]>
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)) { | |
3972bf59 AD |
21 | $subscribers = db_escape_string($link, $line["subscribers"]); |
22 | $feed_url = db_escape_string($link, $line["feed_url"]); | |
23 | $title = db_escape_string($link, $line["title"]); | |
24 | $site_url = db_escape_string($link, $line["site_url"]); | |
79178062 AD |
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 | |
ee0542ce | 80 | AND ttrss_user_prefs.value != '-1' |
2c08214a AD |
81 | AND ttrss_feeds.last_updated < NOW() - CAST((ttrss_user_prefs.value || ' minutes') AS INTERVAL) |
82 | ) OR ( | |
83 | ttrss_feeds.update_interval > 0 | |
84 | AND ttrss_feeds.last_updated < NOW() - CAST((ttrss_feeds.update_interval || ' minutes') AS INTERVAL) | |
941e48a4 AD |
85 | ) OR ttrss_feeds.last_updated IS NULL |
86 | OR last_updated = '1970-01-01 00:00:00')"; | |
2c08214a AD |
87 | } else { |
88 | $update_limit_qpart = "AND (( | |
89 | ttrss_feeds.update_interval = 0 | |
ee0542ce | 90 | AND ttrss_user_prefs.value != '-1' |
2c08214a AD |
91 | AND ttrss_feeds.last_updated < DATE_SUB(NOW(), INTERVAL CONVERT(ttrss_user_prefs.value, SIGNED INTEGER) MINUTE) |
92 | ) OR ( | |
93 | ttrss_feeds.update_interval > 0 | |
94 | AND ttrss_feeds.last_updated < DATE_SUB(NOW(), INTERVAL ttrss_feeds.update_interval MINUTE) | |
941e48a4 AD |
95 | ) OR ttrss_feeds.last_updated IS NULL |
96 | OR last_updated = '1970-01-01 00:00:00')"; | |
2c08214a AD |
97 | } |
98 | ||
99 | // Test if feed is currently being updated by another process. | |
100 | if (DB_TYPE == "pgsql") { | |
101 | $updstart_thresh_qpart = "AND (ttrss_feeds.last_update_started IS NULL OR ttrss_feeds.last_update_started < NOW() - INTERVAL '5 minutes')"; | |
102 | } else { | |
103 | $updstart_thresh_qpart = "AND (ttrss_feeds.last_update_started IS NULL OR ttrss_feeds.last_update_started < DATE_SUB(NOW(), INTERVAL 5 MINUTE))"; | |
104 | } | |
105 | ||
106 | // Test if there is a limit to number of updated feeds | |
107 | $query_limit = ""; | |
108 | if($limit) $query_limit = sprintf("LIMIT %d", $limit); | |
109 | ||
110 | $random_qpart = sql_random_function(); | |
111 | ||
112 | // We search for feed needing update. | |
e81610d9 | 113 | $result = db_query($link, "SELECT DISTINCT ttrss_feeds.feed_url,$random_qpart |
2c08214a AD |
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 | |
1c4421fc | 121 | $updstart_thresh_qpart |
e81610d9 | 122 | ORDER BY $random_qpart $query_limit"); |
2c08214a AD |
123 | |
124 | $user_prefs_cache = array(); | |
125 | ||
ca5ff2d9 | 126 | if($debug) _debug(sprintf("Scheduled %d feeds to update...", db_num_rows($result))); |
2c08214a AD |
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($result)) { | |
1c4421fc | 131 | array_push($feeds_to_update, db_escape_string($link, $line['feed_url'])); |
2c08214a AD |
132 | } |
133 | ||
134 | // We update the feed last update started date before anything else. | |
135 | // There is no lag due to feed contents downloads | |
136 | // It prevent an other process to update the same feed. | |
1c4421fc AD |
137 | |
138 | if(count($feeds_to_update) > 0) { | |
139 | $feeds_quoted = array(); | |
140 | ||
141 | foreach ($feeds_to_update as $feed) { | |
142 | array_push($feeds_quoted, "'" . db_escape_string($link, $feed) . "'"); | |
143 | } | |
144 | ||
2c08214a | 145 | db_query($link, sprintf("UPDATE ttrss_feeds SET last_update_started = NOW() |
1c4421fc | 146 | WHERE feed_url IN (%s)", implode(',', $feeds_quoted))); |
2c08214a AD |
147 | } |
148 | ||
3c696512 | 149 | expire_cached_files($debug); |
2a91b6ff | 150 | expire_lock_files($debug); |
3c696512 | 151 | |
8292d05b AD |
152 | $nf = 0; |
153 | ||
2c08214a | 154 | // For each feed, we call the feed update function. |
1c4421fc AD |
155 | foreach ($feeds_to_update as $feed) { |
156 | if($debug) _debug("Base feed: $feed"); | |
2c08214a | 157 | |
1c4421fc | 158 | //update_rss_feed($link, $line["id"], true); |
2c08214a | 159 | |
1c4421fc AD |
160 | // since we have the data cached, we can deal with other feeds with the same url |
161 | ||
162 | $tmp_result = db_query($link, "SELECT ttrss_feeds.feed_url,ttrss_feeds.id,last_updated | |
ee0542ce AD |
163 | FROM ttrss_feeds, ttrss_users, ttrss_user_prefs WHERE |
164 | ttrss_user_prefs.owner_uid = ttrss_feeds.owner_uid AND | |
165 | ttrss_users.id = ttrss_user_prefs.owner_uid AND | |
166 | ttrss_user_prefs.pref_name = 'DEFAULT_UPDATE_INTERVAL' AND | |
30ac8d1f | 167 | feed_url = '".db_escape_string($link, $feed)."' AND |
ee0542ce AD |
168 | (ttrss_feeds.update_interval > 0 OR |
169 | ttrss_user_prefs.value != '-1') | |
1c4421fc AD |
170 | $login_thresh_qpart |
171 | ORDER BY feed_url $query_limit"); | |
172 | ||
173 | if (db_num_rows($tmp_result) > 0) { | |
174 | while ($tline = db_fetch_assoc($tmp_result)) { | |
32b05702 | 175 | if($debug) _debug(" => " . $tline["last_updated"] . ", " . $tline["id"]); |
1c4421fc | 176 | update_rss_feed($link, $tline["id"], true); |
8292d05b | 177 | ++$nf; |
1c4421fc AD |
178 | } |
179 | } | |
2c08214a AD |
180 | } |
181 | ||
f1611683 AD |
182 | require_once "digest.php"; |
183 | ||
2c08214a | 184 | // Send feed digests by email if needed. |
8437c066 | 185 | send_headlines_digests($link, $debug); |
2c08214a | 186 | |
8292d05b AD |
187 | return $nf; |
188 | ||
2c08214a AD |
189 | } // function update_daemon_common |
190 | ||
87764a50 | 191 | // ignore_daemon is not used |
0e4a7d7a | 192 | function update_rss_feed($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, |
5321e775 | 205 | mark_unread_on_update, owner_uid, |
87764a50 AD |
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")); | |
2c08214a AD |
220 | $pubsub_state = db_fetch_result($result, 0, "pubsub_state"); |
221 | ||
222 | db_query($link, "UPDATE ttrss_feeds SET last_update_started = NOW() | |
223 | WHERE id = '$feed'"); | |
224 | ||
225 | $auth_login = db_fetch_result($result, 0, "auth_login"); | |
226 | $auth_pass = db_fetch_result($result, 0, "auth_pass"); | |
227 | ||
