]>
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 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) | |
84 | ) OR ttrss_feeds.last_updated IS NULL | |
85 | OR last_updated = '1970-01-01 00:00:00')"; | |
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) | |
93 | ) OR ttrss_feeds.last_updated IS NULL | |
94 | OR last_updated = '1970-01-01 00:00:00')"; | |
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 DISTINCT ttrss_feeds.feed_url | |
112 | FROM | |
113 | ttrss_feeds, ttrss_users, ttrss_user_prefs | |
114 | WHERE | |
115 | ttrss_feeds.owner_uid = ttrss_users.id | |
116 | AND ttrss_users.id = ttrss_user_prefs.owner_uid | |
117 | AND ttrss_user_prefs.pref_name = 'DEFAULT_UPDATE_INTERVAL' | |
118 | $login_thresh_qpart $update_limit_qpart | |
119 | $updstart_thresh_qpart | |
120 | ORDER BY feed_url $query_limit"); | |
121 | ||
122 | $user_prefs_cache = array(); | |
123 | ||
124 | if($debug) _debug(sprintf("Scheduled %d feeds to update...", db_num_rows($result))); | |
125 | ||
126 | // Here is a little cache magic in order to minimize risk of double feed updates. | |
127 | $feeds_to_update = array(); | |
128 | while ($line = db_fetch_assoc($result)) { | |
129 | array_push($feeds_to_update, db_escape_string($link, $line['feed_url'])); | |
130 | } | |
131 | ||
132 | // We update the feed last update started date before anything else. | |
133 | // There is no lag due to feed contents downloads | |
134 | // It prevent an other process to update the same feed. | |
135 | ||
136 | if(count($feeds_to_update) > 0) { | |
137 | $feeds_quoted = array(); | |
138 | ||
139 | foreach ($feeds_to_update as $feed) { | |
140 | array_push($feeds_quoted, "'" . db_escape_string($link, $feed) . "'"); | |
141 | } | |
142 | ||
143 | db_query($link, sprintf("UPDATE ttrss_feeds SET last_update_started = NOW() | |
144 | WHERE feed_url IN (%s)", implode(',', $feeds_quoted))); | |
145 | } | |
146 | ||
147 | expire_cached_files($debug); | |
148 | expire_lock_files($debug); | |
149 | ||
150 | // For each feed, we call the feed update function. | |
151 | foreach ($feeds_to_update as $feed) { | |
152 | if($debug) _debug("Base feed: $feed"); | |
153 | ||
154 | //update_rss_feed($link, $line["id"], true); | |
155 | ||
156 | // since we have the data cached, we can deal with other feeds with the same url | |
157 | ||
158 | $tmp_result = db_query($link, "SELECT ttrss_feeds.feed_url,ttrss_feeds.id,last_updated | |
159 | FROM ttrss_feeds, ttrss_users WHERE | |
160 | ttrss_users.id = ttrss_feeds.owner_uid AND | |
161 | feed_url = '".db_escape_string($link, $feed)."' AND | |
162 | ttrss_feeds.update_interval != -1 | |
163 | $login_thresh_qpart | |
164 | ORDER BY feed_url $query_limit"); | |
165 | ||
166 | if (db_num_rows($tmp_result) > 0) { | |
167 | while ($tline = db_fetch_assoc($tmp_result)) { | |
168 | if($debug) _debug(" => " . $tline["feed_url"] . ", " . $tline["last_updated"] . ", " . $tline["id"]); | |
169 | update_rss_feed($link, $tline["id"], true); | |
170 | } | |
171 | } | |
172 | } | |
173 | ||
174 | require_once "digest.php"; | |
175 | ||
176 | // Send feed digests by email if needed. | |
177 | send_headlines_digests($link, $debug); | |
178 | ||
179 | } // function update_daemon_common | |
180 | ||
181 | // ignore_daemon is not used | |
182 | function update_rss_feed($link, $feed, $ignore_daemon = false, $no_cache = false, | |
183 | $override_url = false) { | |
184 | ||
185 | require_once "lib/simplepie/simplepie.inc"; | |
186 | ||
187 | $debug_enabled = defined('DAEMON_EXTENDED_DEBUG') || $_REQUEST['xdebug']; | |
188 | ||
189 | if ($debug_enabled) { | |
190 | _debug("update_rss_feed: start"); | |
191 | } | |
192 | ||
193 | $result = db_query($link, "SELECT id,update_interval,auth_login, | |
194 | feed_url,auth_pass,cache_images,last_updated, | |
195 | mark_unread_on_update, owner_uid, | |
196 | pubsub_state | |
197 | FROM ttrss_feeds WHERE id = '$feed'"); | |
198 | ||
199 | if (db_num_rows($result) == 0) { | |
200 | if ($debug_enabled) { | |
201 | _debug("update_rss_feed: feed $feed NOT FOUND/SKIPPED"); | |
202 | } | |
203 | return false; | |
204 | } | |
205 | ||
206 | $last_updated = db_fetch_result($result, 0, "last_updated"); | |
207 | $owner_uid = db_fetch_result($result, 0, "owner_uid"); | |
208 | $mark_unread_on_update = sql_bool_to_bool(db_fetch_result($result, | |
209 | 0, "mark_unread_on_update")); | |
210 | $pubsub_state = db_fetch_result($result, 0, "pubsub_state"); | |
211 | ||
212 | db_query($link, "UPDATE ttrss_feeds SET last_update_started = NOW() | |
213 | WHERE id = '$feed'"); | |
214 | ||
215 | $auth_login = db_fetch_result($result, 0, "auth_login"); | |
216 | $auth_pass = db_fetch_result($result, 0, "auth_pass"); | |
217 | ||
218 | $cache_images = sql_bool_to_bool(db_fetch_result($result, 0, "cache_images")); | |
219 | $fetch_url = db_fetch_result($result, 0, "feed_url"); | |
220 | ||
221 | $feed = db_escape_string($link, $feed); | |
222 | ||
223 | if ($override_url) $fetch_url = $override_url; | |
224 | ||
225 | $date_feed_processed = date('Y-m-d H:i'); | |
226 | ||
227 | $cache_filename = CACHE_DIR . "/simplepie/" . sha1($fetch_url) . ".feed"; | |
228 | ||
229 | // Ignore cache if new feed or manual update. | |
230 | $cache_age = ($no_cache || is_null($last_updated) || $last_updated == '1970-01-01 00:00:00') ? | |
