]> git.wh0rd.org - tt-rss.git/blame - include/rssfuncs.php
af_unburn: set useragent
[tt-rss.git] / include / rssfuncs.php
CommitLineData
2c08214a 1<?php
09e8bdfd
AD
2 define('DAEMON_UPDATE_LOGIN_LIMIT', 30);
3 define('DAEMON_FEED_LIMIT', 100);
0f556e3e 4 define('DAEMON_SLEEP_INTERVAL', 60);
09e8bdfd 5
79178062
AD
6 function update_feedbrowser_cache($link) {
7
8 $result = db_query($link, "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($link, "BEGIN");
15
16 db_query($link, "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 $tmp_result = db_query($link, "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($link, "INSERT INTO ttrss_feedbrowser_cache
32 (feed_url, site_url, title, subscribers) VALUES ('$feed_url',
33 '$site_url', '$title', '$subscribers')");
34
35 ++$count;
36
37 }
38
39 }
40
41 db_query($link, "COMMIT");
42
43 return $count;
44
45 }
46
47
2c08214a
AD
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 $debug Set to false to disable debug output. Default to true.
58 * @return void
59 */
60 function update_daemon_common($link, $limit = DAEMON_FEED_LIMIT, $from_http = false, $debug = true) {
61 // Process all other feeds using last_updated and interval parameters
62
61c1812f
AD
63 define('PREFS_NO_CACHE', true);
64
2c08214a 65 // Test if the user has loggued in recently. If not, it does not update its feeds.
09e8bdfd 66 if (!SINGLE_USER_MODE && DAEMON_UPDATE_LOGIN_LIMIT > 0) {
2c08214a
AD
67 if (DB_TYPE == "pgsql") {
68 $login_thresh_qpart = "AND ttrss_users.last_login >= NOW() - INTERVAL '".DAEMON_UPDATE_LOGIN_LIMIT." days'";
69 } else {
70 $login_thresh_qpart = "AND ttrss_users.last_login >= DATE_SUB(NOW(), INTERVAL ".DAEMON_UPDATE_LOGIN_LIMIT." DAY)";
71 }
72 } else {
73 $login_thresh_qpart = "";
74 }
75
76 // Test if the feed need a update (update interval exceded).
77 if (DB_TYPE == "pgsql") {
78 $update_limit_qpart = "AND ((
79 ttrss_feeds.update_interval = 0
80 AND ttrss_feeds.last_updated < NOW() - CAST((ttrss_user_prefs.value || ' minutes') AS INTERVAL)
81 ) OR (
82 ttrss_feeds.update_interval > 0
83 AND ttrss_feeds.last_updated < NOW() - CAST((ttrss_feeds.update_interval || ' minutes') AS INTERVAL)
941e48a4
AD
84 ) OR ttrss_feeds.last_updated IS NULL
85 OR last_updated = '1970-01-01 00:00:00')";
2c08214a
AD
86 } else {
87 $update_limit_qpart = "AND ((
88 ttrss_feeds.update_interval = 0
89 AND ttrss_feeds.last_updated < DATE_SUB(NOW(), INTERVAL CONVERT(ttrss_user_prefs.value, SIGNED INTEGER) MINUTE)
90 ) OR (
91 ttrss_feeds.update_interval > 0
92 AND ttrss_feeds.last_updated < DATE_SUB(NOW(), INTERVAL ttrss_feeds.update_interval MINUTE)
941e48a4
AD
93 ) OR ttrss_feeds.last_updated IS NULL
94 OR last_updated = '1970-01-01 00:00:00')";
2c08214a
AD
95 }
96
97 // Test if feed is currently being updated by another process.
98 if (DB_TYPE == "pgsql") {
99 $updstart_thresh_qpart = "AND (ttrss_feeds.last_update_started IS NULL OR ttrss_feeds.last_update_started < NOW() - INTERVAL '5 minutes')";
100 } else {
101 $updstart_thresh_qpart = "AND (ttrss_feeds.last_update_started IS NULL OR ttrss_feeds.last_update_started < DATE_SUB(NOW(), INTERVAL 5 MINUTE))";
102 }
103
104 // Test if there is a limit to number of updated feeds
105 $query_limit = "";
106 if($limit) $query_limit = sprintf("LIMIT %d", $limit);
107
108 $random_qpart = sql_random_function();
109
110 // We search for feed needing update.
111 $result = db_query($link, "SELECT ttrss_feeds.feed_url,ttrss_feeds.id, ttrss_feeds.owner_uid,
112 ".SUBSTRING_FOR_DATE."(ttrss_feeds.last_updated,1,19) AS last_updated,
113 ttrss_feeds.update_interval
114 FROM
115 ttrss_feeds, ttrss_users, ttrss_user_prefs
116 WHERE
117 ttrss_feeds.owner_uid = ttrss_users.id
118 AND ttrss_users.id = ttrss_user_prefs.owner_uid
119 AND ttrss_user_prefs.pref_name = 'DEFAULT_UPDATE_INTERVAL'
120 $login_thresh_qpart $update_limit_qpart
121 $updstart_thresh_qpart
122 ORDER BY $random_qpart $query_limit");
123
124 $user_prefs_cache = array();
125
126 if($debug) _debug(sprintf("Scheduled %d feeds to update...\n", db_num_rows($result)));
127
128 // Here is a little cache magic in order to minimize risk of double feed updates.
129 $feeds_to_update = array();
130 while ($line = db_fetch_assoc($result)) {
131 $feeds_to_update[$line['id']] = $line;
132 }
133
134 // We update the feed last update started date before anything else.
135 // There is no lag due to feed contents downloads
136 // It prevent an other process to update the same feed.
137 $feed_ids = array_keys($feeds_to_update);
138 if($feed_ids) {
139 db_query($link, sprintf("UPDATE ttrss_feeds SET last_update_started = NOW()
140 WHERE id IN (%s)", implode(',', $feed_ids)));
141 }
142
3c696512 143 expire_cached_files($debug);
2a91b6ff 144 expire_lock_files($debug);
3c696512 145
2c08214a
AD
146 // For each feed, we call the feed update function.
