]>
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 | $subscribers = db_escape_string($line["subscribers"]); | |
22 | $feed_url = db_escape_string($line["feed_url"]); | |
23 | $title = db_escape_string($line["title"]); | |
24 | $site_url = db_escape_string($line["site_url"]); | |
25 | ||
26 | $tmp_result = db_query($link, "SELECT subscribers FROM | |
27 | ttrss_feedbrowser_cache WHERE feed_url = '$feed_url'"); | |
28 | ||
29 | if (db_num_rows($tmp_result) == 0) { | |
30 | ||
31 | db_query($link, "INSERT INTO ttrss_feedbrowser_cache | |
32 | (feed_url, site_url, title, subscribers) VALUES ('$feed_url', | |
33 | '$site_url', '$title', '$subscribers')"); | |
34 | ||
35 | ++$count; | |
36 | ||
37 | } | |
38 | ||
39 | } | |
40 | ||
41 | db_query($link, "COMMIT"); | |
42 | ||
43 | return $count; | |
44 | ||
45 | } | |
46 | ||
47 | ||
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 ttrss_feeds.feed_url,ttrss_feeds.id, ttrss_feeds.owner_uid, | |
112 | ".SUBSTRING_FOR_DATE."(ttrss_feeds.last_updated,1,19) AS last_updated, | |
113 | ttrss_feeds.update_interval | |
114 | FROM | |
115 | ttrss_feeds, ttrss_users, ttrss_user_prefs | |
116 | WHERE | |
117 | ttrss_feeds.owner_uid = ttrss_users.id | |
118 | AND ttrss_users.id = ttrss_user_prefs.owner_uid | |
119 | AND ttrss_user_prefs.pref_name = 'DEFAULT_UPDATE_INTERVAL' | |
120 | $login_thresh_qpart $update_limit_qpart | |
121 | $updstart_thresh_qpart | |
122 | ORDER BY $random_qpart $query_limit"); | |
123 | ||
124 | $user_prefs_cache = array(); | |
125 | ||
126 | if($debug) _debug(sprintf("Scheduled %d feeds to update...\n", db_num_rows($result))); | |
127 | ||
128 | // Here is a little cache magic in order to minimize risk of double feed updates. | |
129 | $feeds_to_update = array(); | |
130 | while ($line = db_fetch_assoc($result)) { | |
131 | $feeds_to_update[$line['id']] = $line; | |
132 | } | |
133 | ||
134 | // We update the feed last update started date before anything else. | |
135 | // There is no lag due to feed contents downloads | |
136 | // It prevent an other process to update the same feed. | |
137 | $feed_ids = array_keys($feeds_to_update); | |
138 | if($feed_ids) { | |
139 | db_query($link, sprintf("UPDATE ttrss_feeds SET last_update_started = NOW() | |
140 | WHERE id IN (%s)", implode(',', $feed_ids))); | |
141 | } | |
142 | ||
143 | expire_cached_files($debug); | |
144 | expire_lock_files($debug); | |
145 | ||
146 | // For each feed, we call the feed update function. | |
147 | while ($line = array_pop($feeds_to_update)) { | |
148 | ||
149 | if($debug) _debug("Feed: " . $line["feed_url"] . ", " . $line["last_updated"]); | |
150 | ||
151 | update_rss_feed($link, $line["id"], true); | |
152 | ||
153 | sleep(1); // prevent flood (FIXME make this an option?) | |
154 | } | |
155 | ||
156 | // Send feed digests by email if needed. | |
157 | send_headlines_digests($link, $debug); | |
158 | ||
159 | } // function update_daemon_common | |
160 | ||
161 | // ignore_daemon is not used | |
162 | function update_rss_feed($link, $feed, $ignore_daemon = false, $no_cache = false, | |
163 | $override_url = false) { | |
164 | ||
165 | require_once "lib/simplepie/simplepie.inc"; | |
166 | require_once "lib/magpierss/rss_fetch.inc"; | |
167 | require_once 'lib/magpierss/rss_utils.inc'; | |
168 | ||
169 | $debug_enabled = defined('DAEMON_EXTENDED_DEBUG') || $_REQUEST['xdebug']; | |
170 | ||
171 | if ($debug_enabled) { | |
172 | _debug("update_rss_feed: start"); | |
173 | } | |
174 | ||
175 | $result = db_query($link, "SELECT id,update_interval,auth_login, | |
176 | feed_url,auth_pass,cache_images,update_method,last_updated,cache_content, | |
177 | mark_unread_on_update, owner_uid, update_on_checksum_change, | |
178 | pubsub_state | |
179 | FROM ttrss_feeds WHERE id = '$feed'"); | |
180 | ||
181 | if (db_num_rows($result) == 0) { | |
182 | if ($debug_enabled) { | |
183 | _debug("update_rss_feed: feed $feed NOT FOUND/SKIPPED"); | |
184 | } | |
185 | return false; | |
186 | } | |
187 | ||
188 | $update_method = db_fetch_result($result, 0, "update_method"); | |
189 | $last_updated = db_fetch_result($result, 0, "last_updated"); | |
190 | $owner_uid = db_fetch_result($result, 0, "owner_uid"); | |
191 | $mark_unread_on_update = sql_bool_to_bool(db_fetch_result($result, | |
192 | 0, "mark_unread_on_update")); | |
193 | $update_on_checksum_change = sql_bool_to_bool(db_fetch_result($result, | |
194 | 0, "update_on_checksum_change")); | |
195 | $pubsub_state = db_fetch_result($result, 0, "pubsub_state"); | |
196 | ||
197 | db_query($link, "UPDATE ttrss_feeds SET last_update_started = NOW() | |
198 | WHERE id = '$feed'"); | |
199 | ||
200 | $auth_login = db_fetch_result($result, 0, "auth_login"); | |
201 | $auth_pass = db_fetch_result($result, 0, "auth_pass"); | |
202 | ||
203 | if ($update_method == 0) | |
204 | $update_method = DEFAULT_UPDATE_METHOD + 1; | |
205 | ||
206 | // 1 - Magpie | |
207 | // 2 - SimplePie | |
208 | // 3 - Twitter OAuth | |
209 | ||
210 | if ($update_method == 2) | |
211 | $use_simplepie = true; | |
212 | else | |
213 | $use_simplepie = false; | |
214 | ||
215 | if ($debug_enabled) { | |
216 | _debug("update method: $update_method (feed setting: $update_method) (use simplepie: $use_simplepie)\n"); | |
217 | } | |
218 | ||
219 | if ($update_method == 1) { | |
220 | $auth_login = urlencode($auth_login); | |
221 | $auth_pass = urlencode($auth_pass); | |
222 | } | |
223 | ||
224 | $cache_images = sql_bool_to_bool(db_fetch_result($result, 0, "cache_images")); | |
225 | $cache_content = sql_bool_to_bool(db_fetch_result($result, 0, "cache_content")); | |
226 | $fetch_url = db_fetch_result($result, 0, "feed_url"); | |
227 | ||
228 | $feed = db_escape_string($feed); | |
229 | ||
230 | if ($auth_login && $auth_pass ){ | |
231 | $url_parts = array(); | |
232 | preg_match("/(^[^:]*):\/\/(.*)/", $fetch_url, $url_parts); | |
233 | ||
234 | if ($url_parts[1] && $url_parts[2]) { | |
235 | $fetch_url = $url_parts[1] . "://$auth_login:$auth_pass@" . $url_parts[2]; | |
236 | } | |
237 | ||
238 | } | |
239 | ||
240 | if ($override_url) | |
241 | $fetch_url = $override_url; | |
242 | ||
243 | if ($debug_enabled) { | |
244 | _debug("update_rss_feed: fetching [$fetch_url]..."); | |
245 | } | |
246 | ||
247 | // Ignore cache if new feed or manual update. | |
