();
594
595 $entry_link = rewrite_relative_url($site_url, $item->get_link());
596
597 _debug("title $entry_title", $debug_enabled);
598 _debug("link $entry_link", $debug_enabled);
599
600 if (!$entry_title) $entry_title = date("Y-m-d H:i:s", $entry_timestamp);;
601
602 $entry_content = $item->get_content();
603 if (!$entry_content) $entry_content = $item->get_description();
604
605 if ($_REQUEST["xdebug"] == 2) {
606 print "content: ";
607 print $entry_content;
608 print "\n";
609 }
610
611 $entry_language = "";
612
613 if (DETECT_ARTICLE_LANGUAGE) {
614 $entry_language = $lang->detect($entry_title . " " . $entry_content, 1);
615
616 if (count($entry_language) > 0) {
617 $entry_language = array_keys($entry_language);
618
619 // the fuck?
620 if (is_array($entry_language))
621 $entry_language = "";
622 else
623 $entry_language = db_escape_string(substr($entry_language[0], 0, 2));
624
625 _debug("detected language: $entry_language", $debug_enabled);
626 } else {
627 $entry_language = "";
628 }
629 }
630
631 $entry_comments = $item->get_comments_url();
632 $entry_author = $item->get_author();
633
634 $entry_guid = db_escape_string(mb_substr($entry_guid, 0, 245));
635
636 $entry_comments = db_escape_string(mb_substr(trim($entry_comments), 0, 245));
637 $entry_author = db_escape_string(mb_substr(trim($entry_author), 0, 245));
638
639 $num_comments = (int) $item->get_comments_count();
640
641 _debug("author $entry_author", $debug_enabled);
642 _debug("num_comments: $num_comments", $debug_enabled);
643 _debug("looking for tags...", $debug_enabled);
644
645 // parse <category> entries into tags
646
647 $additional_tags = array();
648
649 $additional_tags_src = $item->get_categories();
650
651 if (is_array($additional_tags_src)) {
652 foreach ($additional_tags_src as $tobj) {
653 array_push($additional_tags, $tobj);
654 }
655 }
656
657 $entry_tags = array_unique($additional_tags);
658
659 for ($i = 0; $i < count($entry_tags); $i++)
660 $entry_tags[$i] = mb_strtolower($entry_tags[$i], 'utf-8');
661
662 _debug("tags found: " . join(",", $entry_tags), $debug_enabled);
663
664 _debug("done collecting data.", $debug_enabled);
665
666 $result = db_query("SELECT id, content_hash FROM ttrss_entries
667 WHERE guid = '".db_escape_string($entry_guid)."' OR guid = '$entry_guid_hashed'");
668
669 if (db_num_rows($result) != 0) {
670 $base_entry_id = db_fetch_result($result, 0, "id");
671 $entry_stored_hash = db_fetch_result($result, 0, "content_hash");
672 } else {
673 $base_entry_id = false;
674 $entry_stored_hash = "";
675 }
676
677 $article = array("owner_uid" => $owner_uid, // read only
678 "guid" => $entry_guid, // read only
679 "title" => $entry_title,
680 "content" => $entry_content,
681 "link" => $entry_link,
682 "tags" => $entry_tags,
683 "author" => $entry_author,
684 "feed" => array("id" => $feed,
685 "fetch_url" => $fetch_url,
686 "site_url" => $site_url)
687 );
688
689 $entry_plugin_data = "";
690 $entry_current_hash = calculate_article_hash($article, $pluginhost);
691
692 _debug("article hash: $entry_current_hash [stored=$entry_stored_hash]", $debug_enabled);
693
694 if ($entry_current_hash == $entry_stored_hash) {
695 _debug("stored article seems up to date [IID: $base_entry_id], updating timestamp only", $debug_enabled);
696
697 // we keep encountering the entry in feeds, so we need to
698 // update date_updated column so that we don't get horrible
699 // dupes when the entry gets purged and reinserted again e.g.
