]>
Commit | Line | Data |
---|---|---|
2c08214a | 1 | <?php |
09e8bdfd AD |
2 | define('DAEMON_UPDATE_LOGIN_LIMIT', 30); |
3 | define('DAEMON_FEED_LIMIT', 100); | |
0f556e3e | 4 | define('DAEMON_SLEEP_INTERVAL', 60); |
09e8bdfd | 5 | |
79178062 AD |
6 | function update_feedbrowser_cache($link) { |
7 | ||
8 | $result = db_query($link, "SELECT feed_url, site_url, title, COUNT(id) AS subscribers | |
9 | FROM ttrss_feeds WHERE (SELECT COUNT(id) = 0 FROM ttrss_feeds AS tf | |
10 | WHERE tf.feed_url = ttrss_feeds.feed_url | |
11 | AND (private IS true OR auth_login != '' OR auth_pass != '' OR feed_url LIKE '%:%@%/%')) | |
12 | GROUP BY feed_url, site_url, title ORDER BY subscribers DESC LIMIT 1000"); | |
13 | ||
14 | db_query($link, "BEGIN"); | |
15 | ||
16 | db_query($link, "DELETE FROM ttrss_feedbrowser_cache"); | |
17 | ||
18 | $count = 0; | |
19 | ||
20 | while ($line = db_fetch_assoc($result)) { | |
21 | $subscribers = db_escape_string($line["subscribers"]); | |
22 | $feed_url = db_escape_string($line["feed_url"]); | |
23 | $title = db_escape_string($line["title"]); | |
24 | $site_url = db_escape_string($line["site_url"]); | |
25 | ||
26 | $tmp_result = db_query($link, "SELECT subscribers FROM | |
27 | ttrss_feedbrowser_cache WHERE feed_url = '$feed_url'"); | |
28 | ||
29 | if (db_num_rows($tmp_result) == 0) { | |
30 | ||
31 | db_query($link, "INSERT INTO ttrss_feedbrowser_cache | |
32 | (feed_url, site_url, title, subscribers) VALUES ('$feed_url', | |
33 | '$site_url', '$title', '$subscribers')"); | |
34 | ||
35 | ++$count; | |
36 | ||
37 | } | |
38 | ||
39 | } | |
40 | ||
41 | db_query($link, "COMMIT"); | |
42 | ||
43 | return $count; | |
44 | ||
45 | } | |
46 | ||
47 | ||
2c08214a AD |
48 | /** |
49 | * Update a feed batch. | |
50 | * Used by daemons to update n feeds by run. | |
51 | * Only update feed needing a update, and not being processed | |
52 | * by another process. | |
53 | * | |
54 | * @param mixed $link Database link | |
55 | * @param integer $limit Maximum number of feeds in update batch. Default to DAEMON_FEED_LIMIT. | |
56 | * @param boolean $from_http Set to true if you call this function from http to disable cli specific code. | |
57 | * @param boolean $debug Set to false to disable debug output. Default to true. | |
58 | * @return void | |
59 | */ | |
60 | function update_daemon_common($link, $limit = DAEMON_FEED_LIMIT, $from_http = false, $debug = true) { | |
61 | // Process all other feeds using last_updated and interval parameters | |
62 | ||
61c1812f AD |
63 | define('PREFS_NO_CACHE', true); |
64 | ||
2c08214a | 65 | // Test if the user has loggued in recently. If not, it does not update its feeds. |
09e8bdfd | 66 | if (!SINGLE_USER_MODE && DAEMON_UPDATE_LOGIN_LIMIT > 0) { |
2c08214a AD |
67 | if (DB_TYPE == "pgsql") { |
68 | $login_thresh_qpart = "AND ttrss_users.last_login >= NOW() - INTERVAL '".DAEMON_UPDATE_LOGIN_LIMIT." days'"; | |
69 | } else { | |
70 | $login_thresh_qpart = "AND ttrss_users.last_login >= DATE_SUB(NOW(), INTERVAL ".DAEMON_UPDATE_LOGIN_LIMIT." DAY)"; | |
71 | } | |
72 | } else { | |
73 | $login_thresh_qpart = ""; | |
74 | } | |
75 | ||
76 | // Test if the feed need a update (update interval exceded). | |
77 | if (DB_TYPE == "pgsql") { | |
78 | $update_limit_qpart = "AND (( | |
79 | ttrss_feeds.update_interval = 0 | |
80 | AND ttrss_feeds.last_updated < NOW() - CAST((ttrss_user_prefs.value || ' minutes') AS INTERVAL) | |
81 | ) OR ( | |
82 | ttrss_feeds.update_interval > 0 | |
83 | AND ttrss_feeds.last_updated < NOW() - CAST((ttrss_feeds.update_interval || ' minutes') AS INTERVAL) | |
941e48a4 AD |
84 | ) OR ttrss_feeds.last_updated IS NULL |
85 | OR last_updated = '1970-01-01 00:00:00')"; | |
2c08214a AD |
86 | } else { |
87 | $update_limit_qpart = "AND (( | |
88 | ttrss_feeds.update_interval = 0 | |
89 | AND ttrss_feeds.last_updated < DATE_SUB(NOW(), INTERVAL CONVERT(ttrss_user_prefs.value, SIGNED INTEGER) MINUTE) | |
90 | ) OR ( | |
91 | ttrss_feeds.update_interval > 0 | |
92 | AND ttrss_feeds.last_updated < DATE_SUB(NOW(), INTERVAL ttrss_feeds.update_interval MINUTE) | |
941e48a4 AD |
93 | ) OR ttrss_feeds.last_updated IS NULL |
94 | OR last_updated = '1970-01-01 00:00:00')"; | |
2c08214a AD |
95 | } |
96 | ||
97 | // Test if feed is currently being updated by another process. | |
98 | if (DB_TYPE == "pgsql") { | |
99 | $updstart_thresh_qpart = "AND (ttrss_feeds.last_update_started IS NULL OR ttrss_feeds.last_update_started < NOW() - INTERVAL '5 minutes')"; | |
100 | } else { | |
101 | $updstart_thresh_qpart = "AND (ttrss_feeds.last_update_started IS NULL OR ttrss_feeds.last_update_started < DATE_SUB(NOW(), INTERVAL 5 MINUTE))"; | |
102 | } | |
103 | ||
104 | // Test if there is a limit to number of updated feeds | |
105 | $query_limit = ""; | |
106 | if($limit) $query_limit = sprintf("LIMIT %d", $limit); | |
107 | ||
108 | $random_qpart = sql_random_function(); | |
109 | ||
110 | // We search for feed needing update. | |
111 | $result = db_query($link, "SELECT ttrss_feeds.feed_url,ttrss_feeds.id, ttrss_feeds.owner_uid, | |
112 | ".SUBSTRING_FOR_DATE."(ttrss_feeds.last_updated,1,19) AS last_updated, | |
113 | ttrss_feeds.update_interval | |
114 | FROM | |
115 | ttrss_feeds, ttrss_users, ttrss_user_prefs | |
116 | WHERE | |
117 | ttrss_feeds.owner_uid = ttrss_users.id | |
118 | AND ttrss_users.id = ttrss_user_prefs.owner_uid | |
119 | AND ttrss_user_prefs.pref_name = 'DEFAULT_UPDATE_INTERVAL' | |
120 | $login_thresh_qpart $update_limit_qpart | |
121 | $updstart_thresh_qpart | |
122 | ORDER BY $random_qpart $query_limit"); | |
123 | ||
124 | $user_prefs_cache = array(); | |
125 | ||
126 | if($debug) _debug(sprintf("Scheduled %d feeds to update...\n", db_num_rows($result))); | |
127 | ||
128 | // Here is a little cache magic in order to minimize risk of double feed updates. | |
129 | $feeds_to_update = array(); | |
130 | while ($line = db_fetch_assoc($result)) { | |
131 | $feeds_to_update[$line['id']] = $line; | |
132 | } | |
133 | ||
134 | // We update the feed last update started date before anything else. | |
135 | // There is no lag due to feed contents downloads | |
136 | // It prevent an other process to update the same feed. | |