248 | $cache_age = (is_null($last_updated) || $last_updated == '1970-01-01 00:00:00') ? | |
249 | -1 : get_feed_update_interval($link, $feed) * 60; | |
250 | ||
251 | if ($update_method == 1) { | |
252 | ||
253 | define('MAGPIE_CACHE_AGE', $cache_age); | |
254 | define('MAGPIE_CACHE_ON', !$no_cache); | |
255 | define('MAGPIE_FETCH_TIME_OUT', $no_cache ? 15 : 60); | |
256 | define('MAGPIE_CACHE_DIR', CACHE_DIR . "/magpie"); | |
257 | ||
258 | $rss = @fetch_rss($fetch_url); | |
259 | } else { | |
260 | $simplepie_cache_dir = CACHE_DIR . "/simplepie"; | |
261 | ||
262 | if (!is_dir($simplepie_cache_dir)) { | |
263 | mkdir($simplepie_cache_dir); | |
264 | } | |
265 | ||
266 | $rss = new SimplePie(); | |
267 | $rss->set_useragent(SELF_USER_AGENT); | |
268 | $rss->set_timeout($no_cache ? 15 : 60); | |
269 | $rss->set_feed_url($fetch_url); | |
270 | $rss->set_output_encoding('UTF-8'); | |
271 | //$rss->force_feed(true); | |
272 | ||
273 | if ($debug_enabled) { | |
274 | _debug("feed update interval (sec): " . | |
275 | get_feed_update_interval($link, $feed)*60); | |
276 | } | |
277 | ||
278 | $rss->enable_cache(!$no_cache); | |
279 | ||
280 | if (!$no_cache) { | |
281 | $rss->set_cache_location($simplepie_cache_dir); | |
282 | $rss->set_cache_duration($cache_age); | |
283 | } | |
284 | ||
285 | $rss->init(); | |
286 | } | |
287 | ||
288 | // print_r($rss); | |
289 | ||
290 | if ($debug_enabled) { | |
291 | _debug("update_rss_feed: fetch done, parsing..."); | |
292 | } | |
293 | ||
294 | $feed = db_escape_string($feed); | |
295 | ||
296 | if ($update_method == 2) { | |
297 | $fetch_ok = !$rss->error(); | |
298 | } else { | |
299 | $fetch_ok = !!$rss; | |
300 | } | |
301 | ||
302 | if ($fetch_ok) { | |
303 | ||
304 | if ($debug_enabled) { | |
305 | _debug("update_rss_feed: processing feed data..."); | |
306 | } | |
307 | ||
308 | // db_query($link, "BEGIN"); | |
309 | ||
310 | if (DB_TYPE == "pgsql") { | |
311 | $favicon_interval_qpart = "favicon_last_checked < NOW() - INTERVAL '12 hour'"; | |
312 | } else { | |
313 | $favicon_interval_qpart = "favicon_last_checked < DATE_SUB(NOW(), INTERVAL 12 HOUR)"; | |
314 | } | |
315 | ||
316 | $result = db_query($link, "SELECT title,icon_url,site_url,owner_uid, | |
317 | (favicon_last_checked IS NULL OR $favicon_interval_qpart) AS | |
318 | favicon_needs_check | |
319 | FROM ttrss_feeds WHERE id = '$feed'"); | |
320 | ||
321 | $registered_title = db_fetch_result($result, 0, "title"); | |
322 | $orig_icon_url = db_fetch_result($result, 0, "icon_url"); | |
323 | $orig_site_url = db_fetch_result($result, 0, "site_url"); | |
324 | $favicon_needs_check = sql_bool_to_bool(db_fetch_result($result, 0, | |
325 | "favicon_needs_check")); | |
326 | ||
327 | $owner_uid = db_fetch_result($result, 0, "owner_uid"); | |
328 | ||
329 | if ($use_simplepie) { | |
330 | $site_url = db_escape_string(trim($rss->get_link())); | |
331 | } else { | |
332 | $site_url = db_escape_string(trim($rss->channel["link"])); | |
333 | } | |
334 | ||
335 | // weird, weird Magpie | |
336 | if (!$use_simplepie) { | |
337 | if (!$site_url) $site_url = db_escape_string($rss->channel["link_"]); | |
338 | } | |
339 | ||
340 | $site_url = rewrite_relative_url($fetch_url, $site_url); | |
341 | $site_url = substr($site_url, 0, 250); | |
342 | ||
343 | if ($debug_enabled) { | |
344 | _debug("update_rss_feed: checking favicon..."); | |
345 | } | |
346 | ||
347 | if ($favicon_needs_check) { | |
348 | check_feed_favicon($site_url, $feed, $link); | |
349 | ||
350 | db_query($link, "UPDATE ttrss_feeds SET favicon_last_checked = NOW() | |
351 | WHERE id = '$feed'"); | |
352 | } | |
353 | ||
354 | if (!$registered_title || $registered_title == "[Unknown]") { | |
355 | ||
356 | if ($use_simplepie) { | |
357 | $feed_title = db_escape_string($rss->get_title()); | |
358 | } else { | |
359 | $feed_title = db_escape_string($rss->channel["title"]); | |
360 | } | |
361 | ||
362 | if ($debug_enabled) { | |
363 | _debug("update_rss_feed: registering title: $feed_title"); | |
364 | } | |
365 | ||
366 | db_query($link, "UPDATE ttrss_feeds SET | |
367 | title = '$feed_title' WHERE id = '$feed'"); | |
368 | } | |
369 | ||
370 | if ($site_url && $orig_site_url != $site_url) { | |
371 | db_query($link, "UPDATE ttrss_feeds SET | |
372 | site_url = '$site_url' WHERE id = '$feed'"); | |
373 | } | |
374 | ||
375 | // print "I: " . $rss->channel["image"]["url"]; | |
376 | ||
377 | if (!$use_simplepie) { | |
378 | $icon_url = db_escape_string(trim($rss->image["url"])); | |
379 | } else { | |
380 | $icon_url = db_escape_string(trim($rss->get_image_url())); | |
381 | } | |
382 | ||
383 | $icon_url = rewrite_relative_url($fetch_url, $icon_url); | |
384 | $icon_url = substr($icon_url, 0, 250); | |
385 | ||
386 | if ($icon_url && $orig_icon_url != $icon_url) { | |
387 | db_query($link, "UPDATE ttrss_feeds SET icon_url = '$icon_url' WHERE id = '$feed'"); | |
388 | } | |
389 | ||
390 | if ($debug_enabled) { | |
391 | _debug("update_rss_feed: loading filters & labels..."); | |
392 | } | |
393 | ||
394 | $filters = load_filters($link, $feed, $owner_uid); | |
395 | $labels = get_all_labels($link, $owner_uid); | |
396 | ||
397 | if ($debug_enabled) { | |
398 | //print_r($filters); | |
399 | _debug("update_rss_feed: " . count($filters) . " filters loaded."); | |
400 | } | |
401 | ||
402 | if ($use_simplepie) { | |
403 | $iterator = $rss->get_items(); | |
404 | } else { | |
405 | $iterator = $rss->items; | |
406 | if (!$iterator || !is_array($iterator)) $iterator = $rss->entries; | |
407 | if (!$iterator || !is_array($iterator)) $iterator = $rss; | |
408 | } | |
409 | ||
410 | if (!is_array($iterator)) { | |
411 | /* db_query($link, "UPDATE ttrss_feeds | |
412 | SET last_error = 'Parse error: can\'t find any articles.' | |
413 | WHERE id = '$feed'"); */ | |
414 | ||
415 | // clear any errors and mark feed as updated if fetched okay | |
416 | // even if it's blank | |
417 | ||
418 | if ($debug_enabled) { | |
419 | _debug("update_rss_feed: entry iterator is not an array, no articles?"); | |
420 | } | |
421 | ||
422 | db_query($link, "UPDATE ttrss_feeds | |
423 | SET last_updated = NOW(), last_error = '' WHERE id = '$feed'"); | |
424 | ||
425 | return; // no articles | |
426 | } | |
427 | ||
428 | if ($pubsub_state != 2 && PUBSUBHUBBUB_ENABLED) { | |
429 | ||
430 | if ($debug_enabled) _debug("update_rss_feed: checking for PUSH hub..."); | |
