]>
Commit | Line | Data |
---|---|---|
1 | <?php | |
2 | define('DAEMON_UPDATE_LOGIN_LIMIT', 30); | |
3 | define('DAEMON_FEED_LIMIT', 100); | |
4 | define('DAEMON_SLEEP_INTERVAL', 60); | |
5 | ||
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 | $subscribers = db_escape_string($link, $line["subscribers"]); | |
22 | $feed_url = db_escape_string($link, $line["feed_url"]); | |
23 | $title = db_escape_string($link, $line["title"]); | |
24 | $site_url = db_escape_string($link, $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 | ||
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 | ||
63 | define('PREFS_NO_CACHE', true); | |
64 | ||
65 | // Test if the user has loggued in recently. If not, it does not update its feeds. | |
66 | if (!SINGLE_USER_MODE && DAEMON_UPDATE_LOGIN_LIMIT > 0) { | |
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 CAST(ttrss_user_prefs.value AS INTEGER) != -1 | |
81 | AND ttrss_feeds.last_updated < NOW() - CAST((ttrss_user_prefs.value || ' minutes') AS INTERVAL) | |
82 | ) OR ( | |
83 | ttrss_feeds.update_interval > 0 | |
84 | AND ttrss_feeds.last_updated < NOW() - CAST((ttrss_feeds.update_interval || ' minutes') AS INTERVAL) | |
85 | ) OR ttrss_feeds.last_updated IS NULL | |
86 | OR last_updated = '1970-01-01 00:00:00')"; | |
87 | } else { | |
88 | $update_limit_qpart = "AND (( | |
89 | ttrss_feeds.update_interval = 0 | |
90 | AND CONVERT(ttrss_user_prefs.value, SIGNED INTEGER) != -1 | |
91 | AND ttrss_feeds.last_updated < DATE_SUB(NOW(), INTERVAL CONVERT(ttrss_user_prefs.value, SIGNED INTEGER) MINUTE) | |
92 | ) OR ( | |
93 | ttrss_feeds.update_interval > 0 | |
94 | AND ttrss_feeds.last_updated < DATE_SUB(NOW(), INTERVAL ttrss_feeds.update_interval MINUTE) | |
95 | ) OR ttrss_feeds.last_updated IS NULL | |
96 | OR last_updated = '1970-01-01 00:00:00')"; | |
97 | } | |
98 | ||
99 | // Test if feed is currently being updated by another process. | |
100 | if (DB_TYPE == "pgsql") { | |
101 | $updstart_thresh_qpart = "AND (ttrss_feeds.last_update_started IS NULL OR ttrss_feeds.last_update_started < NOW() - INTERVAL '5 minutes')"; | |
102 | } else { | |
103 | $updstart_thresh_qpart = "AND (ttrss_feeds.last_update_started IS NULL OR ttrss_feeds.last_update_started < DATE_SUB(NOW(), INTERVAL 5 MINUTE))"; | |
104 | } | |
105 | ||
106 | // Test if there is a limit to number of updated feeds | |
107 | $query_limit = ""; | |
108 | if($limit) $query_limit = sprintf("LIMIT %d", $limit); | |
109 | ||
110 | $random_qpart = sql_random_function(); | |
111 | ||
112 | // We search for feed needing update. | |
113 | $result = db_query($link, "SELECT DISTINCT ttrss_feeds.feed_url,$random_qpart | |
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...", 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($result)) { | |
131 | array_push($feeds_to_update, db_escape_string($link, $line['feed_url'])); | |
132 | } | |
133 | ||
134 | // We update the feed last update started date before anything else. | |
135 | // There is no lag due to feed contents downloads | |
136 | // It prevent an other process to update the same feed. | |
137 | ||
138 | if(count($feeds_to_update) > 0) { | |
139 | $feeds_quoted = array(); | |
140 | ||
141 | foreach ($feeds_to_update as $feed) { | |
142 | array_push($feeds_quoted, "'" . db_escape_string($link, $feed) . "'"); | |
143 | } | |
144 | ||
145 | db_query($link, sprintf("UPDATE ttrss_feeds SET last_update_started = NOW() | |
146 | WHERE feed_url IN (%s)", implode(',', $feeds_quoted))); | |
147 | } | |
148 | ||
149 | expire_cached_files($debug); | |
150 | expire_lock_files($debug); | |
151 | ||
152 | $nf = 0; | |
153 | ||
154 | // For each feed, we call the feed update function. | |
155 | foreach ($feeds_to_update as $feed) { | |
156 | if($debug) _debug("Base feed: $feed"); | |
157 | ||
158 | //update_rss_feed($link, $line["id"], true); | |
159 | ||
160 | // since we have the data cached, we can deal with other feeds with the same url | |
161 | ||
162 | $tmp_result = db_query($link, "SELECT ttrss_feeds.feed_url,ttrss_feeds.id,last_updated | |
163 | FROM ttrss_feeds, ttrss_users WHERE | |
164 | ttrss_users.id = ttrss_feeds.owner_uid AND | |
165 | feed_url = '".db_escape_string($link, $feed)."' AND | |
166 | ttrss_feeds.update_interval != -1 | |
167 | $login_thresh_qpart | |
168 | ORDER BY feed_url $query_limit"); | |
169 | ||
170 | if (db_num_rows($tmp_result) > 0) { | |
171 | while ($tline = db_fetch_assoc($tmp_result)) { | |
172 | if($debug) _debug(" => " . $tline["last_updated"] . ", " . $tline["id"]); | |
173 | update_rss_feed($link, $tline["id"], true); | |
174 | ++$nf; | |
175 | } | |
176 | } | |
177 | } | |
178 | ||
179 | require_once "digest.php"; | |
180 | ||
181 | // Send feed digests by email if needed. | |
182 | send_headlines_digests($link, $debug); | |
183 | ||
184 | return $nf; | |
185 | ||
186 | } // function update_daemon_common | |
187 | ||
188 | // ignore_daemon is not used | |
189 | function update_rss_feed($link, $feed, $ignore_daemon = false, $no_cache = false, | |
190 | $override_url = false) { | |
191 | ||
192 | require_once "lib/simplepie/simplepie.inc"; | |
193 | ||
194 | $debug_enabled = defined('DAEMON_EXTENDED_DEBUG') || $_REQUEST['xdebug']; | |
195 | ||
196 | if ($debug_enabled) { | |
197 | _debug("update_rss_feed: start"); | |
198 | } | |
199 | ||
200 | $result = db_query($link, "SELECT id,update_interval,auth_login, | |
201 | feed_url,auth_pass,cache_images,last_updated, | |
202 | mark_unread_on_update, owner_uid, | |
203 | pubsub_state | |
204 | FROM ttrss_feeds WHERE id = '$feed'"); | |
205 | ||
206 | if (db_num_rows($result) == 0) { | |