2c08214a AD |
228 | $cache_images = sql_bool_to_bool(db_fetch_result($result, 0, "cache_images")); |
229 | $fetch_url = db_fetch_result($result, 0, "feed_url"); | |
230 | ||
3972bf59 | 231 | $feed = db_escape_string($link, $feed); |
2c08214a | 232 | |
f074ffe9 | 233 | if ($override_url) $fetch_url = $override_url; |
2c08214a | 234 | |
f074ffe9 | 235 | $date_feed_processed = date('Y-m-d H:i'); |
2c08214a | 236 | |
4f9cbdff | 237 | $cache_filename = CACHE_DIR . "/simplepie/" . sha1($fetch_url) . ".feed"; |
f074ffe9 AD |
238 | |
239 | // Ignore cache if new feed or manual update. | |
240 | $cache_age = ($no_cache || is_null($last_updated) || $last_updated == '1970-01-01 00:00:00') ? | |
241 | 30 : get_feed_update_interval($link, $feed) * 60; | |
2c08214a AD |
242 | |
243 | if ($debug_enabled) { | |
f074ffe9 AD |
244 | _debug("update_rss_feed: cache filename: $cache_filename exists: " . file_exists($cache_filename)); |
245 | _debug("update_rss_feed: cache age: $cache_age; no cache: $no_cache"); | |
2c08214a AD |
246 | } |
247 | ||
f074ffe9 | 248 | $cached_feed_data_hash = false; |
2c08214a | 249 | |
4f9cbdff AD |
250 | $rss = false; |
251 | $rss_hash = false; | |
7a01dc77 | 252 | $cache_timestamp = file_exists($cache_filename) ? filemtime($cache_filename) : 0; |
31623bfa | 253 | $last_updated_timestamp = strtotime($last_updated); |
4f9cbdff | 254 | |
f074ffe9 AD |
255 | if (file_exists($cache_filename) && |
256 | is_readable($cache_filename) && | |
257 | !$auth_login && !$auth_pass && | |
258 | filemtime($cache_filename) > time() - $cache_age) { | |
2c08214a | 259 | |
f074ffe9 AD |
260 | if ($debug_enabled) { |
261 | _debug("update_rss_feed: using local cache."); | |
262 | } | |
be574731 | 263 | |
17e74b21 AD |
264 | if ($cache_timestamp > $last_updated_timestamp) { |
265 | @$rss_data = file_get_contents($cache_filename); | |
f074ffe9 | 266 | |
17e74b21 AD |
267 | if ($rss_data) { |
268 | $rss_hash = sha1($rss_data); | |
269 | @$rss = unserialize($rss_data); | |
270 | } | |
271 | } else { | |
272 | if ($debug_enabled) { | |
273 | _debug("update_rss_feed: local cache valid and older than last_updated, nothing to do."); | |
274 | } | |
275 | return; | |
4f9cbdff | 276 | } |
f074ffe9 | 277 | } |
017401dd | 278 | |
4f9cbdff | 279 | if (!$rss) { |
017401dd | 280 | |
4f9cbdff AD |
281 | if (!$feed_data) { |
282 | if ($debug_enabled) { | |
31623bfa | 283 | _debug("update_rss_feed: fetching [$fetch_url] (ts: $cache_timestamp/$last_updated_timestamp)"); |
4f9cbdff | 284 | } |
017401dd | 285 | |
4f9cbdff | 286 | $feed_data = fetch_file_contents($fetch_url, false, |
31623bfa AD |
287 | $auth_login, $auth_pass, false, $no_cache ? 15 : 45, |
288 | max($last_updated_timestamp, $cache_timestamp)); | |
96f98cb0 | 289 | |
4f9cbdff AD |
290 | if ($debug_enabled) { |
291 | _debug("update_rss_feed: fetch done."); | |
292 | } | |
017401dd | 293 | |
4f9cbdff | 294 | } |
017401dd | 295 | |
4f9cbdff AD |
296 | if (!$feed_data) { |
297 | global $fetch_last_error; | |
7a01dc77 | 298 | global $fetch_last_error_code; |
f074ffe9 AD |
299 | |
300 | if ($debug_enabled) { | |
7a01dc77 | 301 | _debug("update_rss_feed: unable to fetch: $fetch_last_error [$fetch_last_error_code]"); |
f074ffe9 AD |
302 | } |
303 | ||
7a01dc77 AD |
304 | $error_escaped = ''; |
305 | ||
306 | // If-Modified-Since | |
307 | if ($fetch_last_error_code != 304) { | |
308 | $error_escaped = db_escape_string($link, $fetch_last_error); | |
309 | } else { | |
310 | if ($debug_enabled) { | |
311 | _debug("update_rss_feed: source claims data not modified, nothing to do."); | |
312 | } | |
313 | } | |
4f9cbdff AD |
314 | |
315 | db_query($link, | |
316 | "UPDATE ttrss_feeds SET last_error = '$error_escaped', | |
317 | last_updated = NOW() WHERE id = '$feed'"); | |
318 | ||
319 | return; | |
f074ffe9 AD |
320 | } |
321 | } | |
322 | ||
017401dd | 323 | $pluginhost = new PluginHost($link); |
be178857 | 324 | $pluginhost->set_debug($debug_enabled); |
ab457a9c AD |
325 | $user_plugins = get_pref($link, "_ENABLED_PLUGINS", $owner_uid); |
326 | ||
327 | $pluginhost->load(PLUGINS, $pluginhost::KIND_ALL); | |
328 | $pluginhost->load($user_plugins, $pluginhost::KIND_USER, $owner_uid); | |
329 | $pluginhost->load_data(); | |
017401dd AD |
330 | |
331 | foreach ($pluginhost->get_hooks($pluginhost::HOOK_FEED_FETCHED) as $plugin) { | |
332 | $feed_data = $plugin->hook_feed_fetched($feed_data); | |
333 | } | |
334 | ||
4f9cbdff AD |
335 | if (!$rss) { |
336 | $rss = new SimplePie(); | |
337 | $rss->set_sanitize_class("SanitizeDummy"); | |
338 | // simplepie ignores the above and creates default sanitizer anyway, | |
339 | // so let's override it... | |
340 | $rss->sanitize = new SanitizeDummy(); | |
341 | $rss->set_output_encoding('UTF-8'); | |
342 | $rss->set_raw_data($feed_data); | |
343 | $rss->enable_cache(false); | |
344 | ||
345 | @$rss->init(); | |
5de51df7 AD |
346 | } |
347 | ||
2c08214a AD |
348 | // print_r($rss); |
349 | ||
3972bf59 | 350 | $feed = db_escape_string($link, $feed); |
2c08214a | 351 | |
19b3992b | 352 | if (!$rss->error()) { |
2c08214a | 353 | |
4f9cbdff AD |
354 | // cache data for later |
355 | if (!$auth_pass && !$auth_login && is_writable(CACHE_DIR . "/simplepie")) { | |
356 | $rss_data = serialize($rss); | |
357 | $new_rss_hash = sha1($rss_data); | |
358 | ||
359 | if ($new_rss_hash != $rss_hash) { | |
360 | if ($debug_enabled) { | |
361 | _debug("update_rss_feed: saving $cache_filename"); | |
362 | } | |
363 | @file_put_contents($cache_filename, serialize($rss)); | |
364 | } | |
365 | } | |
366 | ||
d2a421e3 | 367 | // We use local pluginhost here because we need to load different per-user feed plugins |
4412b877 AD |
368 | $pluginhost->run_hooks($pluginhost::HOOK_FEED_PARSED, "hook_feed_parsed", $rss); |
369 | ||
2c08214a AD |
370 | if ($debug_enabled) { |
371 | _debug("update_rss_feed: processing feed data..."); | |
372 | } | |
373 | ||
374 | // db_query($link, "BEGIN"); | |
375 | ||
382268c6 AD |
