]>
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 | |
6322ac79 | 6 | function update_feedbrowser_cache() { |
79178062 | 7 | |
a42c55f0 | 8 | $result = db_query("SELECT feed_url, site_url, title, COUNT(id) AS subscribers |
79178062 AD |
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 | ||
a42c55f0 | 14 | db_query("BEGIN"); |
79178062 | 15 | |
a42c55f0 | 16 | db_query("DELETE FROM ttrss_feedbrowser_cache"); |
79178062 AD |
17 | |
18 | $count = 0; | |
19 | ||
20 | while ($line = db_fetch_assoc($result)) { | |
a42c55f0 AD |
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"]); | |
79178062 | 25 | |
a42c55f0 | 26 | $tmp_result = db_query("SELECT subscribers FROM |
79178062 AD |
27 | ttrss_feedbrowser_cache WHERE feed_url = '$feed_url'"); |
28 | ||
29 | if (db_num_rows($tmp_result) == 0) { | |
30 | ||
a42c55f0 | 31 | db_query("INSERT INTO ttrss_feedbrowser_cache |
79178062 AD |
32 | (feed_url, site_url, title, subscribers) VALUES ('$feed_url', |
33 | '$site_url', '$title', '$subscribers')"); | |
34 | ||
35 | ++$count; | |
36 | ||
37 | } | |
38 | ||
39 | } | |
40 | ||
a42c55f0 | 41 | db_query("COMMIT"); |
79178062 AD |
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 $debug Set to false to disable debug output. Default to true. | |
58 | * @return void | |
59 | */ | |
a42c55f0 | 60 | function update_daemon_common($limit = DAEMON_FEED_LIMIT, $from_http = false, $debug = true) { |
2c08214a AD |
61 | // Process all other feeds using last_updated and interval parameters |
62 | ||
6322ac79 | 63 | $schema_version = get_schema_version(); |
857efe49 AD |
64 | |
65 | if ($schema_version != SCHEMA_VERSION) { | |
66 | die("Schema version is wrong, please upgrade the database.\n"); | |
67 | } | |
68 | ||
61c1812f AD |
69 | define('PREFS_NO_CACHE', true); |
70 | ||
2c08214a | 71 | // Test if the user has loggued in recently. If not, it does not update its feeds. |
09e8bdfd | 72 | if (!SINGLE_USER_MODE && DAEMON_UPDATE_LOGIN_LIMIT > 0) { |
2c08214a AD |
73 | if (DB_TYPE == "pgsql") { |
74 | $login_thresh_qpart = "AND ttrss_users.last_login >= NOW() - INTERVAL '".DAEMON_UPDATE_LOGIN_LIMIT." days'"; | |
75 | } else { | |
76 | $login_thresh_qpart = "AND ttrss_users.last_login >= DATE_SUB(NOW(), INTERVAL ".DAEMON_UPDATE_LOGIN_LIMIT." DAY)"; | |
77 | } | |
78 | } else { | |
79 | $login_thresh_qpart = ""; | |
80 | } | |
81 | ||
82 | // Test if the feed need a update (update interval exceded). | |
83 | if (DB_TYPE == "pgsql") { | |
84 | $update_limit_qpart = "AND (( | |
85 | ttrss_feeds.update_interval = 0 | |
ee0542ce | 86 | AND ttrss_user_prefs.value != '-1' |
2c08214a AD |
87 | AND ttrss_feeds.last_updated < NOW() - CAST((ttrss_user_prefs.value || ' minutes') AS INTERVAL) |
88 | ) OR ( | |
89 | ttrss_feeds.update_interval > 0 | |
90 | AND ttrss_feeds.last_updated < NOW() - CAST((ttrss_feeds.update_interval || ' minutes') AS INTERVAL) | |
941e48a4 AD |
91 | ) OR ttrss_feeds.last_updated IS NULL |
92 | OR last_updated = '1970-01-01 00:00:00')"; | |
2c08214a AD |
93 | } else { |
94 | $update_limit_qpart = "AND (( | |
95 | ttrss_feeds.update_interval = 0 | |
ee0542ce | 96 | AND ttrss_user_prefs.value != '-1' |
2c08214a AD |
97 | AND ttrss_feeds.last_updated < DATE_SUB(NOW(), INTERVAL CONVERT(ttrss_user_prefs.value, SIGNED INTEGER) MINUTE) |
98 | ) OR ( | |
99 | ttrss_feeds.update_interval > 0 | |
100 | AND ttrss_feeds.last_updated < DATE_SUB(NOW(), INTERVAL ttrss_feeds.update_interval MINUTE) | |
941e48a4 AD |
101 | ) OR ttrss_feeds.last_updated IS NULL |
102 | OR last_updated = '1970-01-01 00:00:00')"; | |
2c08214a AD |
103 | } |
104 | ||
105 | // Test if feed is currently being updated by another process. | |
106 | if (DB_TYPE == "pgsql") { | |
107 | $updstart_thresh_qpart = "AND (ttrss_feeds.last_update_started IS NULL OR ttrss_feeds.last_update_started < NOW() - INTERVAL '5 minutes')"; | |
108 | } else { | |
109 | $updstart_thresh_qpart = "AND (ttrss_feeds.last_update_started IS NULL OR ttrss_feeds.last_update_started < DATE_SUB(NOW(), INTERVAL 5 MINUTE))"; | |
110 | } | |
111 | ||
112 | // Test if there is a limit to number of updated feeds | |
113 | $query_limit = ""; | |
114 | if($limit) $query_limit = sprintf("LIMIT %d", $limit); | |
115 | ||
116 | $random_qpart = sql_random_function(); | |
117 | ||
118 | // We search for feed needing update. | |
a42c55f0 | 119 | $result = db_query("SELECT DISTINCT ttrss_feeds.feed_url,$random_qpart |
2c08214a AD |
120 | FROM |
121 | ttrss_feeds, ttrss_users, ttrss_user_prefs | |
122 | WHERE | |
123 | ttrss_feeds.owner_uid = ttrss_users.id | |
124 | AND ttrss_users.id = ttrss_user_prefs.owner_uid | |
125 | AND ttrss_user_prefs.pref_name = 'DEFAULT_UPDATE_INTERVAL' | |
126 | $login_thresh_qpart $update_limit_qpart | |
1c4421fc | 127 | $updstart_thresh_qpart |
e81610d9 | 128 | ORDER BY $random_qpart $query_limit"); |
2c08214a | 129 | |
ca5ff2d9 | 130 | if($debug) _debug(sprintf("Scheduled %d feeds to update...", db_num_rows($result))); |
2c08214a AD |
131 | |
132 | // Here is a little cache magic in order to minimize risk of double feed updates. | |
133 | $feeds_to_update = array(); | |
134 | while ($line = db_fetch_assoc($result)) { | |
a42c55f0 | 135 | array_push($feeds_to_update, db_escape_string($line['feed_url'])); |
2c08214a AD |
136 | } |
137 | ||
138 | // We update the feed last update started date before anything else. | |
139 | // There is no lag due to feed contents downloads | |
140 | // It prevent an other process to update the same feed. | |
1c4421fc AD |
141 | |
142 | if(count($feeds_to_update) > 0) { | |
143 | $feeds_quoted = array(); | |
144 | ||
145 | foreach ($feeds_to_update as $feed) { | |
a42c55f0 | 146 | array_push($feeds_quoted, "'" . db_escape_string($feed) . "'"); |
1c4421fc AD |
147 | } |
148 | ||
a42c55f0 | 149 | db_query(sprintf("UPDATE ttrss_feeds SET last_update_started = NOW() |
1c4421fc | 150 | WHERE feed_url IN (%s)", implode(',', $feeds_quoted))); |
2c08214a AD |
151 | } |
152 | ||
3c696512 | 153 | expire_cached_files($debug); |
2a91b6ff | 154 | expire_lock_files($debug); |
a42c55f0 | 155 | expire_error_log($debug); |
3c696512 | 156 | |
8292d05b AD |
157 | $nf = 0; |
158 | ||
2c08214a | 159 | // For each feed, we call the feed update function. |
1c4421fc AD |
160 | foreach ($feeds_to_update as $feed) { |
161 | if($debug) _debug("Base feed: $feed"); | |
2c08214a | 162 | |
a42c55f0 | 163 | //update_rss_feed($line["id"], true); |
2c08214a | 164 | |
1c4421fc AD |
165 | // since we have the data cached, we can deal with other feeds with the same url |
166 | ||
a42c55f0 | 167 | $tmp_result = db_query("SELECT DISTINCT ttrss_feeds.id,last_updated |
ee0542ce AD |
168 | FROM ttrss_feeds, ttrss_users, ttrss_user_prefs WHERE |
169 | ttrss_user_prefs.owner_uid = ttrss_feeds.owner_uid AND | |