207 | if ($debug_enabled) { | |
208 | _debug("update_rss_feed: feed $feed NOT FOUND/SKIPPED"); | |
209 | } | |
210 | return false; | |
211 | } | |
212 | ||
213 | $last_updated = db_fetch_result($result, 0, "last_updated"); | |
214 | $owner_uid = db_fetch_result($result, 0, "owner_uid"); | |
215 | $mark_unread_on_update = sql_bool_to_bool(db_fetch_result($result, | |
216 | 0, "mark_unread_on_update")); | |
217 | $pubsub_state = db_fetch_result($result, 0, "pubsub_state"); | |
218 | ||
219 | db_query($link, "UPDATE ttrss_feeds SET last_update_started = NOW() | |
220 | WHERE id = '$feed'"); | |
221 | ||
222 | $auth_login = db_fetch_result($result, 0, "auth_login"); | |
223 | $auth_pass = db_fetch_result($result, 0, "auth_pass"); | |
224 | ||
225 | $cache_images = sql_bool_to_bool(db_fetch_result($result, 0, "cache_images")); | |
226 | $fetch_url = db_fetch_result($result, 0, "feed_url"); | |
227 | ||
228 | $feed = db_escape_string($link, $feed); | |
229 | ||
230 | if ($override_url) $fetch_url = $override_url; | |
231 | ||
232 | $date_feed_processed = date('Y-m-d H:i'); | |
233 | ||
234 | $cache_filename = CACHE_DIR . "/simplepie/" . sha1($fetch_url) . ".feed"; | |
235 | ||
236 | // Ignore cache if new feed or manual update. | |
237 | $cache_age = ($no_cache || is_null($last_updated) || $last_updated == '1970-01-01 00:00:00') ? | |
238 | 30 : get_feed_update_interval($link, $feed) * 60; | |
239 | ||
240 | if ($debug_enabled) { | |
241 | _debug("update_rss_feed: cache filename: $cache_filename exists: " . file_exists($cache_filename)); | |
242 | _debug("update_rss_feed: cache age: $cache_age; no cache: $no_cache"); | |
243 | } | |
244 | ||
245 | $cached_feed_data_hash = false; | |
246 | ||
247 | $rss = false; | |
248 | $rss_hash = false; | |
249 | $cache_timestamp = file_exists($cache_filename) ? filemtime($cache_filename) : 0; | |
250 | $last_updated_timestamp = strtotime($last_updated); | |
251 | ||
252 | if (file_exists($cache_filename) && | |
253 | is_readable($cache_filename) && | |
254 | !$auth_login && !$auth_pass && | |
255 | filemtime($cache_filename) > time() - $cache_age) { | |
256 | ||
257 | if ($debug_enabled) { | |
258 | _debug("update_rss_feed: using local cache."); | |
259 | } | |
260 | ||
261 | if ($cache_timestamp > $last_updated_timestamp) { | |
262 | @$rss_data = file_get_contents($cache_filename); | |
263 | ||
264 | if ($rss_data) { | |
265 | $rss_hash = sha1($rss_data); | |
266 | @$rss = unserialize($rss_data); | |
267 | } | |
268 | } else { | |
269 | if ($debug_enabled) { | |
270 | _debug("update_rss_feed: local cache valid and older than last_updated, nothing to do."); | |
271 | } | |
272 | return; | |
273 | } | |
274 | } | |
275 | ||
276 | if (!$rss) { | |
277 | ||
278 | if (!$feed_data) { | |
279 | if ($debug_enabled) { | |
280 | _debug("update_rss_feed: fetching [$fetch_url] (ts: $cache_timestamp/$last_updated_timestamp)"); | |
281 | } | |
282 | ||
283 | $feed_data = fetch_file_contents($fetch_url, false, | |
284 | $auth_login, $auth_pass, false, $no_cache ? 15 : 45, | |
285 | max($last_updated_timestamp, $cache_timestamp)); | |
286 | ||
287 | if ($debug_enabled) { | |
288 | _debug("update_rss_feed: fetch done."); | |
289 | } | |
290 | ||
291 | } | |
292 | ||
293 | if (!$feed_data) { | |
294 | global $fetch_last_error; | |
295 | global $fetch_last_error_code; | |
296 | ||
297 | if ($debug_enabled) { | |
298 | _debug("update_rss_feed: unable to fetch: $fetch_last_error [$fetch_last_error_code]"); | |
299 | } | |
300 | ||
301 | $error_escaped = ''; | |
302 | ||
303 | // If-Modified-Since | |
304 | if ($fetch_last_error_code != 304) { | |
305 | $error_escaped = db_escape_string($link, $fetch_last_error); | |
306 | } else { | |
307 | if ($debug_enabled) { | |
308 | _debug("update_rss_feed: source claims data not modified, nothing to do."); | |
309 | } | |
310 | } | |
311 | ||
312 | db_query($link, | |
313 | "UPDATE ttrss_feeds SET last_error = '$error_escaped', | |
314 | last_updated = NOW() WHERE id = '$feed'"); | |
315 | ||
316 | return; | |
317 | } | |
318 | } | |
319 | ||
320 | $pluginhost = new PluginHost($link); | |
321 | $pluginhost->set_debug($debug_enabled); | |
322 | $user_plugins = get_pref($link, "_ENABLED_PLUGINS", $owner_uid); | |
323 | ||
324 | $pluginhost->load(PLUGINS, $pluginhost::KIND_ALL); | |
325 | $pluginhost->load($user_plugins, $pluginhost::KIND_USER, $owner_uid); | |
326 | $pluginhost->load_data(); | |
327 | ||
328 | foreach ($pluginhost->get_hooks($pluginhost::HOOK_FEED_FETCHED) as $plugin) { | |
329 | $feed_data = $plugin->hook_feed_fetched($feed_data); | |
330 | } | |
331 | ||
332 | if (!$rss) { | |
333 | $rss = new SimplePie(); | |
334 | $rss->set_sanitize_class("SanitizeDummy"); | |
335 | // simplepie ignores the above and creates default sanitizer anyway, | |
336 | // so let's override it... | |
337 | $rss->sanitize = new SanitizeDummy(); | |
338 | $rss->set_output_encoding('UTF-8'); | |
339 | $rss->set_raw_data($feed_data); | |
340 | $rss->enable_cache(false); | |
341 | ||
342 | @$rss->init(); | |
343 | } | |
344 | ||
345 | // print_r($rss); | |
346 | ||
347 | $feed = db_escape_string($link, $feed); | |
348 | ||
349 | if (!$rss->error()) { | |
350 | ||
351 | // cache data for later | |
352 | if (!$auth_pass && !$auth_login && is_writable(CACHE_DIR . "/simplepie")) { | |
353 | $rss_data = serialize($rss); | |
354 | $new_rss_hash = sha1($rss_data); | |
355 | ||
356 | if ($new_rss_hash != $rss_hash) { | |
357 | if ($debug_enabled) { | |
358 | _debug("update_rss_feed: saving $cache_filename"); | |
359 | } | |
360 | @file_put_contents($cache_filename, serialize($rss)); | |