431 | ||
432 | $feed_hub_url = false; | |
433 | if ($use_simplepie) { | |
434 | $links = $rss->get_links('hub'); | |
435 | ||
436 | if ($links && is_array($links)) { | |
437 | foreach ($links as $l) { | |
438 | $feed_hub_url = $l; | |
439 | break; | |
440 | } | |
441 | } | |
442 | ||
443 | } else { | |
444 | $atom = $rss->channel['atom']; | |
445 | ||
446 | if ($atom) { | |
447 | if ($atom['link@rel'] == 'hub') { | |
448 | $feed_hub_url = $atom['link@href']; | |
449 | } | |
450 | ||
451 | if (!$feed_hub_url && $atom['link#'] > 1) { | |
452 | for ($i = 2; $i <= $atom['link#']; $i++) { | |
453 | if ($atom["link#$i@rel"] == 'hub') { | |
454 | $feed_hub_url = $atom["link#$i@href"]; | |
455 | break; | |
456 | } | |
457 | } | |
458 | } | |
459 | } else { | |
460 | $feed_hub_url = $rss->channel['link_hub']; | |
461 | } | |
462 | } | |
463 | ||
464 | if ($debug_enabled) _debug("update_rss_feed: feed hub url: $feed_hub_url"); | |
465 | ||
466 | if ($feed_hub_url && function_exists('curl_init') && | |
467 | !ini_get("open_basedir")) { | |
468 | ||
469 | require_once 'lib/pubsubhubbub/subscriber.php'; | |
470 | ||
471 | $callback_url = get_self_url_prefix() . | |
472 | "/public.php?op=pubsub&id=$feed"; | |
473 | ||
474 | $s = new Subscriber($feed_hub_url, $callback_url); | |
475 | ||
476 | $rc = $s->subscribe($fetch_url); | |
477 | ||
478 | if ($debug_enabled) | |
479 | _debug("update_rss_feed: feed hub url found, subscribe request sent."); | |
480 | ||
481 | db_query($link, "UPDATE ttrss_feeds SET pubsub_state = 1 | |
482 | WHERE id = '$feed'"); | |
483 | } | |
484 | } | |
485 | ||
486 | if ($debug_enabled) { | |
487 | _debug("update_rss_feed: processing articles..."); | |
488 | } | |
489 | ||
490 | foreach ($iterator as $item) { | |
491 | if ($_REQUEST['xdebug'] == 2) { | |
492 | print_r($item); | |
493 | } | |
494 | ||
495 | if ($use_simplepie) { | |
496 | $entry_guid = $item->get_id(); | |
497 | if (!$entry_guid) $entry_guid = $item->get_link(); | |
498 | if (!$entry_guid) $entry_guid = make_guid_from_title($item->get_title()); | |
499 | ||
500 | } else { | |
501 | ||
502 | $entry_guid = $item["id"]; | |
503 | ||
504 | if (!$entry_guid) $entry_guid = $item["guid"]; | |
505 | if (!$entry_guid) $entry_guid = $item["about"]; | |
506 | if (!$entry_guid) $entry_guid = $item["link"]; | |
507 | if (!$entry_guid) $entry_guid = make_guid_from_title($item["title"]); | |
508 | } | |
509 | ||
510 | if ($cache_content) { | |
511 | $entry_guid = "ccache:$entry_guid"; | |
512 | } | |
513 | ||
514 | if ($auth_login || $auth_pass) { | |
515 | $entry_guid = "auth,$owner_uid:$entry_guid"; | |
516 | } | |
517 | ||
518 | if ($debug_enabled) { | |
519 | _debug("update_rss_feed: guid $entry_guid"); | |
520 | } | |
521 | ||
522 | if (!$entry_guid) continue; | |
523 | ||
524 | $entry_timestamp = ""; | |
525 | ||
526 | if ($use_simplepie) { | |
527 | $entry_timestamp = strtotime($item->get_date()); | |
528 | } else { | |
529 | $rss_2_date = $item['pubdate']; | |
530 | $rss_1_date = $item['dc']['date']; | |
531 | $atom_date = $item['issued']; | |
532 | if (!$atom_date) $atom_date = $item['updated']; | |
533 | ||
534 | if ($atom_date != "") $entry_timestamp = parse_w3cdtf($atom_date); | |
535 | if ($rss_1_date != "") $entry_timestamp = parse_w3cdtf($rss_1_date); | |
536 | if ($rss_2_date != "") $entry_timestamp = strtotime($rss_2_date); | |
537 | ||
538 | } | |
539 | ||
540 | if ($entry_timestamp == "" || $entry_timestamp == -1 || !$entry_timestamp) { | |
541 | $entry_timestamp = time(); | |
542 | $no_orig_date = 'true'; | |
543 | } else { | |
544 | $no_orig_date = 'false'; | |
545 | } | |
546 | ||
547 | $entry_timestamp_fmt = strftime("%Y/%m/%d %H:%M:%S", $entry_timestamp); | |
548 | ||
549 | if ($debug_enabled) { | |
550 | _debug("update_rss_feed: date $entry_timestamp [$entry_timestamp_fmt]"); | |
551 | } | |
552 | ||
553 | if ($use_simplepie) { | |
554 | $entry_title = $item->get_title(); | |
555 | } else { | |
556 | $entry_title = trim(strip_tags($item["title"])); | |
557 | } | |
558 | ||
559 | if ($use_simplepie) { | |
560 | $entry_link = $item->get_link(); | |
561 | } else { | |
562 | // strange Magpie workaround | |
563 | $entry_link = $item["link_"]; | |
564 | if (!$entry_link) $entry_link = $item["link"]; | |
565 | } | |
566 | ||
567 | $entry_link = rewrite_relative_url($site_url, $entry_link); | |
568 | ||
569 | if ($debug_enabled) { | |
570 | _debug("update_rss_feed: title $entry_title"); | |
571 | _debug("update_rss_feed: link $entry_link"); | |
572 | } | |
573 | ||
574 | if (!$entry_title) $entry_title = date("Y-m-d H:i:s", $entry_timestamp);; | |
575 | ||
576 | $entry_link = strip_tags($entry_link); | |
577 | ||
578 | if ($use_simplepie) { | |
579 | $entry_content = $item->get_content(); | |
580 | if (!$entry_content) $entry_content = $item->get_description(); | |
581 | } else { | |
582 | $entry_content = $item["content:escaped"]; | |
583 | ||
584 | if (!$entry_content) $entry_content = $item["content:encoded"]; | |
585 | if (!$entry_content && is_array($entry_content)) $entry_content = $item["content"]["encoded"]; | |
586 | if (!$entry_content) $entry_content = $item["content"]; | |
587 | ||
588 | if (is_array($entry_content)) $entry_content = $entry_content[0]; | |
589 | ||
590 | // Magpie bugs are getting ridiculous | |
591 | if (trim($entry_content) == "Array") $entry_content = false; | |
592 | ||
593 | if (!$entry_content) $entry_content = $item["atom_content"]; | |
594 | if (!$entry_content) $entry_content = $item["summary"]; | |
595 | ||
596 | if (!$entry_content || | |
597 | strlen($entry_content) < strlen($item["description"])) { | |
598 | $entry_content = $item["description"]; | |
599 | }; | |
600 | ||
601 | // WTF | |
602 | if (is_array($entry_content)) { | |
603 | $entry_content = $entry_content["encoded"]; | |
604 | if (!$entry_content) $entry_content = $entry_content["escaped"]; | |
605 | } | |
606 | } | |
607 | ||
608 | if ($cache_images && is_writable(CACHE_DIR . '/images')) | |
609 | $entry_content = cache_images($entry_content, $site_url, $debug_enabled); | |
610 | ||
611 | if ($_REQUEST["xdebug"] == 2) { | |
612 | print "update_rss_feed: content: "; | |
613 | print $entry_content; | |
614 | print "\n"; | |
615 | } | |
616 | ||
617 | $entry_cached_content = ""; | |
618 | ||
619 | if ($use_simplepie) { | |