170 | ttrss_users.id = ttrss_user_prefs.owner_uid AND | |
171 | ttrss_user_prefs.pref_name = 'DEFAULT_UPDATE_INTERVAL' AND | |
a42c55f0 | 172 | feed_url = '".db_escape_string($feed)."' AND |
ee0542ce AD |
173 | (ttrss_feeds.update_interval > 0 OR |
174 | ttrss_user_prefs.value != '-1') | |
1c4421fc | 175 | $login_thresh_qpart |
5929a0c1 | 176 | ORDER BY ttrss_feeds.id $query_limit"); |
1c4421fc AD |
177 | |
178 | if (db_num_rows($tmp_result) > 0) { | |
179 | while ($tline = db_fetch_assoc($tmp_result)) { | |
32b05702 | 180 | if($debug) _debug(" => " . $tline["last_updated"] . ", " . $tline["id"]); |
a42c55f0 | 181 | update_rss_feed($tline["id"], true); |
8292d05b | 182 | ++$nf; |
1c4421fc AD |
183 | } |
184 | } | |
2c08214a AD |
185 | } |
186 | ||
f1611683 AD |
187 | require_once "digest.php"; |
188 | ||
2c08214a | 189 | // Send feed digests by email if needed. |
a42c55f0 | 190 | send_headlines_digests($debug); |
2c08214a | 191 | |
8292d05b AD |
192 | return $nf; |
193 | ||
2c08214a AD |
194 | } // function update_daemon_common |
195 | ||
87764a50 | 196 | // ignore_daemon is not used |
a42c55f0 | 197 | function update_rss_feed($feed, $ignore_daemon = false, $no_cache = false, |
2c08214a AD |
198 | $override_url = false) { |
199 | ||
200 | require_once "lib/simplepie/simplepie.inc"; | |
2c08214a | 201 | |
2c08214a AD |
202 | $debug_enabled = defined('DAEMON_EXTENDED_DEBUG') || $_REQUEST['xdebug']; |
203 | ||
2c08214a AD |
204 | if ($debug_enabled) { |
205 | _debug("update_rss_feed: start"); | |
206 | } | |
207 | ||
a42c55f0 | 208 | $result = db_query("SELECT id,update_interval,auth_login, |
0a3fd79b | 209 | feed_url,auth_pass,cache_images,last_updated, |
5321e775 | 210 | mark_unread_on_update, owner_uid, |
044cff2d | 211 | pubsub_state, auth_pass_encrypted |
87764a50 | 212 | FROM ttrss_feeds WHERE id = '$feed'"); |
2c08214a AD |
213 | |
214 | if (db_num_rows($result) == 0) { | |
215 | if ($debug_enabled) { | |
216 | _debug("update_rss_feed: feed $feed NOT FOUND/SKIPPED"); | |
217 | } | |
218 | return false; | |
219 | } | |
220 | ||
2c08214a AD |
221 | $last_updated = db_fetch_result($result, 0, "last_updated"); |
222 | $owner_uid = db_fetch_result($result, 0, "owner_uid"); | |
223 | $mark_unread_on_update = sql_bool_to_bool(db_fetch_result($result, | |
224 | 0, "mark_unread_on_update")); | |
2c08214a | 225 | $pubsub_state = db_fetch_result($result, 0, "pubsub_state"); |
044cff2d AD |
226 | $auth_pass_encrypted = sql_bool_to_bool(db_fetch_result($result, |
227 | 0, "auth_pass_encrypted")); | |
2c08214a | 228 | |
a42c55f0 | 229 | db_query("UPDATE ttrss_feeds SET last_update_started = NOW() |
2c08214a AD |
230 | WHERE id = '$feed'"); |
231 | ||
232 | $auth_login = db_fetch_result($result, 0, "auth_login"); | |
233 | $auth_pass = db_fetch_result($result, 0, "auth_pass"); | |
234 | ||
044cff2d AD |
235 | if ($auth_pass_encrypted) { |
236 | require_once "crypt.php"; | |
237 | $auth_pass = decrypt_string($auth_pass); | |
238 | } | |
239 | ||
2c08214a AD |
240 | $cache_images = sql_bool_to_bool(db_fetch_result($result, 0, "cache_images")); |
241 | $fetch_url = db_fetch_result($result, 0, "feed_url"); | |
242 | ||
a42c55f0 | 243 | $feed = db_escape_string($feed); |
2c08214a | 244 | |
f074ffe9 | 245 | if ($override_url) $fetch_url = $override_url; |
2c08214a | 246 | |
f074ffe9 | 247 | $date_feed_processed = date('Y-m-d H:i'); |
2c08214a | 248 | |
4f9cbdff | 249 | $cache_filename = CACHE_DIR . "/simplepie/" . sha1($fetch_url) . ".feed"; |
f074ffe9 AD |
250 | |
251 | // Ignore cache if new feed or manual update. | |
252 | $cache_age = ($no_cache || is_null($last_updated) || $last_updated == '1970-01-01 00:00:00') ? | |
a42c55f0 | 253 | 30 : get_feed_update_interval($feed) * 60; |
2c08214a AD |
254 | |
255 | if ($debug_enabled) { | |
f074ffe9 AD |
256 | _debug("update_rss_feed: cache filename: $cache_filename exists: " . file_exists($cache_filename)); |
257 | _debug("update_rss_feed: cache age: $cache_age; no cache: $no_cache"); | |
2c08214a AD |
258 | } |
259 | ||
f074ffe9 | 260 | $cached_feed_data_hash = false; |
2c08214a | 261 | |
4f9cbdff AD |
262 | $rss = false; |
263 | $rss_hash = false; | |
7a01dc77 | 264 | $cache_timestamp = file_exists($cache_filename) ? filemtime($cache_filename) : 0; |
31623bfa | 265 | $last_updated_timestamp = strtotime($last_updated); |
4f9cbdff | 266 | |
687a4f59 AD |
267 | $force_refetch = isset($_REQUEST["force_refetch"]); |
268 | ||
f074ffe9 AD |
269 | if (file_exists($cache_filename) && |
270 | is_readable($cache_filename) && | |
271 | !$auth_login && !$auth_pass && | |
272 | filemtime($cache_filename) > time() - $cache_age) { | |
2c08214a | 273 | |
f074ffe9 AD |
274 | if ($debug_enabled) { |
275 | _debug("update_rss_feed: using local cache."); | |
276 | } | |
be574731 | 277 | |
17e74b21 AD |
278 | if ($cache_timestamp > $last_updated_timestamp) { |
279 | @$rss_data = file_get_contents($cache_filename); | |
f074ffe9 | 280 | |
17e74b21 AD |
281 | if ($rss_data) { |
282 | $rss_hash = sha1($rss_data); | |
283 | @$rss = unserialize($rss_data); | |
284 | } | |
687a4f59 | 285 | } else if (!$force_refetch) { |
17e74b21 AD |
286 | if ($debug_enabled) { |
287 | _debug("update_rss_feed: local cache valid and older than last_updated, nothing to do."); | |
288 | } | |
289 | return; | |
4f9cbdff | 290 | } |
f074ffe9 | 291 | } |
017401dd | 292 | |
4f9cbdff | 293 | if (!$rss) { |
017401dd | 294 | |
4f9cbdff AD |
295 | if (!$feed_data) { |
296 | if ($debug_enabled) { | |
31623bfa | 297 | _debug("update_rss_feed: fetching [$fetch_url] (ts: $cache_timestamp/$last_updated_timestamp)"); |
4f9cbdff | 298 | } |
017401dd | 299 | |
4f9cbdff | 300 | $feed_data = fetch_file_contents($fetch_url, false, |
5929a0c1 | 301 | $auth_login, $auth_pass, false, |
8401101d | 302 | $no_cache ? FEED_FETCH_NO_CACHE_TIMEOUT : FEED_FETCH_TIMEOUT, |
5e736e45 | 303 | $force_refetch ? 0 : max($last_updated_timestamp, $cache_timestamp)); |
96f98cb0 | 304 | |
3f6f0857 AD |
305 | global $fetch_curl_used; |
306 | ||
307 | if (!$fetch_curl_used) { | |
308 | $tmp = @gzdecode($feed_data); | |
309 | ||
310 | if ($tmp) $feed_data = $tmp; | |
311 | } | |
312 | ||
4f9cbdff AD |
313 | if ($debug_enabled) { |
314 | _debug("update_rss_feed: fetch done."); | |
315 | } | |
017401dd | 316 | |
4f9cbdff | 317 | } |
017401dd | 318 | |
4f9cbdff AD |
319 | if (!$feed_data) { |
320 | global $fetch_last_error; | |
7a01dc77 | 321 | global $fetch_last_error_code; |
f074ffe9 AD |
322 | |
323 | if ($debug_enabled) { | |
7a01dc77 | 324 | _debug("update_rss_feed: unable to fetch: $fetch_last_error [$fetch_last_error_code]"); |
f074ffe9 AD |
325 | } |
326 | ||
7a01dc77 AD |
327 | $error_escaped = ''; |
328 | ||
329 | // If-Modified-Since | |
330 | if ($fetch_last_error_code != 304) { | |