376 | if (DB_TYPE == "pgsql") { |
377 | $favicon_interval_qpart = "favicon_last_checked < NOW() - INTERVAL '12 hour'"; | |
378 | } else { | |
379 | $favicon_interval_qpart = "favicon_last_checked < DATE_SUB(NOW(), INTERVAL 12 HOUR)"; | |
380 | } | |
381 | ||
750cfcd2 | 382 | $result = db_query($link, "SELECT title,site_url,owner_uid, |
382268c6 AD |
383 | (favicon_last_checked IS NULL OR $favicon_interval_qpart) AS |
384 | favicon_needs_check | |
2c08214a AD |
385 | FROM ttrss_feeds WHERE id = '$feed'"); |
386 | ||
387 | $registered_title = db_fetch_result($result, 0, "title"); | |
2c08214a | 388 | $orig_site_url = db_fetch_result($result, 0, "site_url"); |
382268c6 AD |
389 | $favicon_needs_check = sql_bool_to_bool(db_fetch_result($result, 0, |
390 | "favicon_needs_check")); | |
2c08214a AD |
391 | |
392 | $owner_uid = db_fetch_result($result, 0, "owner_uid"); | |
393 | ||
3972bf59 | 394 | $site_url = db_escape_string($link, mb_substr(rewrite_relative_url($fetch_url, $rss->get_link()), 0, 245)); |
2c08214a AD |
395 | |
396 | if ($debug_enabled) { | |
397 | _debug("update_rss_feed: checking favicon..."); | |
398 | } | |
399 | ||
f2798eb6 AD |
400 | if ($favicon_needs_check) { |
401 | check_feed_favicon($site_url, $feed, $link); | |
402 | ||
403 | db_query($link, "UPDATE ttrss_feeds SET favicon_last_checked = NOW() | |
404 | WHERE id = '$feed'"); | |
405 | } | |
2c08214a AD |
406 | |
407 | if (!$registered_title || $registered_title == "[Unknown]") { | |
408 | ||
3972bf59 | 409 | $feed_title = db_escape_string($link, $rss->get_title()); |
2c08214a AD |
410 | |
411 | if ($debug_enabled) { | |
412 | _debug("update_rss_feed: registering title: $feed_title"); | |
413 | } | |
414 | ||
415 | db_query($link, "UPDATE ttrss_feeds SET | |
416 | title = '$feed_title' WHERE id = '$feed'"); | |
417 | } | |
418 | ||
0cf81637 | 419 | if ($site_url && $orig_site_url != $site_url) { |
2c08214a AD |
420 | db_query($link, "UPDATE ttrss_feeds SET |
421 | site_url = '$site_url' WHERE id = '$feed'"); | |
422 | } | |
423 | ||
2c08214a | 424 | if ($debug_enabled) { |
b24504b1 | 425 | _debug("update_rss_feed: loading filters & labels..."); |
2c08214a AD |
426 | } |
427 | ||
428 | $filters = load_filters($link, $feed, $owner_uid); | |
b24504b1 | 429 | $labels = get_all_labels($link, $owner_uid); |
2c08214a | 430 | |
67bd0b1f AD |
431 | if ($debug_enabled) { |
432 | //print_r($filters); | |
433 | _debug("update_rss_feed: " . count($filters) . " filters loaded."); | |
434 | } | |
2c08214a | 435 | |
19b3992b | 436 | $items = $rss->get_items(); |
2c08214a | 437 | |
19b3992b | 438 | if (!is_array($items)) { |
2c08214a | 439 | if ($debug_enabled) { |
19b3992b | 440 | _debug("update_rss_feed: no articles found."); |
2c08214a AD |
441 | } |
442 | ||
443 | db_query($link, "UPDATE ttrss_feeds | |
444 | SET last_updated = NOW(), last_error = '' WHERE id = '$feed'"); | |
445 | ||
446 | return; // no articles | |
447 | } | |
448 | ||
449 | if ($pubsub_state != 2 && PUBSUBHUBBUB_ENABLED) { | |
450 | ||
451 | if ($debug_enabled) _debug("update_rss_feed: checking for PUSH hub..."); | |
452 | ||
453 | $feed_hub_url = false; | |
2c08214a | 454 | |
19b3992b | 455 | $links = $rss->get_links('hub'); |
2c08214a | 456 | |
19b3992b AD |
457 | if ($links && is_array($links)) { |
458 | foreach ($links as $l) { | |
459 | $feed_hub_url = $l; | |
460 | break; | |
2c08214a AD |
461 | } |
462 | } | |
463 | ||
464 | if ($debug_enabled) _debug("update_rss_feed: feed hub url: $feed_hub_url"); | |
465 | ||
466 | if ($feed_hub_url && function_exists('curl_init') && | |
467 | !ini_get("open_basedir")) { | |
468 | ||
469 | require_once 'lib/pubsubhubbub/subscriber.php'; | |
470 | ||
471 | $callback_url = get_self_url_prefix() . | |
472 | "/public.php?op=pubsub&id=$feed"; | |
473 | ||
474 | $s = new Subscriber($feed_hub_url, $callback_url); | |
475 | ||
476 | $rc = $s->subscribe($fetch_url); | |
477 | ||
478 | if ($debug_enabled) | |
479 | _debug("update_rss_feed: feed hub url found, subscribe request sent."); | |
480 | ||
481 | db_query($link, "UPDATE ttrss_feeds SET pubsub_state = 1 | |
482 | WHERE id = '$feed'"); | |
483 | } | |
484 | } | |
485 | ||
486 | if ($debug_enabled) { | |
487 | _debug("update_rss_feed: processing articles..."); | |
488 | } | |
489 | ||
19b3992b | 490 | foreach ($items as $item) { |
5d56d100 | 491 | if ($_REQUEST['xdebug'] == 3) { |
2c08214a AD |
492 | print_r($item); |
493 | } | |
494 | ||
19b3992b AD |
495 | $entry_guid = $item->get_id(); |
496 | if (!$entry_guid) $entry_guid = $item->get_link(); | |
497 | if (!$entry_guid) $entry_guid = make_guid_from_title($item->get_title()); | |
2c08214a AD |
498 | |
499 | if ($debug_enabled) { | |
500 | _debug("update_rss_feed: guid $entry_guid"); | |
501 | } | |
502 | ||
503 | if (!$entry_guid) continue; | |
504 | ||
3a4c8973 AD |
505 | $entry_guid = "$owner_uid,$entry_guid"; |
506 | ||
2c08214a AD |
507 | $entry_timestamp = ""; |
508 | ||
19b3992b | 509 | $entry_timestamp = strtotime($item->get_date()); |
2c08214a | 510 | |
30123fe6 | 511 | if ($entry_timestamp == -1 || !$entry_timestamp || $entry_timestamp > time()) { |
2c08214a AD |
512 | $entry_timestamp = time(); |
513 | $no_orig_date = 'true'; | |
514 | } else { | |
515 | $no_orig_date = 'false'; | |
516 | } | |
517 | ||
518 | $entry_timestamp_fmt = strftime("%Y/%m/%d %H:%M:%S", $entry_timestamp); | |
519 | ||
520 | if ($debug_enabled) { | |
521 | _debug("update_rss_feed: date $entry_timestamp [$entry_timestamp_fmt]"); | |
522 | } | |
523 | ||
19b3992b | 524 | $entry_title = $item->get_title(); |
1b35d30c | 525 | |
5d56d100 | 526 | $entry_link = rewrite_relative_url($site_url, $item->get_link()); |
2c08214a AD |
527 | |
528 | if ($debug_enabled) { | |
529 | _debug("update_rss_feed: title $entry_title"); | |
530 | _debug("update_rss_feed: link $entry_link"); | |
531 | } | |
532 | ||
533 | if (!$entry_title) $entry_title = date("Y-m-d H:i:s", $entry_timestamp);; | |
534 | ||
19b3992b AD |