361 | } | |
362 | } | |
363 | ||
364 | // We use local pluginhost here because we need to load different per-user feed plugins | |
365 | $pluginhost->run_hooks($pluginhost::HOOK_FEED_PARSED, "hook_feed_parsed", $rss); | |
366 | ||
367 | if ($debug_enabled) { | |
368 | _debug("update_rss_feed: processing feed data..."); | |
369 | } | |
370 | ||
371 | // db_query($link, "BEGIN"); | |
372 | ||
373 | if (DB_TYPE == "pgsql") { | |
374 | $favicon_interval_qpart = "favicon_last_checked < NOW() - INTERVAL '12 hour'"; | |
375 | } else { | |
376 | $favicon_interval_qpart = "favicon_last_checked < DATE_SUB(NOW(), INTERVAL 12 HOUR)"; | |
377 | } | |
378 | ||
379 | $result = db_query($link, "SELECT title,site_url,owner_uid, | |
380 | (favicon_last_checked IS NULL OR $favicon_interval_qpart) AS | |
381 | favicon_needs_check | |
382 | FROM ttrss_feeds WHERE id = '$feed'"); | |
383 | ||
384 | $registered_title = db_fetch_result($result, 0, "title"); | |
385 | $orig_site_url = db_fetch_result($result, 0, "site_url"); | |
386 | $favicon_needs_check = sql_bool_to_bool(db_fetch_result($result, 0, | |
387 | "favicon_needs_check")); | |
388 | ||
389 | $owner_uid = db_fetch_result($result, 0, "owner_uid"); | |
390 | ||
391 | $site_url = db_escape_string($link, mb_substr(rewrite_relative_url($fetch_url, $rss->get_link()), 0, 245)); | |
392 | ||
393 | if ($debug_enabled) { | |
394 | _debug("update_rss_feed: checking favicon..."); | |
395 | } | |
396 | ||
397 | if ($favicon_needs_check) { | |
398 | check_feed_favicon($site_url, $feed, $link); | |
399 | ||
400 | db_query($link, "UPDATE ttrss_feeds SET favicon_last_checked = NOW() | |
401 | WHERE id = '$feed'"); | |
402 | } | |
403 | ||
404 | if (!$registered_title || $registered_title == "[Unknown]") { | |
405 | ||
406 | $feed_title = db_escape_string($link, $rss->get_title()); | |
407 | ||
408 | if ($debug_enabled) { | |
409 | _debug("update_rss_feed: registering title: $feed_title"); | |
410 | } | |
411 | ||
412 | db_query($link, "UPDATE ttrss_feeds SET | |
413 | title = '$feed_title' WHERE id = '$feed'"); | |
414 | } | |
415 | ||
416 | if ($site_url && $orig_site_url != $site_url) { | |
417 | db_query($link, "UPDATE ttrss_feeds SET | |
418 | site_url = '$site_url' WHERE id = '$feed'"); | |
419 | } | |
420 | ||
421 | if ($debug_enabled) { | |
422 | _debug("update_rss_feed: loading filters & labels..."); | |
423 | } | |
424 | ||
425 | $filters = load_filters($link, $feed, $owner_uid); | |
426 | $labels = get_all_labels($link, $owner_uid); | |
427 | ||
428 | if ($debug_enabled) { | |
429 | //print_r($filters); | |
430 | _debug("update_rss_feed: " . count($filters) . " filters loaded."); | |
431 | } | |
432 | ||
433 | $items = $rss->get_items(); | |
434 | ||
435 | if (!is_array($items)) { | |
436 | if ($debug_enabled) { | |
437 | _debug("update_rss_feed: no articles found."); | |
438 | } | |
439 | ||
440 | db_query($link, "UPDATE ttrss_feeds | |
441 | SET last_updated = NOW(), last_error = '' WHERE id = '$feed'"); | |
442 | ||
443 | return; // no articles | |
444 | } | |
445 | ||
446 | if ($pubsub_state != 2 && PUBSUBHUBBUB_ENABLED) { | |
447 | ||
448 | if ($debug_enabled) _debug("update_rss_feed: checking for PUSH hub..."); | |
449 | ||
450 | $feed_hub_url = false; | |
451 | ||
452 | $links = $rss->get_links('hub'); | |
453 | ||
454 | if ($links && is_array($links)) { | |
455 | foreach ($links as $l) { | |
456 | $feed_hub_url = $l; | |
457 | break; | |
458 | } | |
459 | } | |
460 | ||
461 | if ($debug_enabled) _debug("update_rss_feed: feed hub url: $feed_hub_url"); | |
462 | ||
463 | if ($feed_hub_url && function_exists('curl_init') && | |
464 | !ini_get("open_basedir")) { | |
465 | ||
466 | require_once 'lib/pubsubhubbub/subscriber.php'; | |
467 | ||
468 | $callback_url = get_self_url_prefix() . | |
469 | "/public.php?op=pubsub&id=$feed"; | |
470 | ||
471 | $s = new Subscriber($feed_hub_url, $callback_url); | |
472 | ||
473 | $rc = $s->subscribe($fetch_url); | |
474 | ||
475 | if ($debug_enabled) | |
476 | _debug("update_rss_feed: feed hub url found, subscribe request sent."); | |
477 | ||
478 | db_query($link, "UPDATE ttrss_feeds SET pubsub_state = 1 | |
479 | WHERE id = '$feed'"); | |
480 | } | |
481 | } | |
482 | ||
483 | if ($debug_enabled) { | |
484 | _debug("update_rss_feed: processing articles..."); | |
485 | } | |
486 | ||
487 | foreach ($items as $item) { | |
488 | if ($_REQUEST['xdebug'] == 3) { | |
489 | print_r($item); | |
490 | } | |
491 | ||
492 | $entry_guid = $item->get_id(); | |
493 | if (!$entry_guid) $entry_guid = $item->get_link(); | |
494 | if (!$entry_guid) $entry_guid = make_guid_from_title($item->get_title()); | |
495 | ||
496 | if ($debug_enabled) { | |
497 | _debug("update_rss_feed: guid $entry_guid"); | |
498 | } | |
499 | ||
500 | if (!$entry_guid) continue; | |
501 | ||
502 | $entry_guid = "$owner_uid,$entry_guid"; | |
503 | ||
504 | $entry_timestamp = ""; | |
505 | ||
506 | $entry_timestamp = strtotime($item->get_date()); | |
507 | ||
508 | if ($entry_timestamp == -1 || !$entry_timestamp || $entry_timestamp > time()) { | |
509 | $entry_timestamp = time(); | |
510 | $no_orig_date = 'true'; | |
511 | } else { | |
512 | $no_orig_date = 'false'; | |
513 | } | |
514 | ||
515 | $entry_timestamp_fmt = strftime("%Y/%m/%d %H:%M:%S", $entry_timestamp); | |
516 | ||
517 | if ($debug_enabled) { | |
518 | _debug("update_rss_feed: date $entry_timestamp [$entry_timestamp_fmt]"); | |
519 | } | |
520 | ||
521 | $entry_title = $item->get_title(); | |
522 | ||
523 | $entry_link = rewrite_relative_url($site_url, $item->get_link()); | |
524 | ||