a42c55f0 | 331 | $error_escaped = db_escape_string($fetch_last_error); |
7a01dc77 AD |
332 | } else { |
333 | if ($debug_enabled) { | |
334 | _debug("update_rss_feed: source claims data not modified, nothing to do."); | |
335 | } | |
336 | } | |
4f9cbdff | 337 | |
6322ac79 | 338 | db_query( |
4f9cbdff AD |
339 | "UPDATE ttrss_feeds SET last_error = '$error_escaped', |
340 | last_updated = NOW() WHERE id = '$feed'"); | |
341 | ||
342 | return; | |
f074ffe9 AD |
343 | } |
344 | } | |
345 | ||
1ffe3391 | 346 | $pluginhost = new PluginHost(); |
be178857 | 347 | $pluginhost->set_debug($debug_enabled); |
a42c55f0 | 348 | $user_plugins = get_pref("_ENABLED_PLUGINS", $owner_uid); |
ab457a9c | 349 | |
1ffe3391 AD |
350 | $pluginhost->load(PLUGINS, PluginHost::KIND_ALL); |
351 | $pluginhost->load($user_plugins, PluginHost::KIND_USER, $owner_uid); | |
ab457a9c | 352 | $pluginhost->load_data(); |
017401dd | 353 | |
1ffe3391 | 354 | foreach ($pluginhost->get_hooks(PluginHost::HOOK_FEED_FETCHED) as $plugin) { |
017401dd AD |
355 | $feed_data = $plugin->hook_feed_fetched($feed_data); |
356 | } | |
357 | ||
4f9cbdff AD |
358 | if (!$rss) { |
359 | $rss = new SimplePie(); | |
360 | $rss->set_sanitize_class("SanitizeDummy"); | |
361 | // simplepie ignores the above and creates default sanitizer anyway, | |
362 | // so let's override it... | |
363 | $rss->sanitize = new SanitizeDummy(); | |
364 | $rss->set_output_encoding('UTF-8'); | |
365 | $rss->set_raw_data($feed_data); | |
366 | $rss->enable_cache(false); | |
367 | ||
368 | @$rss->init(); | |
5de51df7 AD |
369 | } |
370 | ||
2c08214a AD |
371 | // print_r($rss); |
372 | ||
a42c55f0 | 373 | $feed = db_escape_string($feed); |
2c08214a | 374 | |
19b3992b | 375 | if (!$rss->error()) { |
2c08214a | 376 | |
4f9cbdff AD |
377 | // cache data for later |
378 | if (!$auth_pass && !$auth_login && is_writable(CACHE_DIR . "/simplepie")) { | |
379 | $rss_data = serialize($rss); | |
380 | $new_rss_hash = sha1($rss_data); | |
381 | ||
382 | if ($new_rss_hash != $rss_hash) { | |
383 | if ($debug_enabled) { | |
384 | _debug("update_rss_feed: saving $cache_filename"); | |
385 | } | |
386 | @file_put_contents($cache_filename, serialize($rss)); | |
387 | } | |
388 | } | |
389 | ||
d2a421e3 | 390 | // We use local pluginhost here because we need to load different per-user feed plugins |
1ffe3391 | 391 | $pluginhost->run_hooks(PluginHost::HOOK_FEED_PARSED, "hook_feed_parsed", $rss); |
4412b877 | 392 | |
2c08214a AD |
393 | if ($debug_enabled) { |
394 | _debug("update_rss_feed: processing feed data..."); | |
395 | } | |
396 | ||
a42c55f0 | 397 | // db_query("BEGIN"); |
2c08214a | 398 | |
382268c6 AD |
399 | if (DB_TYPE == "pgsql") { |
400 | $favicon_interval_qpart = "favicon_last_checked < NOW() - INTERVAL '12 hour'"; | |
401 | } else { | |
402 | $favicon_interval_qpart = "favicon_last_checked < DATE_SUB(NOW(), INTERVAL 12 HOUR)"; | |
403 | } | |
404 | ||
a42c55f0 | 405 | $result = db_query("SELECT title,site_url,owner_uid, |
382268c6 AD |
406 | (favicon_last_checked IS NULL OR $favicon_interval_qpart) AS |
407 | favicon_needs_check | |
2c08214a AD |
408 | FROM ttrss_feeds WHERE id = '$feed'"); |
409 | ||
410 | $registered_title = db_fetch_result($result, 0, "title"); | |
2c08214a | 411 | $orig_site_url = db_fetch_result($result, 0, "site_url"); |
382268c6 AD |
412 | $favicon_needs_check = sql_bool_to_bool(db_fetch_result($result, 0, |
413 | "favicon_needs_check")); | |
2c08214a AD |
414 | |
415 | $owner_uid = db_fetch_result($result, 0, "owner_uid"); | |
416 | ||
a42c55f0 | 417 | $site_url = db_escape_string(mb_substr(rewrite_relative_url($fetch_url, $rss->get_link()), 0, 245)); |
2c08214a | 418 | |
687a4f59 AD |
419 | if ($favicon_needs_check || $force_refetch) { |
420 | if ($debug_enabled) { | |
421 | _debug("update_rss_feed: checking favicon..."); | |
422 | } | |
423 | ||
424 | check_feed_favicon($site_url, $feed, $link); | |
425 | $favicon_file = ICONS_DIR . "/$feed.ico"; | |
426 | ||
427 | if (file_exists($favicon_file)) { | |
bc7a144d | 428 | require_once "colors.php"; |
687a4f59 | 429 | |
6322ac79 | 430 | $favicon_color = db_escape_string( |
bc7a144d | 431 | calculate_avg_color($favicon_file)); |
63c323f7 | 432 | |
bc7a144d | 433 | $favicon_colorstring = ",favicon_avg_color = '".$favicon_color."'"; |
6ac722d5 | 434 | } |
687a4f59 | 435 | |
a42c55f0 | 436 | db_query("UPDATE ttrss_feeds SET favicon_last_checked = NOW() |
bc7a144d | 437 | $favicon_colorstring |
f2798eb6 AD |
438 | WHERE id = '$feed'"); |
439 | } | |
2c08214a AD |
440 | |
441 | if (!$registered_title || $registered_title == "[Unknown]") { | |
442 | ||
a42c55f0 | 443 | $feed_title = db_escape_string($rss->get_title()); |
2c08214a AD |
444 | |
445 | if ($debug_enabled) { | |
446 | _debug("update_rss_feed: registering title: $feed_title"); | |
447 | } | |
448 | ||
a42c55f0 | 449 | db_query("UPDATE ttrss_feeds SET |
2c08214a AD |
450 | title = '$feed_title' WHERE id = '$feed'"); |
451 | } | |
452 | ||
0cf81637 | 453 | if ($site_url && $orig_site_url != $site_url) { |
a42c55f0 | 454 | db_query("UPDATE ttrss_feeds SET |
2c08214a AD |
455 | site_url = '$site_url' WHERE id = '$feed'"); |
456 | } | |
457 | ||
2c08214a | 458 | if ($debug_enabled) { |
b24504b1 | 459 | _debug("update_rss_feed: loading filters & labels..."); |
2c08214a AD |
460 | } |
461 | ||
a42c55f0 AD |
462 | $filters = load_filters($feed, $owner_uid); |
463 | $labels = get_all_labels($owner_uid); | |
2c08214a | 464 | |
67bd0b1f AD |
465 | if ($debug_enabled) { |
466 | //print_r($filters); | |
467 | _debug("update_rss_feed: " . count($filters) . " filters loaded."); | |
468 | } | |
2c08214a | 469 | |
19b3992b | 470 | $items = $rss->get_items(); |
2c08214a | 471 | |
19b3992b | 472 | if (!is_array($items)) { |
2c08214a | 473 | if ($debug_enabled) { |
19b3992b | 474 | _debug("update_rss_feed: no articles found."); |
2c08214a AD |
475 | } |
476 | ||
a42c55f0 | 477 | db_query("UPDATE ttrss_feeds |
2c08214a AD |
478 | SET last_updated = NOW(), last_error = '' WHERE id = '$feed'"); |
479 | ||
480 | return; // no articles | |
481 | } | |
482 | ||
483 | if ($pubsub_state != 2 && PUBSUBHUBBUB_ENABLED) { | |
484 | ||
485 | if ($debug_enabled) _debug("update_rss_feed: checking for PUSH hub..."); | |
486 | ||
487 | $feed_hub_url = false; | |
2c08214a | 488 | |
19b3992b | 489 | $links = $rss->get_links('hub'); |
2c08214a | 490 | |
19b3992b AD |
491 | if ($links && is_array($links)) { |
492 | foreach ($links as $l) { | |
493 | $feed_hub_url = $l; | |
494 | break; | |
2c08214a AD |
495 | } |
496 | } | |
497 | ||
498 | if ($debug_enabled) _debug("update_rss_feed: feed hub url: $feed_hub_url"); | |
499 | ||
500 | if ($feed_hub_url && function_exists('curl_init') && | |
501 | !ini_get("open_basedir")) { | |
502 | ||
503 | require_once 'lib/pubsubhubbub/subscriber.php'; | |