137 | $feed_ids = array_keys($feeds_to_update); | |
138 | if($feed_ids) { | |
139 | db_query($link, sprintf("UPDATE ttrss_feeds SET last_update_started = NOW() | |
140 | WHERE id IN (%s)", implode(',', $feed_ids))); | |
141 | } | |
142 | ||
3c696512 | 143 | expire_cached_files($debug); |
2a91b6ff | 144 | expire_lock_files($debug); |
3c696512 | 145 | |
2c08214a AD |
146 | // For each feed, we call the feed update function. |
147 | while ($line = array_pop($feeds_to_update)) { | |
148 | ||
149 | if($debug) _debug("Feed: " . $line["feed_url"] . ", " . $line["last_updated"]); | |
150 | ||
151 | update_rss_feed($link, $line["id"], true); | |
152 | ||
153 | sleep(1); // prevent flood (FIXME make this an option?) | |
154 | } | |
155 | ||
f1611683 AD |
156 | require_once "digest.php"; |
157 | ||
2c08214a | 158 | // Send feed digests by email if needed. |
8437c066 | 159 | send_headlines_digests($link, $debug); |
2c08214a | 160 | |
2ea09bde AD |
161 | // Purge feeds with stale data OR not being updated for a while to keep DB size down |
162 | ||
163 | if ($debug) _debug("Purging inactive feeds..."); | |
164 | ||
165 | if (DB_TYPE == "pgsql") { | |
166 | $interval_qpart = "NOW() - INTERVAL '1 month'"; | |
167 | } else { | |
168 | $interval_qpart = "DATE_SUB(NOW(), INTERVAL 1 MONTH)"; | |
169 | } | |
170 | ||
171 | $result = db_query($link, "SELECT id, feed_url FROM ttrss_feeds WHERE | |
172 | (SELECT MAX(updated) FROM ttrss_entries, ttrss_user_entries WHERE | |
173 | ttrss_entries.id = ref_id AND | |
174 | ttrss_user_entries.feed_id = ttrss_feeds.id) < $interval_qpart OR | |
175 | last_updated < $interval_qpart"); | |
176 | ||
177 | $feeds_purged = 0; | |
178 | $articles_removed = 0; | |
179 | ||
180 | while ($line = db_fetch_assoc($result)) { | |
181 | $articles_removed += purge_feed($link, $line["id"], 0, false); | |
182 | ++$feeds_purged; | |
183 | } | |
184 | ||
185 | if ($debug && $articles_removed > 0) | |
186 | _debug(sprintf("Purged %d feeds (%d articles).", $feeds_purged, | |
187 | $articles_removed)); | |
188 | ||
2c08214a AD |
189 | } // function update_daemon_common |
190 | ||
87764a50 | 191 | // ignore_daemon is not used |
0e4a7d7a | 192 | function update_rss_feed($link, $feed, $ignore_daemon = false, $no_cache = false, |
2c08214a AD |
193 | $override_url = false) { |
194 | ||
195 | require_once "lib/simplepie/simplepie.inc"; | |
2c08214a | 196 | |
2c08214a AD |
197 | $debug_enabled = defined('DAEMON_EXTENDED_DEBUG') || $_REQUEST['xdebug']; |
198 | ||
2c08214a AD |
199 | if ($debug_enabled) { |
200 | _debug("update_rss_feed: start"); | |
201 | } | |
202 | ||
87764a50 | 203 | $result = db_query($link, "SELECT id,update_interval,auth_login, |
0a3fd79b | 204 | feed_url,auth_pass,cache_images,last_updated, |
87764a50 AD |
205 | mark_unread_on_update, owner_uid, update_on_checksum_change, |
206 | pubsub_state | |
207 | FROM ttrss_feeds WHERE id = '$feed'"); | |
2c08214a AD |
208 | |
209 | if (db_num_rows($result) == 0) { | |
210 | if ($debug_enabled) { | |
211 | _debug("update_rss_feed: feed $feed NOT FOUND/SKIPPED"); | |
212 | } | |
213 | return false; | |
214 | } | |
215 | ||
2c08214a AD |
216 | $last_updated = db_fetch_result($result, 0, "last_updated"); |
217 | $owner_uid = db_fetch_result($result, 0, "owner_uid"); | |
218 | $mark_unread_on_update = sql_bool_to_bool(db_fetch_result($result, | |
219 | 0, "mark_unread_on_update")); | |
220 | $update_on_checksum_change = sql_bool_to_bool(db_fetch_result($result, | |
221 | 0, "update_on_checksum_change")); | |
222 | $pubsub_state = db_fetch_result($result, 0, "pubsub_state"); | |
223 | ||
224 | db_query($link, "UPDATE ttrss_feeds SET last_update_started = NOW() | |
225 | WHERE id = '$feed'"); | |
226 | ||
227 | $auth_login = db_fetch_result($result, 0, "auth_login"); | |
228 | $auth_pass = db_fetch_result($result, 0, "auth_pass"); | |
229 | ||
2c08214a AD |
230 | $cache_images = sql_bool_to_bool(db_fetch_result($result, 0, "cache_images")); |
231 | $fetch_url = db_fetch_result($result, 0, "feed_url"); | |
232 | ||
2c08214a AD |
233 | $feed = db_escape_string($feed); |
234 | ||
017401dd | 235 | /* if ($auth_login && $auth_pass ){ |
2c08214a AD |
236 | $url_parts = array(); |
237 | preg_match("/(^[^:]*):\/\/(.*)/", $fetch_url, $url_parts); | |
238 | ||
239 | if ($url_parts[1] && $url_parts[2]) { | |
240 | $fetch_url = $url_parts[1] . "://$auth_login:$auth_pass@" . $url_parts[2]; | |
241 | } | |
017401dd | 242 | } */ |
2c08214a AD |
243 | |
244 | if ($override_url) | |
245 | $fetch_url = $override_url; | |
246 | ||
247 | if ($debug_enabled) { | |
248 | _debug("update_rss_feed: fetching [$fetch_url]..."); | |
249 | } | |
250 | ||
0e4a7d7a AD |
251 | // Ignore cache if new feed or manual update. |
252 | $cache_age = (is_null($last_updated) || $last_updated == '1970-01-01 00:00:00') ? | |
253 | -1 : get_feed_update_interval($link, $feed) * 60; | |
2c08214a | 254 | |
19b3992b | 255 | $simplepie_cache_dir = CACHE_DIR . "/simplepie"; |
2c08214a | 256 | |
19b3992b AD |
257 | if (!is_dir($simplepie_cache_dir)) { |
258 | mkdir($simplepie_cache_dir); | |
259 | } | |
2c08214a | 260 | |
017401dd AD |
261 | $feed_data = fetch_file_contents($fetch_url, false, |
262 | $auth_login, $auth_pass, false, $no_cache ? 15 : 45); | |
263 | ||
264 | if (!$feed_data) { | |
265 | global $fetch_last_error; | |
266 | ||
267 | if ($debug_enabled) { | |
268 | _debug("update_rss_feed: unable to fetch: $fetch_last_error"); | |
269 | } | |
270 | ||
271 | db_query($link, | |
272 | "UPDATE ttrss_feeds SET last_error = '$fetch_last_error', | |
273 | last_updated = NOW() WHERE id = '$feed'"); | |
274 | ||
275 | return; | |
276 | } | |
277 | ||
278 | $pluginhost = new PluginHost($link); | |
ab457a9c AD |
279 | $user_plugins = get_pref($link, "_ENABLED_PLUGINS", $owner_uid); |
280 | ||
281 | $pluginhost->load(PLUGINS, $pluginhost::KIND_ALL); | |
282 | $pluginhost->load($user_plugins, $pluginhost::KIND_USER, $owner_uid); | |
283 | $pluginhost->load_data(); | |
017401dd AD |
284 | |
285 | foreach ($pluginhost->get_hooks($pluginhost::HOOK_FEED_FETCHED) as $plugin) { | |
286 | $feed_data = $plugin->hook_feed_fetched($feed_data); | |
287 | } | |
288 | ||