535 | $entry_content = $item->get_content(); |
536 | if (!$entry_content) $entry_content = $item->get_description(); | |
2c08214a AD |
537 | |
538 | if ($_REQUEST["xdebug"] == 2) { | |
539 | print "update_rss_feed: content: "; | |
487f0750 | 540 | print $entry_content; |
3c696512 | 541 | print "\n"; |
2c08214a AD |
542 | } |
543 | ||
19b3992b | 544 | $entry_comments = $item->data["comments"]; |
2c08214a | 545 | |
19b3992b AD |
546 | if ($item->get_author()) { |
547 | $entry_author_item = $item->get_author(); | |
548 | $entry_author = $entry_author_item->get_name(); | |
549 | if (!$entry_author) $entry_author = $entry_author_item->get_email(); | |
2c08214a | 550 | |
3972bf59 | 551 | $entry_author = db_escape_string($link, $entry_author); |
2c08214a AD |
552 | } |
553 | ||
3972bf59 | 554 | $entry_guid = db_escape_string($link, mb_substr($entry_guid, 0, 245)); |
2c08214a | 555 | |
3972bf59 AD |
556 | $entry_comments = db_escape_string($link, mb_substr($entry_comments, 0, 245)); |
557 | $entry_author = db_escape_string($link, mb_substr($entry_author, 0, 245)); | |
2c08214a | 558 | |
19b3992b | 559 | $num_comments = $item->get_item_tags('http://purl.org/rss/1.0/modules/slash/', 'comments'); |
83e6e313 | 560 | |
19b3992b AD |
561 | if (is_array($num_comments) && is_array($num_comments[0])) { |
562 | $num_comments = (int) $num_comments[0]["data"]; | |
2c08214a | 563 | } else { |
19b3992b | 564 | $num_comments = 0; |
2c08214a AD |
565 | } |
566 | ||
2c08214a | 567 | if ($debug_enabled) { |
83e6e313 | 568 | _debug("update_rss_feed: num_comments: $num_comments"); |
2c08214a AD |
569 | _debug("update_rss_feed: looking for tags [1]..."); |
570 | } | |
571 | ||
572 | // parse <category> entries into tags | |
573 | ||
574 | $additional_tags = array(); | |
575 | ||
19b3992b | 576 | $additional_tags_src = $item->get_categories(); |
2c08214a | 577 | |
19b3992b AD |
578 | if (is_array($additional_tags_src)) { |
579 | foreach ($additional_tags_src as $tobj) { | |
580 | array_push($additional_tags, $tobj->get_term()); | |
2c08214a | 581 | } |
19b3992b | 582 | } |
2c08214a | 583 | |
19b3992b AD |
584 | if ($debug_enabled) { |
585 | _debug("update_rss_feed: category tags:"); | |
586 | print_r($additional_tags); | |
2c08214a AD |
587 | } |
588 | ||
589 | if ($debug_enabled) { | |
590 | _debug("update_rss_feed: looking for tags [2]..."); | |
591 | } | |
592 | ||
fa6fbd36 | 593 | $entry_tags = array_unique($additional_tags); |
2c08214a AD |
594 | |
595 | for ($i = 0; $i < count($entry_tags); $i++) | |
596 | $entry_tags[$i] = mb_strtolower($entry_tags[$i], 'utf-8'); | |
597 | ||
598 | if ($debug_enabled) { | |
6aff7845 AD |
599 | //_debug("update_rss_feed: unfiltered tags found:"); |
600 | //print_r($entry_tags); | |
2c08214a AD |
601 | } |
602 | ||
2c08214a | 603 | if ($debug_enabled) { |
70caff48 | 604 | _debug("update_rss_feed: done collecting data."); |
2c08214a AD |
605 | } |
606 | ||
cc85704f | 607 | // TODO: less memory-hungry implementation |
19c73507 | 608 | |
19b3992b AD |
609 | if ($debug_enabled) { |
610 | _debug("update_rss_feed: applying plugin filters.."); | |
611 | } | |
cc85704f | 612 | |
e02555c1 AD |
613 | // FIXME not sure if owner_uid is a good idea here, we may have a base entry without user entry (?) |
614 | $result = db_query($link, "SELECT plugin_data,title,content,link,tag_cache,author FROM ttrss_entries, ttrss_user_entries | |
3972bf59 | 615 | WHERE ref_id = id AND guid = '".db_escape_string($link, $entry_guid)."' AND owner_uid = $owner_uid"); |
b30abdad AD |
616 | |
617 | if (db_num_rows($result) != 0) { | |
618 | $entry_plugin_data = db_fetch_result($result, 0, "plugin_data"); | |
e02555c1 AD |
619 | $stored_article = array("title" => db_fetch_result($result, 0, "title"), |
620 | "content" => db_fetch_result($result, 0, "content"), | |
621 | "link" => db_fetch_result($result, 0, "link"), | |
622 | "tags" => explode(",", db_fetch_result($result, 0, "tag_cache")), | |
623 | "author" => db_fetch_result($result, 0, "author")); | |
b30abdad AD |
624 | } else { |
625 | $entry_plugin_data = ""; | |
e02555c1 | 626 | $stored_article = array(); |
b30abdad AD |
627 | } |
628 | ||
455b1401 | 629 | $article = array("owner_uid" => $owner_uid, // read only |
b30abdad | 630 | "guid" => $entry_guid, // read only |
19b3992b AD |
631 | "title" => $entry_title, |
632 | "content" => $entry_content, | |
633 | "link" => $entry_link, | |
634 | "tags" => $entry_tags, | |
b30abdad | 635 | "plugin_data" => $entry_plugin_data, |
e02555c1 AD |
636 | "author" => $entry_author, |
637 | "stored" => $stored_article); | |
cc85704f | 638 | |
19b3992b AD |
639 | foreach ($pluginhost->get_hooks($pluginhost::HOOK_ARTICLE_FILTER) as $plugin) { |
640 | $article = $plugin->hook_article_filter($article); | |
cc85704f AD |
641 | } |
642 | ||
19b3992b | 643 | $entry_tags = $article["tags"]; |
3972bf59 AD |
644 | $entry_guid = db_escape_string($link, $entry_guid); |
645 | $entry_title = db_escape_string($link, $article["title"]); | |
646 | $entry_author = db_escape_string($link, $article["author"]); | |
647 | $entry_link = db_escape_string($link, $article["link"]); | |
648 | $entry_plugin_data = db_escape_string($link, $article["plugin_data"]); | |
f935d98e AD |
649 | $entry_content = $article["content"]; // escaped below |
650 | ||
b30abdad AD |
651 | |
652 | if ($debug_enabled) { | |
653 | _debug("update_rss_feed: plugin data: $entry_plugin_data"); | |
654 | } | |
2bbd6994 | 655 | |
0a3fd79b | 656 | if ($cache_images && is_writable(CACHE_DIR . '/images')) |
f0bd8e65 | 657 | cache_images($entry_content, $site_url, $debug_enabled); |
0a3fd79b | 658 | |
3972bf59 | 659 | $entry_content = db_escape_string($link, $entry_content, false); |
1f45c857 | 660 | |
9e222305 | 661 | $content_hash = "SHA1:" . sha1($entry_content); |
cc85704f | 662 | |
2c08214a AD |
663 | db_query($link, "BEGIN"); |
664 | ||
9e222305 AD |
665 | $result = db_query($link, "SELECT id FROM ttrss_entries |
666 | WHERE guid = '$entry_guid'"); | |