525 | if ($debug_enabled) { | |
526 | _debug("update_rss_feed: title $entry_title"); | |
527 | _debug("update_rss_feed: link $entry_link"); | |
528 | } | |
529 | ||
530 | if (!$entry_title) $entry_title = date("Y-m-d H:i:s", $entry_timestamp);; | |
531 | ||
532 | $entry_content = $item->get_content(); | |
533 | if (!$entry_content) $entry_content = $item->get_description(); | |
534 | ||
535 | if ($_REQUEST["xdebug"] == 2) { | |
536 | print "update_rss_feed: content: "; | |
537 | print $entry_content; | |
538 | print "\n"; | |
539 | } | |
540 | ||
541 | $entry_comments = $item->data["comments"]; | |
542 | ||
543 | if ($item->get_author()) { | |
544 | $entry_author_item = $item->get_author(); | |
545 | $entry_author = $entry_author_item->get_name(); | |
546 | if (!$entry_author) $entry_author = $entry_author_item->get_email(); | |
547 | ||
548 | $entry_author = db_escape_string($link, $entry_author); | |
549 | } | |
550 | ||
551 | $entry_guid = db_escape_string($link, mb_substr($entry_guid, 0, 245)); | |
552 | ||
553 | $entry_comments = db_escape_string($link, mb_substr($entry_comments, 0, 245)); | |
554 | $entry_author = db_escape_string($link, mb_substr($entry_author, 0, 245)); | |
555 | ||
556 | $num_comments = $item->get_item_tags('http://purl.org/rss/1.0/modules/slash/', 'comments'); | |
557 | ||
558 | if (is_array($num_comments) && is_array($num_comments[0])) { | |
559 | $num_comments = (int) $num_comments[0]["data"]; | |
560 | } else { | |
561 | $num_comments = 0; | |
562 | } | |
563 | ||
564 | if ($debug_enabled) { | |
565 | _debug("update_rss_feed: num_comments: $num_comments"); | |
566 | _debug("update_rss_feed: looking for tags [1]..."); | |
567 | } | |
568 | ||
569 | // parse <category> entries into tags | |
570 | ||
571 | $additional_tags = array(); | |
572 | ||
573 | $additional_tags_src = $item->get_categories(); | |
574 | ||
575 | if (is_array($additional_tags_src)) { | |
576 | foreach ($additional_tags_src as $tobj) { | |
577 | array_push($additional_tags, $tobj->get_term()); | |
578 | } | |
579 | } | |
580 | ||
581 | if ($debug_enabled) { | |
582 | _debug("update_rss_feed: category tags:"); | |
583 | print_r($additional_tags); | |
584 | } | |
585 | ||
586 | if ($debug_enabled) { | |
587 | _debug("update_rss_feed: looking for tags [2]..."); | |
588 | } | |
589 | ||
590 | $entry_tags = array_unique($additional_tags); | |
591 | ||
592 | for ($i = 0; $i < count($entry_tags); $i++) | |
593 | $entry_tags[$i] = mb_strtolower($entry_tags[$i], 'utf-8'); | |
594 | ||
595 | if ($debug_enabled) { | |
596 | //_debug("update_rss_feed: unfiltered tags found:"); | |
597 | //print_r($entry_tags); | |
598 | } | |
599 | ||
600 | if ($debug_enabled) { | |
601 | _debug("update_rss_feed: done collecting data."); | |
602 | } | |
603 | ||
604 | // TODO: less memory-hungry implementation | |
605 | ||
606 | if ($debug_enabled) { | |
607 | _debug("update_rss_feed: applying plugin filters.."); | |
608 | } | |
609 | ||
610 | // FIXME not sure if owner_uid is a good idea here, we may have a base entry without user entry (?) | |
611 | $result = db_query($link, "SELECT plugin_data,title,content,link,tag_cache,author FROM ttrss_entries, ttrss_user_entries | |
612 | WHERE ref_id = id AND guid = '".db_escape_string($link, $entry_guid)."' AND owner_uid = $owner_uid"); | |
613 | ||
614 | if (db_num_rows($result) != 0) { | |
615 | $entry_plugin_data = db_fetch_result($result, 0, "plugin_data"); | |
616 | $stored_article = array("title" => db_fetch_result($result, 0, "title"), | |
617 | "content" => db_fetch_result($result, 0, "content"), | |
618 | "link" => db_fetch_result($result, 0, "link"), | |
619 | "tags" => explode(",", db_fetch_result($result, 0, "tag_cache")), | |
620 | "author" => db_fetch_result($result, 0, "author")); | |
621 | } else { | |
622 | $entry_plugin_data = ""; | |
623 | $stored_article = array(); | |
624 | } | |
625 | ||
626 | $article = array("owner_uid" => $owner_uid, // read only | |
627 | "guid" => $entry_guid, // read only | |
628 | "title" => $entry_title, | |
629 | "content" => $entry_content, | |
630 | "link" => $entry_link, | |
631 | "tags" => $entry_tags, | |
632 | "plugin_data" => $entry_plugin_data, | |
633 | "author" => $entry_author, | |
634 | "stored" => $stored_article); | |
635 | ||
636 | foreach ($pluginhost->get_hooks($pluginhost::HOOK_ARTICLE_FILTER) as $plugin) { | |
637 | $article = $plugin->hook_article_filter($article); | |
638 | } | |
639 | ||
640 | $entry_tags = $article["tags"]; | |
641 | $entry_guid = db_escape_string($link, $entry_guid); | |
642 | $entry_title = db_escape_string($link, $article["title"]); | |
643 | $entry_author = db_escape_string($link, $article["author"]); | |
644 | $entry_link = db_escape_string($link, $article["link"]); | |
645 | $entry_plugin_data = db_escape_string($link, $article["plugin_data"]); | |
646 | $entry_content = $article["content"]; // escaped below | |
647 | ||
648 | ||
649 | if ($debug_enabled) { | |
650 | _debug("update_rss_feed: plugin data: $entry_plugin_data"); | |
651 | } | |
652 | ||
653 | if ($cache_images && is_writable(CACHE_DIR . '/images')) | |
654 | cache_images($entry_content, $site_url, $debug_enabled); | |
655 | ||
656 | $entry_content = db_escape_string($link, $entry_content, false); | |
657 | ||
658 | $content_hash = "SHA1:" . sha1($entry_content); | |
659 | ||
660 | db_query($link, "BEGIN"); | |
661 | ||
662 | $result = db_query($link, "SELECT id FROM ttrss_entries | |
663 | WHERE guid = '$entry_guid'"); | |
664 | ||
665 | if (db_num_rows($result) == 0) { | |
666 | ||
667 | if ($debug_enabled) { | |