700 // in the case of SLOW SLOW OMG SLOW updating feeds
701
702 $base_entry_id = db_fetch_result($result, 0, "id");
703
704 db_query("UPDATE ttrss_entries SET date_updated = NOW()
705 WHERE id = '$base_entry_id'");
706
707 continue;
708 }
709
710 _debug("hash differs, applying plugin filters:", $debug_enabled);
711
712 foreach ($pluginhost->get_hooks(PluginHost::HOOK_ARTICLE_FILTER) as $plugin) {
713 _debug("... " . get_class($plugin), $debug_enabled);
714
715 $start = microtime(true);
716 $article = $plugin->hook_article_filter($article);
717
718 _debug("=== " . sprintf("%.4f (sec)", microtime(true) - $start), $debug_enabled);
719
720 $entry_plugin_data .= mb_strtolower(get_class($plugin)) . ",";
721 }
722
723 $entry_plugin_data = db_escape_string($entry_plugin_data);
724
725 _debug("plugin data: $entry_plugin_data", $debug_enabled);
726
727 $entry_tags = $article["tags"];
728 $entry_guid = db_escape_string($entry_guid);
729 $entry_title = db_escape_string($article["title"]);
730 $entry_author = db_escape_string($article["author"]);
731 $entry_link = db_escape_string($article["link"]);
732 $entry_content = $article["content"]; // escaped below
733
734 if ($cache_images && is_writable(CACHE_DIR . '/images'))
735 cache_images($entry_content, $site_url, $debug_enabled);
736
737 $entry_content = db_escape_string($entry_content, false);
738
739 db_query("BEGIN");
740
741 $result = db_query("SELECT id FROM ttrss_entries
742 WHERE (guid = '$entry_guid' OR guid = '$entry_guid_hashed')");
743
744 if (db_num_rows($result) == 0) {
745
746 _debug("base guid [$entry_guid] not found", $debug_enabled);
747
748 // base post entry does not exist, create it
749
750 $result = db_query(
751 "INSERT INTO ttrss_entries
752 (title,
753 guid,
754 link,
755 updated,
756 content,
757 content_hash,
758 no_orig_date,
759 date_updated,
760 date_entered,
761 comments,
762 num_comments,
763 plugin_data,
764 lang,
765 author)
766 VALUES
767 ('$entry_title',
768 '$entry_guid_hashed',
769 '$entry_link',
770 '$entry_timestamp_fmt',
771 '$entry_content',
772 '$entry_current_hash',
773 false,
774 NOW(),
775 '$date_feed_processed',
776 '$entry_comments',
777 '$num_comments',
778 '$entry_plugin_data',
779 '$entry_language',
780 '$entry_author')");
781
782 $article_labels = array();
783
784 } else {
785 $base_entry_id = db_fetch_result($result, 0, "id");
786
787 $article_labels = get_article_labels($base_entry_id, $owner_uid);
788 }
789
790 // now it should exist, if not - bad luck then
791
792 $result = db_query("SELECT id FROM ttrss_entries
793 WHERE guid = '$entry_guid' OR guid = '$entry_guid_hashed'");
794
795 $entry_ref_id = 0;
796 $entry_int_id = 0;
797
798 if (db_num_rows($result) == 1) {
799
800 _debug("base guid found, checking for user record", $debug_enabled);
801
802 $ref_id = db_fetch_result($result, 0, "id");
803 $entry_ref_id = $ref_id;
804
805 /* $stored_guid = db_fetch_result($result, 0, "guid");
806 if ($stored_guid != $entry_guid_hashed) {
807 if ($debug_enabled) _debug("upgrading compat guid to hashed one", $debug_enabled);
808
809 db_query("UPDATE ttrss_entries SET guid = '$entry_guid_hashed' WHERE
810 id = '$ref_id'");
811 } */
812
813 // check for user post link to main table
814
815 // do we allow duplicate posts with same GUID in different feeds?