667 | ||
2c08214a AD |
668 | if (db_num_rows($result) == 0) { |
669 | ||
670 | if ($debug_enabled) { | |
9e222305 | 671 | _debug("update_rss_feed: base guid [$entry_guid] not found"); |
2c08214a AD |
672 | } |
673 | ||
674 | // base post entry does not exist, create it | |
675 | ||
676 | $result = db_query($link, | |
677 | "INSERT INTO ttrss_entries | |
678 | (title, | |
679 | guid, | |
680 | link, | |
681 | updated, | |
682 | content, | |
683 | content_hash, | |
87764a50 | 684 | cached_content, |
2c08214a AD |
685 | no_orig_date, |
686 | date_updated, | |
687 | date_entered, | |
688 | comments, | |
689 | num_comments, | |
b30abdad | 690 | plugin_data, |
2c08214a AD |
691 | author) |
692 | VALUES | |
693 | ('$entry_title', | |
694 | '$entry_guid', | |
695 | '$entry_link', | |
696 | '$entry_timestamp_fmt', | |
697 | '$entry_content', | |
698 | '$content_hash', | |
0a3fd79b | 699 | '', |
2c08214a AD |
700 | $no_orig_date, |
701 | NOW(), | |
be574731 | 702 | '$date_feed_processed', |
2c08214a AD |
703 | '$entry_comments', |
704 | '$num_comments', | |
b30abdad | 705 | '$entry_plugin_data', |
2c08214a | 706 | '$entry_author')"); |
e8291805 AD |
707 | |
708 | $article_labels = array(); | |
709 | ||
2c08214a AD |
710 | } else { |
711 | // we keep encountering the entry in feeds, so we need to | |
712 | // update date_updated column so that we don't get horrible | |
713 | // dupes when the entry gets purged and reinserted again e.g. | |
714 | // in the case of SLOW SLOW OMG SLOW updating feeds | |
715 | ||
716 | $base_entry_id = db_fetch_result($result, 0, "id"); | |
717 | ||
718 | db_query($link, "UPDATE ttrss_entries SET date_updated = NOW() | |
719 | WHERE id = '$base_entry_id'"); | |
e8291805 AD |
720 | |
721 | $article_labels = get_article_labels($link, $base_entry_id, $owner_uid); | |
2c08214a AD |
722 | } |
723 | ||
724 | // now it should exist, if not - bad luck then | |
725 | ||
726 | $result = db_query($link, "SELECT | |
b30abdad | 727 | id,content_hash,no_orig_date,title,plugin_data, |
2c08214a AD |
728 | ".SUBSTRING_FOR_DATE."(date_updated,1,19) as date_updated, |
729 | ".SUBSTRING_FOR_DATE."(updated,1,19) as updated, | |
0a3fd79b | 730 | num_comments |
2c08214a AD |
731 | FROM |
732 | ttrss_entries | |
733 | WHERE guid = '$entry_guid'"); | |
734 | ||
735 | $entry_ref_id = 0; | |
736 | $entry_int_id = 0; | |
737 | ||
738 | if (db_num_rows($result) == 1) { | |
739 | ||
740 | if ($debug_enabled) { | |
9e222305 | 741 | _debug("update_rss_feed: base guid [$entry_guid] found, checking for user record"); |
2c08214a AD |
742 | } |
743 | ||
744 | // this will be used below in update handler | |
745 | $orig_content_hash = db_fetch_result($result, 0, "content_hash"); | |
746 | $orig_title = db_fetch_result($result, 0, "title"); | |
747 | $orig_num_comments = db_fetch_result($result, 0, "num_comments"); | |
748 | $orig_date_updated = strtotime(db_fetch_result($result, | |
749 | 0, "date_updated")); | |
b30abdad | 750 | $orig_plugin_data = db_fetch_result($result, 0, "plugin_data"); |
2c08214a AD |
751 | |
752 | $ref_id = db_fetch_result($result, 0, "id"); | |
753 | $entry_ref_id = $ref_id; | |
754 | ||
755 | // check for user post link to main table | |
756 | ||
757 | // do we allow duplicate posts with same GUID in different feeds? | |
758 | if (get_pref($link, "ALLOW_DUPLICATE_POSTS", $owner_uid, false)) { | |
759 | $dupcheck_qpart = "AND (feed_id = '$feed' OR feed_id IS NULL)"; | |
760 | } else { | |
761 | $dupcheck_qpart = ""; | |
762 | } | |
763 | ||
764 | /* Collect article tags here so we could filter by them: */ | |
765 | ||
766 | $article_filters = get_article_filters($filters, $entry_title, | |
4021d61a | 767 | $entry_content, $entry_link, $entry_timestamp, $entry_author, |
2c08214a AD |
768 | $entry_tags); |
769 | ||
770 | if ($debug_enabled) { | |
771 | _debug("update_rss_feed: article filters: "); | |
772 | if (count($article_filters) != 0) { | |
773 | print_r($article_filters); | |
774 | } | |
775 | } | |
776 | ||
777 | if (find_article_filter($article_filters, "filter")) { | |
778 | db_query($link, "COMMIT"); // close transaction in progress | |
779 | continue; | |
780 | } | |
781 | ||
782 | $score = calculate_article_score($article_filters); | |
783 | ||
784 | if ($debug_enabled) { | |
785 | _debug("update_rss_feed: initial score: $score"); | |
786 | } | |
787 | ||
788 | $query = "SELECT ref_id, int_id FROM ttrss_user_entries WHERE | |
789 | ref_id = '$ref_id' AND owner_uid = '$owner_uid' | |
790 | $dupcheck_qpart"; | |
791 | ||
792 | // if ($_REQUEST["xdebug"]) print "$query\n"; | |
793 | ||
794 | $result = db_query($link, $query); | |
795 | ||
796 | // okay it doesn't exist - create user entry | |
797 | if (db_num_rows($result) == 0) { | |
798 | ||
799 | if ($debug_enabled) { | |
800 | _debug("update_rss_feed: user record not found, creating..."); | |
801 | } | |
802 | ||
803 | if ($score >= -500 && !find_article_filter($article_filters, 'catchup')) { | |
804 | $unread = 'true'; | |
805 | $last_read_qpart = 'NULL'; | |
806 | } else { | |
807 | $unread = 'false'; | |
808 | $last_read_qpart = 'NOW()'; | |
809 | } | |
810 | ||
811 | if (find_article_filter($article_filters, 'mark') || $score > 1000) { | |
812 | $marked = 'true'; | |
813 | } else { | |
814 | $marked = 'false'; | |
815 | } | |
816 | ||
817 | if (find_article_filter($article_filters, 'publish')) { | |
818 | $published = 'true'; | |
819 | } else { | |
820 | $published = 'false'; | |
821 | } | |
822 | ||
2ea9bbfd AD |
823 | // N-grams |
824 | ||
825 | if (DB_TYPE == "pgsql" and defined('_NGRAM_TITLE_DUPLICATE_THRESHOLD')) { | |
826 | ||
827 | $result = db_query($link, "SELECT COUNT(*) AS similar FROM | |
828 | ttrss_entries,ttrss_user_entries | |
829 | WHERE ref_id = id AND updated >= NOW() - INTERVAL '7 day' | |
830 | AND similarity(title, '$entry_title') >= "._NGRAM_TITLE_DUPLICATE_THRESHOLD." | |
831 | AND owner_uid = $owner_uid"); | |
832 | ||
833 | $ngram_similar = db_fetch_result($result, 0, "similar"); | |