147 while ($line = array_pop($feeds_to_update)) {
148
149 if($debug) _debug("Feed: " . $line["feed_url"] . ", " . $line["last_updated"]);
150
151 update_rss_feed($link, $line["id"], true);
152
153 sleep(1); // prevent flood (FIXME make this an option?)
154 }
155
f1611683
AD
156 require_once "digest.php";
157
2c08214a 158 // Send feed digests by email if needed.
8437c066 159 send_headlines_digests($link, $debug);
2c08214a
AD
160
161 } // function update_daemon_common
162
87764a50 163 // ignore_daemon is not used
0e4a7d7a 164 function update_rss_feed($link, $feed, $ignore_daemon = false, $no_cache = false,
2c08214a
AD
165 $override_url = false) {
166
167 require_once "lib/simplepie/simplepie.inc";
2c08214a 168
2c08214a
AD
169 $debug_enabled = defined('DAEMON_EXTENDED_DEBUG') || $_REQUEST['xdebug'];
170
2c08214a
AD
171 if ($debug_enabled) {
172 _debug("update_rss_feed: start");
173 }
174
87764a50 175 $result = db_query($link, "SELECT id,update_interval,auth_login,
0a3fd79b 176 feed_url,auth_pass,cache_images,last_updated,
87764a50
AD
177 mark_unread_on_update, owner_uid, update_on_checksum_change,
178 pubsub_state
179 FROM ttrss_feeds WHERE id = '$feed'");
2c08214a
AD
180
181 if (db_num_rows($result) == 0) {
182 if ($debug_enabled) {
183 _debug("update_rss_feed: feed $feed NOT FOUND/SKIPPED");
184 }
185 return false;
186 }
187
2c08214a
AD
188 $last_updated = db_fetch_result($result, 0, "last_updated");
189 $owner_uid = db_fetch_result($result, 0, "owner_uid");
190 $mark_unread_on_update = sql_bool_to_bool(db_fetch_result($result,
191 0, "mark_unread_on_update"));
192 $update_on_checksum_change = sql_bool_to_bool(db_fetch_result($result,
193 0, "update_on_checksum_change"));
194 $pubsub_state = db_fetch_result($result, 0, "pubsub_state");
195
196 db_query($link, "UPDATE ttrss_feeds SET last_update_started = NOW()
197 WHERE id = '$feed'");
198
199 $auth_login = db_fetch_result($result, 0, "auth_login");
200 $auth_pass = db_fetch_result($result, 0, "auth_pass");
201
2c08214a
AD
202 $cache_images = sql_bool_to_bool(db_fetch_result($result, 0, "cache_images"));
203 $fetch_url = db_fetch_result($result, 0, "feed_url");
204
2c08214a
AD
205 $feed = db_escape_string($feed);
206
207 if ($auth_login && $auth_pass ){
208 $url_parts = array();
209 preg_match("/(^[^:]*):\/\/(.*)/", $fetch_url, $url_parts);
210
211 if ($url_parts[1] && $url_parts[2]) {
212 $fetch_url = $url_parts[1] . "://$auth_login:$auth_pass@" . $url_parts[2];
213 }
214
215 }
216
217 if ($override_url)
218 $fetch_url = $override_url;
219
220 if ($debug_enabled) {
221 _debug("update_rss_feed: fetching [$fetch_url]...");
222 }
223
0e4a7d7a
AD
224 // Ignore cache if new feed or manual update.
225 $cache_age = (is_null($last_updated) || $last_updated == '1970-01-01 00:00:00') ?
226 -1 : get_feed_update_interval($link, $feed) * 60;
2c08214a 227
19b3992b 228 $simplepie_cache_dir = CACHE_DIR . "/simplepie";
2c08214a 229
19b3992b
AD
230 if (!is_dir($simplepie_cache_dir)) {
231 mkdir($simplepie_cache_dir);
232 }
2c08214a 233
19b3992b
AD
234 $rss = new SimplePie();
235 $rss->set_useragent(SELF_USER_AGENT);
236 $rss->set_timeout($no_cache ? 15 : 60);
237 $rss->set_feed_url($fetch_url);
238 $rss->set_output_encoding('UTF-8');
ba946ba2 239 $rss->force_feed(true);
2c08214a 240
19b3992b
AD
241 if ($debug_enabled) {
242 _debug("feed update interval (sec): " .