231 | 30 : get_feed_update_interval($link, $feed) * 60; | |
232 | ||
233 | if ($debug_enabled) { | |
234 | _debug("update_rss_feed: cache filename: $cache_filename exists: " . file_exists($cache_filename)); | |
235 | _debug("update_rss_feed: cache age: $cache_age; no cache: $no_cache"); | |
236 | } | |
237 | ||
238 | $cached_feed_data_hash = false; | |
239 | ||
240 | $rss = false; | |
241 | $rss_hash = false; | |
242 | ||
243 | if (file_exists($cache_filename) && | |
244 | is_readable($cache_filename) && | |
245 | !$auth_login && !$auth_pass && | |
246 | filemtime($cache_filename) > time() - $cache_age) { | |
247 | ||
248 | if ($debug_enabled) { | |
249 | _debug("update_rss_feed: using local cache."); | |
250 | } | |
251 | ||
252 | @$rss_data = file_get_contents($cache_filename); | |
253 | ||
254 | if ($rss_data) { | |
255 | $rss_hash = sha1($rss_data); | |
256 | @$rss = unserialize($rss_data); | |
257 | } | |
258 | } | |
259 | ||
260 | if (!$rss) { | |
261 | ||
262 | if (!$feed_data) { | |
263 | if ($debug_enabled) { | |
264 | _debug("update_rss_feed: fetching [$fetch_url]..."); | |
265 | } | |
266 | ||
267 | $feed_data = fetch_file_contents($fetch_url, false, | |
268 | $auth_login, $auth_pass, false, $no_cache ? 15 : 45); | |
269 | ||
270 | if ($debug_enabled) { | |
271 | _debug("update_rss_feed: fetch done."); | |
272 | } | |
273 | ||
274 | } | |
275 | ||
276 | if (!$feed_data) { | |
277 | global $fetch_last_error; | |
278 | ||
279 | if ($debug_enabled) { | |
280 | _debug("update_rss_feed: unable to fetch: $fetch_last_error"); | |
281 | } | |
282 | ||
283 | $error_escaped = db_escape_string($link, $fetch_last_error); | |
284 | ||
285 | db_query($link, | |
286 | "UPDATE ttrss_feeds SET last_error = '$error_escaped', | |
287 | last_updated = NOW() WHERE id = '$feed'"); | |
288 | ||
289 | return; | |
290 | } | |
291 | } | |
292 | ||
293 | $pluginhost = new PluginHost($link); | |
294 | $pluginhost->set_debug($debug_enabled); | |
295 | $user_plugins = get_pref($link, "_ENABLED_PLUGINS", $owner_uid); | |
296 | ||
297 | $pluginhost->load(PLUGINS, $pluginhost::KIND_ALL); | |
298 | $pluginhost->load($user_plugins, $pluginhost::KIND_USER, $owner_uid); | |
299 | $pluginhost->load_data(); | |
300 | ||
301 | foreach ($pluginhost->get_hooks($pluginhost::HOOK_FEED_FETCHED) as $plugin) { | |
302 | $feed_data = $plugin->hook_feed_fetched($feed_data); | |
303 | } | |
304 | ||
305 | if (!$rss) { | |
306 | $rss = new SimplePie(); | |
307 | $rss->set_sanitize_class("SanitizeDummy"); | |
308 | // simplepie ignores the above and creates default sanitizer anyway, | |
309 | // so let's override it... | |
310 | $rss->sanitize = new SanitizeDummy(); | |
311 | $rss->set_output_encoding('UTF-8'); | |
312 | $rss->set_raw_data($feed_data); | |
313 | $rss->enable_cache(false); | |
314 | ||
315 | @$rss->init(); | |
316 | } | |
317 | ||
318 | // print_r($rss); | |
319 | ||
320 | $feed = db_escape_string($link, $feed); | |
321 | ||
322 | if (!$rss->error()) { | |
323 | ||
324 | // cache data for later | |
325 | if (!$auth_pass && !$auth_login && is_writable(CACHE_DIR . "/simplepie")) { | |
326 | $rss_data = serialize($rss); | |
327 | $new_rss_hash = sha1($rss_data); | |
328 | ||
329 | if ($new_rss_hash != $rss_hash) { | |
330 | if ($debug_enabled) { | |
331 | _debug("update_rss_feed: saving $cache_filename"); | |
332 | } | |
333 | @file_put_contents($cache_filename, serialize($rss)); | |
334 | } | |
335 | } | |
336 | ||
337 | // We use local pluginhost here because we need to load different per-user feed plugins | |
338 | $pluginhost->run_hooks($pluginhost::HOOK_FEED_PARSED, "hook_feed_parsed", $rss); | |
339 | ||
340 | if ($debug_enabled) { | |
341 | _debug("update_rss_feed: processing feed data..."); | |
342 | } | |
343 | ||
344 | // db_query($link, "BEGIN"); | |
345 | ||
346 | if (DB_TYPE == "pgsql") { | |
347 | $favicon_interval_qpart = "favicon_last_checked < NOW() - INTERVAL '12 hour'"; | |
348 | } else { | |
349 | $favicon_interval_qpart = "favicon_last_checked < DATE_SUB(NOW(), INTERVAL 12 HOUR)"; | |
350 | } | |
351 | ||
352 | $result = db_query($link, "SELECT title,site_url,owner_uid, | |
353 | (favicon_last_checked IS NULL OR $favicon_interval_qpart) AS | |
354 | favicon_needs_check | |
355 | FROM ttrss_feeds WHERE id = '$feed'"); | |
356 | ||
357 | $registered_title = db_fetch_result($result, 0, "title"); | |
358 | $orig_site_url = db_fetch_result($result, 0, "site_url"); | |
359 | $favicon_needs_check = sql_bool_to_bool(db_fetch_result($result, 0, | |
360 | "favicon_needs_check")); | |
361 | ||
362 | $owner_uid = db_fetch_result($result, 0, "owner_uid"); | |
363 | ||
364 | $site_url = db_escape_string($link, mb_substr(rewrite_relative_url($fetch_url, $rss->get_link()), 0, 245)); | |
365 | ||
366 | if ($debug_enabled) { | |
367 | _debug("update_rss_feed: checking favicon..."); | |
368 | } | |
369 | ||
370 | if ($favicon_needs_check) { | |
371 | check_feed_favicon($site_url, $feed, $link); | |
372 | ||
373 | db_query($link, "UPDATE ttrss_feeds SET favicon_last_checked = NOW() | |
374 | WHERE id = '$feed'"); | |
375 | } | |
376 | ||
377 | if (!$registered_title || $registered_title == "[Unknown]") { | |
378 | ||
379 | $feed_title = db_escape_string($link, $rss->get_title()); | |
380 | ||