5de51df7 AD |
289 | if ($debug_enabled) { |
290 | _debug("update_rss_feed: fetch done, parsing..."); | |
291 | } | |
292 | ||
19b3992b | 293 | $rss = new SimplePie(); |
5d56d100 AD |
294 | $rss->set_sanitize_class("SanitizeDummy"); |
295 | // simplepie ignores the above and creates default sanitizer anyway, | |
296 | // so let's override it... | |
297 | $rss->sanitize = new SanitizeDummy(); | |
19b3992b | 298 | $rss->set_output_encoding('UTF-8'); |
017401dd | 299 | $rss->set_raw_data($feed_data); |
2c08214a | 300 | |
19b3992b AD |
301 | if ($debug_enabled) { |
302 | _debug("feed update interval (sec): " . | |
303 | get_feed_update_interval($link, $feed)*60); | |
304 | } | |
2c08214a | 305 | |
19b3992b | 306 | $rss->enable_cache(!$no_cache); |
2c08214a | 307 | |
19b3992b AD |
308 | if (!$no_cache) { |
309 | $rss->set_cache_location($simplepie_cache_dir); | |
310 | $rss->set_cache_duration($cache_age); | |
2c08214a AD |
311 | } |
312 | ||
ada103f2 | 313 | @$rss->init(); |
19b3992b | 314 | |
2c08214a AD |
315 | // print_r($rss); |
316 | ||
2c08214a AD |
317 | $feed = db_escape_string($feed); |
318 | ||
19b3992b | 319 | if (!$rss->error()) { |
2c08214a | 320 | |
d2a421e3 | 321 | // We use local pluginhost here because we need to load different per-user feed plugins |
4412b877 AD |
322 | $pluginhost->run_hooks($pluginhost::HOOK_FEED_PARSED, "hook_feed_parsed", $rss); |
323 | ||
2c08214a AD |
324 | if ($debug_enabled) { |
325 | _debug("update_rss_feed: processing feed data..."); | |
326 | } | |
327 | ||
328 | // db_query($link, "BEGIN"); | |
329 | ||
382268c6 AD |
330 | if (DB_TYPE == "pgsql") { |
331 | $favicon_interval_qpart = "favicon_last_checked < NOW() - INTERVAL '12 hour'"; | |
332 | } else { | |
333 | $favicon_interval_qpart = "favicon_last_checked < DATE_SUB(NOW(), INTERVAL 12 HOUR)"; | |
334 | } | |
335 | ||
750cfcd2 | 336 | $result = db_query($link, "SELECT title,site_url,owner_uid, |
382268c6 AD |
337 | (favicon_last_checked IS NULL OR $favicon_interval_qpart) AS |
338 | favicon_needs_check | |
2c08214a AD |
339 | FROM ttrss_feeds WHERE id = '$feed'"); |
340 | ||
341 | $registered_title = db_fetch_result($result, 0, "title"); | |
2c08214a | 342 | $orig_site_url = db_fetch_result($result, 0, "site_url"); |
382268c6 AD |
343 | $favicon_needs_check = sql_bool_to_bool(db_fetch_result($result, 0, |
344 | "favicon_needs_check")); | |
2c08214a AD |
345 | |
346 | $owner_uid = db_fetch_result($result, 0, "owner_uid"); | |
347 | ||
bb566dc2 | 348 | $site_url = db_escape_string(mb_substr(rewrite_relative_url($fetch_url, $rss->get_link()), 0, 245)); |
2c08214a AD |
349 | |
350 | if ($debug_enabled) { | |
351 | _debug("update_rss_feed: checking favicon..."); | |
352 | } | |
353 | ||
f2798eb6 AD |
354 | if ($favicon_needs_check) { |
355 | check_feed_favicon($site_url, $feed, $link); | |
356 | ||
357 | db_query($link, "UPDATE ttrss_feeds SET favicon_last_checked = NOW() | |
358 | WHERE id = '$feed'"); | |
359 | } | |
2c08214a AD |
360 | |
361 | if (!$registered_title || $registered_title == "[Unknown]") { | |
362 | ||
19b3992b | 363 | $feed_title = db_escape_string($rss->get_title()); |
2c08214a AD |
364 | |
365 | if ($debug_enabled) { | |
366 | _debug("update_rss_feed: registering title: $feed_title"); | |
367 | } | |
368 | ||
369 | db_query($link, "UPDATE ttrss_feeds SET | |
370 | title = '$feed_title' WHERE id = '$feed'"); | |
371 | } | |
372 | ||
0cf81637 | 373 | if ($site_url && $orig_site_url != $site_url) { |
2c08214a AD |
374 | db_query($link, "UPDATE ttrss_feeds SET |
375 | site_url = '$site_url' WHERE id = '$feed'"); | |
376 | } | |
377 | ||
2c08214a | 378 | if ($debug_enabled) { |
b24504b1 | 379 | _debug("update_rss_feed: loading filters & labels..."); |
2c08214a AD |
380 | } |
381 | ||
382 | $filters = load_filters($link, $feed, $owner_uid); | |
b24504b1 | 383 | $labels = get_all_labels($link, $owner_uid); |
2c08214a | 384 | |
67bd0b1f AD |
385 | if ($debug_enabled) { |
386 | //print_r($filters); | |
387 | _debug("update_rss_feed: " . count($filters) . " filters loaded."); | |
388 | } | |
2c08214a | 389 | |
19b3992b | 390 | $items = $rss->get_items(); |
2c08214a | 391 | |
19b3992b | 392 | if (!is_array($items)) { |
2c08214a | 393 | if ($debug_enabled) { |
19b3992b | 394 | _debug("update_rss_feed: no articles found."); |
2c08214a AD |
395 | } |
396 | ||
397 | db_query($link, "UPDATE ttrss_feeds | |
398 | SET last_updated = NOW(), last_error = '' WHERE id = '$feed'"); | |
399 | ||
400 | return; // no articles | |
401 | } | |
402 | ||
403 | if ($pubsub_state != 2 && PUBSUBHUBBUB_ENABLED) { | |
404 | ||
405 | if ($debug_enabled) _debug("update_rss_feed: checking for PUSH hub..."); | |
406 | ||
407 | $feed_hub_url = false; | |
2c08214a | 408 | |
19b3992b | 409 | $links = $rss->get_links('hub'); |
2c08214a | 410 | |
19b3992b AD |
411 | if ($links && is_array($links)) { |
412 | foreach ($links as $l) { | |
413 | $feed_hub_url = $l; | |
414 | break; | |
2c08214a AD |
415 | } |
416 | } | |
417 | ||
418 | if ($debug_enabled) _debug("update_rss_feed: feed hub url: $feed_hub_url"); | |
419 | ||
420 | if ($feed_hub_url && function_exists('curl_init') && | |
421 | !ini_get("open_basedir")) { | |
422 | ||
423 | require_once 'lib/pubsubhubbub/subscriber.php'; | |
424 | ||
425 | $callback_url = get_self_url_prefix() . | |
426 | "/public.php?op=pubsub&id=$feed"; | |
427 | ||
428 | $s = new Subscriber($feed_hub_url, $callback_url); | |
429 | ||
430 | $rc = $s->subscribe($fetch_url); | |
431 | ||
432 | if ($debug_enabled) | |
433 | _debug("update_rss_feed: feed hub url found, subscribe request sent."); | |
434 | ||
435 | db_query($link, "UPDATE ttrss_feeds SET pubsub_state = 1 | |
436 | WHERE id = '$feed'"); | |
437 | } | |
438 | } | |
439 | ||
440 | if ($debug_enabled) { | |
441 | _debug("update_rss_feed: processing articles..."); | |
442 | } | |
443 | ||
19b3992b | 444 | foreach ($items as $item) { |
5d56d100 | 445 | if ($_REQUEST['xdebug'] == 3) { |
2c08214a AD |
446 | print_r($item); |
447 | } | |
448 | ||
19b3992b AD |
449 | $entry_guid = $item->get_id(); |
450 | if (!$entry_guid) $entry_guid = $item->get_link(); | |
451 | if (!$entry_guid) $entry_guid = make_guid_from_title($item->get_title()); | |