504 | ||
505 | $callback_url = get_self_url_prefix() . | |
506 | "/public.php?op=pubsub&id=$feed"; | |
507 | ||
508 | $s = new Subscriber($feed_hub_url, $callback_url); | |
509 | ||
510 | $rc = $s->subscribe($fetch_url); | |
511 | ||
512 | if ($debug_enabled) | |
513 | _debug("update_rss_feed: feed hub url found, subscribe request sent."); | |
514 | ||
a42c55f0 | 515 | db_query("UPDATE ttrss_feeds SET pubsub_state = 1 |
2c08214a AD |
516 | WHERE id = '$feed'"); |
517 | } | |
518 | } | |
519 | ||
520 | if ($debug_enabled) { | |
521 | _debug("update_rss_feed: processing articles..."); | |
522 | } | |
523 | ||
19b3992b | 524 | foreach ($items as $item) { |
5d56d100 | 525 | if ($_REQUEST['xdebug'] == 3) { |
2c08214a AD |
526 | print_r($item); |
527 | } | |
528 | ||
19b3992b AD |
529 | $entry_guid = $item->get_id(); |
530 | if (!$entry_guid) $entry_guid = $item->get_link(); | |
531 | if (!$entry_guid) $entry_guid = make_guid_from_title($item->get_title()); | |
2c08214a | 532 | |
2c08214a AD |
533 | if (!$entry_guid) continue; |
534 | ||
3a4c8973 AD |
535 | $entry_guid = "$owner_uid,$entry_guid"; |
536 | ||
a42c55f0 | 537 | $entry_guid_hashed = db_escape_string('SHA1:' . sha1($entry_guid)); |
5e3d5480 AD |
538 | |
539 | if ($debug_enabled) { | |
540 | _debug("update_rss_feed: guid $entry_guid / $entry_guid_hashed"); | |
541 | } | |
542 | ||
2c08214a AD |
543 | $entry_timestamp = ""; |
544 | ||
19b3992b | 545 | $entry_timestamp = strtotime($item->get_date()); |
2c08214a | 546 | |
30123fe6 | 547 | if ($entry_timestamp == -1 || !$entry_timestamp || $entry_timestamp > time()) { |
2c08214a AD |
548 | $entry_timestamp = time(); |
549 | $no_orig_date = 'true'; | |
550 | } else { | |
551 | $no_orig_date = 'false'; | |
552 | } | |
553 | ||
554 | $entry_timestamp_fmt = strftime("%Y/%m/%d %H:%M:%S", $entry_timestamp); | |
555 | ||
556 | if ($debug_enabled) { | |
557 | _debug("update_rss_feed: date $entry_timestamp [$entry_timestamp_fmt]"); | |
558 | } | |
559 | ||
c5569ff8 | 560 | $entry_title = html_entity_decode($item->get_title()); |
1b35d30c | 561 | |
5d56d100 | 562 | $entry_link = rewrite_relative_url($site_url, $item->get_link()); |
2c08214a AD |
563 | |
564 | if ($debug_enabled) { | |
565 | _debug("update_rss_feed: title $entry_title"); | |
566 | _debug("update_rss_feed: link $entry_link"); | |
567 | } | |
568 | ||
569 | if (!$entry_title) $entry_title = date("Y-m-d H:i:s", $entry_timestamp);; | |
570 | ||
19b3992b AD |
571 | $entry_content = $item->get_content(); |
572 | if (!$entry_content) $entry_content = $item->get_description(); | |
2c08214a AD |
573 | |
574 | if ($_REQUEST["xdebug"] == 2) { | |
575 | print "update_rss_feed: content: "; | |
487f0750 | 576 | print $entry_content; |
3c696512 | 577 | print "\n"; |
2c08214a AD |
578 | } |
579 | ||
19b3992b | 580 | $entry_comments = $item->data["comments"]; |
2c08214a | 581 | |
19b3992b AD |
582 | if ($item->get_author()) { |
583 | $entry_author_item = $item->get_author(); | |
584 | $entry_author = $entry_author_item->get_name(); | |
585 | if (!$entry_author) $entry_author = $entry_author_item->get_email(); | |
2c08214a | 586 | |
a42c55f0 | 587 | $entry_author = db_escape_string($entry_author); |
2c08214a AD |
588 | } |
589 | ||
a42c55f0 | 590 | $entry_guid = db_escape_string(mb_substr($entry_guid, 0, 245)); |
2c08214a | 591 | |
a42c55f0 AD |
592 | $entry_comments = db_escape_string(mb_substr($entry_comments, 0, 245)); |
593 | $entry_author = db_escape_string(mb_substr($entry_author, 0, 245)); | |
2c08214a | 594 | |
19b3992b | 595 | $num_comments = $item->get_item_tags('http://purl.org/rss/1.0/modules/slash/', 'comments'); |
83e6e313 | 596 | |
19b3992b AD |
597 | if (is_array($num_comments) && is_array($num_comments[0])) { |
598 | $num_comments = (int) $num_comments[0]["data"]; | |
2c08214a | 599 | } else { |
19b3992b | 600 | $num_comments = 0; |
2c08214a AD |
601 | } |
602 | ||
2c08214a | 603 | if ($debug_enabled) { |
48cb2536 | 604 | _debug("update_rss_feed: author $entry_author"); |
83e6e313 | 605 | _debug("update_rss_feed: num_comments: $num_comments"); |
2c08214a AD |
606 | _debug("update_rss_feed: looking for tags [1]..."); |
607 | } | |
608 | ||
609 | // parse <category> entries into tags | |
610 | ||
611 | $additional_tags = array(); | |
612 | ||
19b3992b | 613 | $additional_tags_src = $item->get_categories(); |
2c08214a | 614 | |
19b3992b AD |
615 | if (is_array($additional_tags_src)) { |
616 | foreach ($additional_tags_src as $tobj) { | |
617 | array_push($additional_tags, $tobj->get_term()); | |
2c08214a | 618 | } |
19b3992b | 619 | } |
2c08214a | 620 | |
19b3992b AD |
621 | if ($debug_enabled) { |
622 | _debug("update_rss_feed: category tags:"); | |
623 | print_r($additional_tags); | |
2c08214a AD |
624 | } |
625 | ||
626 | if ($debug_enabled) { | |
627 | _debug("update_rss_feed: looking for tags [2]..."); | |
628 | } | |
629 | ||
fa6fbd36 | 630 | $entry_tags = array_unique($additional_tags); |
2c08214a AD |
631 | |
632 | for ($i = 0; $i < count($entry_tags); $i++) | |
633 | $entry_tags[$i] = mb_strtolower($entry_tags[$i], 'utf-8'); | |
634 | ||
635 | if ($debug_enabled) { | |
6aff7845 AD |
636 | //_debug("update_rss_feed: unfiltered tags found:"); |
637 | //print_r($entry_tags); | |
2c08214a AD |
638 | } |
639 | ||
2c08214a | 640 | if ($debug_enabled) { |
70caff48 | 641 | _debug("update_rss_feed: done collecting data."); |
2c08214a AD |
642 | } |
643 | ||
cc85704f | 644 | // TODO: less memory-hungry implementation |
19c73507 | 645 | |
19b3992b AD |
646 | if ($debug_enabled) { |
647 | _debug("update_rss_feed: applying plugin filters.."); | |
648 | } | |
cc85704f | 649 | |
e02555c1 | 650 | // FIXME not sure if owner_uid is a good idea here, we may have a base entry without user entry (?) |
a42c55f0 AD |
651 | $result = db_query("SELECT plugin_data,title,content,link,tag_cache,author FROM ttrss_entries, ttrss_user_entries |
652 | WHERE ref_id = id AND (guid = '".db_escape_string($entry_guid)."' OR guid = '$entry_guid_hashed') AND owner_uid = $owner_uid"); | |
b30abdad AD |
653 | |
654 | if (db_num_rows($result) != 0) { | |
655 | $entry_plugin_data = db_fetch_result($result, 0, "plugin_data"); | |
e02555c1 AD |
656 | $stored_article = array("title" => db_fetch_result($result, 0, "title"), |
657 | "content" => db_fetch_result($result, 0, "content"), | |
658 | "link" => db_fetch_result($result, 0, "link"), | |
659 | "tags" => explode(",", db_fetch_result($result, 0, "tag_cache")), | |
660 | "author" => db_fetch_result($result, 0, "author")); | |
b30abdad AD |
661 | } else { |
662 | $entry_plugin_data = ""; | |
e02555c1 | 663 | $stored_article = array(); |
b30abdad AD |
664 | } |
665 | ||
455b1401 | 666 | $article = array("owner_uid" => $owner_uid, // read only |
b30abdad | 667 | "guid" => $entry_guid, // read only |