381 | if ($debug_enabled) { | |
382 | _debug("update_rss_feed: registering title: $feed_title"); | |
383 | } | |
384 | ||
385 | db_query($link, "UPDATE ttrss_feeds SET | |
386 | title = '$feed_title' WHERE id = '$feed'"); | |
387 | } | |
388 | ||
389 | if ($site_url && $orig_site_url != $site_url) { | |
390 | db_query($link, "UPDATE ttrss_feeds SET | |
391 | site_url = '$site_url' WHERE id = '$feed'"); | |
392 | } | |
393 | ||
394 | if ($debug_enabled) { | |
395 | _debug("update_rss_feed: loading filters & labels..."); | |
396 | } | |
397 | ||
398 | $filters = load_filters($link, $feed, $owner_uid); | |
399 | $labels = get_all_labels($link, $owner_uid); | |
400 | ||
401 | if ($debug_enabled) { | |
402 | //print_r($filters); | |
403 | _debug("update_rss_feed: " . count($filters) . " filters loaded."); | |
404 | } | |
405 | ||
406 | $items = $rss->get_items(); | |
407 | ||
408 | if (!is_array($items)) { | |
409 | if ($debug_enabled) { | |
410 | _debug("update_rss_feed: no articles found."); | |
411 | } | |
412 | ||
413 | db_query($link, "UPDATE ttrss_feeds | |
414 | SET last_updated = NOW(), last_error = '' WHERE id = '$feed'"); | |
415 | ||
416 | return; // no articles | |
417 | } | |
418 | ||
419 | if ($pubsub_state != 2 && PUBSUBHUBBUB_ENABLED) { | |
420 | ||
421 | if ($debug_enabled) _debug("update_rss_feed: checking for PUSH hub..."); | |
422 | ||
423 | $feed_hub_url = false; | |
424 | ||
425 | $links = $rss->get_links('hub'); | |
426 | ||
427 | if ($links && is_array($links)) { | |
428 | foreach ($links as $l) { | |
429 | $feed_hub_url = $l; | |
430 | break; | |
431 | } | |
432 | } | |
433 | ||
434 | if ($debug_enabled) _debug("update_rss_feed: feed hub url: $feed_hub_url"); | |
435 | ||
436 | if ($feed_hub_url && function_exists('curl_init') && | |
437 | !ini_get("open_basedir")) { | |
438 | ||
439 | require_once 'lib/pubsubhubbub/subscriber.php'; | |
440 | ||
441 | $callback_url = get_self_url_prefix() . | |
442 | "/public.php?op=pubsub&id=$feed"; | |
443 | ||
444 | $s = new Subscriber($feed_hub_url, $callback_url); | |
445 | ||
446 | $rc = $s->subscribe($fetch_url); | |
447 | ||
448 | if ($debug_enabled) | |
449 | _debug("update_rss_feed: feed hub url found, subscribe request sent."); | |
450 | ||
451 | db_query($link, "UPDATE ttrss_feeds SET pubsub_state = 1 | |
452 | WHERE id = '$feed'"); | |
453 | } | |
454 | } | |
455 | ||
456 | if ($debug_enabled) { | |
457 | _debug("update_rss_feed: processing articles..."); | |
458 | } | |
459 | ||
460 | foreach ($items as $item) { | |
461 | if ($_REQUEST['xdebug'] == 3) { | |
462 | print_r($item); | |
463 | } | |
464 | ||
465 | $entry_guid = $item->get_id(); | |
466 | if (!$entry_guid) $entry_guid = $item->get_link(); | |
467 | if (!$entry_guid) $entry_guid = make_guid_from_title($item->get_title()); | |
468 | ||
469 | if ($debug_enabled) { | |
470 | _debug("update_rss_feed: guid $entry_guid"); | |
471 | } | |
472 | ||
473 | if (!$entry_guid) continue; | |
474 | ||
475 | $entry_guid = "$owner_uid,$entry_guid"; | |
476 | ||
477 | $entry_timestamp = ""; | |
478 | ||
479 | $entry_timestamp = strtotime($item->get_date()); | |
480 | ||
481 | if ($entry_timestamp == -1 || !$entry_timestamp || $entry_timestamp > time()) { | |
482 | $entry_timestamp = time(); | |
483 | $no_orig_date = 'true'; | |
484 | } else { | |
485 | $no_orig_date = 'false'; | |
486 | } | |
487 | ||
488 | $entry_timestamp_fmt = strftime("%Y/%m/%d %H:%M:%S", $entry_timestamp); | |
489 | ||
490 | if ($debug_enabled) { | |
491 | _debug("update_rss_feed: date $entry_timestamp [$entry_timestamp_fmt]"); | |
492 | } | |
493 | ||
494 | $entry_title = $item->get_title(); | |
495 | ||
496 | $entry_link = rewrite_relative_url($site_url, $item->get_link()); | |
497 | ||
498 | if ($debug_enabled) { | |
499 | _debug("update_rss_feed: title $entry_title"); | |
500 | _debug("update_rss_feed: link $entry_link"); | |
501 | } | |
502 | ||
503 | if (!$entry_title) $entry_title = date("Y-m-d H:i:s", $entry_timestamp);; | |
504 | ||
505 | $entry_content = $item->get_content(); | |
506 | if (!$entry_content) $entry_content = $item->get_description(); | |
507 | ||
508 | if ($_REQUEST["xdebug"] == 2) { | |
509 | print "update_rss_feed: content: "; | |
510 | print $entry_content; | |
511 | print "\n"; | |
512 | } | |
513 | ||
514 | $entry_comments = $item->data["comments"]; | |
515 | ||
516 | if ($item->get_author()) { | |
517 | $entry_author_item = $item->get_author(); | |
518 | $entry_author = $entry_author_item->get_name(); | |
519 | if (!$entry_author) $entry_author = $entry_author_item->get_email(); | |
520 | ||
521 | $entry_author = db_escape_string($link, $entry_author); | |
522 | } | |
523 | ||
524 | $entry_guid = db_escape_string($link, mb_substr($entry_guid, 0, 245)); | |
525 | ||
526 | $entry_comments = db_escape_string($link, mb_substr($entry_comments, 0, 245)); | |
527 | $entry_author = db_escape_string($link, mb_substr($entry_author, 0, 245)); | |
528 | ||
529 | $num_comments = $item->get_item_tags('http://purl.org/rss/1.0/modules/slash/', 'comments'); | |
530 | ||
531 | if (is_array($num_comments) && is_array($num_comments[0])) { | |
532 | $num_comments = (int) $num_comments[0]["data"]; | |
533 | } else { | |
534 | $num_comments = 0; | |
535 | } | |
536 | ||
537 | if ($debug_enabled) { | |