816 if (get_pref("ALLOW_DUPLICATE_POSTS", $owner_uid, false)) {
817 $dupcheck_qpart = "AND (feed_id = '$feed' OR feed_id IS NULL)";
818 } else {
819 $dupcheck_qpart = "";
820 }
821
822 /* Collect article tags here so we could filter by them: */
823
824 $article_filters = get_article_filters($filters, $entry_title,
825 $entry_content, $entry_link, $entry_timestamp, $entry_author,
826 $entry_tags);
827
828 if ($debug_enabled) {
829 _debug("article filters: ", $debug_enabled);
830 if (count($article_filters) != 0) {
831 print_r($article_filters);
832 }
833 }
834
835 if (find_article_filter($article_filters, "filter")) {
836 db_query("COMMIT"); // close transaction in progress
837 continue;
838 }
839
840 $score = calculate_article_score($article_filters);
841
842 _debug("initial score: $score", $debug_enabled);
843
844 $query = "SELECT ref_id, int_id FROM ttrss_user_entries WHERE
845 ref_id = '$ref_id' AND owner_uid = '$owner_uid'
846 $dupcheck_qpart";
847
848 // if ($_REQUEST["xdebug"]) print "$query\n";
849
850 $result = db_query($query);
851
852 // okay it doesn't exist - create user entry
853 if (db_num_rows($result) == 0) {
854
855 _debug("user record not found, creating...", $debug_enabled);
856
857 if ($score >= -500 && !find_article_filter($article_filters, 'catchup')) {
858 $unread = 'true';
859 $last_read_qpart = 'NULL';
860 } else {
861 $unread = 'false';
862 $last_read_qpart = 'NOW()';
863 }
864
865 if (find_article_filter($article_filters, 'mark') || $score > 1000) {
866 $marked = 'true';
867 } else {
868 $marked = 'false';
869 }
870
871 if (find_article_filter($article_filters, 'publish')) {
872 $published = 'true';
873 } else {
874 $published = 'false';
875 }
876
877 // N-grams
878
879 if (DB_TYPE == "pgsql" and defined('_NGRAM_TITLE_DUPLICATE_THRESHOLD')) {
880
881 $result = db_query("SELECT COUNT(*) AS similar FROM
882 ttrss_entries,ttrss_user_entries
883 WHERE ref_id = id AND updated >= NOW() - INTERVAL '7 day'
884 AND similarity(title, '$entry_title') >= "._NGRAM_TITLE_DUPLICATE_THRESHOLD."
885 AND owner_uid = $owner_uid");
886
887 $ngram_similar = db_fetch_result($result, 0, "similar");
888
889 _debug("N-gram similar results: $ngram_similar", $debug_enabled);
890
891 if ($ngram_similar > 0) {
892 $unread = 'false';
893 }
894 }
895
896 $last_marked = ($marked == 'true') ? 'NOW()' : 'NULL';
897 $last_published = ($published == 'true') ? 'NOW()' : 'NULL';
898
899 $result = db_query(
900 "INSERT INTO ttrss_user_entries
901 (ref_id, owner_uid, feed_id, unread, last_read, marked,
902 published, score, tag_cache, label_cache, uuid,
903 last_marked, last_published)
904 VALUES ('$ref_id', '$owner_uid', '$feed', $unread,
905 $last_read_qpart, $marked, $published, '$score', '', '',
906 '', $last_marked, $last_published)");
907
908 if (PUBSUBHUBBUB_HUB && $published == 'true') {
909 $rss_link = get_self_url_prefix() .
910 "/public.php?op=rss&id=-2&key=" .
