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