538 | _debug("update_rss_feed: num_comments: $num_comments"); | |
539 | _debug("update_rss_feed: looking for tags [1]..."); | |
540 | } | |
541 | ||
542 | // parse <category> entries into tags | |
543 | ||
544 | $additional_tags = array(); | |
545 | ||
546 | $additional_tags_src = $item->get_categories(); | |
547 | ||
548 | if (is_array($additional_tags_src)) { | |
549 | foreach ($additional_tags_src as $tobj) { | |
550 | array_push($additional_tags, $tobj->get_term()); | |
551 | } | |
552 | } | |
553 | ||
554 | if ($debug_enabled) { | |
555 | _debug("update_rss_feed: category tags:"); | |
556 | print_r($additional_tags); | |
557 | } | |
558 | ||
559 | if ($debug_enabled) { | |
560 | _debug("update_rss_feed: looking for tags [2]..."); | |
561 | } | |
562 | ||
563 | $entry_tags = array_unique($additional_tags); | |
564 | ||
565 | for ($i = 0; $i < count($entry_tags); $i++) | |
566 | $entry_tags[$i] = mb_strtolower($entry_tags[$i], 'utf-8'); | |
567 | ||
568 | if ($debug_enabled) { | |
569 | //_debug("update_rss_feed: unfiltered tags found:"); | |
570 | //print_r($entry_tags); | |
571 | } | |
572 | ||
573 | if ($debug_enabled) { | |
574 | _debug("update_rss_feed: done collecting data."); | |
575 | } | |
576 | ||
577 | // TODO: less memory-hungry implementation | |
578 | ||
579 | if ($debug_enabled) { | |
580 | _debug("update_rss_feed: applying plugin filters.."); | |
581 | } | |
582 | ||
583 | // FIXME not sure if owner_uid is a good idea here, we may have a base entry without user entry (?) | |
584 | $result = db_query($link, "SELECT plugin_data,title,content,link,tag_cache,author FROM ttrss_entries, ttrss_user_entries | |
585 | WHERE ref_id = id AND guid = '".db_escape_string($link, $entry_guid)."' AND owner_uid = $owner_uid"); | |
586 | ||
587 | if (db_num_rows($result) != 0) { | |
588 | $entry_plugin_data = db_fetch_result($result, 0, "plugin_data"); | |
589 | $stored_article = array("title" => db_fetch_result($result, 0, "title"), | |
590 | "content" => db_fetch_result($result, 0, "content"), | |
591 | "link" => db_fetch_result($result, 0, "link"), | |
592 | "tags" => explode(",", db_fetch_result($result, 0, "tag_cache")), | |
593 | "author" => db_fetch_result($result, 0, "author")); | |
594 | } else { | |
595 | $entry_plugin_data = ""; | |
596 | $stored_article = array(); | |
597 | } | |
598 | ||
599 | $article = array("owner_uid" => $owner_uid, // read only | |
600 | "guid" => $entry_guid, // read only | |
601 | "title" => $entry_title, | |
602 | "content" => $entry_content, | |
603 | "link" => $entry_link, | |
604 | "tags" => $entry_tags, | |
605 | "plugin_data" => $entry_plugin_data, | |
606 | "author" => $entry_author, | |
607 | "stored" => $stored_article); | |
608 | ||
609 | foreach ($pluginhost->get_hooks($pluginhost::HOOK_ARTICLE_FILTER) as $plugin) { | |
610 | $article = $plugin->hook_article_filter($article); | |
611 | } | |
612 | ||
613 | $entry_tags = $article["tags"]; | |
614 | $entry_guid = db_escape_string($link, $entry_guid); | |
615 | $entry_title = db_escape_string($link, $article["title"]); | |
616 | $entry_author = db_escape_string($link, $article["author"]); | |
617 | $entry_link = db_escape_string($link, $article["link"]); | |
618 | $entry_plugin_data = db_escape_string($link, $article["plugin_data"]); | |
619 | $entry_content = $article["content"]; // escaped below | |
620 | ||
621 | ||
622 | if ($debug_enabled) { | |
623 | _debug("update_rss_feed: plugin data: $entry_plugin_data"); | |
624 | } | |
625 | ||
626 | if ($cache_images && is_writable(CACHE_DIR . '/images')) | |
627 | cache_images($entry_content, $site_url, $debug_enabled); | |
628 | ||
629 | $entry_content = db_escape_string($link, $entry_content, false); | |
630 | ||
631 | $content_hash = "SHA1:" . sha1($entry_content); | |
632 | ||
633 | db_query($link, "BEGIN"); | |
634 | ||
635 | $result = db_query($link, "SELECT id FROM ttrss_entries | |
636 | WHERE guid = '$entry_guid'"); | |
637 | ||
638 | if (db_num_rows($result) == 0) { | |
639 | ||
640 | if ($debug_enabled) { | |
641 | _debug("update_rss_feed: base guid [$entry_guid] not found"); | |
642 | } | |
643 | ||
644 | // base post entry does not exist, create it | |
645 | ||
646 | $result = db_query($link, | |
647 | "INSERT INTO ttrss_entries | |
648 | (title, | |
649 | guid, | |
650 | link, | |
651 | updated, | |
652 | content, | |
653 | content_hash, | |
654 | cached_content, | |
655 | no_orig_date, | |
656 | date_updated, | |
657 | date_entered, | |
658 | comments, | |
659 | num_comments, | |
660 | plugin_data, | |
661 | author) | |
662 | VALUES | |
663 | ('$entry_title', | |
664 | '$entry_guid', | |
665 | '$entry_link', | |
666 | '$entry_timestamp_fmt', | |
667 | '$entry_content', | |
668 | '$content_hash', | |
669 | '', | |
670 | $no_orig_date, | |
671 | NOW(), | |
672 | '$date_feed_processed', | |
673 | '$entry_comments', | |
674 | '$num_comments', | |
675 | '$entry_plugin_data', | |
676 | '$entry_author')"); | |
677 | ||
678 | $article_labels = array(); | |
679 | ||
680 | } else { | |
681 | // we keep encountering the entry in feeds, so we need to | |
682 | // update date_updated column so that we don't get horrible | |
683 | // dupes when the entry gets purged and reinserted again e.g. | |
684 | // in the case of SLOW SLOW OMG SLOW updating feeds | |
685 | ||