243 get_feed_update_interval($link, $feed)*60);
244 }
2c08214a 245
19b3992b 246 $rss->enable_cache(!$no_cache);
2c08214a 247
19b3992b
AD
248 if (!$no_cache) {
249 $rss->set_cache_location($simplepie_cache_dir);
250 $rss->set_cache_duration($cache_age);
2c08214a
AD
251 }
252
19b3992b
AD
253 $rss->init();
254
2c08214a
AD
255// print_r($rss);
256
257 if ($debug_enabled) {
258 _debug("update_rss_feed: fetch done, parsing...");
259 }
260
261 $feed = db_escape_string($feed);
262
19b3992b 263 if (!$rss->error()) {
2c08214a 264
d2a421e3
AD
265 // We use local pluginhost here because we need to load different per-user feed plugins
266 $user_plugins = get_pref($link, "_ENABLED_PLUGINS", $owner_uid);
267
268 $pluginhost = new PluginHost($link);
269
270 $pluginhost->load(PLUGINS, $pluginhost::KIND_ALL);
455b1401
AD
271 $pluginhost->load($user_plugins, $pluginhost::KIND_USER, $owner_uid);
272 $pluginhost->load_data();
d2a421e3 273
4412b877
AD
274 $pluginhost->run_hooks($pluginhost::HOOK_FEED_PARSED, "hook_feed_parsed", $rss);
275
2c08214a
AD
276 if ($debug_enabled) {
277 _debug("update_rss_feed: processing feed data...");
278 }
279
280// db_query($link, "BEGIN");
281
382268c6
AD
282 if (DB_TYPE == "pgsql") {
283 $favicon_interval_qpart = "favicon_last_checked < NOW() - INTERVAL '12 hour'";
284 } else {
285 $favicon_interval_qpart = "favicon_last_checked < DATE_SUB(NOW(), INTERVAL 12 HOUR)";
286 }
287
750cfcd2 288 $result = db_query($link, "SELECT title,site_url,owner_uid,
382268c6
AD
289 (favicon_last_checked IS NULL OR $favicon_interval_qpart) AS
290 favicon_needs_check
2c08214a
AD
291 FROM ttrss_feeds WHERE id = '$feed'");
292
293 $registered_title = db_fetch_result($result, 0, "title");
2c08214a 294 $orig_site_url = db_fetch_result($result, 0, "site_url");
382268c6
AD
295 $favicon_needs_check = sql_bool_to_bool(db_fetch_result($result, 0,
296 "favicon_needs_check"));
2c08214a
AD
297
298 $owner_uid = db_fetch_result($result, 0, "owner_uid");
299
bb566dc2 300 $site_url = db_escape_string(mb_substr(rewrite_relative_url($fetch_url, $rss->get_link()), 0, 245));
2c08214a
AD
301
302 if ($debug_enabled) {
303 _debug("update_rss_feed: checking favicon...");
304 }
305
f2798eb6
AD
306 if ($favicon_needs_check) {
307 check_feed_favicon($site_url, $feed, $link);
308
309 db_query($link, "UPDATE ttrss_feeds SET favicon_last_checked = NOW()
310 WHERE id = '$feed'");
311 }
2c08214a
AD
312
313 if (!$registered_title || $registered_title == "[Unknown]") {
314
19b3992b 315 $feed_title = db_escape_string($rss->get_title());
2c08214a
AD
316
317 if ($debug_enabled) {
318 _debug("update_rss_feed: registering title: $feed_title");
319 }
320
321 db_query($link, "UPDATE ttrss_feeds SET
322 title = '$feed_title' WHERE id = '$feed'");
323 }
324
0cf81637 325 if ($site_url && $orig_site_url != $site_url) {
2c08214a
AD
326 db_query($link, "UPDATE ttrss_feeds SET
327 site_url = '$site_url' WHERE id = '$feed'");
328 }
329
2c08214a 330 if ($debug_enabled) {
b24504b1 331 _debug("update_rss_feed: loading filters & labels...");
2c08214a
AD
332 }
333
334 $filters = load_filters($link, $feed, $owner_uid);
b24504b1 335 $labels = get_all_labels($link, $owner_uid);
2c08214a 336
67bd0b1f
AD
337 if ($debug_enabled) {
338 //print_r($filters);
339 _debug("update_rss_feed: " . count($filters) . " filters loaded.");
340 }
2c08214a 341
19b3992b 342 $items = $rss->get_items();
2c08214a 343
19b3992b 344 if (!is_array($items)) {
2c08214a 345 if ($debug_enabled) {
19b3992b 346 _debug("update_rss_feed: no articles found.");
2c08214a
AD
347 }
348
349 db_query($link, "UPDATE ttrss_feeds
350 SET last_updated = NOW(), last_error = '' WHERE id = '$feed'");
351
352 return; // no articles
353 }
354
355 if ($pubsub_state != 2 && PUBSUBHUBBUB_ENABLED) {
356
357 if ($debug_enabled) _debug("update_rss_feed: checking for PUSH hub...");
358
359 $feed_hub_url = false;
2c08214a 360
19b3992b 361 $links = $rss->get_links('hub');
2c08214a 362
19b3992b
AD
363 if ($links && is_array($links)) {
364 foreach ($links as $l) {
365 $feed_hub_url = $l;
366 break;
2c08214a
AD
367 }
368 }
369
370 if ($debug_enabled) _debug("update_rss_feed: feed hub url: $feed_hub_url");
371
372 if ($feed_hub_url && function_exists('curl_init') &&
373 !ini_get("open_basedir")) {
374
375 require_once 'lib/pubsubhubbub/subscriber.php';
376
377 $callback_url = get_self_url_prefix() .
378 "/public.php?op=pubsub&id=$feed";
379
380 $s = new Subscriber($feed_hub_url, $callback_url);
381
382 $rc = $s->subscribe($fetch_url);
383
384 if ($debug_enabled)
385 _debug("update_rss_feed: feed hub url found, subscribe request sent.");
386
387 db_query($link, "UPDATE ttrss_feeds SET pubsub_state = 1
388 WHERE id = '$feed'");
389 }
390 }
391
392 if ($debug_enabled) {
393 _debug("update_rss_feed: processing articles...");
394 }
395
19b3992b 396 foreach ($items as $item) {
2c08214a
AD
397 if ($_REQUEST['xdebug'] == 2) {
398 print_r($item);
399 }
400
19b3992b
AD
401 $entry_guid = $item->get_id();
402 if (!$entry_guid) $entry_guid = $item->get_link();
403 if (!$entry_guid) $entry_guid = make_guid_from_title($item->get_title());
2c08214a
AD
404
405 if ($debug_enabled) {
406 _debug("update_rss_feed: guid $entry_guid");
407 }
408
409 if (!$entry_guid) continue;
410
3a4c8973
AD
411 $entry_guid = "$owner_uid,$entry_guid";
412
2c08214a
AD
413 $entry_timestamp = "";
414
19b3992b 415 $entry_timestamp = strtotime($item->get_date());
2c08214a 416
19b3992b 417 if ($entry_timestamp == -1 || !$entry_timestamp) {
2c08214a
AD
418 $entry_timestamp = time();
419 $no_orig_date = 'true';
420 } else {
421 $no_orig_date = 'false';
422 }
423
424 $entry_timestamp_fmt = strftime("%Y/%m/%d %H:%M:%S", $entry_timestamp);
425
426 if ($debug_enabled) {
427 _debug("update_rss_feed: date $entry_timestamp [$entry_timestamp_fmt]");
428 }
429
19b3992b 430 $entry_title = $item->get_title();
1b35d30c
AD
431