668 | _debug("update_rss_feed: base guid [$entry_guid] not found"); | |
669 | } | |
670 | ||
671 | // base post entry does not exist, create it | |
672 | ||
673 | $result = db_query($link, | |
674 | "INSERT INTO ttrss_entries | |
675 | (title, | |
676 | guid, | |
677 | link, | |
678 | updated, | |
679 | content, | |
680 | content_hash, | |
681 | cached_content, | |
682 | no_orig_date, | |
683 | date_updated, | |
684 | date_entered, | |
685 | comments, | |
686 | num_comments, | |
687 | plugin_data, | |
688 | author) | |
689 | VALUES | |
690 | ('$entry_title', | |
691 | '$entry_guid', | |
692 | '$entry_link', | |
693 | '$entry_timestamp_fmt', | |
694 | '$entry_content', | |
695 | '$content_hash', | |
696 | '', | |
697 | $no_orig_date, | |
698 | NOW(), | |
699 | '$date_feed_processed', | |
700 | '$entry_comments', | |
701 | '$num_comments', | |
702 | '$entry_plugin_data', | |
703 | '$entry_author')"); | |
704 | ||
705 | $article_labels = array(); | |
706 | ||
707 | } else { | |
708 | // we keep encountering the entry in feeds, so we need to | |
709 | // update date_updated column so that we don't get horrible | |
710 | // dupes when the entry gets purged and reinserted again e.g. | |
711 | // in the case of SLOW SLOW OMG SLOW updating feeds | |
712 | ||
713 | $base_entry_id = db_fetch_result($result, 0, "id"); | |
714 | ||
715 | db_query($link, "UPDATE ttrss_entries SET date_updated = NOW() | |
716 | WHERE id = '$base_entry_id'"); | |
717 | ||
718 | $article_labels = get_article_labels($link, $base_entry_id, $owner_uid); | |
719 | } | |
720 | ||
721 | // now it should exist, if not - bad luck then | |
722 | ||
723 | $result = db_query($link, "SELECT | |
724 | id,content_hash,no_orig_date,title,plugin_data, | |
725 | ".SUBSTRING_FOR_DATE."(date_updated,1,19) as date_updated, | |
726 | ".SUBSTRING_FOR_DATE."(updated,1,19) as updated, | |
727 | num_comments | |
728 | FROM | |
729 | ttrss_entries | |
730 | WHERE guid = '$entry_guid'"); | |
731 | ||
732 | $entry_ref_id = 0; | |
733 | $entry_int_id = 0; | |
734 | ||
735 | if (db_num_rows($result) == 1) { | |
736 | ||
737 | if ($debug_enabled) { | |
738 | _debug("update_rss_feed: base guid [$entry_guid] found, checking for user record"); | |
739 | } | |
740 | ||
741 | // this will be used below in update handler | |
742 | $orig_content_hash = db_fetch_result($result, 0, "content_hash"); | |
743 | $orig_title = db_fetch_result($result, 0, "title"); | |
744 | $orig_num_comments = db_fetch_result($result, 0, "num_comments"); | |
745 | $orig_date_updated = strtotime(db_fetch_result($result, | |
746 | 0, "date_updated")); | |
747 | $orig_plugin_data = db_fetch_result($result, 0, "plugin_data"); | |
748 | ||
749 | $ref_id = db_fetch_result($result, 0, "id"); | |
750 | $entry_ref_id = $ref_id; | |
751 | ||
752 | // check for user post link to main table | |
753 | ||
754 | // do we allow duplicate posts with same GUID in different feeds? | |
755 | if (get_pref($link, "ALLOW_DUPLICATE_POSTS", $owner_uid, false)) { | |
756 | $dupcheck_qpart = "AND (feed_id = '$feed' OR feed_id IS NULL)"; | |
757 | } else { | |
758 | $dupcheck_qpart = ""; | |
759 | } | |
760 | ||
761 | /* Collect article tags here so we could filter by them: */ | |
762 | ||
763 | $article_filters = get_article_filters($filters, $entry_title, | |
764 | $entry_content, $entry_link, $entry_timestamp, $entry_author, | |
765 | $entry_tags); | |
766 | ||
767 | if ($debug_enabled) { | |
768 | _debug("update_rss_feed: article filters: "); | |
769 | if (count($article_filters) != 0) { | |
770 | print_r($article_filters); | |
771 | } | |
772 | } | |
773 | ||
774 | if (find_article_filter($article_filters, "filter")) { | |
775 | db_query($link, "COMMIT"); // close transaction in progress | |
776 | continue; | |
777 | } | |
778 | ||
779 | $score = calculate_article_score($article_filters); | |
780 | ||
781 | if ($debug_enabled) { | |
782 | _debug("update_rss_feed: initial score: $score"); | |
783 | } | |
784 | ||
785 | $query = "SELECT ref_id, int_id FROM ttrss_user_entries WHERE | |
786 | ref_id = '$ref_id' AND owner_uid = '$owner_uid' | |
787 | $dupcheck_qpart"; | |
788 | ||
789 | // if ($_REQUEST["xdebug"]) print "$query\n"; | |
790 | ||
791 | $result = db_query($link, $query); | |
792 | ||
793 | // okay it doesn't exist - create user entry | |
794 | if (db_num_rows($result) == 0) { | |
795 | ||
796 | if ($debug_enabled) { | |
797 | _debug("update_rss_feed: user record not found, creating..."); | |
798 | } | |
799 | ||
800 | if ($score >= -500 && !find_article_filter($article_filters, 'catchup')) { | |
801 | $unread = 'true'; | |
802 | $last_read_qpart = 'NULL'; | |
803 | } else { | |
804 | $unread = 'false'; | |
805 | $last_read_qpart = 'NOW()'; | |
806 | } | |
807 | ||
808 | if (find_article_filter($article_filters, 'mark') || $score > 1000) { | |
809 | $marked = 'true'; | |
810 | } else { | |
811 | $marked = 'false'; | |
812 | } | |
813 | ||
814 | if (find_article_filter($article_filters, 'publish')) { | |
815 | $published = 'true'; | |
816 | } else { | |
817 | $published = 'false'; | |
818 | } | |
819 | ||
820 | // N-grams | |
821 | ||
822 | if (DB_TYPE == "pgsql" and defined('_NGRAM_TITLE_DUPLICATE_THRESHOLD')) { | |
823 | ||
824 | $result = db_query($link, "SELECT COUNT(*) AS similar FROM | |
825 | ttrss_entries,ttrss_user_entries | |
826 | WHERE ref_id = id AND updated >= NOW() - INTERVAL '7 day' | |
827 | AND similarity(title, '$entry_title') >= "._NGRAM_TITLE_DUPLICATE_THRESHOLD." | |
828 | AND owner_uid = $owner_uid"); | |
829 | ||
830 | $ngram_similar = db_fetch_result($result, 0, "similar"); | |
831 | ||
832 | if ($debug_enabled) { | |