620 | $entry_comments = strip_tags($item->data["comments"]); | |
621 | if ($item->get_author()) { | |
622 | $entry_author_item = $item->get_author(); | |
623 | $entry_author = $entry_author_item->get_name(); | |
624 | if (!$entry_author) $entry_author = $entry_author_item->get_email(); | |
625 | ||
626 | $entry_author = db_escape_string($entry_author); | |
627 | } | |
628 | } else { | |
629 | $entry_comments = strip_tags($item["comments"]); | |
630 | ||
631 | $entry_author = db_escape_string(strip_tags($item['dc']['creator'])); | |
632 | ||
633 | if ($item['author']) { | |
634 | ||
635 | if (is_array($item['author'])) { | |
636 | ||
637 | if (!$entry_author) { | |
638 | $entry_author = db_escape_string(strip_tags($item['author']['name'])); | |
639 | } | |
640 | ||
641 | if (!$entry_author) { | |
642 | $entry_author = db_escape_string(strip_tags($item['author']['email'])); | |
643 | } | |
644 | } | |
645 | ||
646 | if (!$entry_author) { | |
647 | $entry_author = db_escape_string(strip_tags($item['author'])); | |
648 | } | |
649 | } | |
650 | } | |
651 | ||
652 | if (preg_match('/^[\t\n\r ]*$/', $entry_author)) $entry_author = ''; | |
653 | ||
654 | $entry_guid = db_escape_string(strip_tags($entry_guid)); | |
655 | $entry_guid = mb_substr($entry_guid, 0, 250); | |
656 | ||
657 | $result = db_query($link, "SELECT id FROM ttrss_entries | |
658 | WHERE guid = '$entry_guid'"); | |
659 | ||
660 | $entry_comments = mb_substr(db_escape_string($entry_comments), 0, 250); | |
661 | $entry_author = mb_substr($entry_author, 0, 250); | |
662 | ||
663 | if ($use_simplepie) { | |
664 | $num_comments = $item->get_item_tags('http://purl.org/rss/1.0/modules/slash/', 'comments'); | |
665 | ||
666 | if (is_array($num_comments) && is_array($num_comments[0])) { | |
667 | $num_comments = (int) $num_comments[0]["data"]; | |
668 | } else { | |
669 | $num_comments = 0; | |
670 | } | |
671 | } else { | |
672 | $num_comments = (int) $item["slash"]["comments"]; | |
673 | } | |
674 | ||
675 | if ($debug_enabled) { | |
676 | _debug("update_rss_feed: num_comments: $num_comments"); | |
677 | _debug("update_rss_feed: looking for tags [1]..."); | |
678 | } | |
679 | ||
680 | // parse <category> entries into tags | |
681 | ||
682 | $additional_tags = array(); | |
683 | ||
684 | if ($use_simplepie) { | |
685 | ||
686 | $additional_tags_src = $item->get_categories(); | |
687 | ||
688 | if (is_array($additional_tags_src)) { | |
689 | foreach ($additional_tags_src as $tobj) { | |
690 | array_push($additional_tags, $tobj->get_term()); | |
691 | } | |
692 | } | |
693 | ||
694 | if ($debug_enabled) { | |
695 | _debug("update_rss_feed: category tags:"); | |
696 | print_r($additional_tags); | |
697 | } | |
698 | ||
699 | } else { | |
700 | ||
701 | $t_ctr = $item['category#']; | |
702 | ||
703 | if ($t_ctr == 0) { | |
704 | $additional_tags = array(); | |
705 | } else if ($t_ctr > 0) { | |
706 | $additional_tags = array($item['category']); | |
707 | ||
708 | if ($item['category@term']) { | |
709 | array_push($additional_tags, $item['category@term']); | |
710 | } | |
711 | ||
712 | for ($i = 0; $i <= $t_ctr; $i++ ) { | |
713 | if ($item["category#$i"]) { | |
714 | array_push($additional_tags, $item["category#$i"]); | |
715 | } | |
716 | ||
717 | if ($item["category#$i@term"]) { | |
718 | array_push($additional_tags, $item["category#$i@term"]); | |
719 | } | |
720 | } | |
721 | } | |
722 | ||
723 | // parse <dc:subject> elements | |
724 | ||
725 | $t_ctr = $item['dc']['subject#']; | |
726 | ||
727 | if ($t_ctr > 0) { | |
728 | array_push($additional_tags, $item['dc']['subject']); | |
729 | ||
730 | for ($i = 0; $i <= $t_ctr; $i++ ) { | |
731 | if ($item['dc']["subject#$i"]) { | |
732 | array_push($additional_tags, $item['dc']["subject#$i"]); | |
733 | } | |
734 | } | |
735 | } | |
736 | } | |
737 | ||
738 | if ($debug_enabled) { | |
739 | _debug("update_rss_feed: looking for tags [2]..."); | |
740 | } | |
741 | ||
742 | /* taaaags */ | |
743 | // <a href="..." rel="tag">Xorg</a>, // | |
744 | ||
745 | $entry_tags = null; | |
746 | ||
747 | preg_match_all("/<a.*?rel=['\"]tag['\"].*?\>([^<]+)<\/a>/i", | |
748 | $entry_content, $entry_tags); | |
749 | ||
750 | $entry_tags = $entry_tags[1]; | |
751 | ||
752 | $entry_tags = array_merge($entry_tags, $additional_tags); | |
753 | $entry_tags = array_unique($entry_tags); | |
754 | ||
755 | for ($i = 0; $i < count($entry_tags); $i++) | |
756 | $entry_tags[$i] = mb_strtolower($entry_tags[$i], 'utf-8'); | |
757 | ||
758 | if ($debug_enabled) { | |
759 | //_debug("update_rss_feed: unfiltered tags found:"); | |
760 | //print_r($entry_tags); | |
761 | } | |
762 | ||
763 | if ($debug_enabled) { | |
764 | _debug("update_rss_feed: done collecting data [TITLE:$entry_title]"); | |
765 | } | |
766 | ||
767 | // TODO: less memory-hungry implementation | |
768 | global $pluginhost; | |
769 | ||
770 | foreach ($pluginhost->get_hooks($pluginhost::HOOK_ARTICLE_FILTER) as $p) { | |
771 | if ($debug_enabled) { | |
772 | _debug("update_rss_feed: applying plugin filters..."); | |
773 | } | |
774 | ||
775 | $article = array("owner_uid" => $owner_uid, | |
776 | "title" => $entry_title, | |
777 | "content" => $entry_content, | |
778 | "link" => $entry_link, | |
779 | "tags" => $entry_tags, | |
780 | "author" => $entry_author); | |
781 | ||
782 | foreach ($filter_plugins as $plugin) { | |
783 | $article = $plugin->hook_article_filter($article); | |
784 | } | |
785 | ||
786 | $entry_title = $article["title"]; | |
787 | $entry_content = $article["content"]; | |
788 | $entry_tags = $article["tags"]; | |
789 | $entry_author = $article["author"]; | |
790 | } | |
791 | ||
792 | $entry_content = db_escape_string($entry_content, false); | |
793 | $entry_title = db_escape_string($entry_title); | |
794 | $entry_author = db_escape_string($entry_author); | |
795 | $entry_link = db_escape_string($entry_link); | |
796 | ||
797 | $content_hash = "SHA1:" . sha1(strip_tags($entry_content)); | |
798 | ||
799 | db_query($link, "BEGIN"); | |
800 | ||
801 | if (db_num_rows($result) == 0) { | |
802 | ||
803 | if ($debug_enabled) { | |
804 | _debug("update_rss_feed: base guid not found"); | |
805 | } | |
806 | ||
807 | if ($cache_content) { | |
808 | if ($debug_enabled) { | |
809 | _debug("update_rss_feed: caching content (initial)..."); | |
810 | } | |
811 | ||
