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