432 $entry_link = rewrite_relative_url($site_url, htmlspecialchars_decode($item->get_link()));
2c08214a
AD
433
434 if ($debug_enabled) {
435 _debug("update_rss_feed: title $entry_title");
436 _debug("update_rss_feed: link $entry_link");
437 }
438
439 if (!$entry_title) $entry_title = date("Y-m-d H:i:s", $entry_timestamp);;
440
19b3992b
AD
441 $entry_content = $item->get_content();
442 if (!$entry_content) $entry_content = $item->get_description();
2c08214a
AD
443
444 if ($_REQUEST["xdebug"] == 2) {
445 print "update_rss_feed: content: ";
487f0750 446 print $entry_content;
3c696512 447 print "\n";
2c08214a
AD
448 }
449
19b3992b 450 $entry_comments = $item->data["comments"];
2c08214a 451
19b3992b
AD
452 if ($item->get_author()) {
453 $entry_author_item = $item->get_author();
454 $entry_author = $entry_author_item->get_name();
455 if (!$entry_author) $entry_author = $entry_author_item->get_email();
2c08214a 456
19b3992b 457 $entry_author = db_escape_string($entry_author);
2c08214a
AD
458 }
459
bb566dc2 460 $entry_guid = db_escape_string(mb_substr($entry_guid, 0, 245));
2c08214a 461
bb566dc2
AD
462 $entry_comments = db_escape_string(mb_substr($entry_comments, 0, 245));
463 $entry_author = db_escape_string(mb_substr($entry_author, 0, 245));
2c08214a 464
19b3992b 465 $num_comments = $item->get_item_tags('http://purl.org/rss/1.0/modules/slash/', 'comments');
83e6e313 466
19b3992b
AD
467 if (is_array($num_comments) && is_array($num_comments[0])) {
468 $num_comments = (int) $num_comments[0]["data"];
2c08214a 469 } else {
19b3992b 470 $num_comments = 0;
2c08214a
AD
471 }
472
2c08214a 473 if ($debug_enabled) {
83e6e313 474 _debug("update_rss_feed: num_comments: $num_comments");
2c08214a
AD
475 _debug("update_rss_feed: looking for tags [1]...");
476 }
477
478 // parse <category> entries into tags
479
480 $additional_tags = array();
481
19b3992b 482 $additional_tags_src = $item->get_categories();
2c08214a 483
19b3992b
AD
484 if (is_array($additional_tags_src)) {
485 foreach ($additional_tags_src as $tobj) {
486 array_push($additional_tags, $tobj->get_term());
2c08214a 487 }
19b3992b 488 }
2c08214a 489
19b3992b
AD
490 if ($debug_enabled) {
491 _debug("update_rss_feed: category tags:");
492 print_r($additional_tags);
2c08214a
AD
493 }
494
495 if ($debug_enabled) {
496 _debug("update_rss_feed: looking for tags [2]...");
497 }
498
fa6fbd36 499 $entry_tags = array_unique($additional_tags);
2c08214a
AD
500
501 for ($i = 0; $i < count($entry_tags); $i++)
502 $entry_tags[$i] = mb_strtolower($entry_tags[$i], 'utf-8');
503
504 if ($debug_enabled) {
6aff7845
AD
505 //_debug("update_rss_feed: unfiltered tags found:");
506 //print_r($entry_tags);
2c08214a
AD
507 }
508
2c08214a
AD
509 if ($debug_enabled) {
510 _debug("update_rss_feed: done collecting data [TITLE:$entry_title]");
511 }
512
cc85704f 513 // TODO: less memory-hungry implementation
19c73507 514
19b3992b
AD
515 if ($debug_enabled) {
516 _debug("update_rss_feed: applying plugin filters..");
517 }
cc85704f 518
b30abdad
AD
519 // Todo unify with id checking below
520 $result = db_query($link, "SELECT plugin_data FROM ttrss_entries
521 WHERE guid = '".db_escape_string($entry_guid)."'");
522
523 if (db_num_rows($result) != 0) {
524 $entry_plugin_data = db_fetch_result($result, 0, "plugin_data");
525 } else {
526 $entry_plugin_data = "";
527 }
528
455b1401 529 $article = array("owner_uid" => $owner_uid, // read only
b30abdad 530 "guid" => $entry_guid, // read only
19b3992b
AD
531 "title" => $entry_title,
532 "content" => $entry_content,
533 "link" => $entry_link,
534 "tags" => $entry_tags,
b30abdad 535 "plugin_data" => $entry_plugin_data,
19b3992b 536 "author" => $entry_author);
cc85704f 537
19b3992b
AD
538 foreach ($pluginhost->get_hooks($pluginhost::HOOK_ARTICLE_FILTER) as $plugin) {
539 $article = $plugin->hook_article_filter($article);
cc85704f
AD
540 }
541
19b3992b 542 $entry_tags = $article["tags"];
b30abdad 543 $entry_guid = db_escape_string($entry_guid);
1b35d30c
AD
544 $entry_content = db_escape_string($article["content"], false);
545 $entry_title = db_escape_string($article["title"]);
546 $entry_author = db_escape_string($article["author"]);
547 $entry_link = db_escape_string($article["link"]);
b30abdad
AD
548 $entry_plugin_data = db_escape_string($article["plugin_data"]);
549
550 if ($debug_enabled) {
551 _debug("update_rss_feed: plugin data: $entry_plugin_data");
552 }
2bbd6994 553
0a3fd79b
AD
554 if ($cache_images && is_writable(CACHE_DIR . '/images'))
555 $entry_content = cache_images($entry_content, $site_url, $debug_enabled);
556
9e222305 557 $content_hash = "SHA1:" . sha1($entry_content);
cc85704f 558
2c08214a
AD
559 db_query($link, "BEGIN");
560
9e222305
AD
561 $result = db_query($link, "SELECT id FROM ttrss_entries
562 WHERE guid = '$entry_guid'");
563
2c08214a
AD
564 if (db_num_rows($result) == 0) {
565
566 if ($debug_enabled) {
9e222305 567 _debug("update_rss_feed: base guid [$entry_guid] not found");
2c08214a
AD
568 }
569
570 // base post entry does not exist, create it
571
572 $result = db_query($link,
573 "INSERT INTO ttrss_entries
574 (title,
575 guid,
576 link,
577 updated,
578 content,
579 content_hash,
87764a50 580 cached_content,
2c08214a
AD
581 no_orig_date,
582 date_updated,
583 date_entered,
584 comments,
585 num_comments,
b30abdad 586 plugin_data,
2c08214a
AD
587 author)
588 VALUES
589 ('$entry_title',
590 '$entry_guid',
591 '$entry_link',
592 '$entry_timestamp_fmt',
593 '$entry_content',
594 '$content_hash',
0a3fd79b 595 '',
2c08214a
AD
596 $no_orig_date,
597 NOW(),
598 NOW(),
599 '$entry_comments',
600 '$num_comments',
b30abdad 601 '$entry_plugin_data',
2c08214a 602 '$entry_author')");
e8291805
AD
603
604 $article_labels = array();
605
2c08214a
AD
606 } else {
607 // we keep encountering the entry in feeds, so we need to
608 // update date_updated column so that we don't get horrible
609 // dupes when the entry gets purged and reinserted again e.g.