686 | $base_entry_id = db_fetch_result($result, 0, "id"); | |
687 | ||
688 | db_query($link, "UPDATE ttrss_entries SET date_updated = NOW() | |
689 | WHERE id = '$base_entry_id'"); | |
690 | ||
691 | $article_labels = get_article_labels($link, $base_entry_id, $owner_uid); | |
692 | } | |
693 | ||
694 | // now it should exist, if not - bad luck then | |
695 | ||
696 | $result = db_query($link, "SELECT | |
697 | id,content_hash,no_orig_date,title,plugin_data, | |
698 | ".SUBSTRING_FOR_DATE."(date_updated,1,19) as date_updated, | |
699 | ".SUBSTRING_FOR_DATE."(updated,1,19) as updated, | |
700 | num_comments | |
701 | FROM | |
702 | ttrss_entries | |
703 | WHERE guid = '$entry_guid'"); | |
704 | ||
705 | $entry_ref_id = 0; | |
706 | $entry_int_id = 0; | |
707 | ||
708 | if (db_num_rows($result) == 1) { | |
709 | ||
710 | if ($debug_enabled) { | |
711 | _debug("update_rss_feed: base guid [$entry_guid] found, checking for user record"); | |
712 | } | |
713 | ||
714 | // this will be used below in update handler | |
715 | $orig_content_hash = db_fetch_result($result, 0, "content_hash"); | |
716 | $orig_title = db_fetch_result($result, 0, "title"); | |
717 | $orig_num_comments = db_fetch_result($result, 0, "num_comments"); | |
718 | $orig_date_updated = strtotime(db_fetch_result($result, | |
719 | 0, "date_updated")); | |
720 | $orig_plugin_data = db_fetch_result($result, 0, "plugin_data"); | |
721 | ||
722 | $ref_id = db_fetch_result($result, 0, "id"); | |
723 | $entry_ref_id = $ref_id; | |
724 | ||
725 | // check for user post link to main table | |
726 | ||
727 | // do we allow duplicate posts with same GUID in different feeds? | |
728 | if (get_pref($link, "ALLOW_DUPLICATE_POSTS", $owner_uid, false)) { | |
729 | $dupcheck_qpart = "AND (feed_id = '$feed' OR feed_id IS NULL)"; | |
730 | } else { | |
731 | $dupcheck_qpart = ""; | |
732 | } | |
733 | ||
734 | /* Collect article tags here so we could filter by them: */ | |
735 | ||
736 | $article_filters = get_article_filters($filters, $entry_title, | |
737 | $entry_content, $entry_link, $entry_timestamp, $entry_author, | |
738 | $entry_tags); | |
739 | ||
740 | if ($debug_enabled) { | |
741 | _debug("update_rss_feed: article filters: "); | |
742 | if (count($article_filters) != 0) { | |
743 | print_r($article_filters); | |
744 | } | |
745 | } | |
746 | ||
747 | if (find_article_filter($article_filters, "filter")) { | |
748 | db_query($link, "COMMIT"); // close transaction in progress | |
749 | continue; | |
750 | } | |
751 | ||
752 | $score = calculate_article_score($article_filters); | |
753 | ||
754 | if ($debug_enabled) { | |
755 | _debug("update_rss_feed: initial score: $score"); | |
756 | } | |
757 | ||
758 | $query = "SELECT ref_id, int_id FROM ttrss_user_entries WHERE | |
759 | ref_id = '$ref_id' AND owner_uid = '$owner_uid' | |
760 | $dupcheck_qpart"; | |
761 | ||
762 | // if ($_REQUEST["xdebug"]) print "$query\n"; | |
763 | ||
764 | $result = db_query($link, $query); | |
765 | ||
766 | // okay it doesn't exist - create user entry | |
767 | if (db_num_rows($result) == 0) { | |
768 | ||
769 | if ($debug_enabled) { | |
770 | _debug("update_rss_feed: user record not found, creating..."); | |
771 | } | |
772 | ||
773 | if ($score >= -500 && !find_article_filter($article_filters, 'catchup')) { | |
774 | $unread = 'true'; | |
775 | $last_read_qpart = 'NULL'; | |
776 | } else { | |
777 | $unread = 'false'; | |
778 | $last_read_qpart = 'NOW()'; | |
779 | } | |
780 | ||
781 | if (find_article_filter($article_filters, 'mark') || $score > 1000) { | |
782 | $marked = 'true'; | |
783 | } else { | |
784 | $marked = 'false'; | |
785 | } | |
786 | ||
787 | if (find_article_filter($article_filters, 'publish')) { | |
788 | $published = 'true'; | |
789 | } else { | |
790 | $published = 'false'; | |
791 | } | |
792 | ||
793 | // N-grams | |
794 | ||
795 | if (DB_TYPE == "pgsql" and defined('_NGRAM_TITLE_DUPLICATE_THRESHOLD')) { | |
796 | ||
797 | $result = db_query($link, "SELECT COUNT(*) AS similar FROM | |
798 | ttrss_entries,ttrss_user_entries | |
799 | WHERE ref_id = id AND updated >= NOW() - INTERVAL '7 day' | |
800 | AND similarity(title, '$entry_title') >= "._NGRAM_TITLE_DUPLICATE_THRESHOLD." | |
801 | AND owner_uid = $owner_uid"); | |
802 | ||
803 | $ngram_similar = db_fetch_result($result, 0, "similar"); | |
804 | ||
805 | if ($debug_enabled) { | |
806 | _debug("update_rss_feed: N-gram similar results: $ngram_similar"); | |
807 | } | |
808 | ||
809 | if ($ngram_similar > 0) { | |
810 | $unread = 'false'; | |
811 | } | |
812 | } | |
813 | ||
814 | $last_marked = ($marked == 'true') ? 'NOW()' : 'NULL'; | |
815 | $last_published = ($published == 'true') ? 'NOW()' : 'NULL'; | |
816 | ||
817 | $result = db_query($link, | |
818 | "INSERT INTO ttrss_user_entries | |
819 | (ref_id, owner_uid, feed_id, unread, last_read, marked, | |
820 | published, score, tag_cache, label_cache, uuid, | |
821 | last_marked, last_published) | |
822 | VALUES ('$ref_id', '$owner_uid', '$feed', $unread, | |
823 | $last_read_qpart, $marked, $published, '$score', '', '', | |
824 | '', $last_marked, $last_published)"); | |
825 | ||
826 | if (PUBSUBHUBBUB_HUB && $published == 'true') { | |
827 | $rss_link = get_self_url_prefix() . | |
828 | "/public.php?op=rss&id=-2&key=" . | |