833 | _debug("update_rss_feed: N-gram similar results: $ngram_similar"); | |
834 | } | |
835 | ||
836 | if ($ngram_similar > 0) { | |
837 | $unread = 'false'; | |
838 | } | |
839 | } | |
840 | ||
841 | $last_marked = ($marked == 'true') ? 'NOW()' : 'NULL'; | |
842 | $last_published = ($published == 'true') ? 'NOW()' : 'NULL'; | |
843 | ||
844 | $result = db_query($link, | |
845 | "INSERT INTO ttrss_user_entries | |
846 | (ref_id, owner_uid, feed_id, unread, last_read, marked, | |
847 | published, score, tag_cache, label_cache, uuid, | |
848 | last_marked, last_published) | |
849 | VALUES ('$ref_id', '$owner_uid', '$feed', $unread, | |
850 | $last_read_qpart, $marked, $published, '$score', '', '', | |
851 | '', $last_marked, $last_published)"); | |
852 | ||
853 | if (PUBSUBHUBBUB_HUB && $published == 'true') { | |
854 | $rss_link = get_self_url_prefix() . | |
855 | "/public.php?op=rss&id=-2&key=" . | |
856 | get_feed_access_key($link, -2, false, $owner_uid); | |
857 | ||
858 | $p = new Publisher(PUBSUBHUBBUB_HUB); | |
859 | ||
860 | $pubsub_result = $p->publish_update($rss_link); | |
861 | } | |
862 | ||
863 | $result = db_query($link, | |
864 | "SELECT int_id FROM ttrss_user_entries WHERE | |
865 | ref_id = '$ref_id' AND owner_uid = '$owner_uid' AND | |
866 | feed_id = '$feed' LIMIT 1"); | |
867 | ||
868 | if (db_num_rows($result) == 1) { | |
869 | $entry_int_id = db_fetch_result($result, 0, "int_id"); | |
870 | } | |
871 | } else { | |
872 | if ($debug_enabled) { | |
873 | _debug("update_rss_feed: user record FOUND"); | |
874 | } | |
875 | ||
876 | $entry_ref_id = db_fetch_result($result, 0, "ref_id"); | |
877 | $entry_int_id = db_fetch_result($result, 0, "int_id"); | |
878 | } | |
879 | ||
880 | if ($debug_enabled) { | |
881 | _debug("update_rss_feed: RID: $entry_ref_id, IID: $entry_int_id"); | |
882 | } | |
883 | ||
884 | $post_needs_update = false; | |
885 | $update_insignificant = false; | |
886 | ||
887 | if ($orig_num_comments != $num_comments) { | |
888 | $post_needs_update = true; | |
889 | $update_insignificant = true; | |
890 | } | |
891 | ||
892 | if ($entry_plugin_data != $orig_plugin_data) { | |
893 | $post_needs_update = true; | |
894 | $update_insignificant = true; | |
895 | } | |
896 | ||
897 | if ($content_hash != $orig_content_hash) { | |
898 | $post_needs_update = true; | |
899 | $update_insignificant = false; | |
900 | } | |
901 | ||
902 | if (db_escape_string($link, $orig_title) != $entry_title) { | |
903 | $post_needs_update = true; | |
904 | $update_insignificant = false; | |
905 | } | |
906 | ||
907 | // if post needs update, update it and mark all user entries | |
908 | // linking to this post as updated | |
909 | if ($post_needs_update) { | |
910 | ||
911 | if (defined('DAEMON_EXTENDED_DEBUG')) { | |
912 | _debug("update_rss_feed: post $entry_guid needs update..."); | |
913 | } | |
914 | ||
915 | // print "<!-- post $orig_title needs update : $post_needs_update -->"; | |
916 | ||
917 | db_query($link, "UPDATE ttrss_entries | |
918 | SET title = '$entry_title', content = '$entry_content', | |
919 | content_hash = '$content_hash', | |
920 | updated = '$entry_timestamp_fmt', | |
921 | num_comments = '$num_comments', | |
922 | plugin_data = '$entry_plugin_data' | |
923 | WHERE id = '$ref_id'"); | |
924 | ||
925 | if (!$update_insignificant) { | |
926 | if ($mark_unread_on_update) { | |
927 | db_query($link, "UPDATE ttrss_user_entries | |
928 | SET last_read = null, unread = true WHERE ref_id = '$ref_id'"); | |
929 | } | |
930 | } | |
931 | } | |
932 | } | |
933 | ||
934 | db_query($link, "COMMIT"); | |
935 | ||
936 | if ($debug_enabled) { | |
937 | _debug("update_rss_feed: assigning labels..."); | |
938 | } | |
939 | ||
940 | assign_article_to_label_filters($link, $entry_ref_id, $article_filters, | |
941 | $owner_uid, $article_labels); | |
942 | ||
943 | if ($debug_enabled) { | |
944 | _debug("update_rss_feed: looking for enclosures..."); | |
945 | } | |
946 | ||
947 | // enclosures | |
948 | ||
949 | $enclosures = array(); | |
950 | ||
951 | $encs = $item->get_enclosures(); | |
952 | ||
953 | if (is_array($encs)) { | |
954 | foreach ($encs as $e) { | |
955 | $e_item = array( | |
956 | $e->link, $e->type, $e->length); | |
957 | array_push($enclosures, $e_item); | |
958 | } | |
959 | } | |
960 | ||
961 | if ($debug_enabled) { | |
962 | _debug("update_rss_feed: article enclosures:"); | |
963 | print_r($enclosures); | |
964 | } | |
965 | ||
966 | db_query($link, "BEGIN"); | |
967 | ||
968 | foreach ($enclosures as $enc) { | |
969 | $enc_url = db_escape_string($link, $enc[0]); | |
970 | $enc_type = db_escape_string($link, $enc[1]); | |
971 | $enc_dur = db_escape_string($link, $enc[2]); | |
972 | ||
973 | $result = db_query($link, "SELECT id FROM ttrss_enclosures | |
974 | WHERE content_url = '$enc_url' AND post_id = '$entry_ref_id'"); | |
975 | ||
976 | if (db_num_rows($result) == 0) { | |
977 | db_query($link, "INSERT INTO ttrss_enclosures | |
978 | (content_url, content_type, title, duration, post_id) VALUES | |
979 | ('$enc_url', '$enc_type', '', '$enc_dur', '$entry_ref_id')"); | |
980 | } | |
981 | } | |
982 | ||
983 | db_query($link, "COMMIT"); | |
984 | ||
985 | // check for manual tags (we have to do it here since they're loaded from filters) | |
986 | ||
987 | foreach ($article_filters as $f) { | |
988 | if ($f["type"] == "tag") { | |
989 | ||
990 | $manual_tags = trim_array(explode(",", $f["param"])); | |
991 | ||
992 | foreach ($manual_tags as $tag) { | |
993 | if (tag_is_valid($tag)) { | |
994 | array_push($entry_tags, $tag); | |
995 | } | |
996 | } | |
997 | } | |
998 | } | |
999 | ||