610 // in the case of SLOW SLOW OMG SLOW updating feeds
611
612 $base_entry_id = db_fetch_result($result, 0, "id");
613
614 db_query($link, "UPDATE ttrss_entries SET date_updated = NOW()
615 WHERE id = '$base_entry_id'");
e8291805
AD
616
617 $article_labels = get_article_labels($link, $base_entry_id, $owner_uid);
2c08214a
AD
618 }
619
620 // now it should exist, if not - bad luck then
621
622 $result = db_query($link, "SELECT
b30abdad 623 id,content_hash,no_orig_date,title,plugin_data,
2c08214a
AD
624 ".SUBSTRING_FOR_DATE."(date_updated,1,19) as date_updated,
625 ".SUBSTRING_FOR_DATE."(updated,1,19) as updated,
0a3fd79b 626 num_comments
2c08214a
AD
627 FROM
628 ttrss_entries
629 WHERE guid = '$entry_guid'");
630
631 $entry_ref_id = 0;
632 $entry_int_id = 0;
633
634 if (db_num_rows($result) == 1) {
635
636 if ($debug_enabled) {
9e222305 637 _debug("update_rss_feed: base guid [$entry_guid] found, checking for user record");
2c08214a
AD
638 }
639
640 // this will be used below in update handler
641 $orig_content_hash = db_fetch_result($result, 0, "content_hash");
642 $orig_title = db_fetch_result($result, 0, "title");
643 $orig_num_comments = db_fetch_result($result, 0, "num_comments");
644 $orig_date_updated = strtotime(db_fetch_result($result,
645 0, "date_updated"));
b30abdad 646 $orig_plugin_data = db_fetch_result($result, 0, "plugin_data");
2c08214a
AD
647
648 $ref_id = db_fetch_result($result, 0, "id");
649 $entry_ref_id = $ref_id;
650
651 // check for user post link to main table
652
653 // do we allow duplicate posts with same GUID in different feeds?
654 if (get_pref($link, "ALLOW_DUPLICATE_POSTS", $owner_uid, false)) {
655 $dupcheck_qpart = "AND (feed_id = '$feed' OR feed_id IS NULL)";
656 } else {
657 $dupcheck_qpart = "";
658 }
659
660 /* Collect article tags here so we could filter by them: */
661
662 $article_filters = get_article_filters($filters, $entry_title,
4021d61a 663 $entry_content, $entry_link, $entry_timestamp, $entry_author,
2c08214a
AD
664 $entry_tags);
665
666 if ($debug_enabled) {
667 _debug("update_rss_feed: article filters: ");
668 if (count($article_filters) != 0) {
669 print_r($article_filters);
670 }
671 }
672
673 if (find_article_filter($article_filters, "filter")) {
674 db_query($link, "COMMIT"); // close transaction in progress
675 continue;
676 }
677
678 $score = calculate_article_score($article_filters);
679
680 if ($debug_enabled) {
681 _debug("update_rss_feed: initial score: $score");
682 }
683
684 $query = "SELECT ref_id, int_id FROM ttrss_user_entries WHERE
685 ref_id = '$ref_id' AND owner_uid = '$owner_uid'
686 $dupcheck_qpart";
687
688// if ($_REQUEST["xdebug"]) print "$query\n";
689
690 $result = db_query($link, $query);
691
692 // okay it doesn't exist - create user entry
693 if (db_num_rows($result) == 0) {
694
695 if ($debug_enabled) {
696 _debug("update_rss_feed: user record not found, creating...");
697 }
698
699 if ($score >= -500 && !find_article_filter($article_filters, 'catchup')) {
700 $unread = 'true';
701 $last_read_qpart = 'NULL';
702 } else {
703 $unread = 'false';
704 $last_read_qpart = 'NOW()';
705 }
706
707 if (find_article_filter($article_filters, 'mark') || $score > 1000) {
708 $marked = 'true';
709 } else {
710 $marked = 'false';
711 }
712
713 if (find_article_filter($article_filters, 'publish')) {
714 $published = 'true';
715 } else {
716 $published = 'false';
717 }
718
2ea9bbfd
AD
719 // N-grams
720
721 if (DB_TYPE == "pgsql" and defined('_NGRAM_TITLE_DUPLICATE_THRESHOLD')) {
722
723 $result = db_query($link, "SELECT COUNT(*) AS similar FROM
724 ttrss_entries,ttrss_user_entries
725 WHERE ref_id = id AND updated >= NOW() - INTERVAL '7 day'
726 AND similarity(title, '$entry_title') >= "._NGRAM_TITLE_DUPLICATE_THRESHOLD."
727 AND owner_uid = $owner_uid");
728
729 $ngram_similar = db_fetch_result($result, 0, "similar");
730
731 if ($debug_enabled) {
732 _debug("update_rss_feed: N-gram similar results: $ngram_similar");
733 }
734
735 if ($ngram_similar > 0) {
736 $unread = 'false';
737 }
738 }
739
2c08214a
AD
740 $result = db_query($link,
741 "INSERT INTO ttrss_user_entries
742 (ref_id, owner_uid, feed_id, unread, last_read, marked,
743 published, score, tag_cache, label_cache, uuid)
744 VALUES ('$ref_id', '$owner_uid', '$feed', $unread,
745 $last_read_qpart, $marked, $published, '$score', '', '', '')");
746
747 if (PUBSUBHUBBUB_HUB && $published == 'true') {
748 $rss_link = get_self_url_prefix() .