834 | ||
835 | if ($debug_enabled) { | |
836 | _debug("update_rss_feed: N-gram similar results: $ngram_similar"); | |
837 | } | |
838 | ||
839 | if ($ngram_similar > 0) { | |
840 | $unread = 'false'; | |
841 | } | |
842 | } | |
843 | ||
7873d588 AD |
844 | $last_marked = ($marked == 'true') ? 'NOW()' : 'NULL'; |
845 | $last_published = ($published == 'true') ? 'NOW()' : 'NULL'; | |
846 | ||
2c08214a AD |
847 | $result = db_query($link, |
848 | "INSERT INTO ttrss_user_entries | |
849 | (ref_id, owner_uid, feed_id, unread, last_read, marked, | |
7873d588 AD |
850 | published, score, tag_cache, label_cache, uuid, |
851 | last_marked, last_published) | |
2c08214a | 852 | VALUES ('$ref_id', '$owner_uid', '$feed', $unread, |
7873d588 AD |
853 | $last_read_qpart, $marked, $published, '$score', '', '', |
854 | '', $last_marked, $last_published)"); | |
2c08214a AD |
855 | |
856 | if (PUBSUBHUBBUB_HUB && $published == 'true') { | |
857 | $rss_link = get_self_url_prefix() . | |
858 | "/public.php?op=rss&id=-2&key=" . | |
859 | get_feed_access_key($link, -2, false, $owner_uid); | |
860 | ||
861 | $p = new Publisher(PUBSUBHUBBUB_HUB); | |
862 | ||
863 | $pubsub_result = $p->publish_update($rss_link); | |
864 | } | |
865 | ||
866 | $result = db_query($link, | |
867 | "SELECT int_id FROM ttrss_user_entries WHERE | |
868 | ref_id = '$ref_id' AND owner_uid = '$owner_uid' AND | |
869 | feed_id = '$feed' LIMIT 1"); | |
870 | ||
871 | if (db_num_rows($result) == 1) { | |
872 | $entry_int_id = db_fetch_result($result, 0, "int_id"); | |
873 | } | |
874 | } else { | |
875 | if ($debug_enabled) { | |
876 | _debug("update_rss_feed: user record FOUND"); | |
877 | } | |
878 | ||
879 | $entry_ref_id = db_fetch_result($result, 0, "ref_id"); | |
880 | $entry_int_id = db_fetch_result($result, 0, "int_id"); | |
881 | } | |
882 | ||
883 | if ($debug_enabled) { | |
884 | _debug("update_rss_feed: RID: $entry_ref_id, IID: $entry_int_id"); | |
885 | } | |
886 | ||
887 | $post_needs_update = false; | |
888 | $update_insignificant = false; | |
889 | ||
890 | if ($orig_num_comments != $num_comments) { | |
891 | $post_needs_update = true; | |
892 | $update_insignificant = true; | |
893 | } | |
894 | ||
b30abdad AD |
895 | if ($entry_plugin_data != $orig_plugin_data) { |
896 | $post_needs_update = true; | |
897 | $update_insignificant = true; | |
898 | } | |
899 | ||
2c08214a AD |
900 | if ($content_hash != $orig_content_hash) { |
901 | $post_needs_update = true; | |
902 | $update_insignificant = false; | |
903 | } | |
904 | ||
3972bf59 | 905 | if (db_escape_string($link, $orig_title) != $entry_title) { |
2c08214a AD |
906 | $post_needs_update = true; |
907 | $update_insignificant = false; | |
908 | } | |
909 | ||
910 | // if post needs update, update it and mark all user entries | |
911 | // linking to this post as updated | |
912 | if ($post_needs_update) { | |
913 | ||
914 | if (defined('DAEMON_EXTENDED_DEBUG')) { | |
915 | _debug("update_rss_feed: post $entry_guid needs update..."); | |
916 | } | |
917 | ||
918 | // print "<!-- post $orig_title needs update : $post_needs_update -->"; | |
919 | ||
920 | db_query($link, "UPDATE ttrss_entries | |
921 | SET title = '$entry_title', content = '$entry_content', | |
922 | content_hash = '$content_hash', | |
923 | updated = '$entry_timestamp_fmt', | |
b30abdad AD |
924 | num_comments = '$num_comments', |
925 | plugin_data = '$entry_plugin_data' | |
2c08214a AD |
926 | WHERE id = '$ref_id'"); |
927 | ||
928 | if (!$update_insignificant) { | |
929 | if ($mark_unread_on_update) { | |
930 | db_query($link, "UPDATE ttrss_user_entries | |
931 | SET last_read = null, unread = true WHERE ref_id = '$ref_id'"); | |
2c08214a AD |
932 | } |
933 | } | |
934 | } | |
935 | } | |
936 | ||
937 | db_query($link, "COMMIT"); | |
938 | ||
939 | if ($debug_enabled) { | |
940 | _debug("update_rss_feed: assigning labels..."); | |
941 | } | |
942 | ||
92c14e9d | 943 | assign_article_to_label_filters($link, $entry_ref_id, $article_filters, |
b24504b1 | 944 | $owner_uid, $article_labels); |
2c08214a AD |
945 | |
946 | if ($debug_enabled) { | |
947 | _debug("update_rss_feed: looking for enclosures..."); | |
948 | } | |
949 | ||
950 | // enclosures | |
951 | ||
952 | $enclosures = array(); | |
953 | ||
19b3992b | 954 | $encs = $item->get_enclosures(); |
2c08214a | 955 | |
19b3992b AD |
956 | if (is_array($encs)) { |
957 | foreach ($encs as $e) { | |
958 | $e_item = array( | |
959 | $e->link, $e->type, $e->length); | |
2c08214a | 960 | array_push($enclosures, $e_item); |
2c08214a AD |
961 | } |
962 | } | |
963 | ||
2c08214a AD |
964 | if ($debug_enabled) { |
965 | _debug("update_rss_feed: article enclosures:"); | |
966 | print_r($enclosures); | |
967 | } | |
968 | ||
969 | db_query($link, "BEGIN"); | |
970 | ||
971 | foreach ($enclosures as $enc) { | |
3972bf59 AD |
972 | $enc_url = db_escape_string($link, $enc[0]); |
973 | $enc_type = db_escape_string($link, $enc[1]); | |
974 | $enc_dur = db_escape_string($link, $enc[2]); | |
2c08214a AD |
975 | |
976 | $result = db_query($link, "SELECT id FROM ttrss_enclosures | |
977 | WHERE content_url = '$enc_url' AND post_id = '$entry_ref_id'"); | |
978 | ||
979 | if (db_num_rows($result) == 0) { | |
980 | db_query($link, "INSERT INTO ttrss_enclosures | |
981 | (content_url, content_type, title, duration, post_id) VALUES | |
982 | ('$enc_url', '$enc_type', '', '$enc_dur', '$entry_ref_id')"); | |
983 | } | |
984 | } | |
985 | ||
986 | db_query($link, "COMMIT"); | |
987 | ||
988 | // check for manual tags (we have to do it here since they're loaded from filters) | |
989 | ||
990 | foreach ($article_filters as $f) { | |
6aff7845 | 991 | if ($f["type"] == "tag") { |
2c08214a | 992 | |
6aff7845 | 993 | $manual_tags = trim_array(explode(",", $f["param"])); |
2c08214a AD |
994 | |
995 | foreach ($manual_tags as $tag) { | |
996 | if (tag_is_valid($tag)) { | |
997 | array_push($entry_tags, $tag); | |
998 | } | |
999 | } | |