2c08214a AD |
452 | |
453 | if ($debug_enabled) { | |
454 | _debug("update_rss_feed: guid $entry_guid"); | |
455 | } | |
456 | ||
457 | if (!$entry_guid) continue; | |
458 | ||
3a4c8973 AD |
459 | $entry_guid = "$owner_uid,$entry_guid"; |
460 | ||
2c08214a AD |
461 | $entry_timestamp = ""; |
462 | ||
19b3992b | 463 | $entry_timestamp = strtotime($item->get_date()); |
2c08214a | 464 | |
19b3992b | 465 | if ($entry_timestamp == -1 || !$entry_timestamp) { |
2c08214a AD |
466 | $entry_timestamp = time(); |
467 | $no_orig_date = 'true'; | |
468 | } else { | |
469 | $no_orig_date = 'false'; | |
470 | } | |
471 | ||
472 | $entry_timestamp_fmt = strftime("%Y/%m/%d %H:%M:%S", $entry_timestamp); | |
473 | ||
474 | if ($debug_enabled) { | |
475 | _debug("update_rss_feed: date $entry_timestamp [$entry_timestamp_fmt]"); | |
476 | } | |
477 | ||
19b3992b | 478 | $entry_title = $item->get_title(); |
1b35d30c | 479 | |
5d56d100 | 480 | $entry_link = rewrite_relative_url($site_url, $item->get_link()); |
2c08214a AD |
481 | |
482 | if ($debug_enabled) { | |
483 | _debug("update_rss_feed: title $entry_title"); | |
484 | _debug("update_rss_feed: link $entry_link"); | |
485 | } | |
486 | ||
487 | if (!$entry_title) $entry_title = date("Y-m-d H:i:s", $entry_timestamp);; | |
488 | ||
19b3992b AD |
489 | $entry_content = $item->get_content(); |
490 | if (!$entry_content) $entry_content = $item->get_description(); | |
2c08214a AD |
491 | |
492 | if ($_REQUEST["xdebug"] == 2) { | |
493 | print "update_rss_feed: content: "; | |
487f0750 | 494 | print $entry_content; |
3c696512 | 495 | print "\n"; |
2c08214a AD |
496 | } |
497 | ||
19b3992b | 498 | $entry_comments = $item->data["comments"]; |
2c08214a | 499 | |
19b3992b AD |
500 | if ($item->get_author()) { |
501 | $entry_author_item = $item->get_author(); | |
502 | $entry_author = $entry_author_item->get_name(); | |
503 | if (!$entry_author) $entry_author = $entry_author_item->get_email(); | |
2c08214a | 504 | |
19b3992b | 505 | $entry_author = db_escape_string($entry_author); |
2c08214a AD |
506 | } |
507 | ||
bb566dc2 | 508 | $entry_guid = db_escape_string(mb_substr($entry_guid, 0, 245)); |
2c08214a | 509 | |
bb566dc2 AD |
510 | $entry_comments = db_escape_string(mb_substr($entry_comments, 0, 245)); |
511 | $entry_author = db_escape_string(mb_substr($entry_author, 0, 245)); | |
2c08214a | 512 | |
19b3992b | 513 | $num_comments = $item->get_item_tags('http://purl.org/rss/1.0/modules/slash/', 'comments'); |
83e6e313 | 514 | |
19b3992b AD |
515 | if (is_array($num_comments) && is_array($num_comments[0])) { |
516 | $num_comments = (int) $num_comments[0]["data"]; | |
2c08214a | 517 | } else { |
19b3992b | 518 | $num_comments = 0; |
2c08214a AD |
519 | } |
520 | ||
2c08214a | 521 | if ($debug_enabled) { |
83e6e313 | 522 | _debug("update_rss_feed: num_comments: $num_comments"); |
2c08214a AD |
523 | _debug("update_rss_feed: looking for tags [1]..."); |
524 | } | |
525 | ||
526 | // parse <category> entries into tags | |
527 | ||
528 | $additional_tags = array(); | |
529 | ||
19b3992b | 530 | $additional_tags_src = $item->get_categories(); |
2c08214a | 531 | |
19b3992b AD |
532 | if (is_array($additional_tags_src)) { |
533 | foreach ($additional_tags_src as $tobj) { | |
534 | array_push($additional_tags, $tobj->get_term()); | |
2c08214a | 535 | } |
19b3992b | 536 | } |
2c08214a | 537 | |
19b3992b AD |
538 | if ($debug_enabled) { |
539 | _debug("update_rss_feed: category tags:"); | |
540 | print_r($additional_tags); | |
2c08214a AD |
541 | } |
542 | ||
543 | if ($debug_enabled) { | |
544 | _debug("update_rss_feed: looking for tags [2]..."); | |
545 | } | |
546 | ||
fa6fbd36 | 547 | $entry_tags = array_unique($additional_tags); |
2c08214a AD |
548 | |
549 | for ($i = 0; $i < count($entry_tags); $i++) | |
550 | $entry_tags[$i] = mb_strtolower($entry_tags[$i], 'utf-8'); | |
551 | ||
552 | if ($debug_enabled) { | |
6aff7845 AD |
553 | //_debug("update_rss_feed: unfiltered tags found:"); |
554 | //print_r($entry_tags); | |
2c08214a AD |
555 | } |
556 | ||
2c08214a AD |
557 | if ($debug_enabled) { |
558 | _debug("update_rss_feed: done collecting data [TITLE:$entry_title]"); | |
559 | } | |
560 | ||
cc85704f | 561 | // TODO: less memory-hungry implementation |
19c73507 | 562 | |
19b3992b AD |
563 | if ($debug_enabled) { |
564 | _debug("update_rss_feed: applying plugin filters.."); | |
565 | } | |
cc85704f | 566 | |
e02555c1 AD |
567 | // FIXME not sure if owner_uid is a good idea here, we may have a base entry without user entry (?) |
568 | $result = db_query($link, "SELECT plugin_data,title,content,link,tag_cache,author FROM ttrss_entries, ttrss_user_entries | |
569 | WHERE ref_id = id AND guid = '".db_escape_string($entry_guid)."' AND owner_uid = $owner_uid"); | |
b30abdad AD |
570 | |
571 | if (db_num_rows($result) != 0) { | |
572 | $entry_plugin_data = db_fetch_result($result, 0, "plugin_data"); | |
e02555c1 AD |
573 | $stored_article = array("title" => db_fetch_result($result, 0, "title"), |
574 | "content" => db_fetch_result($result, 0, "content"), | |
575 | "link" => db_fetch_result($result, 0, "link"), | |
576 | "tags" => explode(",", db_fetch_result($result, 0, "tag_cache")), | |
577 | "author" => db_fetch_result($result, 0, "author")); | |
b30abdad AD |
578 | } else { |
579 | $entry_plugin_data = ""; | |
e02555c1 | 580 | $stored_article = array(); |
b30abdad AD |
581 | } |
582 | ||
455b1401 | 583 | $article = array("owner_uid" => $owner_uid, // read only |
b30abdad | 584 | "guid" => $entry_guid, // read only |
19b3992b AD |
585 | "title" => $entry_title, |
586 | "content" => $entry_content, | |
587 | "link" => $entry_link, | |
588 | "tags" => $entry_tags, | |
b30abdad | 589 | "plugin_data" => $entry_plugin_data, |
e02555c1 AD |
590 | "author" => $entry_author, |
591 | "stored" => $stored_article); | |
cc85704f | 592 | |
19b3992b AD |
593 | foreach ($pluginhost->get_hooks($pluginhost::HOOK_ARTICLE_FILTER) as $plugin) { |
594 | $article = $plugin->hook_article_filter($article); | |
cc85704f AD |
595 | } |
596 | ||
19b3992b | 597 | $entry_tags = $article["tags"]; |
b30abdad | 598 | $entry_guid = db_escape_string($entry_guid); |