749 "/public.php?op=rss&id=-2&key=" .
750 get_feed_access_key($link, -2, false, $owner_uid);
751
752 $p = new Publisher(PUBSUBHUBBUB_HUB);
753
754 $pubsub_result = $p->publish_update($rss_link);
755 }
756
757 $result = db_query($link,
758 "SELECT int_id FROM ttrss_user_entries WHERE
759 ref_id = '$ref_id' AND owner_uid = '$owner_uid' AND
760 feed_id = '$feed' LIMIT 1");
761
762 if (db_num_rows($result) == 1) {
763 $entry_int_id = db_fetch_result($result, 0, "int_id");
764 }
765 } else {
766 if ($debug_enabled) {
767 _debug("update_rss_feed: user record FOUND");
768 }
769
770 $entry_ref_id = db_fetch_result($result, 0, "ref_id");
771 $entry_int_id = db_fetch_result($result, 0, "int_id");
772 }
773
774 if ($debug_enabled) {
775 _debug("update_rss_feed: RID: $entry_ref_id, IID: $entry_int_id");
776 }
777
778 $post_needs_update = false;
779 $update_insignificant = false;
780
781 if ($orig_num_comments != $num_comments) {
782 $post_needs_update = true;
783 $update_insignificant = true;
784 }
785
b30abdad
AD
786 if ($entry_plugin_data != $orig_plugin_data) {
787 $post_needs_update = true;
788 $update_insignificant = true;
789 }
790
2c08214a
AD
791 if ($content_hash != $orig_content_hash) {
792 $post_needs_update = true;
793 $update_insignificant = false;
794 }
795
796 if (db_escape_string($orig_title) != $entry_title) {
797 $post_needs_update = true;
798 $update_insignificant = false;
799 }
800
801 // if post needs update, update it and mark all user entries
802 // linking to this post as updated
803 if ($post_needs_update) {
804
805 if (defined('DAEMON_EXTENDED_DEBUG')) {
806 _debug("update_rss_feed: post $entry_guid needs update...");
807 }
808
809// print "<!-- post $orig_title needs update : $post_needs_update -->";
810
811 db_query($link, "UPDATE ttrss_entries
812 SET title = '$entry_title', content = '$entry_content',
813 content_hash = '$content_hash',
814 updated = '$entry_timestamp_fmt',
b30abdad
AD
815 num_comments = '$num_comments',
816 plugin_data = '$entry_plugin_data'
2c08214a
AD
817 WHERE id = '$ref_id'");
818
819 if (!$update_insignificant) {
820 if ($mark_unread_on_update) {
821 db_query($link, "UPDATE ttrss_user_entries
822 SET last_read = null, unread = true WHERE ref_id = '$ref_id'");
823 } else if ($update_on_checksum_change) {
824 db_query($link, "UPDATE ttrss_user_entries
825 SET last_read = null WHERE ref_id = '$ref_id'
826 AND unread = false");
827 }
828 }
829 }
830 }
831
832 db_query($link, "COMMIT");
833
834 if ($debug_enabled) {
835 _debug("update_rss_feed: assigning labels...");
836 }
837
92c14e9d 838 assign_article_to_label_filters($link, $entry_ref_id, $article_filters,
b24504b1 839 $owner_uid, $article_labels);
2c08214a
AD
840
841 if ($debug_enabled) {
842 _debug("update_rss_feed: looking for enclosures...");
843 }
844
845 // enclosures
846
847 $enclosures = array();
848
19b3992b 849 $encs = $item->get_enclosures();
2c08214a 850
19b3992b
AD
851 if (is_array($encs)) {
852 foreach ($encs as $e) {
853 $e_item = array(
854 $e->link, $e->type, $e->length);
2c08214a 855 array_push($enclosures, $e_item);
2c08214a
AD
856 }
857 }
858
2c08214a
AD
859 if ($debug_enabled) {
860 _debug("update_rss_feed: article enclosures:");
861 print_r($enclosures);
862 }
863
864 db_query($link, "BEGIN");
865
866 foreach ($enclosures as $enc) {
867 $enc_url = db_escape_string($enc[0]);
868 $enc_type = db_escape_string($enc[1]);
869 $enc_dur = db_escape_string($enc[2]);
870
871 $result = db_query($link, "SELECT id FROM ttrss_enclosures
872 WHERE content_url = '$enc_url' AND post_id = '$entry_ref_id'");
873
874 if (db_num_rows($result) == 0) {
875 db_query($link, "INSERT INTO ttrss_enclosures
876 (content_url, content_type, title, duration, post_id) VALUES
877 ('$enc_url', '$enc_type', '', '$enc_dur', '$entry_ref_id')");
878 }
879 }
880
881 db_query($link, "COMMIT");
882
883 // check for manual tags (we have to do it here since they're loaded from filters)
884
885 foreach ($article_filters as $f) {
6aff7845 886 if ($f["type"] == "tag") {
2c08214a 887
6aff7845 888 $manual_tags = trim_array(explode(",", $f["param"]));
2c08214a
AD
889
890 foreach ($manual_tags as $tag) {
891 if (tag_is_valid($tag)) {
892 array_push($entry_tags, $tag);
893 }
894 }
895 }
896 }
897
898 // Skip boring tags
899