812 | $entry_cached_content = cache_content($link, $entry_link, $auth_login, $auth_pass); | |
813 | ||
814 | if ($cache_images && is_writable(CACHE_DIR . '/images')) | |
815 | $entry_cached_content = cache_images($entry_cached_content, $site_url, $debug_enabled); | |
816 | ||
817 | $entry_cached_content = db_escape_string($entry_cached_content, false); | |
818 | } | |
819 | ||
820 | // base post entry does not exist, create it | |
821 | ||
822 | $result = db_query($link, | |
823 | "INSERT INTO ttrss_entries | |
824 | (title, | |
825 | guid, | |
826 | link, | |
827 | updated, | |
828 | content, | |
829 | content_hash, | |
830 | cached_content, | |
831 | no_orig_date, | |
832 | date_updated, | |
833 | date_entered, | |
834 | comments, | |
835 | num_comments, | |
836 | author) | |
837 | VALUES | |
838 | ('$entry_title', | |
839 | '$entry_guid', | |
840 | '$entry_link', | |
841 | '$entry_timestamp_fmt', | |
842 | '$entry_content', | |
843 | '$content_hash', | |
844 | '$entry_cached_content', | |
845 | $no_orig_date, | |
846 | NOW(), | |
847 | NOW(), | |
848 | '$entry_comments', | |
849 | '$num_comments', | |
850 | '$entry_author')"); | |
851 | ||
852 | $article_labels = array(); | |
853 | ||
854 | } else { | |
855 | // we keep encountering the entry in feeds, so we need to | |
856 | // update date_updated column so that we don't get horrible | |
857 | // dupes when the entry gets purged and reinserted again e.g. | |
858 | // in the case of SLOW SLOW OMG SLOW updating feeds | |
859 | ||
860 | $base_entry_id = db_fetch_result($result, 0, "id"); | |
861 | ||
862 | db_query($link, "UPDATE ttrss_entries SET date_updated = NOW() | |
863 | WHERE id = '$base_entry_id'"); | |
864 | ||
865 | $article_labels = get_article_labels($link, $base_entry_id, $owner_uid); | |
866 | } | |
867 | ||
868 | // now it should exist, if not - bad luck then | |
869 | ||
870 | $result = db_query($link, "SELECT | |
871 | id,content_hash,no_orig_date,title, | |
872 | ".SUBSTRING_FOR_DATE."(date_updated,1,19) as date_updated, | |
873 | ".SUBSTRING_FOR_DATE."(updated,1,19) as updated, | |
874 | num_comments, cached_content | |
875 | FROM | |
876 | ttrss_entries | |
877 | WHERE guid = '$entry_guid'"); | |
878 | ||
879 | $entry_ref_id = 0; | |
880 | $entry_int_id = 0; | |
881 | ||
882 | if (db_num_rows($result) == 1) { | |
883 | ||
884 | if ($debug_enabled) { | |
885 | _debug("update_rss_feed: base guid found, checking for user record"); | |
886 | } | |
887 | ||
888 | // this will be used below in update handler | |
889 | $orig_content_hash = db_fetch_result($result, 0, "content_hash"); | |
890 | $orig_title = db_fetch_result($result, 0, "title"); | |
891 | $orig_num_comments = db_fetch_result($result, 0, "num_comments"); | |
892 | $orig_cached_content = trim(db_fetch_result($result, 0, "cached_content")); | |
893 | $orig_date_updated = strtotime(db_fetch_result($result, | |
894 | 0, "date_updated")); | |
895 | ||
896 | $ref_id = db_fetch_result($result, 0, "id"); | |
897 | $entry_ref_id = $ref_id; | |
898 | ||
899 | // check for user post link to main table | |
900 | ||
901 | // do we allow duplicate posts with same GUID in different feeds? | |
902 | if (get_pref($link, "ALLOW_DUPLICATE_POSTS", $owner_uid, false)) { | |
903 | $dupcheck_qpart = "AND (feed_id = '$feed' OR feed_id IS NULL)"; | |
904 | } else { | |
905 | $dupcheck_qpart = ""; | |
906 | } | |
907 | ||
908 | /* Collect article tags here so we could filter by them: */ | |
909 | ||
910 | $article_filters = get_article_filters($filters, $entry_title, | |
911 | $entry_content, $entry_link, $entry_timestamp, $entry_author, | |
912 | $entry_tags); | |
913 | ||
914 | if ($debug_enabled) { | |
915 | _debug("update_rss_feed: article filters: "); | |
916 | if (count($article_filters) != 0) { | |
917 | print_r($article_filters); | |
918 | } | |
919 | } | |
920 | ||
921 | if (find_article_filter($article_filters, "filter")) { | |
922 | db_query($link, "COMMIT"); // close transaction in progress | |
923 | continue; | |
924 | } | |
925 | ||
926 | $score = calculate_article_score($article_filters); | |
927 | ||
928 | if ($debug_enabled) { | |
929 | _debug("update_rss_feed: initial score: $score"); | |
930 | } | |
931 | ||
932 | $query = "SELECT ref_id, int_id FROM ttrss_user_entries WHERE | |
933 | ref_id = '$ref_id' AND owner_uid = '$owner_uid' | |
934 | $dupcheck_qpart"; | |
935 | ||
936 | // if ($_REQUEST["xdebug"]) print "$query\n"; | |
937 | ||
938 | $result = db_query($link, $query); | |
939 | ||
940 | // okay it doesn't exist - create user entry | |
941 | if (db_num_rows($result) == 0) { | |
942 | ||
943 | if ($debug_enabled) { | |
944 | _debug("update_rss_feed: user record not found, creating..."); | |
945 | } | |
946 | ||
947 | if ($score >= -500 && !find_article_filter($article_filters, 'catchup')) { | |
948 | $unread = 'true'; | |
949 | $last_read_qpart = 'NULL'; | |
950 | } else { | |
951 | $unread = 'false'; | |
952 | $last_read_qpart = 'NOW()'; | |
953 | } | |
954 | ||
955 | if (find_article_filter($article_filters, 'mark') || $score > 1000) { | |
956 | $marked = 'true'; | |
957 | } else { | |
958 | $marked = 'false'; | |
959 | } | |
960 | ||
961 | if (find_article_filter($article_filters, 'publish')) { | |
962 | $published = 'true'; | |
963 | } else { | |
964 | $published = 'false'; | |
965 | } | |
966 | ||
967 | // N-grams | |
968 | ||
969 | if (DB_TYPE == "pgsql" and defined('_NGRAM_TITLE_DUPLICATE_THRESHOLD')) { | |
970 | ||
971 | $result = db_query($link, "SELECT COUNT(*) AS similar FROM | |
972 | ttrss_entries,ttrss_user_entries | |
973 | WHERE ref_id = id AND updated >= NOW() - INTERVAL '7 day' | |
974 | AND similarity(title, '$entry_title') >= "._NGRAM_TITLE_DUPLICATE_THRESHOLD." | |
975 | AND owner_uid = $owner_uid"); | |
976 | ||
977 | $ngram_similar = db_fetch_result($result, 0, "similar"); | |
978 | ||
979 | if ($debug_enabled) { | |
980 | _debug("update_rss_feed: N-gram similar results: $ngram_similar"); | |
981 | } | |
982 | ||
983 | if ($ngram_similar > 0) { | |
984 | $unread = 'false'; | |
985 | } | |
986 | } | |
987 | ||
988 | $result = db_query($link, | |
989 | "INSERT INTO ttrss_user_entries | |
990 | (ref_id, owner_uid, feed_id, unread, last_read, marked, | |
991 | published, score, tag_cache, label_cache, uuid) | |
992 | VALUES ('$ref_id', '$owner_uid', '$feed', $unread, | |
993 | $last_read_qpart, $marked, $published, '$score', '', '', '')"); | |