1b35d30c AD |
599 | $entry_content = db_escape_string($article["content"], false); |
600 | $entry_title = db_escape_string($article["title"]); | |
601 | $entry_author = db_escape_string($article["author"]); | |
602 | $entry_link = db_escape_string($article["link"]); | |
b30abdad AD |
603 | $entry_plugin_data = db_escape_string($article["plugin_data"]); |
604 | ||
605 | if ($debug_enabled) { | |
606 | _debug("update_rss_feed: plugin data: $entry_plugin_data"); | |
607 | } | |
2bbd6994 | 608 | |
0a3fd79b AD |
609 | if ($cache_images && is_writable(CACHE_DIR . '/images')) |
610 | $entry_content = cache_images($entry_content, $site_url, $debug_enabled); | |
611 | ||
9e222305 | 612 | $content_hash = "SHA1:" . sha1($entry_content); |
cc85704f | 613 | |
2c08214a AD |
614 | db_query($link, "BEGIN"); |
615 | ||
9e222305 AD |
616 | $result = db_query($link, "SELECT id FROM ttrss_entries |
617 | WHERE guid = '$entry_guid'"); | |
618 | ||
2c08214a AD |
619 | if (db_num_rows($result) == 0) { |
620 | ||
621 | if ($debug_enabled) { | |
9e222305 | 622 | _debug("update_rss_feed: base guid [$entry_guid] not found"); |
2c08214a AD |
623 | } |
624 | ||
625 | // base post entry does not exist, create it | |
626 | ||
627 | $result = db_query($link, | |
628 | "INSERT INTO ttrss_entries | |
629 | (title, | |
630 | guid, | |
631 | link, | |
632 | updated, | |
633 | content, | |
634 | content_hash, | |
87764a50 | 635 | cached_content, |
2c08214a AD |
636 | no_orig_date, |
637 | date_updated, | |
638 | date_entered, | |
639 | comments, | |
640 | num_comments, | |
b30abdad | 641 | plugin_data, |
2c08214a AD |
642 | author) |
643 | VALUES | |
644 | ('$entry_title', | |
645 | '$entry_guid', | |
646 | '$entry_link', | |
647 | '$entry_timestamp_fmt', | |
648 | '$entry_content', | |
649 | '$content_hash', | |
0a3fd79b | 650 | '', |
2c08214a AD |
651 | $no_orig_date, |
652 | NOW(), | |
653 | NOW(), | |
654 | '$entry_comments', | |
655 | '$num_comments', | |
b30abdad | 656 | '$entry_plugin_data', |
2c08214a | 657 | '$entry_author')"); |
e8291805 AD |
658 | |
659 | $article_labels = array(); | |
660 | ||
2c08214a AD |
661 | } else { |
662 | // we keep encountering the entry in feeds, so we need to | |
663 | // update date_updated column so that we don't get horrible | |
664 | // dupes when the entry gets purged and reinserted again e.g. | |
665 | // in the case of SLOW SLOW OMG SLOW updating feeds | |
666 | ||
667 | $base_entry_id = db_fetch_result($result, 0, "id"); | |
668 | ||
669 | db_query($link, "UPDATE ttrss_entries SET date_updated = NOW() | |
670 | WHERE id = '$base_entry_id'"); | |
e8291805 AD |
671 | |
672 | $article_labels = get_article_labels($link, $base_entry_id, $owner_uid); | |
2c08214a AD |
673 | } |
674 | ||
675 | // now it should exist, if not - bad luck then | |
676 | ||
677 | $result = db_query($link, "SELECT | |
b30abdad | 678 | id,content_hash,no_orig_date,title,plugin_data, |
2c08214a AD |
679 | ".SUBSTRING_FOR_DATE."(date_updated,1,19) as date_updated, |
680 | ".SUBSTRING_FOR_DATE."(updated,1,19) as updated, | |
0a3fd79b | 681 | num_comments |
2c08214a AD |
682 | FROM |
683 | ttrss_entries | |
684 | WHERE guid = '$entry_guid'"); | |
685 | ||
686 | $entry_ref_id = 0; | |
687 | $entry_int_id = 0; | |
688 | ||
689 | if (db_num_rows($result) == 1) { | |
690 | ||
691 | if ($debug_enabled) { | |
9e222305 | 692 | _debug("update_rss_feed: base guid [$entry_guid] found, checking for user record"); |
2c08214a AD |
693 | } |
694 | ||
695 | // this will be used below in update handler | |
696 | $orig_content_hash = db_fetch_result($result, 0, "content_hash"); | |
697 | $orig_title = db_fetch_result($result, 0, "title"); | |
698 | $orig_num_comments = db_fetch_result($result, 0, "num_comments"); | |
699 | $orig_date_updated = strtotime(db_fetch_result($result, | |
700 | 0, "date_updated")); | |
b30abdad | 701 | $orig_plugin_data = db_fetch_result($result, 0, "plugin_data"); |
2c08214a AD |
702 | |
703 | $ref_id = db_fetch_result($result, 0, "id"); | |
704 | $entry_ref_id = $ref_id; | |
705 | ||
706 | // check for user post link to main table | |
707 | ||
708 | // do we allow duplicate posts with same GUID in different feeds? | |
709 | if (get_pref($link, "ALLOW_DUPLICATE_POSTS", $owner_uid, false)) { | |
710 | $dupcheck_qpart = "AND (feed_id = '$feed' OR feed_id IS NULL)"; | |
711 | } else { | |
712 | $dupcheck_qpart = ""; | |
713 | } | |
714 | ||
715 | /* Collect article tags here so we could filter by them: */ | |
716 | ||
717 | $article_filters = get_article_filters($filters, $entry_title, | |
4021d61a | 718 | $entry_content, $entry_link, $entry_timestamp, $entry_author, |
2c08214a AD |
719 | $entry_tags); |
720 | ||
721 | if ($debug_enabled) { | |
722 | _debug("update_rss_feed: article filters: "); | |
723 | if (count($article_filters) != 0) { | |
724 | print_r($article_filters); | |
725 | } | |
726 | } | |
727 | ||
728 | if (find_article_filter($article_filters, "filter")) { | |
729 | db_query($link, "COMMIT"); // close transaction in progress | |
730 | continue; | |
731 | } | |
732 | ||
733 | $score = calculate_article_score($article_filters); | |
734 | ||
735 | if ($debug_enabled) { | |
736 | _debug("update_rss_feed: initial score: $score"); | |
737 | } | |
738 | ||
739 | $query = "SELECT ref_id, int_id FROM ttrss_user_entries WHERE | |
740 | ref_id = '$ref_id' AND owner_uid = '$owner_uid' | |
741 | $dupcheck_qpart"; | |
742 | ||
743 | // if ($_REQUEST["xdebug"]) print "$query\n"; | |
744 | ||
745 | $result = db_query($link, $query); | |
746 | ||
747 | // okay it doesn't exist - create user entry | |
748 | if (db_num_rows($result) == 0) { | |
749 | ||
750 | if ($debug_enabled) { | |
751 | _debug("update_rss_feed: user record not found, creating..."); | |
752 | } | |
753 | ||
754 | if ($score >= -500 && !find_article_filter($article_filters, 'catchup')) { | |
755 | $unread = 'true'; | |
756 | $last_read_qpart = 'NULL'; | |
757 | } else { | |
758 | $unread = 'false'; | |
759 | $last_read_qpart = 'NOW()'; | |
760 | } | |
761 | ||
762 | if (find_article_filter($article_filters, 'mark') || $score > 1000) { | |
763 | $marked = 'true'; | |
764 | } else { | |
765 | $marked = 'false'; | |