900 $boring_tags = trim_array(explode(",", mb_strtolower(get_pref($link,
901 'BLACKLISTED_TAGS', $owner_uid, ''), 'utf-8')));
902
903 $filtered_tags = array();
904 $tags_to_cache = array();
905
906 if ($entry_tags && is_array($entry_tags)) {
907 foreach ($entry_tags as $tag) {
908 if (array_search($tag, $boring_tags) === false) {
909 array_push($filtered_tags, $tag);
910 }
911 }
912 }
913
914 $filtered_tags = array_unique($filtered_tags);
915
916 if ($debug_enabled) {
917 _debug("update_rss_feed: filtered article tags:");
918 print_r($filtered_tags);
919 }
920
921 // Save article tags in the database
922
923 if (count($filtered_tags) > 0) {
924
925 db_query($link, "BEGIN");
926
927 foreach ($filtered_tags as $tag) {
928
929 $tag = sanitize_tag($tag);
930 $tag = db_escape_string($tag);
931
932 if (!tag_is_valid($tag)) continue;
933
934 $result = db_query($link, "SELECT id FROM ttrss_tags
935 WHERE tag_name = '$tag' AND post_int_id = '$entry_int_id' AND
936 owner_uid = '$owner_uid' LIMIT 1");
937
938 if ($result && db_num_rows($result) == 0) {
939
940 db_query($link, "INSERT INTO ttrss_tags
941 (owner_uid,tag_name,post_int_id)
942 VALUES ('$owner_uid','$tag', '$entry_int_id')");
943 }
944
945 array_push($tags_to_cache, $tag);
946 }
947
948 /* update the cache */
949
950 $tags_to_cache = array_unique($tags_to_cache);
951
952 $tags_str = db_escape_string(join(",", $tags_to_cache));
953
954 db_query($link, "UPDATE ttrss_user_entries
955 SET tag_cache = '$tags_str' WHERE ref_id = '$entry_ref_id'
956 AND owner_uid = $owner_uid");
957
958 db_query($link, "COMMIT");
959 }
960
b24504b1
AD
961 if (get_pref($link, "AUTO_ASSIGN_LABELS", $owner_uid, false)) {
962 if ($debug_enabled) {
963 _debug("update_rss_feed: auto-assigning labels...");
964 }
965
966 foreach ($labels as $label) {
967 $caption = $label["caption"];
968
3fc6e71a 969 if (preg_match("/\b$caption\b/i", "$tags_str " . strip_tags($entry_content) . " $entry_title")) {
b24504b1
AD
970 if (!labels_contains_caption($article_labels, $caption)) {
971 label_add_article($link, $entry_ref_id, $caption, $owner_uid);
972 }
973 }
974 }
975 }
976
2c08214a
AD
977 if ($debug_enabled) {
978 _debug("update_rss_feed: article processed");
979 }
980 }
981
982 if (!$last_updated) {
983 if ($debug_enabled) {
984 _debug("update_rss_feed: new feed, catching it up...");
985 }
986 catchup_feed($link, $feed, false, $owner_uid);
987 }
988
989 if ($debug_enabled) {
990 _debug("purging feed...");
991 }
992
993 purge_feed($link, $feed, 0, $debug_enabled);
994
995 db_query($link, "UPDATE ttrss_feeds
f2798eb6 996 SET last_updated = NOW(), last_error = '' WHERE id = '$feed'");
2c08214a
AD
997
998// db_query($link, "COMMIT");
999
1000 } else {
1001
bb566dc2 1002 $error_msg = db_escape_string(mb_substr($rss->error(), 0, 245));
2c08214a
AD
1003
1004 if ($debug_enabled) {
1005 _debug("update_rss_feed: error fetching feed: $error_msg");
1006 }
1007
2c08214a
AD
1008 db_query($link,
1009 "UPDATE ttrss_feeds SET last_error = '$error_msg',
1010 last_updated = NOW() WHERE id = '$feed'");
1011 }
1012
19b3992b 1013 unset($rss);
2c08214a
AD
1014
1015 if ($debug_enabled) {
1016 _debug("update_rss_feed: done");
1017 }
1018
1019 }
1020
3c696512
AD
1021 function cache_images($html, $site_url, $debug) {
1022 $cache_dir = CACHE_DIR . "/images";
1023
1024 libxml_use_internal_errors(true);
1025
1026 $charset_hack = '<head>
1027 <meta http-equiv="Content-Type" content="text/html; charset=utf-8"/>
1028 </head>';
1029
1030 $doc = new DOMDocument();
1031 $doc->loadHTML($charset_hack . $html);
1032 $xpath = new DOMXPath($doc);
1033
1034 $entries = $xpath->query('(//img[@src])');
1035
1036 foreach ($entries as $entry) {
1037 if ($entry->hasAttribute('src')) {
1038 $src = rewrite_relative_url($site_url, $entry->getAttribute('src'));
1039
1040 $local_filename = CACHE_DIR . "/images/" . sha1($src) . ".png";
1041
1042 if ($debug) _debug("cache_images: downloading: $src to $local_filename");
1043
1044 if (!file_exists($local_filename)) {
1045 $file_content = fetch_file_contents($src);
1046
b8379e69 1047 if ($file_content && strlen($file_content) > 1024) {
3c696512
AD
1048 file_put_contents($local_filename, $file_content);
1049 }
1050 }
1051
1052 if (file_exists($local_filename)) {
1053 $entry->setAttribute('src', SELF_URL_PATH . '/image.php?url=' .