1000 | } | |
1001 | } | |
1002 | ||
1003 | // Skip boring tags | |
1004 | ||
1005 | $boring_tags = trim_array(explode(",", mb_strtolower(get_pref($link, | |
1006 | 'BLACKLISTED_TAGS', $owner_uid, ''), 'utf-8'))); | |
1007 | ||
1008 | $filtered_tags = array(); | |
1009 | $tags_to_cache = array(); | |
1010 | ||
1011 | if ($entry_tags && is_array($entry_tags)) { | |
1012 | foreach ($entry_tags as $tag) { | |
1013 | if (array_search($tag, $boring_tags) === false) { | |
1014 | array_push($filtered_tags, $tag); | |
1015 | } | |
1016 | } | |
1017 | } | |
1018 | ||
1019 | $filtered_tags = array_unique($filtered_tags); | |
1020 | ||
1021 | if ($debug_enabled) { | |
1022 | _debug("update_rss_feed: filtered article tags:"); | |
1023 | print_r($filtered_tags); | |
1024 | } | |
1025 | ||
1026 | // Save article tags in the database | |
1027 | ||
1028 | if (count($filtered_tags) > 0) { | |
1029 | ||
1030 | db_query($link, "BEGIN"); | |
1031 | ||
1032 | foreach ($filtered_tags as $tag) { | |
1033 | ||
1034 | $tag = sanitize_tag($tag); | |
3972bf59 | 1035 | $tag = db_escape_string($link, $tag); |
2c08214a AD |
1036 | |
1037 | if (!tag_is_valid($tag)) continue; | |
1038 | ||
1039 | $result = db_query($link, "SELECT id FROM ttrss_tags | |
1040 | WHERE tag_name = '$tag' AND post_int_id = '$entry_int_id' AND | |
1041 | owner_uid = '$owner_uid' LIMIT 1"); | |
1042 | ||
1043 | if ($result && db_num_rows($result) == 0) { | |
1044 | ||
1045 | db_query($link, "INSERT INTO ttrss_tags | |
1046 | (owner_uid,tag_name,post_int_id) | |
1047 | VALUES ('$owner_uid','$tag', '$entry_int_id')"); | |
1048 | } | |
1049 | ||
1050 | array_push($tags_to_cache, $tag); | |
1051 | } | |
1052 | ||
1053 | /* update the cache */ | |
1054 | ||
1055 | $tags_to_cache = array_unique($tags_to_cache); | |
1056 | ||
3972bf59 | 1057 | $tags_str = db_escape_string($link, join(",", $tags_to_cache)); |
2c08214a AD |
1058 | |
1059 | db_query($link, "UPDATE ttrss_user_entries | |
1060 | SET tag_cache = '$tags_str' WHERE ref_id = '$entry_ref_id' | |
1061 | AND owner_uid = $owner_uid"); | |
1062 | ||
1063 | db_query($link, "COMMIT"); | |
1064 | } | |
1065 | ||
b24504b1 AD |
1066 | if (get_pref($link, "AUTO_ASSIGN_LABELS", $owner_uid, false)) { |
1067 | if ($debug_enabled) { | |
1068 | _debug("update_rss_feed: auto-assigning labels..."); | |
1069 | } | |
1070 | ||
1071 | foreach ($labels as $label) { | |
227d5e41 | 1072 | $caption = preg_quote($label["caption"]); |
b24504b1 | 1073 | |
9811276d | 1074 | if ($caption && preg_match("/\b$caption\b/i", "$tags_str " . strip_tags($entry_content) . " $entry_title")) { |
b24504b1 AD |
1075 | if (!labels_contains_caption($article_labels, $caption)) { |
1076 | label_add_article($link, $entry_ref_id, $caption, $owner_uid); | |
1077 | } | |
1078 | } | |
1079 | } | |
1080 | } | |
1081 | ||
2c08214a AD |
1082 | if ($debug_enabled) { |
1083 | _debug("update_rss_feed: article processed"); | |
1084 | } | |
1085 | } | |
1086 | ||
1087 | if (!$last_updated) { | |
1088 | if ($debug_enabled) { | |
1089 | _debug("update_rss_feed: new feed, catching it up..."); | |
1090 | } | |
1091 | catchup_feed($link, $feed, false, $owner_uid); | |
1092 | } | |
1093 | ||
1094 | if ($debug_enabled) { | |
1095 | _debug("purging feed..."); | |
1096 | } | |
1097 | ||
1098 | purge_feed($link, $feed, 0, $debug_enabled); | |
1099 | ||
1100 | db_query($link, "UPDATE ttrss_feeds | |
f2798eb6 | 1101 | SET last_updated = NOW(), last_error = '' WHERE id = '$feed'"); |
2c08214a AD |
1102 | |
1103 | // db_query($link, "COMMIT"); | |
1104 | ||
1105 | } else { | |
1106 | ||
3972bf59 | 1107 | $error_msg = db_escape_string($link, mb_substr($rss->error(), 0, 245)); |
2c08214a AD |
1108 | |
1109 | if ($debug_enabled) { | |
1110 | _debug("update_rss_feed: error fetching feed: $error_msg"); | |
1111 | } | |
1112 | ||
2c08214a AD |
1113 | db_query($link, |
1114 | "UPDATE ttrss_feeds SET last_error = '$error_msg', | |
1115 | last_updated = NOW() WHERE id = '$feed'"); | |
1116 | } | |
1117 | ||
19b3992b | 1118 | unset($rss); |
2c08214a AD |
1119 | |
1120 | if ($debug_enabled) { | |
1121 | _debug("update_rss_feed: done"); | |
1122 | } | |
1123 | ||
1124 | } | |
1125 | ||
3c696512 AD |
1126 | function cache_images($html, $site_url, $debug) { |
1127 | $cache_dir = CACHE_DIR . "/images"; | |
1128 | ||
1129 | libxml_use_internal_errors(true); | |
1130 | ||
1131 | $charset_hack = '<head> | |
1132 | <meta http-equiv="Content-Type" content="text/html; charset=utf-8"/> | |
1133 | </head>'; | |
1134 | ||
1135 | $doc = new DOMDocument(); | |
1136 | $doc->loadHTML($charset_hack . $html); | |
1137 | $xpath = new DOMXPath($doc); | |
1138 | ||
1139 | $entries = $xpath->query('(//img[@src])'); | |
1140 | ||
1141 | foreach ($entries as $entry) { | |
1142 | if ($entry->hasAttribute('src')) { | |
1143 | $src = rewrite_relative_url($site_url, $entry->getAttribute('src')); | |
1144 | ||
1145 | $local_filename = CACHE_DIR . "/images/" . sha1($src) . ".png"; | |
1146 | ||
1147 | if ($debug) _debug("cache_images: downloading: $src to $local_filename"); | |
1148 | ||
1149 | if (!file_exists($local_filename)) { | |
1150 | $file_content = fetch_file_contents($src); | |
1151 | ||
b8379e69 | 1152 | if ($file_content && strlen($file_content) > 1024) { |
3c696512 AD |
1153 | file_put_contents($local_filename, $file_content); |
1154 | } | |
1155 | } | |
1156 | ||
1157 | if (file_exists($local_filename)) { | |
1158 | $entry->setAttribute('src', SELF_URL_PATH . '/image.php?url=' . | |
487f0750 | 1159 | base64_encode($src)); |
3c696512 AD |
1160 | } |
1161 | } | |
1162 | } | |
1163 | ||
1164 | $node = $doc->getElementsByTagName('body')->item(0); | |
1165 | ||
cc38c8e5 | 1166 | return $doc->saveXML($node); |
3c696512 AD |
1167 | } |
1168 | ||
2a91b6ff AD |
1169 | function expire_lock_files($debug) { |
1170 | if ($debug) _debug("Removing old lock files..."); | |
1171 | ||
1172 | $num_deleted = 0; | |
1173 | ||
1174 | if (is_writable(LOCK_DIRECTORY)) { | |