766 | } | |
767 | ||
768 | if (find_article_filter($article_filters, 'publish')) { | |
769 | $published = 'true'; | |
770 | } else { | |
771 | $published = 'false'; | |
772 | } | |
773 | ||
2ea9bbfd AD |
774 | // N-grams |
775 | ||
776 | if (DB_TYPE == "pgsql" and defined('_NGRAM_TITLE_DUPLICATE_THRESHOLD')) { | |
777 | ||
778 | $result = db_query($link, "SELECT COUNT(*) AS similar FROM | |
779 | ttrss_entries,ttrss_user_entries | |
780 | WHERE ref_id = id AND updated >= NOW() - INTERVAL '7 day' | |
781 | AND similarity(title, '$entry_title') >= "._NGRAM_TITLE_DUPLICATE_THRESHOLD." | |
782 | AND owner_uid = $owner_uid"); | |
783 | ||
784 | $ngram_similar = db_fetch_result($result, 0, "similar"); | |
785 | ||
786 | if ($debug_enabled) { | |
787 | _debug("update_rss_feed: N-gram similar results: $ngram_similar"); | |
788 | } | |
789 | ||
790 | if ($ngram_similar > 0) { | |
791 | $unread = 'false'; | |
792 | } | |
793 | } | |
794 | ||
2c08214a AD |
795 | $result = db_query($link, |
796 | "INSERT INTO ttrss_user_entries | |
797 | (ref_id, owner_uid, feed_id, unread, last_read, marked, | |
798 | published, score, tag_cache, label_cache, uuid) | |
799 | VALUES ('$ref_id', '$owner_uid', '$feed', $unread, | |
800 | $last_read_qpart, $marked, $published, '$score', '', '', '')"); | |
801 | ||
802 | if (PUBSUBHUBBUB_HUB && $published == 'true') { | |
803 | $rss_link = get_self_url_prefix() . | |
804 | "/public.php?op=rss&id=-2&key=" . | |
805 | get_feed_access_key($link, -2, false, $owner_uid); | |
806 | ||
807 | $p = new Publisher(PUBSUBHUBBUB_HUB); | |
808 | ||
809 | $pubsub_result = $p->publish_update($rss_link); | |
810 | } | |
811 | ||
812 | $result = db_query($link, | |
813 | "SELECT int_id FROM ttrss_user_entries WHERE | |
814 | ref_id = '$ref_id' AND owner_uid = '$owner_uid' AND | |
815 | feed_id = '$feed' LIMIT 1"); | |
816 | ||
817 | if (db_num_rows($result) == 1) { | |
818 | $entry_int_id = db_fetch_result($result, 0, "int_id"); | |
819 | } | |
820 | } else { | |
821 | if ($debug_enabled) { | |
822 | _debug("update_rss_feed: user record FOUND"); | |
823 | } | |
824 | ||
825 | $entry_ref_id = db_fetch_result($result, 0, "ref_id"); | |
826 | $entry_int_id = db_fetch_result($result, 0, "int_id"); | |
827 | } | |
828 | ||
829 | if ($debug_enabled) { | |
830 | _debug("update_rss_feed: RID: $entry_ref_id, IID: $entry_int_id"); | |
831 | } | |
832 | ||
833 | $post_needs_update = false; | |
834 | $update_insignificant = false; | |
835 | ||
836 | if ($orig_num_comments != $num_comments) { | |
837 | $post_needs_update = true; | |
838 | $update_insignificant = true; | |
839 | } | |
840 | ||
b30abdad AD |
841 | if ($entry_plugin_data != $orig_plugin_data) { |
842 | $post_needs_update = true; | |
843 | $update_insignificant = true; | |
844 | } | |
845 | ||
2c08214a AD |
846 | if ($content_hash != $orig_content_hash) { |
847 | $post_needs_update = true; | |
848 | $update_insignificant = false; | |
849 | } | |
850 | ||
851 | if (db_escape_string($orig_title) != $entry_title) { | |
852 | $post_needs_update = true; | |
853 | $update_insignificant = false; | |
854 | } | |
855 | ||
856 | // if post needs update, update it and mark all user entries | |
857 | // linking to this post as updated | |
858 | if ($post_needs_update) { | |
859 | ||
860 | if (defined('DAEMON_EXTENDED_DEBUG')) { | |
861 | _debug("update_rss_feed: post $entry_guid needs update..."); | |
862 | } | |
863 | ||
864 | // print "<!-- post $orig_title needs update : $post_needs_update -->"; | |
865 | ||
866 | db_query($link, "UPDATE ttrss_entries | |
867 | SET title = '$entry_title', content = '$entry_content', | |
868 | content_hash = '$content_hash', | |
869 | updated = '$entry_timestamp_fmt', | |
b30abdad AD |
870 | num_comments = '$num_comments', |
871 | plugin_data = '$entry_plugin_data' | |
2c08214a AD |
872 | WHERE id = '$ref_id'"); |
873 | ||
874 | if (!$update_insignificant) { | |
875 | if ($mark_unread_on_update) { | |
876 | db_query($link, "UPDATE ttrss_user_entries | |
877 | SET last_read = null, unread = true WHERE ref_id = '$ref_id'"); | |
878 | } else if ($update_on_checksum_change) { | |
879 | db_query($link, "UPDATE ttrss_user_entries | |
880 | SET last_read = null WHERE ref_id = '$ref_id' | |
881 | AND unread = false"); | |
882 | } | |
883 | } | |
884 | } | |
885 | } | |
886 | ||
887 | db_query($link, "COMMIT"); | |
888 | ||
889 | if ($debug_enabled) { | |
890 | _debug("update_rss_feed: assigning labels..."); | |
891 | } | |
892 | ||
92c14e9d | 893 | assign_article_to_label_filters($link, $entry_ref_id, $article_filters, |
b24504b1 | 894 | $owner_uid, $article_labels); |
2c08214a AD |
895 | |
896 | if ($debug_enabled) { | |
897 | _debug("update_rss_feed: looking for enclosures..."); | |
898 | } | |
899 | ||
900 | // enclosures | |
901 | ||
902 | $enclosures = array(); | |
903 | ||
19b3992b | 904 | $encs = $item->get_enclosures(); |
2c08214a | 905 | |
19b3992b AD |
906 | if (is_array($encs)) { |
907 | foreach ($encs as $e) { | |
908 | $e_item = array( | |
909 | $e->link, $e->type, $e->length); | |
2c08214a | 910 | array_push($enclosures, $e_item); |
2c08214a AD |
911 | } |
912 | } | |
913 | ||
2c08214a AD |
914 | if ($debug_enabled) { |
915 | _debug("update_rss_feed: article enclosures:"); | |
916 | print_r($enclosures); | |
917 | } | |
918 | ||
919 | db_query($link, "BEGIN"); | |
920 | ||
921 | foreach ($enclosures as $enc) { | |
922 | $enc_url = db_escape_string($enc[0]); | |
923 | $enc_type = db_escape_string($enc[1]); | |
924 | $enc_dur = db_escape_string($enc[2]); | |
925 | ||
926 | $result = db_query($link, "SELECT id FROM ttrss_enclosures | |
927 | WHERE content_url = '$enc_url' AND post_id = '$entry_ref_id'"); | |
928 | ||
929 | if (db_num_rows($result) == 0) { | |
930 | db_query($link, "INSERT INTO ttrss_enclosures | |
931 | (content_url, content_type, title, duration, post_id) VALUES | |
932 | ('$enc_url', '$enc_type', '', '$enc_dur', '$entry_ref_id')"); | |
933 | } | |
934 | } | |
935 | ||
936 | db_query($link, "COMMIT"); | |
937 | ||
938 | // check for manual tags (we have to do it here since they're loaded from filters) | |
939 | ||
940 | foreach ($article_filters as $f) { | |