829 | get_feed_access_key($link, -2, false, $owner_uid); | |
830 | ||
831 | $p = new Publisher(PUBSUBHUBBUB_HUB); | |
832 | ||
833 | $pubsub_result = $p->publish_update($rss_link); | |
834 | } | |
835 | ||
836 | $result = db_query($link, | |
837 | "SELECT int_id FROM ttrss_user_entries WHERE | |
838 | ref_id = '$ref_id' AND owner_uid = '$owner_uid' AND | |
839 | feed_id = '$feed' LIMIT 1"); | |
840 | ||
841 | if (db_num_rows($result) == 1) { | |
842 | $entry_int_id = db_fetch_result($result, 0, "int_id"); | |
843 | } | |
844 | } else { | |
845 | if ($debug_enabled) { | |
846 | _debug("update_rss_feed: user record FOUND"); | |
847 | } | |
848 | ||
849 | $entry_ref_id = db_fetch_result($result, 0, "ref_id"); | |
850 | $entry_int_id = db_fetch_result($result, 0, "int_id"); | |
851 | } | |
852 | ||
853 | if ($debug_enabled) { | |
854 | _debug("update_rss_feed: RID: $entry_ref_id, IID: $entry_int_id"); | |
855 | } | |
856 | ||
857 | $post_needs_update = false; | |
858 | $update_insignificant = false; | |
859 | ||
860 | if ($orig_num_comments != $num_comments) { | |
861 | $post_needs_update = true; | |
862 | $update_insignificant = true; | |
863 | } | |
864 | ||
865 | if ($entry_plugin_data != $orig_plugin_data) { | |
866 | $post_needs_update = true; | |
867 | $update_insignificant = true; | |
868 | } | |
869 | ||
870 | if ($content_hash != $orig_content_hash) { | |
871 | $post_needs_update = true; | |
872 | $update_insignificant = false; | |
873 | } | |
874 | ||
875 | if (db_escape_string($link, $orig_title) != $entry_title) { | |
876 | $post_needs_update = true; | |
877 | $update_insignificant = false; | |
878 | } | |
879 | ||
880 | // if post needs update, update it and mark all user entries | |
881 | // linking to this post as updated | |
882 | if ($post_needs_update) { | |
883 | ||
884 | if (defined('DAEMON_EXTENDED_DEBUG')) { | |
885 | _debug("update_rss_feed: post $entry_guid needs update..."); | |
886 | } | |
887 | ||
888 | // print "<!-- post $orig_title needs update : $post_needs_update -->"; | |
889 | ||
890 | db_query($link, "UPDATE ttrss_entries | |
891 | SET title = '$entry_title', content = '$entry_content', | |
892 | content_hash = '$content_hash', | |
893 | updated = '$entry_timestamp_fmt', | |
894 | num_comments = '$num_comments', | |
895 | plugin_data = '$entry_plugin_data' | |
896 | WHERE id = '$ref_id'"); | |
897 | ||
898 | if (!$update_insignificant) { | |
899 | if ($mark_unread_on_update) { | |
900 | db_query($link, "UPDATE ttrss_user_entries | |
901 | SET last_read = null, unread = true WHERE ref_id = '$ref_id'"); | |
902 | } | |
903 | } | |
904 | } | |
905 | } | |
906 | ||
907 | db_query($link, "COMMIT"); | |
908 | ||
909 | if ($debug_enabled) { | |
910 | _debug("update_rss_feed: assigning labels..."); | |
911 | } | |
912 | ||
913 | assign_article_to_label_filters($link, $entry_ref_id, $article_filters, | |
914 | $owner_uid, $article_labels); | |
915 | ||
916 | if ($debug_enabled) { | |
917 | _debug("update_rss_feed: looking for enclosures..."); | |
918 | } | |
919 | ||
920 | // enclosures | |
921 | ||
922 | $enclosures = array(); | |
923 | ||
924 | $encs = $item->get_enclosures(); | |
925 | ||
926 | if (is_array($encs)) { | |
927 | foreach ($encs as $e) { | |
928 | $e_item = array( | |
929 | $e->link, $e->type, $e->length); | |
930 | array_push($enclosures, $e_item); | |
931 | } | |
932 | } | |
933 | ||
934 | if ($debug_enabled) { | |
935 | _debug("update_rss_feed: article enclosures:"); | |
936 | print_r($enclosures); | |
937 | } | |
938 | ||
939 | db_query($link, "BEGIN"); | |
940 | ||
941 | foreach ($enclosures as $enc) { | |
942 | $enc_url = db_escape_string($link, $enc[0]); | |
943 | $enc_type = db_escape_string($link, $enc[1]); | |
944 | $enc_dur = db_escape_string($link, $enc[2]); | |
945 | ||
946 | $result = db_query($link, "SELECT id FROM ttrss_enclosures | |
947 | WHERE content_url = '$enc_url' AND post_id = '$entry_ref_id'"); | |
948 | ||
949 | if (db_num_rows($result) == 0) { | |
950 | db_query($link, "INSERT INTO ttrss_enclosures | |
951 | (content_url, content_type, title, duration, post_id) VALUES | |
952 | ('$enc_url', '$enc_type', '', '$enc_dur', '$entry_ref_id')"); | |
953 | } | |
954 | } | |
955 | ||
956 | db_query($link, "COMMIT"); | |
957 | ||
958 | // check for manual tags (we have to do it here since they're loaded from filters) | |
959 | ||
960 | foreach ($article_filters as $f) { | |
961 | if ($f["type"] == "tag") { | |
962 | ||
963 | $manual_tags = trim_array(explode(",", $f["param"])); | |
964 | ||
965 | foreach ($manual_tags as $tag) { | |
966 | if (tag_is_valid($tag)) { | |
967 | array_push($entry_tags, $tag); | |
968 | } | |
969 | } | |
970 | } | |
971 | } | |
972 | ||
973 | // Skip boring tags | |
974 | ||
975 | $boring_tags = trim_array(explode(",", mb_strtolower(get_pref($link, | |
976 | 'BLACKLISTED_TAGS', $owner_uid, ''), 'utf-8'))); | |
977 | ||
978 | $filtered_tags = array(); | |
979 | $tags_to_cache = array(); | |
980 | ||
981 | if ($entry_tags && is_array($entry_tags)) { | |
982 | foreach ($entry_tags as $tag) { | |
983 | if (array_search($tag, $boring_tags) === false) { | |
984 | array_push($filtered_tags, $tag); | |
985 | } | |
986 | } | |
987 | } | |
988 | ||
989 | $filtered_tags = array_unique($filtered_tags); | |
990 | ||
991 | if ($debug_enabled) { | |