994 | ||
995 | if (PUBSUBHUBBUB_HUB && $published == 'true') { | |
996 | $rss_link = get_self_url_prefix() . | |
997 | "/public.php?op=rss&id=-2&key=" . | |
998 | get_feed_access_key($link, -2, false, $owner_uid); | |
999 | ||
1000 | $p = new Publisher(PUBSUBHUBBUB_HUB); | |
1001 | ||
1002 | $pubsub_result = $p->publish_update($rss_link); | |
1003 | } | |
1004 | ||
1005 | $result = db_query($link, | |
1006 | "SELECT int_id FROM ttrss_user_entries WHERE | |
1007 | ref_id = '$ref_id' AND owner_uid = '$owner_uid' AND | |
1008 | feed_id = '$feed' LIMIT 1"); | |
1009 | ||
1010 | if (db_num_rows($result) == 1) { | |
1011 | $entry_int_id = db_fetch_result($result, 0, "int_id"); | |
1012 | } | |
1013 | } else { | |
1014 | if ($debug_enabled) { | |
1015 | _debug("update_rss_feed: user record FOUND"); | |
1016 | } | |
1017 | ||
1018 | $entry_ref_id = db_fetch_result($result, 0, "ref_id"); | |
1019 | $entry_int_id = db_fetch_result($result, 0, "int_id"); | |
1020 | } | |
1021 | ||
1022 | if ($debug_enabled) { | |
1023 | _debug("update_rss_feed: RID: $entry_ref_id, IID: $entry_int_id"); | |
1024 | } | |
1025 | ||
1026 | $post_needs_update = false; | |
1027 | $update_insignificant = false; | |
1028 | $cached_content_needs_update = false; | |
1029 | ||
1030 | if ($orig_num_comments != $num_comments) { | |
1031 | $post_needs_update = true; | |
1032 | $update_insignificant = true; | |
1033 | } | |
1034 | ||
1035 | if ($content_hash != $orig_content_hash) { | |
1036 | $post_needs_update = true; | |
1037 | $update_insignificant = false; | |
1038 | $cached_content_needs_update = true; | |
1039 | } | |
1040 | ||
1041 | if ($cache_content) { | |
1042 | if ($debug_enabled) { | |
1043 | _debug("update_rss_feed: caching content because original checksum changed..."); | |
1044 | } | |
1045 | ||
1046 | $entry_cached_content = cache_content($link, $entry_link, $auth_login, $auth_pass); | |
1047 | ||
1048 | if ($entry_cached_content) { | |
1049 | if ($cache_images && is_writable(CACHE_DIR . '/images')) | |
1050 | $entry_cached_content = cache_images($entry_cached_content, $site_url, $debug_enabled); | |
1051 | ||
1052 | $entry_cached_content = db_escape_string($entry_cached_content, false); | |
1053 | $post_needs_update = true; | |
1054 | } else { | |
1055 | $entry_cached_content = db_escape_string($orig_cached_content); | |
1056 | } | |
1057 | } else { | |
1058 | $entry_cached_content = db_escape_string($orig_cached_content); | |
1059 | } | |
1060 | ||
1061 | if (db_escape_string($orig_title) != $entry_title) { | |
1062 | $post_needs_update = true; | |
1063 | $update_insignificant = false; | |
1064 | } | |
1065 | ||
1066 | // if post needs update, update it and mark all user entries | |
1067 | // linking to this post as updated | |
1068 | if ($post_needs_update) { | |
1069 | ||
1070 | if (defined('DAEMON_EXTENDED_DEBUG')) { | |
1071 | _debug("update_rss_feed: post $entry_guid needs update..."); | |
1072 | } | |
1073 | ||
1074 | // print "<!-- post $orig_title needs update : $post_needs_update -->"; | |
1075 | ||
1076 | db_query($link, "UPDATE ttrss_entries | |
1077 | SET title = '$entry_title', content = '$entry_content', | |
1078 | content_hash = '$content_hash', | |
1079 | cached_content = '$entry_cached_content', | |
1080 | updated = '$entry_timestamp_fmt', | |
1081 | num_comments = '$num_comments' | |
1082 | WHERE id = '$ref_id'"); | |
1083 | ||
1084 | if (!$update_insignificant) { | |
1085 | if ($mark_unread_on_update) { | |
1086 | db_query($link, "UPDATE ttrss_user_entries | |
1087 | SET last_read = null, unread = true WHERE ref_id = '$ref_id'"); | |
1088 | } else if ($update_on_checksum_change) { | |
1089 | db_query($link, "UPDATE ttrss_user_entries | |
1090 | SET last_read = null WHERE ref_id = '$ref_id' | |
1091 | AND unread = false"); | |
1092 | } | |
1093 | } | |
1094 | } | |
1095 | } | |
1096 | ||
1097 | db_query($link, "COMMIT"); | |
1098 | ||
1099 | if ($debug_enabled) { | |
1100 | _debug("update_rss_feed: assigning labels..."); | |
1101 | } | |
1102 | ||
1103 | assign_article_to_label_filters($link, $entry_ref_id, $article_filters, | |
1104 | $owner_uid, $article_labels); | |
1105 | ||
1106 | if ($debug_enabled) { | |
1107 | _debug("update_rss_feed: looking for enclosures..."); | |
1108 | } | |
1109 | ||
1110 | // enclosures | |
1111 | ||
1112 | $enclosures = array(); | |
1113 | ||
1114 | if ($use_simplepie) { | |
1115 | $encs = $item->get_enclosures(); | |
1116 | ||
1117 | if (is_array($encs)) { | |
1118 | foreach ($encs as $e) { | |
1119 | $e_item = array( | |
1120 | $e->link, $e->type, $e->length); | |
1121 | ||
1122 | array_push($enclosures, $e_item); | |
1123 | } | |
1124 | } | |
1125 | ||
1126 | } else { | |
1127 | // <enclosure> | |
1128 | ||
1129 | $e_ctr = $item['enclosure#']; | |
1130 | ||
1131 | if ($e_ctr > 0) { | |
1132 | $e_item = array($item['enclosure@url'], | |
1133 | $item['enclosure@type'], | |
1134 | $item['enclosure@length']); | |
1135 | ||
1136 | array_push($enclosures, $e_item); | |
1137 | ||
1138 | for ($i = 0; $i <= $e_ctr; $i++ ) { | |
1139 | ||
1140 | if ($item["enclosure#$i@url"]) { | |
1141 | $e_item = array($item["enclosure#$i@url"], | |
1142 | $item["enclosure#$i@type"], | |
1143 | $item["enclosure#$i@length"]); | |
1144 | array_push($enclosures, $e_item); | |
1145 | } | |
1146 | } | |
1147 | } | |
1148 | ||
1149 | // <media:content> | |
1150 | // can there be many of those? yes -fox | |
1151 | ||
1152 | $m_ctr = $item['media']['content#']; | |
1153 | ||
1154 | if ($m_ctr > 0) { | |
1155 | $e_item = array($item['media']['content@url'], | |
1156 | $item['media']['content@medium'], | |
1157 | $item['media']['content@length']); | |
1158 | ||
1159 | array_push($enclosures, $e_item); | |
1160 | ||
1161 | for ($i = 0; $i <= $m_ctr; $i++ ) { | |
1162 | ||
1163 | if ($item["media"]["content#$i@url"]) { | |
1164 | $e_item = array($item["media"]["content#$i@url"], | |
1165 | $item["media"]["content#$i@medium"], | |
1166 | $item["media"]["content#$i@length"]); | |
1167 | array_push($enclosures, $e_item); | |
1168 | } | |
1169 | } | |
1170 | ||
1171 | } | |
1172 | } | |
1173 | ||
1174 | ||
1175 | if ($debug_enabled) { | |
1176 | _debug("update_rss_feed: article enclosures:"); | |
1177 | print_r($enclosures); | |
1178 | } | |
1179 | ||
1180 | db_query($link, "BEGIN"); | |
1181 | ||
1182 | foreach ($enclosures as $enc) { | |