6aff7845 | 941 | if ($f["type"] == "tag") { |
2c08214a | 942 | |
6aff7845 | 943 | $manual_tags = trim_array(explode(",", $f["param"])); |
2c08214a AD |
944 | |
945 | foreach ($manual_tags as $tag) { | |
946 | if (tag_is_valid($tag)) { | |
947 | array_push($entry_tags, $tag); | |
948 | } | |
949 | } | |
950 | } | |
951 | } | |
952 | ||
953 | // Skip boring tags | |
954 | ||
955 | $boring_tags = trim_array(explode(",", mb_strtolower(get_pref($link, | |
956 | 'BLACKLISTED_TAGS', $owner_uid, ''), 'utf-8'))); | |
957 | ||
958 | $filtered_tags = array(); | |
959 | $tags_to_cache = array(); | |
960 | ||
961 | if ($entry_tags && is_array($entry_tags)) { | |
962 | foreach ($entry_tags as $tag) { | |
963 | if (array_search($tag, $boring_tags) === false) { | |
964 | array_push($filtered_tags, $tag); | |
965 | } | |
966 | } | |
967 | } | |
968 | ||
969 | $filtered_tags = array_unique($filtered_tags); | |
970 | ||
971 | if ($debug_enabled) { | |
972 | _debug("update_rss_feed: filtered article tags:"); | |
973 | print_r($filtered_tags); | |
974 | } | |
975 | ||
976 | // Save article tags in the database | |
977 | ||
978 | if (count($filtered_tags) > 0) { | |
979 | ||
980 | db_query($link, "BEGIN"); | |
981 | ||
982 | foreach ($filtered_tags as $tag) { | |
983 | ||
984 | $tag = sanitize_tag($tag); | |
985 | $tag = db_escape_string($tag); | |
986 | ||
987 | if (!tag_is_valid($tag)) continue; | |
988 | ||
989 | $result = db_query($link, "SELECT id FROM ttrss_tags | |
990 | WHERE tag_name = '$tag' AND post_int_id = '$entry_int_id' AND | |
991 | owner_uid = '$owner_uid' LIMIT 1"); | |
992 | ||
993 | if ($result && db_num_rows($result) == 0) { | |
994 | ||
995 | db_query($link, "INSERT INTO ttrss_tags | |
996 | (owner_uid,tag_name,post_int_id) | |
997 | VALUES ('$owner_uid','$tag', '$entry_int_id')"); | |
998 | } | |
999 | ||
1000 | array_push($tags_to_cache, $tag); | |
1001 | } | |
1002 | ||
1003 | /* update the cache */ | |
1004 | ||
1005 | $tags_to_cache = array_unique($tags_to_cache); | |
1006 | ||
1007 | $tags_str = db_escape_string(join(",", $tags_to_cache)); | |
1008 | ||
1009 | db_query($link, "UPDATE ttrss_user_entries | |
1010 | SET tag_cache = '$tags_str' WHERE ref_id = '$entry_ref_id' | |
1011 | AND owner_uid = $owner_uid"); | |
1012 | ||
1013 | db_query($link, "COMMIT"); | |
1014 | } | |
1015 | ||
b24504b1 AD |
1016 | if (get_pref($link, "AUTO_ASSIGN_LABELS", $owner_uid, false)) { |
1017 | if ($debug_enabled) { | |
1018 | _debug("update_rss_feed: auto-assigning labels..."); | |
1019 | } | |
1020 | ||
1021 | foreach ($labels as $label) { | |
1022 | $caption = $label["caption"]; | |
1023 | ||
3fc6e71a | 1024 | if (preg_match("/\b$caption\b/i", "$tags_str " . strip_tags($entry_content) . " $entry_title")) { |
b24504b1 AD |
1025 | if (!labels_contains_caption($article_labels, $caption)) { |
1026 | label_add_article($link, $entry_ref_id, $caption, $owner_uid); | |
1027 | } | |
1028 | } | |
1029 | } | |
1030 | } | |
1031 | ||
2c08214a AD |
1032 | if ($debug_enabled) { |
1033 | _debug("update_rss_feed: article processed"); | |
1034 | } | |
1035 | } | |
1036 | ||
1037 | if (!$last_updated) { | |
1038 | if ($debug_enabled) { | |
1039 | _debug("update_rss_feed: new feed, catching it up..."); | |
1040 | } | |
1041 | catchup_feed($link, $feed, false, $owner_uid); | |
1042 | } | |
1043 | ||
1044 | if ($debug_enabled) { | |
1045 | _debug("purging feed..."); | |
1046 | } | |
1047 | ||
1048 | purge_feed($link, $feed, 0, $debug_enabled); | |
1049 | ||
1050 | db_query($link, "UPDATE ttrss_feeds | |
f2798eb6 | 1051 | SET last_updated = NOW(), last_error = '' WHERE id = '$feed'"); |
2c08214a AD |
1052 | |
1053 | // db_query($link, "COMMIT"); | |
1054 | ||
1055 | } else { | |
1056 | ||
bb566dc2 | 1057 | $error_msg = db_escape_string(mb_substr($rss->error(), 0, 245)); |
2c08214a AD |
1058 | |
1059 | if ($debug_enabled) { | |
1060 | _debug("update_rss_feed: error fetching feed: $error_msg"); | |
1061 | } | |
1062 | ||
2c08214a AD |
1063 | db_query($link, |
1064 | "UPDATE ttrss_feeds SET last_error = '$error_msg', | |
1065 | last_updated = NOW() WHERE id = '$feed'"); | |
1066 | } | |
1067 | ||
19b3992b | 1068 | unset($rss); |
2c08214a AD |
1069 | |
1070 | if ($debug_enabled) { | |
1071 | _debug("update_rss_feed: done"); | |
1072 | } | |
1073 | ||
1074 | } | |
1075 | ||
3c696512 AD |
1076 | function cache_images($html, $site_url, $debug) { |
1077 | $cache_dir = CACHE_DIR . "/images"; | |
1078 | ||
1079 | libxml_use_internal_errors(true); | |
1080 | ||
1081 | $charset_hack = '<head> | |
1082 | <meta http-equiv="Content-Type" content="text/html; charset=utf-8"/> | |
1083 | </head>'; | |
1084 | ||
1085 | $doc = new DOMDocument(); | |
1086 | $doc->loadHTML($charset_hack . $html); | |
1087 | $xpath = new DOMXPath($doc); | |
1088 | ||
1089 | $entries = $xpath->query('(//img[@src])'); | |
1090 | ||
1091 | foreach ($entries as $entry) { | |
1092 | if ($entry->hasAttribute('src')) { | |
1093 | $src = rewrite_relative_url($site_url, $entry->getAttribute('src')); | |
1094 | ||
1095 | $local_filename = CACHE_DIR . "/images/" . sha1($src) . ".png"; | |
1096 | ||
1097 | if ($debug) _debug("cache_images: downloading: $src to $local_filename"); | |
1098 | ||
1099 | if (!file_exists($local_filename)) { | |
1100 | $file_content = fetch_file_contents($src); | |
1101 | ||
b8379e69 | 1102 | if ($file_content && strlen($file_content) > 1024) { |
3c696512 AD |
1103 | file_put_contents($local_filename, $file_content); |
1104 | } | |
1105 | } | |
1106 | ||
1107 | if (file_exists($local_filename)) { | |
1108 | $entry->setAttribute('src', SELF_URL_PATH . '/image.php?url=' . | |
487f0750 | 1109 | base64_encode($src)); |
3c696512 AD |
1110 | } |
1111 | } | |
1112 | } | |
1113 | ||
1114 | $node = $doc->getElementsByTagName('body')->item(0); | |
1115 | ||
7b8ff151 | 1116 | return $doc->saveXML($node, LIBXML_NOEMPTYTAG); |
3c696512 AD |
1117 | } |
1118 | ||
2a91b6ff AD |
1119 | function expire_lock_files($debug) { |
1120 | if ($debug) _debug("Removing old lock files..."); | |
1121 | ||
1122 | $num_deleted = 0; | |
1123 | ||
1124 | if (is_writable(LOCK_DIRECTORY)) { | |
1125 | $files = glob(LOCK_DIRECTORY . "/*.lock"); | |
1126 | ||
1127 | if ($files) { | |