992 | _debug("update_rss_feed: filtered article tags:"); | |
993 | print_r($filtered_tags); | |
994 | } | |
995 | ||
996 | // Save article tags in the database | |
997 | ||
998 | if (count($filtered_tags) > 0) { | |
999 | ||
1000 | db_query($link, "BEGIN"); | |
1001 | ||
1002 | foreach ($filtered_tags as $tag) { | |
1003 | ||
1004 | $tag = sanitize_tag($tag); | |
1005 | $tag = db_escape_string($link, $tag); | |
1006 | ||
1007 | if (!tag_is_valid($tag)) continue; | |
1008 | ||
1009 | $result = db_query($link, "SELECT id FROM ttrss_tags | |
1010 | WHERE tag_name = '$tag' AND post_int_id = '$entry_int_id' AND | |
1011 | owner_uid = '$owner_uid' LIMIT 1"); | |
1012 | ||
1013 | if ($result && db_num_rows($result) == 0) { | |
1014 | ||
1015 | db_query($link, "INSERT INTO ttrss_tags | |
1016 | (owner_uid,tag_name,post_int_id) | |
1017 | VALUES ('$owner_uid','$tag', '$entry_int_id')"); | |
1018 | } | |
1019 | ||
1020 | array_push($tags_to_cache, $tag); | |
1021 | } | |
1022 | ||
1023 | /* update the cache */ | |
1024 | ||
1025 | $tags_to_cache = array_unique($tags_to_cache); | |
1026 | ||
1027 | $tags_str = db_escape_string($link, join(",", $tags_to_cache)); | |
1028 | ||
1029 | db_query($link, "UPDATE ttrss_user_entries | |
1030 | SET tag_cache = '$tags_str' WHERE ref_id = '$entry_ref_id' | |
1031 | AND owner_uid = $owner_uid"); | |
1032 | ||
1033 | db_query($link, "COMMIT"); | |
1034 | } | |
1035 | ||
1036 | if (get_pref($link, "AUTO_ASSIGN_LABELS", $owner_uid, false)) { | |
1037 | if ($debug_enabled) { | |
1038 | _debug("update_rss_feed: auto-assigning labels..."); | |
1039 | } | |
1040 | ||
1041 | foreach ($labels as $label) { | |
1042 | $caption = preg_quote($label["caption"]); | |
1043 | ||
1044 | if ($caption && preg_match("/\b$caption\b/i", "$tags_str " . strip_tags($entry_content) . " $entry_title")) { | |
1045 | if (!labels_contains_caption($article_labels, $caption)) { | |
1046 | label_add_article($link, $entry_ref_id, $caption, $owner_uid); | |
1047 | } | |
1048 | } | |
1049 | } | |
1050 | } | |
1051 | ||
1052 | if ($debug_enabled) { | |
1053 | _debug("update_rss_feed: article processed"); | |
1054 | } | |
1055 | } | |
1056 | ||
1057 | if (!$last_updated) { | |
1058 | if ($debug_enabled) { | |
1059 | _debug("update_rss_feed: new feed, catching it up..."); | |
1060 | } | |
1061 | catchup_feed($link, $feed, false, $owner_uid); | |
1062 | } | |
1063 | ||
1064 | if ($debug_enabled) { | |
1065 | _debug("purging feed..."); | |
1066 | } | |
1067 | ||
1068 | purge_feed($link, $feed, 0, $debug_enabled); | |
1069 | ||
1070 | db_query($link, "UPDATE ttrss_feeds | |
1071 | SET last_updated = NOW(), last_error = '' WHERE id = '$feed'"); | |
1072 | ||
1073 | // db_query($link, "COMMIT"); | |
1074 | ||
1075 | } else { | |
1076 | ||
1077 | $error_msg = db_escape_string($link, mb_substr($rss->error(), 0, 245)); | |
1078 | ||
1079 | if ($debug_enabled) { | |
1080 | _debug("update_rss_feed: error fetching feed: $error_msg"); | |
1081 | } | |
1082 | ||
1083 | db_query($link, | |
1084 | "UPDATE ttrss_feeds SET last_error = '$error_msg', | |
1085 | last_updated = NOW() WHERE id = '$feed'"); | |
1086 | } | |
1087 | ||
1088 | unset($rss); | |
1089 | ||
1090 | if ($debug_enabled) { | |
1091 | _debug("update_rss_feed: done"); | |
1092 | } | |
1093 | ||
1094 | } | |
1095 | ||
1096 | function cache_images($html, $site_url, $debug) { | |
1097 | $cache_dir = CACHE_DIR . "/images"; | |
1098 | ||
1099 | libxml_use_internal_errors(true); | |
1100 | ||
1101 | $charset_hack = '<head> | |
1102 | <meta http-equiv="Content-Type" content="text/html; charset=utf-8"/> | |
1103 | </head>'; | |
1104 | ||
1105 | $doc = new DOMDocument(); | |
1106 | $doc->loadHTML($charset_hack . $html); | |
1107 | $xpath = new DOMXPath($doc); | |
1108 | ||
1109 | $entries = $xpath->query('(//img[@src])'); | |
1110 | ||
1111 | foreach ($entries as $entry) { | |
1112 | if ($entry->hasAttribute('src')) { | |
1113 | $src = rewrite_relative_url($site_url, $entry->getAttribute('src')); | |
1114 | ||
1115 | $local_filename = CACHE_DIR . "/images/" . sha1($src) . ".png"; | |
1116 | ||
1117 | if ($debug) _debug("cache_images: downloading: $src to $local_filename"); | |
1118 | ||
1119 | if (!file_exists($local_filename)) { | |
1120 | $file_content = fetch_file_contents($src); | |
1121 | ||
1122 | if ($file_content && strlen($file_content) > 1024) { | |
1123 | file_put_contents($local_filename, $file_content); | |
1124 | } | |
1125 | } | |
1126 | ||
1127 | if (file_exists($local_filename)) { | |
1128 | $entry->setAttribute('src', SELF_URL_PATH . '/image.php?url=' . | |
1129 | base64_encode($src)); | |
1130 | } | |
1131 | } | |
1132 | } | |
1133 | ||
1134 | $node = $doc->getElementsByTagName('body')->item(0); | |
1135 | ||
1136 | return $doc->saveXML($node); | |
1137 | } | |
1138 | ||
1139 | function expire_lock_files($debug) { | |
1140 | if ($debug) _debug("Removing old lock files..."); | |
1141 | ||
1142 | $num_deleted = 0; | |
1143 | ||
1144 | if (is_writable(LOCK_DIRECTORY)) { | |
1145 | $files = glob(LOCK_DIRECTORY . "/*.lock"); | |
1146 | ||
1147 | if ($files) { | |
1148 | foreach ($files as $file) { | |
1149 | if (!file_is_locked($file) && time() - filemtime($file) > 86400*2) { | |
1150 | unlink($file); | |
1151 | ++$num_deleted; | |
1152 | } | |