19b3992b AD |
668 | "title" => $entry_title, |
669 | "content" => $entry_content, | |
670 | "link" => $entry_link, | |
671 | "tags" => $entry_tags, | |
b30abdad | 672 | "plugin_data" => $entry_plugin_data, |
e02555c1 AD |
673 | "author" => $entry_author, |
674 | "stored" => $stored_article); | |
cc85704f | 675 | |
1ffe3391 | 676 | foreach ($pluginhost->get_hooks(PluginHost::HOOK_ARTICLE_FILTER) as $plugin) { |
19b3992b | 677 | $article = $plugin->hook_article_filter($article); |
cc85704f AD |
678 | } |
679 | ||
19b3992b | 680 | $entry_tags = $article["tags"]; |
a42c55f0 AD |
681 | $entry_guid = db_escape_string($entry_guid); |
682 | $entry_title = db_escape_string($article["title"]); | |
683 | $entry_author = db_escape_string($article["author"]); | |
684 | $entry_link = db_escape_string($article["link"]); | |
685 | $entry_plugin_data = db_escape_string($article["plugin_data"]); | |
f935d98e AD |
686 | $entry_content = $article["content"]; // escaped below |
687 | ||
b30abdad AD |
688 | |
689 | if ($debug_enabled) { | |
690 | _debug("update_rss_feed: plugin data: $entry_plugin_data"); | |
691 | } | |
2bbd6994 | 692 | |
0a3fd79b | 693 | if ($cache_images && is_writable(CACHE_DIR . '/images')) |
f0bd8e65 | 694 | cache_images($entry_content, $site_url, $debug_enabled); |
0a3fd79b | 695 | |
a42c55f0 | 696 | $entry_content = db_escape_string($entry_content, false); |
1f45c857 | 697 | |
9e222305 | 698 | $content_hash = "SHA1:" . sha1($entry_content); |
cc85704f | 699 | |
a42c55f0 | 700 | db_query("BEGIN"); |
2c08214a | 701 | |
a42c55f0 | 702 | $result = db_query("SELECT id FROM ttrss_entries |
5e3d5480 | 703 | WHERE (guid = '$entry_guid' OR guid = '$entry_guid_hashed')"); |
9e222305 | 704 | |
2c08214a AD |
705 | if (db_num_rows($result) == 0) { |
706 | ||
707 | if ($debug_enabled) { | |
9e222305 | 708 | _debug("update_rss_feed: base guid [$entry_guid] not found"); |
2c08214a AD |
709 | } |
710 | ||
711 | // base post entry does not exist, create it | |
712 | ||
6322ac79 | 713 | $result = db_query( |
2c08214a AD |
714 | "INSERT INTO ttrss_entries |
715 | (title, | |
716 | guid, | |
717 | link, | |
718 | updated, | |
719 | content, | |
720 | content_hash, | |
87764a50 | 721 | cached_content, |
2c08214a AD |
722 | no_orig_date, |
723 | date_updated, | |
724 | date_entered, | |
725 | comments, | |
726 | num_comments, | |
b30abdad | 727 | plugin_data, |
2c08214a AD |
728 | author) |
729 | VALUES | |
730 | ('$entry_title', | |
5e3d5480 | 731 | '$entry_guid_hashed', |
2c08214a AD |
732 | '$entry_link', |
733 | '$entry_timestamp_fmt', | |
734 | '$entry_content', | |
735 | '$content_hash', | |
0a3fd79b | 736 | '', |
2c08214a AD |
737 | $no_orig_date, |
738 | NOW(), | |
be574731 | 739 | '$date_feed_processed', |
2c08214a AD |
740 | '$entry_comments', |
741 | '$num_comments', | |
b30abdad | 742 | '$entry_plugin_data', |
2c08214a | 743 | '$entry_author')"); |
e8291805 AD |
744 | |
745 | $article_labels = array(); | |
746 | ||
2c08214a AD |
747 | } else { |
748 | // we keep encountering the entry in feeds, so we need to | |
749 | // update date_updated column so that we don't get horrible | |
750 | // dupes when the entry gets purged and reinserted again e.g. | |
751 | // in the case of SLOW SLOW OMG SLOW updating feeds | |
752 | ||
753 | $base_entry_id = db_fetch_result($result, 0, "id"); | |
754 | ||
a42c55f0 | 755 | db_query("UPDATE ttrss_entries SET date_updated = NOW() |
2c08214a | 756 | WHERE id = '$base_entry_id'"); |
e8291805 | 757 | |
a42c55f0 | 758 | $article_labels = get_article_labels($base_entry_id, $owner_uid); |
2c08214a AD |
759 | } |
760 | ||
761 | // now it should exist, if not - bad luck then | |
762 | ||
a42c55f0 | 763 | $result = db_query("SELECT |
5e3d5480 | 764 | id,content_hash,no_orig_date,title,plugin_data,guid, |
2c08214a AD |
765 | ".SUBSTRING_FOR_DATE."(date_updated,1,19) as date_updated, |
766 | ".SUBSTRING_FOR_DATE."(updated,1,19) as updated, | |
0a3fd79b | 767 | num_comments |
2c08214a AD |
768 | FROM |
769 | ttrss_entries | |
5e3d5480 | 770 | WHERE guid = '$entry_guid' OR guid = '$entry_guid_hashed'"); |
2c08214a AD |
771 | |
772 | $entry_ref_id = 0; | |
773 | $entry_int_id = 0; | |
774 | ||
775 | if (db_num_rows($result) == 1) { | |
776 | ||
777 | if ($debug_enabled) { | |
5e3d5480 | 778 | _debug("update_rss_feed: base guid found, checking for user record"); |
2c08214a AD |
779 | } |
780 | ||
781 | // this will be used below in update handler | |
782 | $orig_content_hash = db_fetch_result($result, 0, "content_hash"); | |
783 | $orig_title = db_fetch_result($result, 0, "title"); | |
784 | $orig_num_comments = db_fetch_result($result, 0, "num_comments"); | |
785 | $orig_date_updated = strtotime(db_fetch_result($result, | |
786 | 0, "date_updated")); | |
b30abdad | 787 | $orig_plugin_data = db_fetch_result($result, 0, "plugin_data"); |
2c08214a AD |
788 | |
789 | $ref_id = db_fetch_result($result, 0, "id"); | |
790 | $entry_ref_id = $ref_id; | |
791 | ||
5e3d5480 AD |
792 | /* $stored_guid = db_fetch_result($result, 0, "guid"); |
793 | if ($stored_guid != $entry_guid_hashed) { | |
794 | if ($debug_enabled) _debug("upgrading compat guid to hashed one"); | |
795 | ||
a42c55f0 | 796 | db_query("UPDATE ttrss_entries SET guid = '$entry_guid_hashed' WHERE |
5e3d5480 AD |
797 | id = '$ref_id'"); |
798 | } */ | |
799 | ||
2c08214a AD |
800 | // check for user post link to main table |
801 | ||
802 | // do we allow duplicate posts with same GUID in different feeds? | |
a42c55f0 | 803 | if (get_pref("ALLOW_DUPLICATE_POSTS", $owner_uid, false)) { |
2c08214a AD |
804 | $dupcheck_qpart = "AND (feed_id = '$feed' OR feed_id IS NULL)"; |
805 | } else { | |
806 | $dupcheck_qpart = ""; | |
807 | } | |
808 | ||
809 | /* Collect article tags here so we could filter by them: */ | |
810 | ||
811 | $article_filters = get_article_filters($filters, $entry_title, | |
4021d61a | 812 | $entry_content, $entry_link, $entry_timestamp, $entry_author, |
2c08214a AD |
813 | $entry_tags); |
814 | ||
815 | if ($debug_enabled) { | |
816 | _debug("update_rss_feed: article filters: "); | |
817 | if (count($article_filters) != 0) { | |
818 | print_r($article_filters); | |
819 | } | |
820 | } | |
821 | ||
822 | if (find_article_filter($article_filters, "filter")) { | |
a42c55f0 | 823 | db_query("COMMIT"); // close transaction in progress |
2c08214a AD |
824 | continue; |
825 | } | |
826 | ||
827 | $score = calculate_article_score($article_filters); | |
828 | ||
829 | if ($debug_enabled) { | |
830 | _debug("update_rss_feed: initial score: $score"); | |
831 | } | |
832 | ||
833 | $query = "SELECT ref_id, int_id FROM ttrss_user_entries WHERE | |
834 | ref_id = '$ref_id' AND owner_uid = '$owner_uid' | |
835 | $dupcheck_qpart"; | |
836 | ||
837 | // if ($_REQUEST["xdebug"]) print "$query\n"; | |