1128 | foreach ($files as $file) { | |
1129 | if (!file_is_locked($file) && time() - filemtime($file) > 86400*2) { | |
1130 | unlink($file); | |
1131 | ++$num_deleted; | |
1132 | } | |
1133 | } | |
1134 | } | |
1135 | } | |
1136 | ||
1137 | if ($debug) _debug("Removed $num_deleted files."); | |
1138 | } | |
1139 | ||
3c696512 | 1140 | function expire_cached_files($debug) { |
19b3992b | 1141 | foreach (array("simplepie", "images", "export") as $dir) { |
3c696512 | 1142 | $cache_dir = CACHE_DIR . "/$dir"; |
2c08214a | 1143 | |
3c696512 | 1144 | if ($debug) _debug("Expiring $cache_dir"); |
2c08214a | 1145 | |
3c696512 AD |
1146 | $num_deleted = 0; |
1147 | ||
1148 | if (is_writable($cache_dir)) { | |
1149 | $files = glob("$cache_dir/*"); | |
1150 | ||
2a91b6ff | 1151 | if ($files) { |
2ab20c31 AD |
1152 | foreach ($files as $file) { |
1153 | if (time() - filemtime($file) > 86400*7) { | |
1154 | unlink($file); | |
3c696512 | 1155 | |
2ab20c31 AD |
1156 | ++$num_deleted; |
1157 | } | |
3c696512 AD |
1158 | } |
1159 | } | |
2a91b6ff | 1160 | } |
3c696512 AD |
1161 | |
1162 | if ($debug) _debug("Removed $num_deleted files."); | |
1163 | } | |
1164 | } | |
2c08214a | 1165 | |
a3e0bdcf AD |
1166 | /** |
1167 | * Source: http://www.php.net/manual/en/function.parse-url.php#104527 | |
1168 | * Returns the url query as associative array | |
1169 | * | |
1170 | * @param string query | |
1171 | * @return array params | |
1172 | */ | |
1173 | function convertUrlQuery($query) { | |
1174 | $queryParts = explode('&', $query); | |
1175 | ||
1176 | $params = array(); | |
1177 | ||
1178 | foreach ($queryParts as $param) { | |
1179 | $item = explode('=', $param); | |
1180 | $params[$item[0]] = $item[1]; | |
1181 | } | |
1182 | ||
1183 | return $params; | |
1184 | } | |
92c14e9d AD |
1185 | |
1186 | function get_article_filters($filters, $title, $content, $link, $timestamp, $author, $tags) { | |
1187 | $matches = array(); | |
1188 | ||
1189 | foreach ($filters as $filter) { | |
1190 | $match_any_rule = $filter["match_any_rule"]; | |
1191 | $filter_match = false; | |
1192 | ||
1193 | foreach ($filter["rules"] as $rule) { | |
1194 | $match = false; | |
1195 | $reg_exp = $rule["reg_exp"]; | |
1196 | ||
1197 | if (!$reg_exp) | |
1198 | continue; | |
1199 | ||
1200 | switch ($rule["type"]) { | |
1201 | case "title": | |
1202 | $match = @preg_match("/$reg_exp/i", $title); | |
1203 | break; | |
1204 | case "content": | |
d03ae73e AD |
1205 | // we don't need to deal with multiline regexps |
1206 | $content = preg_replace("/[\r\n\t]/", "", $content); | |
1207 | ||
92c14e9d AD |
1208 | $match = @preg_match("/$reg_exp/i", $content); |
1209 | break; | |
1210 | case "both": | |
d03ae73e AD |
1211 | // we don't need to deal with multiline regexps |
1212 | $content = preg_replace("/[\r\n\t]/", "", $content); | |
1213 | ||
72d1d067 | 1214 | $match = (@preg_match("/$reg_exp/i", $title) || @preg_match("/$reg_exp/i", $content)); |
92c14e9d AD |
1215 | break; |
1216 | case "link": | |
1217 | $match = @preg_match("/$reg_exp/i", $link); | |
1218 | break; | |
1219 | case "author": | |
1220 | $match = @preg_match("/$reg_exp/i", $author); | |
1221 | break; | |
1222 | case "tag": | |
1223 | $tag_string = join(",", $tags); | |
1224 | $match = @preg_match("/$reg_exp/i", $tag_string); | |
1225 | break; | |
1226 | } | |
1227 | ||
1228 | if ($match_any_rule) { | |
1229 | if ($match) { | |
1230 | $filter_match = true; | |
1231 | break; | |
1232 | } | |
1233 | } else { | |
1234 | $filter_match = $match; | |
1235 | if (!$match) { | |
1236 | break; | |
1237 | } | |
1238 | } | |
1239 | } | |
1240 | ||
1241 | if ($filter_match) { | |
1242 | foreach ($filter["actions"] AS $action) { | |
1243 | array_push($matches, $action); | |
1244 | } | |
1245 | } | |
1246 | } | |
1247 | ||
1248 | return $matches; | |
1249 | } | |
1250 | ||
1251 | function find_article_filter($filters, $filter_name) { | |
1252 | foreach ($filters as $f) { | |
1253 | if ($f["type"] == $filter_name) { | |
1254 | return $f; | |
1255 | }; | |
1256 | } | |
1257 | return false; | |
1258 | } | |
1259 | ||
1260 | function find_article_filters($filters, $filter_name) { | |
1261 | $results = array(); | |
1262 | ||
1263 | foreach ($filters as $f) { | |
1264 | if ($f["type"] == $filter_name) { | |
1265 | array_push($results, $f); | |
1266 | }; | |
1267 | } | |
1268 | return $results; | |
1269 | } | |
1270 | ||
1271 | function calculate_article_score($filters) { | |
1272 | $score = 0; | |
1273 | ||
1274 | foreach ($filters as $f) { | |
1275 | if ($f["type"] == "score") { | |
1276 | $score += $f["param"]; | |
1277 | }; | |
1278 | } | |
1279 | return $score; | |
1280 | } | |
1281 | ||
b24504b1 AD |
1282 | function labels_contains_caption($labels, $caption) { |
1283 | foreach ($labels as $label) { | |
1284 | if ($label[1] == $caption) { | |
1285 | return true; | |
1286 | } | |
1287 | } | |
1288 | ||
1289 | return false; | |
1290 | } | |
1291 | ||
1292 | function assign_article_to_label_filters($link, $id, $filters, $owner_uid, $article_labels) { | |
92c14e9d AD |
1293 | foreach ($filters as $f) { |
1294 | if ($f["type"] == "label") { | |
b24504b1 AD |
1295 | if (!labels_contains_caption($article_labels, $f["param"])) { |
1296 | label_add_article($link, $id, $f["param"], $owner_uid); | |
1297 | } | |
1298 | } | |
92c14e9d AD |
1299 | } |
1300 | } | |
87764a50 AD |
1301 | |
1302 | function cache_content($link, $url, $login, $pass) { | |
1303 | ||
1304 | $content = fetch_file_contents($url, $login, $pass); | |
1305 | ||
1306 | if ($content) { | |
1307 | $doc = new DOMDocument(); | |
1308 | @$doc->loadHTML($content); | |
1309 | $xpath = new DOMXPath($doc); | |
1310 | ||
1311 | $node = $doc->getElementsByTagName('body')->item(0); | |
1312 | ||
1313 | if ($node) { | |
1314 | $content = $doc->saveXML($node, LIBXML_NOEMPTYTAG); | |
1315 | ||
1316 | return $content; | |
1317 | } | |
1318 | } | |
1319 | ||
1320 | return ""; | |
1321 | } | |
87d7e850 AD |
1322 | |
1323 | function make_guid_from_title($title) { | |
1324 | return preg_replace("/[ \"\',.:;]/", "-", | |
1325 | mb_strtolower(strip_tags($title), 'utf-8')); | |
1326 | } | |
1327 | ||
1328 | ||
2c08214a | 1329 | ?> |