1000 | // Skip boring tags | |
1001 | ||
1002 | $boring_tags = trim_array(explode(",", mb_strtolower(get_pref($link, | |
1003 | 'BLACKLISTED_TAGS', $owner_uid, ''), 'utf-8'))); | |
1004 | ||
1005 | $filtered_tags = array(); | |
1006 | $tags_to_cache = array(); | |
1007 | ||
1008 | if ($entry_tags && is_array($entry_tags)) { | |
1009 | foreach ($entry_tags as $tag) { | |
1010 | if (array_search($tag, $boring_tags) === false) { | |
1011 | array_push($filtered_tags, $tag); | |
1012 | } | |
1013 | } | |
1014 | } | |
1015 | ||
1016 | $filtered_tags = array_unique($filtered_tags); | |
1017 | ||
1018 | if ($debug_enabled) { | |
1019 | _debug("update_rss_feed: filtered article tags:"); | |
1020 | print_r($filtered_tags); | |
1021 | } | |
1022 | ||
1023 | // Save article tags in the database | |
1024 | ||
1025 | if (count($filtered_tags) > 0) { | |
1026 | ||
1027 | db_query($link, "BEGIN"); | |
1028 | ||
1029 | foreach ($filtered_tags as $tag) { | |
1030 | ||
1031 | $tag = sanitize_tag($tag); | |
1032 | $tag = db_escape_string($link, $tag); | |
1033 | ||
1034 | if (!tag_is_valid($tag)) continue; | |
1035 | ||
1036 | $result = db_query($link, "SELECT id FROM ttrss_tags | |
1037 | WHERE tag_name = '$tag' AND post_int_id = '$entry_int_id' AND | |
1038 | owner_uid = '$owner_uid' LIMIT 1"); | |
1039 | ||
1040 | if ($result && db_num_rows($result) == 0) { | |
1041 | ||
1042 | db_query($link, "INSERT INTO ttrss_tags | |
1043 | (owner_uid,tag_name,post_int_id) | |
1044 | VALUES ('$owner_uid','$tag', '$entry_int_id')"); | |
1045 | } | |
1046 | ||
1047 | array_push($tags_to_cache, $tag); | |
1048 | } | |
1049 | ||
1050 | /* update the cache */ | |
1051 | ||
1052 | $tags_to_cache = array_unique($tags_to_cache); | |
1053 | ||
1054 | $tags_str = db_escape_string($link, join(",", $tags_to_cache)); | |
1055 | ||
1056 | db_query($link, "UPDATE ttrss_user_entries | |
1057 | SET tag_cache = '$tags_str' WHERE ref_id = '$entry_ref_id' | |
1058 | AND owner_uid = $owner_uid"); | |
1059 | ||
1060 | db_query($link, "COMMIT"); | |
1061 | } | |
1062 | ||
1063 | if (get_pref($link, "AUTO_ASSIGN_LABELS", $owner_uid, false)) { | |
1064 | if ($debug_enabled) { | |
1065 | _debug("update_rss_feed: auto-assigning labels..."); | |
1066 | } | |
1067 | ||
1068 | foreach ($labels as $label) { | |
1069 | $caption = preg_quote($label["caption"]); | |
1070 | ||
1071 | if ($caption && preg_match("/\b$caption\b/i", "$tags_str " . strip_tags($entry_content) . " $entry_title")) { | |
1072 | if (!labels_contains_caption($article_labels, $caption)) { | |
1073 | label_add_article($link, $entry_ref_id, $caption, $owner_uid); | |
1074 | } | |
1075 | } | |
1076 | } | |
1077 | } | |
1078 | ||
1079 | if ($debug_enabled) { | |
1080 | _debug("update_rss_feed: article processed"); | |
1081 | } | |
1082 | } | |
1083 | ||
1084 | if (!$last_updated) { | |
1085 | if ($debug_enabled) { | |
1086 | _debug("update_rss_feed: new feed, catching it up..."); | |
1087 | } | |
1088 | catchup_feed($link, $feed, false, $owner_uid); | |
1089 | } | |
1090 | ||
1091 | if ($debug_enabled) { | |
1092 | _debug("purging feed..."); | |
1093 | } | |
1094 | ||
1095 | purge_feed($link, $feed, 0, $debug_enabled); | |
1096 | ||
1097 | db_query($link, "UPDATE ttrss_feeds | |
1098 | SET last_updated = NOW(), last_error = '' WHERE id = '$feed'"); | |
1099 | ||
1100 | // db_query($link, "COMMIT"); | |
1101 | ||
1102 | } else { | |
1103 | ||
1104 | $error_msg = db_escape_string($link, mb_substr($rss->error(), 0, 245)); | |
1105 | ||
1106 | if ($debug_enabled) { | |
1107 | _debug("update_rss_feed: error fetching feed: $error_msg"); | |
1108 | } | |
1109 | ||
1110 | db_query($link, | |
1111 | "UPDATE ttrss_feeds SET last_error = '$error_msg', | |
1112 | last_updated = NOW() WHERE id = '$feed'"); | |
1113 | } | |
1114 | ||
1115 | unset($rss); | |
1116 | ||
1117 | if ($debug_enabled) { | |
1118 | _debug("update_rss_feed: done"); | |
1119 | } | |
1120 | ||
1121 | } | |
1122 | ||
1123 | function cache_images($html, $site_url, $debug) { | |
1124 | $cache_dir = CACHE_DIR . "/images"; | |
1125 | ||
1126 | libxml_use_internal_errors(true); | |
1127 | ||
1128 | $charset_hack = '<head> | |
1129 | <meta http-equiv="Content-Type" content="text/html; charset=utf-8"/> | |
1130 | </head>'; | |
1131 | ||
1132 | $doc = new DOMDocument(); | |
1133 | $doc->loadHTML($charset_hack . $html); | |
1134 | $xpath = new DOMXPath($doc); | |
1135 | ||
1136 | $entries = $xpath->query('(//img[@src])'); | |
1137 | ||
1138 | foreach ($entries as $entry) { | |
1139 | if ($entry->hasAttribute('src')) { | |
1140 | $src = rewrite_relative_url($site_url, $entry->getAttribute('src')); | |
1141 | ||
1142 | $local_filename = CACHE_DIR . "/images/" . sha1($src) . ".png"; | |
1143 | ||
1144 | if ($debug) _debug("cache_images: downloading: $src to $local_filename"); | |
1145 | ||
1146 | if (!file_exists($local_filename)) { | |
1147 | $file_content = fetch_file_contents($src); | |
1148 | ||
1149 | if ($file_content && strlen($file_content) > 1024) { | |
1150 | file_put_contents($local_filename, $file_content); | |
1151 | } | |
1152 | } | |
1153 | ||
1154 | if (file_exists($local_filename)) { | |
1155 | $entry->setAttribute('src', SELF_URL_PATH . '/image.php?url=' . | |
1156 | base64_encode($src)); | |
1157 | } | |
1158 | } | |
1159 | } | |
1160 | ||
1161 | $node = $doc->getElementsByTagName('body')->item(0); | |
1162 | ||
1163 | return $doc->saveXML($node); | |
1164 | } | |
1165 | ||
1166 | function expire_lock_files($debug) { | |
1167 | if ($debug) _debug("Removing old lock files..."); | |
1168 | ||
1169 | $num_deleted = 0; | |
1170 | ||