838 | ||
a42c55f0 | 839 | $result = db_query($query); |
2c08214a AD |
840 | |
841 | // okay it doesn't exist - create user entry | |
842 | if (db_num_rows($result) == 0) { | |
843 | ||
844 | if ($debug_enabled) { | |
845 | _debug("update_rss_feed: user record not found, creating..."); | |
846 | } | |
847 | ||
848 | if ($score >= -500 && !find_article_filter($article_filters, 'catchup')) { | |
849 | $unread = 'true'; | |
850 | $last_read_qpart = 'NULL'; | |
851 | } else { | |
852 | $unread = 'false'; | |
853 | $last_read_qpart = 'NOW()'; | |
854 | } | |
855 | ||
856 | if (find_article_filter($article_filters, 'mark') || $score > 1000) { | |
857 | $marked = 'true'; | |
858 | } else { | |
859 | $marked = 'false'; | |
860 | } | |
861 | ||
862 | if (find_article_filter($article_filters, 'publish')) { | |
863 | $published = 'true'; | |
864 | } else { | |
865 | $published = 'false'; | |
866 | } | |
867 | ||
2ea9bbfd AD |
868 | // N-grams |
869 | ||
870 | if (DB_TYPE == "pgsql" and defined('_NGRAM_TITLE_DUPLICATE_THRESHOLD')) { | |
871 | ||
a42c55f0 | 872 | $result = db_query("SELECT COUNT(*) AS similar FROM |
2ea9bbfd AD |
873 | ttrss_entries,ttrss_user_entries |
874 | WHERE ref_id = id AND updated >= NOW() - INTERVAL '7 day' | |
875 | AND similarity(title, '$entry_title') >= "._NGRAM_TITLE_DUPLICATE_THRESHOLD." | |
876 | AND owner_uid = $owner_uid"); | |
877 | ||
878 | $ngram_similar = db_fetch_result($result, 0, "similar"); | |
879 | ||
880 | if ($debug_enabled) { | |
881 | _debug("update_rss_feed: N-gram similar results: $ngram_similar"); | |
882 | } | |
883 | ||
884 | if ($ngram_similar > 0) { | |
885 | $unread = 'false'; | |
886 | } | |
887 | } | |
888 | ||
7873d588 AD |
889 | $last_marked = ($marked == 'true') ? 'NOW()' : 'NULL'; |
890 | $last_published = ($published == 'true') ? 'NOW()' : 'NULL'; | |
891 | ||
6322ac79 | 892 | $result = db_query( |
2c08214a AD |
893 | "INSERT INTO ttrss_user_entries |
894 | (ref_id, owner_uid, feed_id, unread, last_read, marked, | |
7873d588 AD |
895 | published, score, tag_cache, label_cache, uuid, |
896 | last_marked, last_published) | |
2c08214a | 897 | VALUES ('$ref_id', '$owner_uid', '$feed', $unread, |
7873d588 AD |
898 | $last_read_qpart, $marked, $published, '$score', '', '', |
899 | '', $last_marked, $last_published)"); | |
2c08214a AD |
900 | |
901 | if (PUBSUBHUBBUB_HUB && $published == 'true') { | |
902 | $rss_link = get_self_url_prefix() . | |
903 | "/public.php?op=rss&id=-2&key=" . | |
a42c55f0 | 904 | get_feed_access_key(-2, false, $owner_uid); |
2c08214a AD |
905 | |
906 | $p = new Publisher(PUBSUBHUBBUB_HUB); | |
907 | ||
908 | $pubsub_result = $p->publish_update($rss_link); | |
909 | } | |
910 | ||
6322ac79 | 911 | $result = db_query( |
2c08214a AD |
912 | "SELECT int_id FROM ttrss_user_entries WHERE |
913 | ref_id = '$ref_id' AND owner_uid = '$owner_uid' AND | |
914 | feed_id = '$feed' LIMIT 1"); | |
915 | ||
916 | if (db_num_rows($result) == 1) { | |
917 | $entry_int_id = db_fetch_result($result, 0, "int_id"); | |
918 | } | |
919 | } else { | |
920 | if ($debug_enabled) { | |
921 | _debug("update_rss_feed: user record FOUND"); | |
922 | } | |
923 | ||
924 | $entry_ref_id = db_fetch_result($result, 0, "ref_id"); | |
925 | $entry_int_id = db_fetch_result($result, 0, "int_id"); | |
926 | } | |
927 | ||
928 | if ($debug_enabled) { | |
929 | _debug("update_rss_feed: RID: $entry_ref_id, IID: $entry_int_id"); | |
930 | } | |
931 | ||
932 | $post_needs_update = false; | |
933 | $update_insignificant = false; | |
934 | ||
935 | if ($orig_num_comments != $num_comments) { | |
936 | $post_needs_update = true; | |
937 | $update_insignificant = true; | |
938 | } | |
939 | ||
b30abdad AD |
940 | if ($entry_plugin_data != $orig_plugin_data) { |
941 | $post_needs_update = true; | |
942 | $update_insignificant = true; | |
943 | } | |
944 | ||
2c08214a AD |
945 | if ($content_hash != $orig_content_hash) { |
946 | $post_needs_update = true; | |
947 | $update_insignificant = false; | |
948 | } | |
949 | ||
a42c55f0 | 950 | if (db_escape_string($orig_title) != $entry_title) { |
2c08214a AD |
951 | $post_needs_update = true; |
952 | $update_insignificant = false; | |
953 | } | |
954 | ||
955 | // if post needs update, update it and mark all user entries | |
956 | // linking to this post as updated | |
957 | if ($post_needs_update) { | |
958 | ||
959 | if (defined('DAEMON_EXTENDED_DEBUG')) { | |
5e3d5480 | 960 | _debug("update_rss_feed: post $entry_guid_hashed needs update..."); |
2c08214a AD |
961 | } |
962 | ||
963 | // print "<!-- post $orig_title needs update : $post_needs_update -->"; | |
964 | ||
a42c55f0 | 965 | db_query("UPDATE ttrss_entries |
2c08214a AD |
966 | SET title = '$entry_title', content = '$entry_content', |
967 | content_hash = '$content_hash', | |
968 | updated = '$entry_timestamp_fmt', | |
b30abdad AD |
969 | num_comments = '$num_comments', |
970 | plugin_data = '$entry_plugin_data' | |
2c08214a AD |
971 | WHERE id = '$ref_id'"); |
972 | ||
973 | if (!$update_insignificant) { | |
974 | if ($mark_unread_on_update) { | |
a42c55f0 | 975 | db_query("UPDATE ttrss_user_entries |
2c08214a | 976 | SET last_read = null, unread = true WHERE ref_id = '$ref_id'"); |
2c08214a AD |
977 | } |
978 | } | |
979 | } | |
980 | } | |
981 | ||
a42c55f0 | 982 | db_query("COMMIT"); |
2c08214a AD |
983 | |
984 | if ($debug_enabled) { | |
985 | _debug("update_rss_feed: assigning labels..."); | |
986 | } | |
987 | ||
a42c55f0 | 988 | assign_article_to_label_filters($entry_ref_id, $article_filters, |
b24504b1 | 989 | $owner_uid, $article_labels); |
2c08214a AD |
990 | |
991 | if ($debug_enabled) { | |
992 | _debug("update_rss_feed: looking for enclosures..."); | |
993 | } | |
994 | ||
995 | // enclosures | |
996 | ||
997 | $enclosures = array(); | |
998 | ||
19b3992b | 999 | $encs = $item->get_enclosures(); |
2c08214a | 1000 | |
19b3992b AD |
1001 | if (is_array($encs)) { |
1002 | foreach ($encs as $e) { | |
1003 | $e_item = array( | |
1004 | $e->link, $e->type, $e->length); | |
2c08214a | 1005 | array_push($enclosures, $e_item); |
2c08214a AD |
1006 | } |
1007 | } | |
1008 | ||
2c08214a AD |
1009 | if ($debug_enabled) { |
1010 | _debug("update_rss_feed: article enclosures:"); | |
1011 | print_r($enclosures); | |
1012 | } | |
1013 | ||
a42c55f0 | 1014 | db_query("BEGIN"); |
2c08214a AD |
1015 | |
1016 | foreach ($enclosures as $enc) { | |
a42c55f0 AD |
1017 | $enc_url = db_escape_string($enc[0]); |
1018 | $enc_type = db_escape_string($enc[1]); | |
1019 | $enc_dur = db_escape_string($enc[2]); | |
2c08214a | 1020 | |
a42c55f0 | 1021 | $result = db_query("SELECT id FROM ttrss_enclosures |
2c08214a AD |
1022 | WHERE content_url = '$enc_url' AND post_id = '$entry_ref_id'"); |
1023 | ||
1024 | if (db_num_rows($result) == 0) { | |