1175 | $files = glob(LOCK_DIRECTORY . "/*.lock"); | |
1176 | ||
1177 | if ($files) { | |
1178 | foreach ($files as $file) { | |
1179 | if (!file_is_locked($file) && time() - filemtime($file) > 86400*2) { | |
1180 | unlink($file); | |
1181 | ++$num_deleted; | |
1182 | } | |
1183 | } | |
1184 | } | |
1185 | } | |
1186 | ||
1187 | if ($debug) _debug("Removed $num_deleted files."); | |
1188 | } | |
1189 | ||
3c696512 | 1190 | function expire_cached_files($debug) { |
19b3992b | 1191 | foreach (array("simplepie", "images", "export") as $dir) { |
3c696512 | 1192 | $cache_dir = CACHE_DIR . "/$dir"; |
2c08214a | 1193 | |
3c696512 | 1194 | if ($debug) _debug("Expiring $cache_dir"); |
2c08214a | 1195 | |
3c696512 AD |
1196 | $num_deleted = 0; |
1197 | ||
1198 | if (is_writable($cache_dir)) { | |
1199 | $files = glob("$cache_dir/*"); | |
1200 | ||
2a91b6ff | 1201 | if ($files) { |
2ab20c31 AD |
1202 | foreach ($files as $file) { |
1203 | if (time() - filemtime($file) > 86400*7) { | |
1204 | unlink($file); | |
3c696512 | 1205 | |
2ab20c31 AD |
1206 | ++$num_deleted; |
1207 | } | |
3c696512 AD |
1208 | } |
1209 | } | |
2a91b6ff | 1210 | } |
3c696512 AD |
1211 | |
1212 | if ($debug) _debug("Removed $num_deleted files."); | |
1213 | } | |
1214 | } | |
2c08214a | 1215 | |
a3e0bdcf AD |
1216 | /** |
1217 | * Source: http://www.php.net/manual/en/function.parse-url.php#104527 | |
1218 | * Returns the url query as associative array | |
1219 | * | |
1220 | * @param string query | |
1221 | * @return array params | |
1222 | */ | |
1223 | function convertUrlQuery($query) { | |
1224 | $queryParts = explode('&', $query); | |
1225 | ||
1226 | $params = array(); | |
1227 | ||
1228 | foreach ($queryParts as $param) { | |
1229 | $item = explode('=', $param); | |
1230 | $params[$item[0]] = $item[1]; | |
1231 | } | |
1232 | ||
1233 | return $params; | |
1234 | } | |
92c14e9d AD |
1235 | |
1236 | function get_article_filters($filters, $title, $content, $link, $timestamp, $author, $tags) { | |
1237 | $matches = array(); | |
1238 | ||
1239 | foreach ($filters as $filter) { | |
1240 | $match_any_rule = $filter["match_any_rule"]; | |
a3a896a1 | 1241 | $inverse = $filter["inverse"]; |
92c14e9d AD |
1242 | $filter_match = false; |
1243 | ||
1244 | foreach ($filter["rules"] as $rule) { | |
1245 | $match = false; | |
1246 | $reg_exp = $rule["reg_exp"]; | |
a3a896a1 | 1247 | $rule_inverse = $rule["inverse"]; |
92c14e9d AD |
1248 | |
1249 | if (!$reg_exp) | |
1250 | continue; | |
1251 | ||
1252 | switch ($rule["type"]) { | |
1253 | case "title": | |
1254 | $match = @preg_match("/$reg_exp/i", $title); | |
1255 | break; | |
1256 | case "content": | |
d03ae73e AD |
1257 | // we don't need to deal with multiline regexps |
1258 | $content = preg_replace("/[\r\n\t]/", "", $content); | |
1259 | ||
92c14e9d AD |
1260 | $match = @preg_match("/$reg_exp/i", $content); |
1261 | break; | |
1262 | case "both": | |
d03ae73e AD |
1263 | // we don't need to deal with multiline regexps |
1264 | $content = preg_replace("/[\r\n\t]/", "", $content); | |
1265 | ||
72d1d067 | 1266 | $match = (@preg_match("/$reg_exp/i", $title) || @preg_match("/$reg_exp/i", $content)); |
92c14e9d AD |
1267 | break; |
1268 | case "link": | |
1269 | $match = @preg_match("/$reg_exp/i", $link); | |
1270 | break; | |
1271 | case "author": | |
1272 | $match = @preg_match("/$reg_exp/i", $author); | |
1273 | break; | |
1274 | case "tag": | |
1275 | $tag_string = join(",", $tags); | |
1276 | $match = @preg_match("/$reg_exp/i", $tag_string); | |
1277 | break; | |
1278 | } | |
1279 | ||
a3a896a1 AD |
1280 | if ($rule_inverse) $match = !$match; |
1281 | ||
92c14e9d AD |
1282 | if ($match_any_rule) { |
1283 | if ($match) { | |
1284 | $filter_match = true; | |
1285 | break; | |
1286 | } | |
1287 | } else { | |
1288 | $filter_match = $match; | |
1289 | if (!$match) { | |
1290 | break; | |
1291 | } | |
1292 | } | |
1293 | } | |
1294 | ||
a3a896a1 AD |
1295 | if ($inverse) $filter_match = !$filter_match; |
1296 | ||
92c14e9d AD |
1297 | if ($filter_match) { |
1298 | foreach ($filter["actions"] AS $action) { | |
1299 | array_push($matches, $action); | |
1300 | } | |
1301 | } | |
1302 | } | |
1303 | ||
1304 | return $matches; | |
1305 | } | |
1306 | ||
1307 | function find_article_filter($filters, $filter_name) { | |
1308 | foreach ($filters as $f) { | |
1309 | if ($f["type"] == $filter_name) { | |
1310 | return $f; | |
1311 | }; | |
1312 | } | |
1313 | return false; | |
1314 | } | |
1315 | ||
1316 | function find_article_filters($filters, $filter_name) { | |
1317 | $results = array(); | |
1318 | ||
1319 | foreach ($filters as $f) { | |
1320 | if ($f["type"] == $filter_name) { | |
1321 | array_push($results, $f); | |
1322 | }; | |
1323 | } | |
1324 | return $results; | |
1325 | } | |
1326 | ||
1327 | function calculate_article_score($filters) { | |
1328 | $score = 0; | |
1329 | ||
1330 | foreach ($filters as $f) { | |
1331 | if ($f["type"] == "score") { | |
1332 | $score += $f["param"]; | |
1333 | }; | |
1334 | } | |
1335 | return $score; | |
1336 | } | |
1337 | ||
b24504b1 AD |
1338 | function labels_contains_caption($labels, $caption) { |
1339 | foreach ($labels as $label) { | |
1340 | if ($label[1] == $caption) { | |
1341 | return true; | |
1342 | } | |
1343 | } | |
1344 | ||
1345 | return false; | |
1346 | } | |
1347 | ||
1348 | function assign_article_to_label_filters($link, $id, $filters, $owner_uid, $article_labels) { | |
92c14e9d AD |
1349 | foreach ($filters as $f) { |
1350 | if ($f["type"] == "label") { | |
b24504b1 AD |
1351 | if (!labels_contains_caption($article_labels, $f["param"])) { |
1352 | label_add_article($link, $id, $f["param"], $owner_uid); | |
1353 | } | |
1354 | } | |
92c14e9d AD |
1355 | } |
1356 | } | |
87764a50 | 1357 | |
87d7e850 AD |
1358 | function make_guid_from_title($title) { |
1359 | return preg_replace("/[ \"\',.:;]/", "-", | |
1360 | mb_strtolower(strip_tags($title), 'utf-8')); | |
1361 | } | |
1362 | ||
1363 | ||
2c08214a | 1364 | ?> |