911 get_feed_access_key(-2, false, $owner_uid);
912
913 $p = new Publisher(PUBSUBHUBBUB_HUB);
914
915 /* $pubsub_result = */ $p->publish_update($rss_link);
916 }
917
918 $result = db_query(
919 "SELECT int_id FROM ttrss_user_entries WHERE
920 ref_id = '$ref_id' AND owner_uid = '$owner_uid' AND
921 feed_id = '$feed' LIMIT 1");
922
923 if (db_num_rows($result) == 1) {
924 $entry_int_id = db_fetch_result($result, 0, "int_id");
925 }
926 } else {
927 _debug("user record FOUND", $debug_enabled);
928
929 $entry_ref_id = db_fetch_result($result, 0, "ref_id");
930 $entry_int_id = db_fetch_result($result, 0, "int_id");
931 }
932
933 _debug("RID: $entry_ref_id, IID: $entry_int_id", $debug_enabled);
934
935 db_query("UPDATE ttrss_entries
936 SET title = '$entry_title',
937 content = '$entry_content',
938 content_hash = '$entry_current_hash',
939 updated = '$entry_timestamp_fmt',
940 num_comments = '$num_comments',
941 plugin_data = '$entry_plugin_data',
942 author = '$entry_author',
943 lang = '$entry_language'
944 WHERE id = '$ref_id'");
945
946 if ($mark_unread_on_update) {
947 db_query("UPDATE ttrss_user_entries
948 SET last_read = null, unread = true WHERE ref_id = '$ref_id'");
949 }
950 }
951
952 db_query("COMMIT");
953
954 _debug("assigning labels...", $debug_enabled);
955
956 assign_article_to_label_filters($entry_ref_id, $article_filters,
957 $owner_uid, $article_labels);
958
959 _debug("looking for enclosures...", $debug_enabled);
960
961 // enclosures
962
963 $enclosures = array();
964
965 $encs = $item->get_enclosures();
966
967 if (is_array($encs)) {
968 foreach ($encs as $e) {
969 $e_item = array(
970 $e->link, $e->type, $e->length, $e->title, $e->width, $e->height);
971 array_push($enclosures, $e_item);
972 }
973 }
974
975 if ($debug_enabled) {
976 _debug("article enclosures:", $debug_enabled);
977 print_r($enclosures);
978 }
979
980 db_query("BEGIN");
981
982 // debugging
983 // db_query("DELETE FROM ttrss_enclosures WHERE post_id = '$entry_ref_id'");
984
985 foreach ($enclosures as $enc) {
986 $enc_url = db_escape_string($enc[0]);
987 $enc_type = db_escape_string($enc[1]);
988 $enc_dur = db_escape_string($enc[2]);
989 $enc_title = db_escape_string($enc[3]);
990 $enc_width = intval($enc[4]);
991 $enc_height = intval($enc[5]);
992
993 $result = db_query("SELECT id FROM ttrss_enclosures
994 WHERE content_url = '$enc_url' AND post_id = '$entry_ref_id'");
995
996 if (db_num_rows($result) == 0) {
997 db_query("INSERT INTO ttrss_enclosures
998 (content_url, content_type, title, duration, post_id, width, height) VALUES
999 ('$enc_url', '$enc_type', '$enc_title', '$enc_dur', '$entry_ref_id', $enc_width, $enc_height)");
1000 }
1001 }
1002
1003 db_query("COMMIT");
1004
1005 // check for manual tags (we have to do it here since they're loaded from filters)
1006
1007 foreach ($article_filters as $f) {
1008 if ($f["type"] == "tag") {
1009
1010 $manual_tags = trim_array(explode(",", $f["param"]));
1011
1012 foreach ($manual_tags as $tag) {
1013 if (tag_is_valid($tag)) {
1014 array_push($entry_tags, $tag);
1015 }
1016 }
1017 }
1018 }
1019
1020 // Skip boring tags
1021
1022 $boring_tags = trim_array(explode(",", mb_strtolower(get_pref(
1023 'BLACKLISTED_TAGS', $owner_uid, ''), 'utf-8')));
1024
1025 $filtered_tags = array();
1026 $tags_to_cache = array();
1027
1028 if ($entry_tags && is_array($entry_tags)) {
1029 foreach ($entry_tags as $tag) {
1030 if (array_search($tag, $boring_tags) === false) {
1031 array_push($filtered_tags, $tag);
1032 }
1033 }
1034 }
1035
1036 $filtered_tags = array_unique($filtered_tags);
1037
1038 if ($debug_enabled) {
1039 _debug("filtered article tags:", $debug_enabled);