a42c55f0 | 1025 | db_query("INSERT INTO ttrss_enclosures |
2c08214a AD |
1026 | (content_url, content_type, title, duration, post_id) VALUES |
1027 | ('$enc_url', '$enc_type', '', '$enc_dur', '$entry_ref_id')"); | |
1028 | } | |
1029 | } | |
1030 | ||
a42c55f0 | 1031 | db_query("COMMIT"); |
2c08214a AD |
1032 | |
1033 | // check for manual tags (we have to do it here since they're loaded from filters) | |
1034 | ||
1035 | foreach ($article_filters as $f) { | |
6aff7845 | 1036 | if ($f["type"] == "tag") { |
2c08214a | 1037 | |
6aff7845 | 1038 | $manual_tags = trim_array(explode(",", $f["param"])); |
2c08214a AD |
1039 | |
1040 | foreach ($manual_tags as $tag) { | |
1041 | if (tag_is_valid($tag)) { | |
1042 | array_push($entry_tags, $tag); | |
1043 | } | |
1044 | } | |
1045 | } | |
1046 | } | |
1047 | ||
1048 | // Skip boring tags | |
1049 | ||
6322ac79 | 1050 | $boring_tags = trim_array(explode(",", mb_strtolower(get_pref( |
2c08214a AD |
1051 | 'BLACKLISTED_TAGS', $owner_uid, ''), 'utf-8'))); |
1052 | ||
1053 | $filtered_tags = array(); | |
1054 | $tags_to_cache = array(); | |
1055 | ||
1056 | if ($entry_tags && is_array($entry_tags)) { | |
1057 | foreach ($entry_tags as $tag) { | |
1058 | if (array_search($tag, $boring_tags) === false) { | |
1059 | array_push($filtered_tags, $tag); | |
1060 | } | |
1061 | } | |
1062 | } | |
1063 | ||
1064 | $filtered_tags = array_unique($filtered_tags); | |
1065 | ||
1066 | if ($debug_enabled) { | |
1067 | _debug("update_rss_feed: filtered article tags:"); | |
1068 | print_r($filtered_tags); | |
1069 | } | |
1070 | ||
1071 | // Save article tags in the database | |
1072 | ||
1073 | if (count($filtered_tags) > 0) { | |
1074 | ||
a42c55f0 | 1075 | db_query("BEGIN"); |
2c08214a AD |
1076 | |
1077 | foreach ($filtered_tags as $tag) { | |
1078 | ||
1079 | $tag = sanitize_tag($tag); | |
a42c55f0 | 1080 | $tag = db_escape_string($tag); |
2c08214a AD |
1081 | |
1082 | if (!tag_is_valid($tag)) continue; | |
1083 | ||
a42c55f0 | 1084 | $result = db_query("SELECT id FROM ttrss_tags |
2c08214a AD |
1085 | WHERE tag_name = '$tag' AND post_int_id = '$entry_int_id' AND |
1086 | owner_uid = '$owner_uid' LIMIT 1"); | |
1087 | ||
1088 | if ($result && db_num_rows($result) == 0) { | |
1089 | ||
a42c55f0 | 1090 | db_query("INSERT INTO ttrss_tags |
2c08214a AD |
1091 | (owner_uid,tag_name,post_int_id) |
1092 | VALUES ('$owner_uid','$tag', '$entry_int_id')"); | |
1093 | } | |
1094 | ||
1095 | array_push($tags_to_cache, $tag); | |
1096 | } | |
1097 | ||
1098 | /* update the cache */ | |
1099 | ||
1100 | $tags_to_cache = array_unique($tags_to_cache); | |
1101 | ||
a42c55f0 | 1102 | $tags_str = db_escape_string(join(",", $tags_to_cache)); |
2c08214a | 1103 | |
a42c55f0 | 1104 | db_query("UPDATE ttrss_user_entries |
2c08214a AD |
1105 | SET tag_cache = '$tags_str' WHERE ref_id = '$entry_ref_id' |
1106 | AND owner_uid = $owner_uid"); | |
1107 | ||
a42c55f0 | 1108 | db_query("COMMIT"); |
2c08214a AD |
1109 | } |
1110 | ||
a42c55f0 | 1111 | if (get_pref("AUTO_ASSIGN_LABELS", $owner_uid, false)) { |
b24504b1 AD |
1112 | if ($debug_enabled) { |
1113 | _debug("update_rss_feed: auto-assigning labels..."); | |
1114 | } | |
1115 | ||
1116 | foreach ($labels as $label) { | |
227d5e41 | 1117 | $caption = preg_quote($label["caption"]); |
b24504b1 | 1118 | |
9811276d | 1119 | if ($caption && preg_match("/\b$caption\b/i", "$tags_str " . strip_tags($entry_content) . " $entry_title")) { |
b24504b1 | 1120 | if (!labels_contains_caption($article_labels, $caption)) { |
a42c55f0 | 1121 | label_add_article($entry_ref_id, $caption, $owner_uid); |
b24504b1 AD |
1122 | } |
1123 | } | |
1124 | } | |
1125 | } | |
1126 | ||
2c08214a AD |
1127 | if ($debug_enabled) { |
1128 | _debug("update_rss_feed: article processed"); | |
1129 | } | |
1130 | } | |
1131 | ||
1132 | if (!$last_updated) { | |
1133 | if ($debug_enabled) { | |
1134 | _debug("update_rss_feed: new feed, catching it up..."); | |
1135 | } | |
a42c55f0 | 1136 | catchup_feed($feed, false, $owner_uid); |
2c08214a AD |
1137 | } |
1138 | ||
1139 | if ($debug_enabled) { | |
1140 | _debug("purging feed..."); | |
1141 | } | |
1142 | ||
a42c55f0 | 1143 | purge_feed($feed, 0, $debug_enabled); |
2c08214a | 1144 | |
a42c55f0 | 1145 | db_query("UPDATE ttrss_feeds |
f2798eb6 | 1146 | SET last_updated = NOW(), last_error = '' WHERE id = '$feed'"); |
2c08214a | 1147 | |
a42c55f0 | 1148 | // db_query("COMMIT"); |
2c08214a AD |
1149 | |
1150 | } else { | |
1151 | ||
a42c55f0 | 1152 | $error_msg = db_escape_string(mb_substr($rss->error(), 0, 245)); |
2c08214a AD |
1153 | |
1154 | if ($debug_enabled) { | |
1155 | _debug("update_rss_feed: error fetching feed: $error_msg"); | |
1156 | } | |
1157 | ||
6322ac79 | 1158 | db_query( |
2c08214a AD |
1159 | "UPDATE ttrss_feeds SET last_error = '$error_msg', |
1160 | last_updated = NOW() WHERE id = '$feed'"); | |
1161 | } | |
1162 | ||
19b3992b | 1163 | unset($rss); |
2c08214a AD |
1164 | |
1165 | if ($debug_enabled) { | |
1166 | _debug("update_rss_feed: done"); | |
1167 | } | |
1168 | ||
1169 | } | |
1170 | ||
3c696512 AD |
1171 | function cache_images($html, $site_url, $debug) { |
1172 | $cache_dir = CACHE_DIR . "/images"; | |
1173 | ||
1174 | libxml_use_internal_errors(true); | |
1175 | ||
1176 | $charset_hack = '<head> | |
1177 | <meta http-equiv="Content-Type" content="text/html; charset=utf-8"/> | |
1178 | </head>'; | |
1179 | ||
1180 | $doc = new DOMDocument(); | |
1181 | $doc->loadHTML($charset_hack . $html); | |
1182 | $xpath = new DOMXPath($doc); | |
1183 | ||
1184 | $entries = $xpath->query('(//img[@src])'); | |
1185 | ||
1186 | foreach ($entries as $entry) { | |
1187 | if ($entry->hasAttribute('src')) { | |
1188 | $src = rewrite_relative_url($site_url, $entry->getAttribute('src')); | |
1189 | ||
1190 | $local_filename = CACHE_DIR . "/images/" . sha1($src) . ".png"; | |
1191 | ||
1192 | if ($debug) _debug("cache_images: downloading: $src to $local_filename"); | |
1193 | ||
1194 | if (!file_exists($local_filename)) { | |
1195 | $file_content = fetch_file_contents($src); | |
1196 | ||
b8379e69 | 1197 | if ($file_content && strlen($file_content) > 1024) { |
3c696512 AD |
1198 | file_put_contents($local_filename, $file_content); |
1199 | } | |
1200 | } | |
1201 | ||
1202 | if (file_exists($local_filename)) { | |
1203 | $entry->setAttribute('src', SELF_URL_PATH . '/image.php?url=' . | |
487f0750 | 1204 | base64_encode($src)); |
3c696512 AD |
1205 | } |
1206 | } | |
1207 | } | |
1208 | ||
1209 | $node = $doc->getElementsByTagName('body')->item(0); | |
1210 | ||
cc38c8e5 | 1211 | return $doc->saveXML($node); |
3c696512 AD |
1212 | } |
1213 | ||
a42c55f0 | 1214 | function expire_error_log($debug) { |
e2261e17 AD |
1215 | if ($debug) _debug("Removing old error log entries..."); |