1183 | $enc_url = db_escape_string($enc[0]); | |
1184 | $enc_type = db_escape_string($enc[1]); | |
1185 | $enc_dur = db_escape_string($enc[2]); | |
1186 | ||
1187 | $result = db_query($link, "SELECT id FROM ttrss_enclosures | |
1188 | WHERE content_url = '$enc_url' AND post_id = '$entry_ref_id'"); | |
1189 | ||
1190 | if (db_num_rows($result) == 0) { | |
1191 | db_query($link, "INSERT INTO ttrss_enclosures | |
1192 | (content_url, content_type, title, duration, post_id) VALUES | |
1193 | ('$enc_url', '$enc_type', '', '$enc_dur', '$entry_ref_id')"); | |
1194 | } | |
1195 | } | |
1196 | ||
1197 | db_query($link, "COMMIT"); | |
1198 | ||
1199 | // check for manual tags (we have to do it here since they're loaded from filters) | |
1200 | ||
1201 | foreach ($article_filters as $f) { | |
1202 | if ($f["type"] == "tag") { | |
1203 | ||
1204 | $manual_tags = trim_array(explode(",", $f["param"])); | |
1205 | ||
1206 | foreach ($manual_tags as $tag) { | |
1207 | if (tag_is_valid($tag)) { | |
1208 | array_push($entry_tags, $tag); | |
1209 | } | |
1210 | } | |
1211 | } | |
1212 | } | |
1213 | ||
1214 | // Skip boring tags | |
1215 | ||
1216 | $boring_tags = trim_array(explode(",", mb_strtolower(get_pref($link, | |
1217 | 'BLACKLISTED_TAGS', $owner_uid, ''), 'utf-8'))); | |
1218 | ||
1219 | $filtered_tags = array(); | |
1220 | $tags_to_cache = array(); | |
1221 | ||
1222 | if ($entry_tags && is_array($entry_tags)) { | |
1223 | foreach ($entry_tags as $tag) { | |
1224 | if (array_search($tag, $boring_tags) === false) { | |
1225 | array_push($filtered_tags, $tag); | |
1226 | } | |
1227 | } | |
1228 | } | |
1229 | ||
1230 | $filtered_tags = array_unique($filtered_tags); | |
1231 | ||
1232 | if ($debug_enabled) { | |
1233 | _debug("update_rss_feed: filtered article tags:"); | |
1234 | print_r($filtered_tags); | |
1235 | } | |
1236 | ||
1237 | // Save article tags in the database | |
1238 | ||
1239 | if (count($filtered_tags) > 0) { | |
1240 | ||
1241 | db_query($link, "BEGIN"); | |
1242 | ||
1243 | foreach ($filtered_tags as $tag) { | |
1244 | ||
1245 | $tag = sanitize_tag($tag); | |
1246 | $tag = db_escape_string($tag); | |
1247 | ||
1248 | if (!tag_is_valid($tag)) continue; | |
1249 | ||
1250 | $result = db_query($link, "SELECT id FROM ttrss_tags | |
1251 | WHERE tag_name = '$tag' AND post_int_id = '$entry_int_id' AND | |
1252 | owner_uid = '$owner_uid' LIMIT 1"); | |
1253 | ||
1254 | if ($result && db_num_rows($result) == 0) { | |
1255 | ||
1256 | db_query($link, "INSERT INTO ttrss_tags | |
1257 | (owner_uid,tag_name,post_int_id) | |
1258 | VALUES ('$owner_uid','$tag', '$entry_int_id')"); | |
1259 | } | |
1260 | ||
1261 | array_push($tags_to_cache, $tag); | |
1262 | } | |
1263 | ||
1264 | /* update the cache */ | |
1265 | ||
1266 | $tags_to_cache = array_unique($tags_to_cache); | |
1267 | ||
1268 | $tags_str = db_escape_string(join(",", $tags_to_cache)); | |
1269 | ||
1270 | db_query($link, "UPDATE ttrss_user_entries | |
1271 | SET tag_cache = '$tags_str' WHERE ref_id = '$entry_ref_id' | |
1272 | AND owner_uid = $owner_uid"); | |
1273 | ||
1274 | db_query($link, "COMMIT"); | |
1275 | } | |
1276 | ||
1277 | if (get_pref($link, "AUTO_ASSIGN_LABELS", $owner_uid, false)) { | |
1278 | if ($debug_enabled) { | |
1279 | _debug("update_rss_feed: auto-assigning labels..."); | |
1280 | } | |
1281 | ||
1282 | foreach ($labels as $label) { | |
1283 | $caption = $label["caption"]; | |
1284 | ||
1285 | if (preg_match("/\b$caption\b/i", "$tags_str " . strip_tags($entry_content) . " $entry_title")) { | |
1286 | if (!labels_contains_caption($article_labels, $caption)) { | |
1287 | label_add_article($link, $entry_ref_id, $caption, $owner_uid); | |
1288 | } | |
1289 | } | |
1290 | } | |
1291 | } | |
1292 | ||
1293 | if ($debug_enabled) { | |
1294 | _debug("update_rss_feed: article processed"); | |
1295 | } | |
1296 | } | |
1297 | ||
1298 | if (!$last_updated) { | |
1299 | if ($debug_enabled) { | |
1300 | _debug("update_rss_feed: new feed, catching it up..."); | |
1301 | } | |
1302 | catchup_feed($link, $feed, false, $owner_uid); | |
1303 | } | |
1304 | ||
1305 | if ($debug_enabled) { | |
1306 | _debug("purging feed..."); | |
1307 | } | |
1308 | ||
1309 | purge_feed($link, $feed, 0, $debug_enabled); | |
1310 | ||
1311 | db_query($link, "UPDATE ttrss_feeds | |
1312 | SET last_updated = NOW(), last_error = '' WHERE id = '$feed'"); | |
1313 | ||
1314 | // db_query($link, "COMMIT"); | |
1315 | ||
1316 | } else { | |
1317 | ||
1318 | if ($use_simplepie) { | |
1319 | $error_msg = mb_substr($rss->error(), 0, 250); | |
1320 | } else { | |
1321 | $error_msg = mb_substr(magpie_error(), 0, 250); | |
1322 | } | |
1323 | ||
1324 | if ($debug_enabled) { | |
1325 | _debug("update_rss_feed: error fetching feed: $error_msg"); | |
1326 | } | |
1327 | ||
1328 | $error_msg = db_escape_string($error_msg); | |
1329 | ||
1330 | db_query($link, | |
1331 | "UPDATE ttrss_feeds SET last_error = '$error_msg', | |
1332 | last_updated = NOW() WHERE id = '$feed'"); | |
1333 | } | |
1334 | ||
1335 | if ($use_simplepie) { | |
1336 | unset($rss); | |
1337 | } | |
1338 | ||
1339 | if ($debug_enabled) { | |
1340 | _debug("update_rss_feed: done"); | |
1341 | } | |
1342 | ||
1343 | } | |
1344 | ||
1345 | function cache_images($html, $site_url, $debug) { | |
1346 | $cache_dir = CACHE_DIR . "/images"; | |
1347 | ||
1348 | libxml_use_internal_errors(true); | |
1349 | ||
1350 | $charset_hack = '<head> | |
1351 | <meta http-equiv="Content-Type" content="text/html; charset=utf-8"/> | |
1352 | </head>'; | |
1353 | ||
1354 | $doc = new DOMDocument(); | |
1355 | $doc->loadHTML($charset_hack . $html); | |
1356 | $xpath = new DOMXPath($doc); | |
1357 | ||
1358 | $entries = $xpath->query('(//img[@src])'); | |
1359 | ||
1360 | foreach ($entries as $entry) { | |
1361 | if ($entry->hasAttribute('src')) { | |
1362 | $src = rewrite_relative_url($site_url, $entry->getAttribute('src')); | |
1363 | ||
1364 | $local_filename = CACHE_DIR . "/images/" . sha1($src) . ".png"; | |
1365 | ||
1366 | if ($debug) _debug("cache_images: downloading: $src to $local_filename"); | |
1367 | ||
1368 | if (!file_exists($local_filename)) { | |
1369 | $file_content = fetch_file_contents($src); | |
1370 | ||
1371 | if ($file_content && strlen($file_content) > 1024) { | |
1372 | file_put_contents($local_filename, $file_content); | |