1153 | } | |
1154 | } | |
1155 | } | |
1156 | ||
1157 | if ($debug) _debug("Removed $num_deleted files."); | |
1158 | } | |
1159 | ||
1160 | function expire_cached_files($debug) { | |
1161 | foreach (array("simplepie", "images", "export") as $dir) { | |
1162 | $cache_dir = CACHE_DIR . "/$dir"; | |
1163 | ||
1164 | if ($debug) _debug("Expiring $cache_dir"); | |
1165 | ||
1166 | $num_deleted = 0; | |
1167 | ||
1168 | if (is_writable($cache_dir)) { | |
1169 | $files = glob("$cache_dir/*"); | |
1170 | ||
1171 | if ($files) { | |
1172 | foreach ($files as $file) { | |
1173 | if (time() - filemtime($file) > 86400*7) { | |
1174 | unlink($file); | |
1175 | ||
1176 | ++$num_deleted; | |
1177 | } | |
1178 | } | |
1179 | } | |
1180 | } | |
1181 | ||
1182 | if ($debug) _debug("Removed $num_deleted files."); | |
1183 | } | |
1184 | } | |
1185 | ||
1186 | /** | |
1187 | * Source: http://www.php.net/manual/en/function.parse-url.php#104527 | |
1188 | * Returns the url query as associative array | |
1189 | * | |
1190 | * @param string query | |
1191 | * @return array params | |
1192 | */ | |
1193 | function convertUrlQuery($query) { | |
1194 | $queryParts = explode('&', $query); | |
1195 | ||
1196 | $params = array(); | |
1197 | ||
1198 | foreach ($queryParts as $param) { | |
1199 | $item = explode('=', $param); | |
1200 | $params[$item[0]] = $item[1]; | |
1201 | } | |
1202 | ||
1203 | return $params; | |
1204 | } | |
1205 | ||
1206 | function get_article_filters($filters, $title, $content, $link, $timestamp, $author, $tags) { | |
1207 | $matches = array(); | |
1208 | ||
1209 | foreach ($filters as $filter) { | |
1210 | $match_any_rule = $filter["match_any_rule"]; | |
1211 | $inverse = $filter["inverse"]; | |
1212 | $filter_match = false; | |
1213 | ||
1214 | foreach ($filter["rules"] as $rule) { | |
1215 | $match = false; | |
1216 | $reg_exp = $rule["reg_exp"]; | |
1217 | $rule_inverse = $rule["inverse"]; | |
1218 | ||
1219 | if (!$reg_exp) | |
1220 | continue; | |
1221 | ||
1222 | switch ($rule["type"]) { | |
1223 | case "title": | |
1224 | $match = @preg_match("/$reg_exp/i", $title); | |
1225 | break; | |
1226 | case "content": | |
1227 | // we don't need to deal with multiline regexps | |
1228 | $content = preg_replace("/[\r\n\t]/", "", $content); | |
1229 | ||
1230 | $match = @preg_match("/$reg_exp/i", $content); | |
1231 | break; | |
1232 | case "both": | |
1233 | // we don't need to deal with multiline regexps | |
1234 | $content = preg_replace("/[\r\n\t]/", "", $content); | |
1235 | ||
1236 | $match = (@preg_match("/$reg_exp/i", $title) || @preg_match("/$reg_exp/i", $content)); | |
1237 | break; | |
1238 | case "link": | |
1239 | $match = @preg_match("/$reg_exp/i", $link); | |
1240 | break; | |
1241 | case "author": | |
1242 | $match = @preg_match("/$reg_exp/i", $author); | |
1243 | break; | |
1244 | case "tag": | |
1245 | $tag_string = join(",", $tags); | |
1246 | $match = @preg_match("/$reg_exp/i", $tag_string); | |
1247 | break; | |
1248 | } | |
1249 | ||
1250 | if ($rule_inverse) $match = !$match; | |
1251 | ||
1252 | if ($match_any_rule) { | |
1253 | if ($match) { | |
1254 | $filter_match = true; | |
1255 | break; | |
1256 | } | |
1257 | } else { | |
1258 | $filter_match = $match; | |
1259 | if (!$match) { | |
1260 | break; | |
1261 | } | |
1262 | } | |
1263 | } | |
1264 | ||
1265 | if ($inverse) $filter_match = !$filter_match; | |
1266 | ||
1267 | if ($filter_match) { | |
1268 | foreach ($filter["actions"] AS $action) { | |
1269 | array_push($matches, $action); | |
1270 | } | |
1271 | } | |
1272 | } | |
1273 | ||
1274 | return $matches; | |
1275 | } | |
1276 | ||
1277 | function find_article_filter($filters, $filter_name) { | |
1278 | foreach ($filters as $f) { | |
1279 | if ($f["type"] == $filter_name) { | |
1280 | return $f; | |
1281 | }; | |
1282 | } | |
1283 | return false; | |
1284 | } | |
1285 | ||
1286 | function find_article_filters($filters, $filter_name) { | |
1287 | $results = array(); | |
1288 | ||
1289 | foreach ($filters as $f) { | |
1290 | if ($f["type"] == $filter_name) { | |
1291 | array_push($results, $f); | |
1292 | }; | |
1293 | } | |
1294 | return $results; | |
1295 | } | |
1296 | ||
1297 | function calculate_article_score($filters) { | |
1298 | $score = 0; | |
1299 | ||
1300 | foreach ($filters as $f) { | |
1301 | if ($f["type"] == "score") { | |
1302 | $score += $f["param"]; | |
1303 | }; | |
1304 | } | |
1305 | return $score; | |
1306 | } | |
1307 | ||
1308 | function labels_contains_caption($labels, $caption) { | |
1309 | foreach ($labels as $label) { | |
1310 | if ($label[1] == $caption) { | |
1311 | return true; | |
1312 | } | |
1313 | } | |
1314 | ||
1315 | return false; | |
1316 | } | |
1317 | ||
1318 | function assign_article_to_label_filters($link, $id, $filters, $owner_uid, $article_labels) { | |
1319 | foreach ($filters as $f) { | |
1320 | if ($f["type"] == "label") { | |
1321 | if (!labels_contains_caption($article_labels, $f["param"])) { | |
1322 | label_add_article($link, $id, $f["param"], $owner_uid); | |
1323 | } | |
1324 | } | |
1325 | } | |
1326 | } | |
1327 | ||
1328 | function make_guid_from_title($title) { | |
1329 | return preg_replace("/[ \"\',.:;]/", "-", | |
1330 | mb_strtolower(strip_tags($title), 'utf-8')); | |
1331 | } | |
1332 | ||
1333 | ||
1334 | ?> |