1216 | ||
1217 | if (DB_TYPE == "pgsql") { | |
a42c55f0 | 1218 | db_query("DELETE FROM ttrss_error_log |
e2261e17 AD |
1219 | WHERE created_at < NOW() - INTERVAL '7 days'"); |
1220 | } else { | |
a42c55f0 | 1221 | db_query("DELETE FROM ttrss_error_log |
e2261e17 AD |
1222 | WHERE created_at < DATE_SUB(NOW(), INTERVAL 7 DAY)"); |
1223 | } | |
1224 | ||
1225 | } | |
1226 | ||
2a91b6ff AD |
1227 | function expire_lock_files($debug) { |
1228 | if ($debug) _debug("Removing old lock files..."); | |
1229 | ||
1230 | $num_deleted = 0; | |
1231 | ||
1232 | if (is_writable(LOCK_DIRECTORY)) { | |
1233 | $files = glob(LOCK_DIRECTORY . "/*.lock"); | |
1234 | ||
1235 | if ($files) { | |
1236 | foreach ($files as $file) { | |
1237 | if (!file_is_locked($file) && time() - filemtime($file) > 86400*2) { | |
1238 | unlink($file); | |
1239 | ++$num_deleted; | |
1240 | } | |
1241 | } | |
1242 | } | |
1243 | } | |
1244 | ||
1245 | if ($debug) _debug("Removed $num_deleted files."); | |
1246 | } | |
1247 | ||
3c696512 | 1248 | function expire_cached_files($debug) { |
3306daec | 1249 | foreach (array("simplepie", "images", "export", "upload") as $dir) { |
3c696512 | 1250 | $cache_dir = CACHE_DIR . "/$dir"; |
2c08214a | 1251 | |
3c696512 | 1252 | if ($debug) _debug("Expiring $cache_dir"); |
2c08214a | 1253 | |
3c696512 AD |
1254 | $num_deleted = 0; |
1255 | ||
1256 | if (is_writable($cache_dir)) { | |
1257 | $files = glob("$cache_dir/*"); | |
1258 | ||
2a91b6ff | 1259 | if ($files) { |
2ab20c31 AD |
1260 | foreach ($files as $file) { |
1261 | if (time() - filemtime($file) > 86400*7) { | |
1262 | unlink($file); | |
3c696512 | 1263 | |
2ab20c31 AD |
1264 | ++$num_deleted; |
1265 | } | |
3c696512 AD |
1266 | } |
1267 | } | |
2a91b6ff | 1268 | } |
3c696512 AD |
1269 | |
1270 | if ($debug) _debug("Removed $num_deleted files."); | |
1271 | } | |
1272 | } | |
2c08214a | 1273 | |
a3e0bdcf AD |
1274 | /** |
1275 | * Source: http://www.php.net/manual/en/function.parse-url.php#104527 | |
1276 | * Returns the url query as associative array | |
1277 | * | |
1278 | * @param string query | |
1279 | * @return array params | |
1280 | */ | |
1281 | function convertUrlQuery($query) { | |
1282 | $queryParts = explode('&', $query); | |
1283 | ||
1284 | $params = array(); | |
1285 | ||
1286 | foreach ($queryParts as $param) { | |
1287 | $item = explode('=', $param); | |
1288 | $params[$item[0]] = $item[1]; | |
1289 | } | |
1290 | ||
1291 | return $params; | |
1292 | } | |
92c14e9d | 1293 | |
48cb2536 | 1294 | function get_article_filters($filters, $title, $content, $link, $timestamp, $author, $tags) { |
92c14e9d AD |
1295 | $matches = array(); |
1296 | ||
1297 | foreach ($filters as $filter) { | |
1298 | $match_any_rule = $filter["match_any_rule"]; | |
a3a896a1 | 1299 | $inverse = $filter["inverse"]; |
92c14e9d AD |
1300 | $filter_match = false; |
1301 | ||
1302 | foreach ($filter["rules"] as $rule) { | |
1303 | $match = false; | |
1304 | $reg_exp = $rule["reg_exp"]; | |
a3a896a1 | 1305 | $rule_inverse = $rule["inverse"]; |
92c14e9d AD |
1306 | |
1307 | if (!$reg_exp) | |
1308 | continue; | |
1309 | ||
1310 | switch ($rule["type"]) { | |
1311 | case "title": | |
1312 | $match = @preg_match("/$reg_exp/i", $title); | |
1313 | break; | |
1314 | case "content": | |
d03ae73e AD |
1315 | // we don't need to deal with multiline regexps |
1316 | $content = preg_replace("/[\r\n\t]/", "", $content); | |
1317 | ||
92c14e9d AD |
1318 | $match = @preg_match("/$reg_exp/i", $content); |
1319 | break; | |
1320 | case "both": | |
d03ae73e AD |
1321 | // we don't need to deal with multiline regexps |
1322 | $content = preg_replace("/[\r\n\t]/", "", $content); | |
1323 | ||
72d1d067 | 1324 | $match = (@preg_match("/$reg_exp/i", $title) || @preg_match("/$reg_exp/i", $content)); |
92c14e9d AD |
1325 | break; |
1326 | case "link": | |
1327 | $match = @preg_match("/$reg_exp/i", $link); | |
1328 | break; | |
1329 | case "author": | |
1330 | $match = @preg_match("/$reg_exp/i", $author); | |
1331 | break; | |
1332 | case "tag": | |
1333 | $tag_string = join(",", $tags); | |
1334 | $match = @preg_match("/$reg_exp/i", $tag_string); | |
1335 | break; | |
1336 | } | |
1337 | ||
a3a896a1 AD |
1338 | if ($rule_inverse) $match = !$match; |
1339 | ||
92c14e9d AD |
1340 | if ($match_any_rule) { |
1341 | if ($match) { | |
1342 | $filter_match = true; | |
1343 | break; | |
1344 | } | |
1345 | } else { | |
1346 | $filter_match = $match; | |
1347 | if (!$match) { | |
1348 | break; | |
1349 | } | |
1350 | } | |
1351 | } | |
1352 | ||
a3a896a1 AD |
1353 | if ($inverse) $filter_match = !$filter_match; |
1354 | ||
92c14e9d AD |
1355 | if ($filter_match) { |
1356 | foreach ($filter["actions"] AS $action) { | |
1357 | array_push($matches, $action); | |
5e736e45 AD |
1358 | |
1359 | // if Stop action encountered, perform no further processing | |
1360 | if ($action["type"] == "stop") return $matches; | |
92c14e9d AD |
1361 | } |
1362 | } | |
1363 | } | |
1364 | ||
1365 | return $matches; | |
1366 | } | |
1367 | ||
1368 | function find_article_filter($filters, $filter_name) { | |
1369 | foreach ($filters as $f) { | |
1370 | if ($f["type"] == $filter_name) { | |
1371 | return $f; | |
1372 | }; | |
1373 | } | |
1374 | return false; | |
1375 | } | |
1376 | ||
1377 | function find_article_filters($filters, $filter_name) { | |
1378 | $results = array(); | |
1379 | ||
1380 | foreach ($filters as $f) { | |
1381 | if ($f["type"] == $filter_name) { | |
1382 | array_push($results, $f); | |
1383 | }; | |
1384 | } | |
1385 | return $results; | |
1386 | } | |
1387 | ||
1388 | function calculate_article_score($filters) { | |
1389 | $score = 0; | |
1390 | ||
1391 | foreach ($filters as $f) { | |
1392 | if ($f["type"] == "score") { | |
1393 | $score += $f["param"]; | |
1394 | }; | |
1395 | } | |
1396 | return $score; | |
1397 | } | |
1398 | ||
b24504b1 AD |
1399 | function labels_contains_caption($labels, $caption) { |
1400 | foreach ($labels as $label) { | |
1401 | if ($label[1] == $caption) { | |
1402 | return true; | |
1403 | } | |
1404 | } | |
1405 | ||
1406 | return false; | |
1407 | } | |
1408 | ||
a42c55f0 | 1409 | function assign_article_to_label_filters($id, $filters, $owner_uid, $article_labels) { |
92c14e9d AD |
1410 | foreach ($filters as $f) { |
1411 | if ($f["type"] == "label") { | |
b24504b1 | 1412 | if (!labels_contains_caption($article_labels, $f["param"])) { |
a42c55f0 | 1413 | label_add_article($id, $f["param"], $owner_uid); |
b24504b1 AD |
1414 | } |
1415 | } | |
92c14e9d AD |
1416 | } |
1417 | } | |
87764a50 | 1418 | |
87d7e850 AD |
1419 | function make_guid_from_title($title) { |
1420 | return preg_replace("/[ \"\',.:;]/", "-", | |
1421 | mb_strtolower(strip_tags($title), 'utf-8')); | |
1422 | } | |
1423 | ||
1424 | ||
2c08214a | 1425 | ?> |