1171 | if (is_writable(LOCK_DIRECTORY)) { | |
1172 | $files = glob(LOCK_DIRECTORY . "/*.lock"); | |
1173 | ||
1174 | if ($files) { | |
1175 | foreach ($files as $file) { | |
1176 | if (!file_is_locked($file) && time() - filemtime($file) > 86400*2) { | |
1177 | unlink($file); | |
1178 | ++$num_deleted; | |
1179 | } | |
1180 | } | |
1181 | } | |
1182 | } | |
1183 | ||
1184 | if ($debug) _debug("Removed $num_deleted files."); | |
1185 | } | |
1186 | ||
1187 | function expire_cached_files($debug) { | |
1188 | foreach (array("simplepie", "images", "export") as $dir) { | |
1189 | $cache_dir = CACHE_DIR . "/$dir"; | |
1190 | ||
1191 | if ($debug) _debug("Expiring $cache_dir"); | |
1192 | ||
1193 | $num_deleted = 0; | |
1194 | ||
1195 | if (is_writable($cache_dir)) { | |
1196 | $files = glob("$cache_dir/*"); | |
1197 | ||
1198 | if ($files) { | |
1199 | foreach ($files as $file) { | |
1200 | if (time() - filemtime($file) > 86400*7) { | |
1201 | unlink($file); | |
1202 | ||
1203 | ++$num_deleted; | |
1204 | } | |
1205 | } | |
1206 | } | |
1207 | } | |
1208 | ||
1209 | if ($debug) _debug("Removed $num_deleted files."); | |
1210 | } | |
1211 | } | |
1212 | ||
1213 | /** | |
1214 | * Source: http://www.php.net/manual/en/function.parse-url.php#104527 | |
1215 | * Returns the url query as associative array | |
1216 | * | |
1217 | * @param string query | |
1218 | * @return array params | |
1219 | */ | |
1220 | function convertUrlQuery($query) { | |
1221 | $queryParts = explode('&', $query); | |
1222 | ||
1223 | $params = array(); | |
1224 | ||
1225 | foreach ($queryParts as $param) { | |
1226 | $item = explode('=', $param); | |
1227 | $params[$item[0]] = $item[1]; | |
1228 | } | |
1229 | ||
1230 | return $params; | |
1231 | } | |
1232 | ||
1233 | function get_article_filters($filters, $title, $content, $link, $timestamp, $author, $tags) { | |
1234 | $matches = array(); | |
1235 | ||
1236 | foreach ($filters as $filter) { | |
1237 | $match_any_rule = $filter["match_any_rule"]; | |
1238 | $inverse = $filter["inverse"]; | |
1239 | $filter_match = false; | |
1240 | ||
1241 | foreach ($filter["rules"] as $rule) { | |
1242 | $match = false; | |
1243 | $reg_exp = $rule["reg_exp"]; | |
1244 | $rule_inverse = $rule["inverse"]; | |
1245 | ||
1246 | if (!$reg_exp) | |
1247 | continue; | |
1248 | ||
1249 | switch ($rule["type"]) { | |
1250 | case "title": | |
1251 | $match = @preg_match("/$reg_exp/i", $title); | |
1252 | break; | |
1253 | case "content": | |
1254 | // we don't need to deal with multiline regexps | |
1255 | $content = preg_replace("/[\r\n\t]/", "", $content); | |
1256 | ||
1257 | $match = @preg_match("/$reg_exp/i", $content); | |
1258 | break; | |
1259 | case "both": | |
1260 | // we don't need to deal with multiline regexps | |
1261 | $content = preg_replace("/[\r\n\t]/", "", $content); | |
1262 | ||
1263 | $match = (@preg_match("/$reg_exp/i", $title) || @preg_match("/$reg_exp/i", $content)); | |
1264 | break; | |
1265 | case "link": | |
1266 | $match = @preg_match("/$reg_exp/i", $link); | |
1267 | break; | |
1268 | case "author": | |
1269 | $match = @preg_match("/$reg_exp/i", $author); | |
1270 | break; | |
1271 | case "tag": | |
1272 | $tag_string = join(",", $tags); | |
1273 | $match = @preg_match("/$reg_exp/i", $tag_string); | |
1274 | break; | |
1275 | } | |
1276 | ||
1277 | if ($rule_inverse) $match = !$match; | |
1278 | ||
1279 | if ($match_any_rule) { | |
1280 | if ($match) { | |
1281 | $filter_match = true; | |
1282 | break; | |
1283 | } | |
1284 | } else { | |
1285 | $filter_match = $match; | |
1286 | if (!$match) { | |
1287 | break; | |
1288 | } | |
1289 | } | |
1290 | } | |
1291 | ||
1292 | if ($inverse) $filter_match = !$filter_match; | |
1293 | ||
1294 | if ($filter_match) { | |
1295 | foreach ($filter["actions"] AS $action) { | |
1296 | array_push($matches, $action); | |
1297 | } | |
1298 | } | |
1299 | } | |
1300 | ||
1301 | return $matches; | |
1302 | } | |
1303 | ||
1304 | function find_article_filter($filters, $filter_name) { | |
1305 | foreach ($filters as $f) { | |
1306 | if ($f["type"] == $filter_name) { | |
1307 | return $f; | |
1308 | }; | |
1309 | } | |
1310 | return false; | |
1311 | } | |
1312 | ||
1313 | function find_article_filters($filters, $filter_name) { | |
1314 | $results = array(); | |
1315 | ||
1316 | foreach ($filters as $f) { | |
1317 | if ($f["type"] == $filter_name) { | |
1318 | array_push($results, $f); | |
1319 | }; | |
1320 | } | |
1321 | return $results; | |
1322 | } | |
1323 | ||
1324 | function calculate_article_score($filters) { | |
1325 | $score = 0; | |
1326 | ||
1327 | foreach ($filters as $f) { | |
1328 | if ($f["type"] == "score") { | |
1329 | $score += $f["param"]; | |
1330 | }; | |
1331 | } | |
1332 | return $score; | |
1333 | } | |
1334 | ||
1335 | function labels_contains_caption($labels, $caption) { | |
1336 | foreach ($labels as $label) { | |
1337 | if ($label[1] == $caption) { | |
1338 | return true; | |
1339 | } | |
1340 | } | |
1341 | ||
1342 | return false; | |
1343 | } | |
1344 | ||
1345 | function assign_article_to_label_filters($link, $id, $filters, $owner_uid, $article_labels) { | |
1346 | foreach ($filters as $f) { | |
1347 | if ($f["type"] == "label") { | |
1348 | if (!labels_contains_caption($article_labels, $f["param"])) { | |
1349 | label_add_article($link, $id, $f["param"], $owner_uid); | |
1350 | } | |
1351 | } | |
1352 | } | |
1353 | } | |
1354 | ||
1355 | function make_guid_from_title($title) { | |
1356 | return preg_replace("/[ \"\',.:;]/", "-", | |
1357 | mb_strtolower(strip_tags($title), 'utf-8')); | |
1358 | } | |
1359 | ||
1360 | ||
1361 | ?> |