1373 | } | |
1374 | } | |
1375 | ||
1376 | if (file_exists($local_filename)) { | |
1377 | $entry->setAttribute('src', SELF_URL_PATH . '/image.php?url=' . | |
1378 | base64_encode($src)); | |
1379 | } | |
1380 | } | |
1381 | } | |
1382 | ||
1383 | $node = $doc->getElementsByTagName('body')->item(0); | |
1384 | ||
1385 | return $doc->saveXML($node, LIBXML_NOEMPTYTAG); | |
1386 | } | |
1387 | ||
1388 | function expire_lock_files($debug) { | |
1389 | if ($debug) _debug("Removing old lock files..."); | |
1390 | ||
1391 | $num_deleted = 0; | |
1392 | ||
1393 | if (is_writable(LOCK_DIRECTORY)) { | |
1394 | $files = glob(LOCK_DIRECTORY . "/*.lock"); | |
1395 | ||
1396 | if ($files) { | |
1397 | foreach ($files as $file) { | |
1398 | if (!file_is_locked($file) && time() - filemtime($file) > 86400*2) { | |
1399 | unlink($file); | |
1400 | ++$num_deleted; | |
1401 | } | |
1402 | } | |
1403 | } | |
1404 | } | |
1405 | ||
1406 | if ($debug) _debug("Removed $num_deleted files."); | |
1407 | } | |
1408 | ||
1409 | function expire_cached_files($debug) { | |
1410 | foreach (array("magpie", "simplepie", "images", "export") as $dir) { | |
1411 | $cache_dir = CACHE_DIR . "/$dir"; | |
1412 | ||
1413 | if ($debug) _debug("Expiring $cache_dir"); | |
1414 | ||
1415 | $num_deleted = 0; | |
1416 | ||
1417 | if (is_writable($cache_dir)) { | |
1418 | $files = glob("$cache_dir/*"); | |
1419 | ||
1420 | if ($files) { | |
1421 | foreach ($files as $file) { | |
1422 | if (time() - filemtime($file) > 86400*7) { | |
1423 | unlink($file); | |
1424 | ||
1425 | ++$num_deleted; | |
1426 | } | |
1427 | } | |
1428 | } | |
1429 | } | |
1430 | ||
1431 | if ($debug) _debug("Removed $num_deleted files."); | |
1432 | } | |
1433 | } | |
1434 | ||
1435 | /** | |
1436 | * Source: http://www.php.net/manual/en/function.parse-url.php#104527 | |
1437 | * Returns the url query as associative array | |
1438 | * | |
1439 | * @param string query | |
1440 | * @return array params | |
1441 | */ | |
1442 | function convertUrlQuery($query) { | |
1443 | $queryParts = explode('&', $query); | |
1444 | ||
1445 | $params = array(); | |
1446 | ||
1447 | foreach ($queryParts as $param) { | |
1448 | $item = explode('=', $param); | |
1449 | $params[$item[0]] = $item[1]; | |
1450 | } | |
1451 | ||
1452 | return $params; | |
1453 | } | |
1454 | ||
1455 | function get_article_filters($filters, $title, $content, $link, $timestamp, $author, $tags) { | |
1456 | $matches = array(); | |
1457 | ||
1458 | foreach ($filters as $filter) { | |
1459 | $match_any_rule = $filter["match_any_rule"]; | |
1460 | $filter_match = false; | |
1461 | ||
1462 | foreach ($filter["rules"] as $rule) { | |
1463 | $match = false; | |
1464 | $reg_exp = $rule["reg_exp"]; | |
1465 | ||
1466 | if (!$reg_exp) | |
1467 | continue; | |
1468 | ||
1469 | switch ($rule["type"]) { | |
1470 | case "title": | |
1471 | $match = @preg_match("/$reg_exp/i", $title); | |
1472 | break; | |
1473 | case "content": | |
1474 | // we don't need to deal with multiline regexps | |
1475 | $content = preg_replace("/[\r\n\t]/", "", $content); | |
1476 | ||
1477 | $match = @preg_match("/$reg_exp/i", $content); | |
1478 | break; | |
1479 | case "both": | |
1480 | // we don't need to deal with multiline regexps | |
1481 | $content = preg_replace("/[\r\n\t]/", "", $content); | |
1482 | ||
1483 | $match = (@preg_match("/$reg_exp/i", $title) || @preg_match("/$reg_exp/i", $content)); | |
1484 | break; | |
1485 | case "link": | |
1486 | $match = @preg_match("/$reg_exp/i", $link); | |
1487 | break; | |
1488 | case "author": | |
1489 | $match = @preg_match("/$reg_exp/i", $author); | |
1490 | break; | |
1491 | case "tag": | |
1492 | $tag_string = join(",", $tags); | |
1493 | $match = @preg_match("/$reg_exp/i", $tag_string); | |
1494 | break; | |
1495 | } | |
1496 | ||
1497 | if ($match_any_rule) { | |
1498 | if ($match) { | |
1499 | $filter_match = true; | |
1500 | break; | |
1501 | } | |
1502 | } else { | |
1503 | $filter_match = $match; | |
1504 | if (!$match) { | |
1505 | break; | |
1506 | } | |
1507 | } | |
1508 | } | |
1509 | ||
1510 | if ($filter_match) { | |
1511 | foreach ($filter["actions"] AS $action) { | |
1512 | array_push($matches, $action); | |
1513 | } | |
1514 | } | |
1515 | } | |
1516 | ||
1517 | return $matches; | |
1518 | } | |
1519 | ||
1520 | function find_article_filter($filters, $filter_name) { | |
1521 | foreach ($filters as $f) { | |
1522 | if ($f["type"] == $filter_name) { | |
1523 | return $f; | |
1524 | }; | |
1525 | } | |
1526 | return false; | |
1527 | } | |
1528 | ||
1529 | function find_article_filters($filters, $filter_name) { | |
1530 | $results = array(); | |
1531 | ||
1532 | foreach ($filters as $f) { | |
1533 | if ($f["type"] == $filter_name) { | |
1534 | array_push($results, $f); | |
1535 | }; | |
1536 | } | |
1537 | return $results; | |
1538 | } | |
1539 | ||
1540 | function calculate_article_score($filters) { | |
1541 | $score = 0; | |
1542 | ||
1543 | foreach ($filters as $f) { | |
1544 | if ($f["type"] == "score") { | |
1545 | $score += $f["param"]; | |
1546 | }; | |
1547 | } | |
1548 | return $score; | |
1549 | } | |
1550 | ||
1551 | function labels_contains_caption($labels, $caption) { | |
1552 | foreach ($labels as $label) { | |
1553 | if ($label[1] == $caption) { | |
1554 | return true; | |
1555 | } | |
1556 | } | |
1557 | ||
1558 | return false; | |
1559 | } | |
1560 | ||
1561 | function assign_article_to_label_filters($link, $id, $filters, $owner_uid, $article_labels) { | |
1562 | foreach ($filters as $f) { | |
1563 | if ($f["type"] == "label") { | |
1564 | if (!labels_contains_caption($article_labels, $f["param"])) { | |
1565 | label_add_article($link, $id, $f["param"], $owner_uid); | |
1566 | } | |
1567 | } | |
1568 | } | |
1569 | } | |
1570 | ||
1571 | function cache_content($link, $url, $login, $pass) { | |
1572 | ||
1573 | $content = fetch_file_contents($url, $login, $pass); | |
1574 | ||
1575 | if ($content) { | |
1576 | $doc = new DOMDocument(); | |
1577 | @$doc->loadHTML($content); | |
1578 | $xpath = new DOMXPath($doc); | |
1579 | ||
1580 | $node = $doc->getElementsByTagName('body')->item(0); | |
1581 | ||
1582 | if ($node) { | |
1583 | $content = $doc->saveXML($node, LIBXML_NOEMPTYTAG); | |
1584 | ||
1585 | return $content; | |
1586 | } | |
1587 | } | |
1588 | ||
1589 | return ""; | |
1590 | } | |
1591 | ?> |