1040 print_r($filtered_tags);
1041 }
1042
1043 // Save article tags in the database
1044
1045 if (count($filtered_tags) > 0) {
1046
1047 db_query("BEGIN");
1048
1049 foreach ($filtered_tags as $tag) {
1050
1051 $tag = sanitize_tag($tag);
1052 $tag = db_escape_string($tag);
1053
1054 if (!tag_is_valid($tag)) continue;
1055
1056 $result = db_query("SELECT id FROM ttrss_tags
1057 WHERE tag_name = '$tag' AND post_int_id = '$entry_int_id' AND
1058 owner_uid = '$owner_uid' LIMIT 1");
1059
1060 if ($result && db_num_rows($result) == 0) {
1061
1062 db_query("INSERT INTO ttrss_tags
1063 (owner_uid,tag_name,post_int_id)
1064 VALUES ('$owner_uid','$tag', '$entry_int_id')");
1065 }
1066
1067 array_push($tags_to_cache, $tag);
1068 }
1069
1070 /* update the cache */
1071
1072 $tags_to_cache = array_unique($tags_to_cache);
1073
1074 $tags_str = db_escape_string(join(",", $tags_to_cache));
1075
1076 db_query("UPDATE ttrss_user_entries
1077 SET tag_cache = '$tags_str' WHERE ref_id = '$entry_ref_id'
1078 AND owner_uid = $owner_uid");
1079
1080 db_query("COMMIT");
1081 }
1082
1083 if (get_pref("AUTO_ASSIGN_LABELS", $owner_uid, false)) {
1084 _debug("auto-assigning labels...", $debug_enabled);
1085
1086 foreach ($labels as $label) {
1087 $caption = preg_quote($label["caption"]);
1088
1089 if ($caption && preg_match("/\b$caption\b/i", "$tags_str " . strip_tags($entry_content) . " $entry_title")) {
1090 if (!labels_contains_caption($article_labels, $caption)) {
1091 label_add_article($entry_ref_id, $caption, $owner_uid);
1092 }
1093 }
1094 }
1095 }
1096
1097 _debug("article processed", $debug_enabled);
1098 }
1099
1100 _debug("purging feed...", $debug_enabled);
1101
1102 purge_feed($feed, 0, $debug_enabled);
1103
1104 db_query("UPDATE ttrss_feeds
1105 SET last_updated = NOW(), last_error = '' WHERE id = '$feed'");
1106
1107 // db_query("COMMIT");
1108
1109 } else {
1110
1111 $error_msg = db_escape_string(mb_substr($rss->error(), 0, 245));
1112
1113 _debug("fetch error: $error_msg", $debug_enabled);
1114
1115 if (count($rss->errors()) > 1) {
1116 foreach ($rss->errors() as $error) {
1117 _debug("+ $error");
1118 }
1119 }
1120
1121 db_query(
1122 "UPDATE ttrss_feeds SET last_error = '$error_msg',
1123 last_updated = NOW() WHERE id = '$feed'");
1124
1125 unset($rss);
1126 }
1127
1128 _debug("done", $debug_enabled);
1129
1130 return $rss;
1131 }
1132
1133 function cache_images($html, $site_url, $debug) {
1134 libxml_use_internal_errors(true);
1135
1136 $charset_hack = '<head>
1137 <meta http-equiv="Content-Type" content="text/html; charset=utf-8"/>
1138 </head>';
1139
1140 $doc = new DOMDocument();
1141 $doc->loadHTML($charset_hack . $html);
1142 $xpath = new DOMXPath($doc);
1143
1144 $entries = $xpath->query('(//img[@src])');
1145
1146 foreach ($entries as $entry) {
1147 if ($entry->hasAttribute('src')) {
1148 $src = rewrite_relative_url($site_url, $entry->getAttribute('src'));
1149
1150 $local_filename = CACHE_DIR . "/images/" . sha1($src) . ".png";
1151
1152 if ($debug) _debug("cache_images: downloading: $src to $local_filename");
1153
1154 if (!file_exists($local_filename)) {
1155 $file_content = fetch_file_contents($src);
1156
1157 if ($file_content && strlen($file_content) > _MIN_CACHE_IMAGE_SIZE) {
1158 file_put_contents($local_filename, $file_content);
1159 }
1160 }
1161
1162 /* if (file_exists($local_filename)) {
1163 $entry->setAttribute('src', SELF_URL_PATH . '/image.php?url=' .
1164 base64_encode($src));
1165 } */
1166 }
1167 }
1168
1169 //$node = $doc->getElementsByTagName('body')->item(0);
1170 //return $doc->saveXML($node);
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/