487f0750 1054 base64_encode($src));
3c696512
AD
1055 }
1056 }
1057 }
1058
1059 $node = $doc->getElementsByTagName('body')->item(0);
1060
7b8ff151 1061 return $doc->saveXML($node, LIBXML_NOEMPTYTAG);
3c696512
AD
1062 }
1063
2a91b6ff
AD
1064 function expire_lock_files($debug) {
1065 if ($debug) _debug("Removing old lock files...");
1066
1067 $num_deleted = 0;
1068
1069 if (is_writable(LOCK_DIRECTORY)) {
1070 $files = glob(LOCK_DIRECTORY . "/*.lock");
1071
1072 if ($files) {
1073 foreach ($files as $file) {
1074 if (!file_is_locked($file) && time() - filemtime($file) > 86400*2) {
1075 unlink($file);
1076 ++$num_deleted;
1077 }
1078 }
1079 }
1080 }
1081
1082 if ($debug) _debug("Removed $num_deleted files.");
1083 }
1084
3c696512 1085 function expire_cached_files($debug) {
19b3992b 1086 foreach (array("simplepie", "images", "export") as $dir) {
3c696512 1087 $cache_dir = CACHE_DIR . "/$dir";
2c08214a 1088
3c696512 1089 if ($debug) _debug("Expiring $cache_dir");
2c08214a 1090
3c696512
AD
1091 $num_deleted = 0;
1092
1093 if (is_writable($cache_dir)) {
1094 $files = glob("$cache_dir/*");
1095
2a91b6ff 1096 if ($files) {
2ab20c31
AD
1097 foreach ($files as $file) {
1098 if (time() - filemtime($file) > 86400*7) {
1099 unlink($file);
3c696512 1100
2ab20c31
AD
1101 ++$num_deleted;
1102 }
3c696512
AD
1103 }
1104 }
2a91b6ff 1105 }
3c696512
AD
1106
1107 if ($debug) _debug("Removed $num_deleted files.");
1108 }
1109 }
2c08214a 1110
a3e0bdcf
AD
1111 /**
1112 * Source: http://www.php.net/manual/en/function.parse-url.php#104527
1113 * Returns the url query as associative array
1114 *
1115 * @param string query
1116 * @return array params
1117 */
1118 function convertUrlQuery($query) {
1119 $queryParts = explode('&', $query);
1120
1121 $params = array();
1122
1123 foreach ($queryParts as $param) {
1124 $item = explode('=', $param);
1125 $params[$item[0]] = $item[1];
1126 }
1127
1128 return $params;
1129 }
92c14e9d
AD
1130
1131 function get_article_filters($filters, $title, $content, $link, $timestamp, $author, $tags) {
1132 $matches = array();
1133
1134 foreach ($filters as $filter) {
1135 $match_any_rule = $filter["match_any_rule"];
1136 $filter_match = false;
1137
1138 foreach ($filter["rules"] as $rule) {
1139 $match = false;
1140 $reg_exp = $rule["reg_exp"];
1141
1142 if (!$reg_exp)
1143 continue;
1144
1145 switch ($rule["type"]) {
1146 case "title":
1147 $match = @preg_match("/$reg_exp/i", $title);
1148 break;
1149 case "content":
d03ae73e
AD
1150 // we don't need to deal with multiline regexps
1151 $content = preg_replace("/[\r\n\t]/", "", $content);
1152
92c14e9d
AD
1153 $match = @preg_match("/$reg_exp/i", $content);
1154 break;
1155 case "both":
d03ae73e
AD
1156 // we don't need to deal with multiline regexps
1157 $content = preg_replace("/[\r\n\t]/", "", $content);
1158
72d1d067 1159 $match = (@preg_match("/$reg_exp/i", $title) || @preg_match("/$reg_exp/i", $content));
92c14e9d
AD
1160 break;
1161 case "link":
1162 $match = @preg_match("/$reg_exp/i", $link);
1163 break;
1164 case "author":
1165 $match = @preg_match("/$reg_exp/i", $author);
1166 break;
1167 case "tag":
1168 $tag_string = join(",", $tags);
1169 $match = @preg_match("/$reg_exp/i", $tag_string);
1170 break;
1171 }
1172
1173 if ($match_any_rule) {
1174 if ($match) {
1175 $filter_match = true;
1176 break;
1177 }
1178 } else {
1179 $filter_match = $match;
1180 if (!$match) {
1181 break;
1182 }
1183 }
1184 }
1185
1186 if ($filter_match) {
1187 foreach ($filter["actions"] AS $action) {
1188 array_push($matches, $action);
1189 }
1190 }
1191 }
1192
1193 return $matches;
1194 }
1195
1196 function find_article_filter($filters, $filter_name) {
1197 foreach ($filters as $f) {
1198 if ($f["type"] == $filter_name) {
1199 return $f;
1200 };
1201 }
1202 return false;
1203 }
1204
1205 function find_article_filters($filters, $filter_name) {
1206 $results = array();
1207
1208 foreach ($filters as $f) {
1209 if ($f["type"] == $filter_name) {
1210 array_push($results, $f);
1211 };
1212 }
1213 return $results;
1214 }
1215
1216 function calculate_article_score($filters) {
1217 $score = 0;
1218
1219 foreach ($filters as $f) {
1220 if ($f["type"] == "score") {
1221 $score += $f["param"];
1222 };
1223 }
1224 return $score;
1225 }
1226
b24504b1
AD
1227 function labels_contains_caption($labels, $caption) {
1228 foreach ($labels as $label) {
1229 if ($label[1] == $caption) {
1230 return true;
1231 }
1232 }
1233
1234 return false;
1235 }
1236
1237 function assign_article_to_label_filters($link, $id, $filters, $owner_uid, $article_labels) {
92c14e9d
AD
1238 foreach ($filters as $f) {
1239 if ($f["type"] == "label") {
b24504b1
AD
1240 if (!labels_contains_caption($article_labels, $f["param"])) {
1241 label_add_article($link, $id, $f["param"], $owner_uid);
1242 }
1243 }
92c14e9d
AD
1244 }
1245 }
87764a50
AD
1246
1247 function cache_content($link, $url, $login, $pass) {
1248
1249 $content = fetch_file_contents($url, $login, $pass);
1250
1251 if ($content) {
1252 $doc = new DOMDocument();
1253 @$doc->loadHTML($content);
1254 $xpath = new DOMXPath($doc);
1255
1256 $node = $doc->getElementsByTagName('body')->item(0);
1257
1258 if ($node) {
1259 $content = $doc->saveXML($node, LIBXML_NOEMPTYTAG);
1260
1261 return $content;
1262 }
1263 }
1264
1265 return "";
1266 }
87d7e850
AD
1267
1268 function make_guid_from_title($title) {
1269 return preg_replace("/[ \"\',.:;]/", "-",
1270 mb_strtolower(strip_tags($title), 'utf-8'));
1271 }
1272
1273
2c08214a 1274?>