]>
Commit | Line | Data |
---|---|---|
2c08214a | 1 | <?php |
fce451a4 AD |
2 | define_default('DAEMON_UPDATE_LOGIN_LIMIT', 30); |
3 | define_default('DAEMON_FEED_LIMIT', 500); | |
e2cf81e2 | 4 | define_default('DAEMON_SLEEP_INTERVAL', 120); |
41bead9b | 5 | define_default('_MIN_CACHE_FILE_SIZE', 1024); |
09e8bdfd | 6 | |
a230bf88 AD |
7 | // TODO: this needs to be removed from global namespace into classes/RSS.php or something |
8 | ||
eb16bd9f | 9 | function calculate_article_hash($article, $pluginhost) { |
af244f92 AD |
10 | $tmp = ""; |
11 | ||
12 | foreach ($article as $k => $v) { | |
13 | if ($k != "feed" && isset($v)) { | |
24e6ff5d AD |
14 | $x = strip_tags(is_array($v) ? implode(",", $v) : $v); |
15 | ||
16 | //_debug("$k:" . sha1($x) . ":" . htmlspecialchars($x), true); | |
17 | ||
18 | $tmp .= sha1("$k:" . sha1($x)); | |
af244f92 AD |
19 | } |
20 | } | |
21 | ||
eb16bd9f | 22 | return sha1(implode(",", $pluginhost->get_plugin_names()) . $tmp); |
b1840673 AD |
23 | } |
24 | ||
6322ac79 | 25 | function update_feedbrowser_cache() { |
79178062 | 26 | |
a42c55f0 | 27 | $result = db_query("SELECT feed_url, site_url, title, COUNT(id) AS subscribers |
45378752 LD |
28 | FROM ttrss_feeds WHERE feed_url NOT IN (SELECT feed_url FROM ttrss_feeds |
29 | WHERE private IS true OR auth_login != '' OR auth_pass != '' OR feed_url LIKE '%:%@%/%') | |
79178062 AD |
30 | GROUP BY feed_url, site_url, title ORDER BY subscribers DESC LIMIT 1000"); |
31 | ||
a42c55f0 | 32 | db_query("BEGIN"); |
79178062 | 33 | |
a42c55f0 | 34 | db_query("DELETE FROM ttrss_feedbrowser_cache"); |
79178062 AD |
35 | |
36 | $count = 0; | |
37 | ||
38 | while ($line = db_fetch_assoc($result)) { | |
a42c55f0 AD |
39 | $subscribers = db_escape_string($line["subscribers"]); |
40 | $feed_url = db_escape_string($line["feed_url"]); | |
41 | $title = db_escape_string($line["title"]); | |
42 | $site_url = db_escape_string($line["site_url"]); | |
79178062 | 43 | |
a42c55f0 | 44 | $tmp_result = db_query("SELECT subscribers FROM |
79178062 AD |
45 | ttrss_feedbrowser_cache WHERE feed_url = '$feed_url'"); |
46 | ||
47 | if (db_num_rows($tmp_result) == 0) { | |
48 | ||
a42c55f0 | 49 | db_query("INSERT INTO ttrss_feedbrowser_cache |
79178062 AD |
50 | (feed_url, site_url, title, subscribers) VALUES ('$feed_url', |
51 | '$site_url', '$title', '$subscribers')"); | |
52 | ||
53 | ++$count; | |
54 | ||
55 | } | |
56 | ||
57 | } | |
58 | ||
a42c55f0 | 59 | db_query("COMMIT"); |
79178062 AD |
60 | |
61 | return $count; | |
62 | ||
63 | } | |
64 | ||
7b55001e | 65 | function update_daemon_common($limit = DAEMON_FEED_LIMIT, $debug = true) { |
2c08214a AD |
66 | // Process all other feeds using last_updated and interval parameters |
67 | ||
6322ac79 | 68 | $schema_version = get_schema_version(); |
857efe49 AD |
69 | |
70 | if ($schema_version != SCHEMA_VERSION) { | |
71 | die("Schema version is wrong, please upgrade the database.\n"); | |
72 | } | |
73 | ||
61c1812f AD |
74 | define('PREFS_NO_CACHE', true); |
75 | ||
2c08214a | 76 | // Test if the user has loggued in recently. If not, it does not update its feeds. |
09e8bdfd | 77 | if (!SINGLE_USER_MODE && DAEMON_UPDATE_LOGIN_LIMIT > 0) { |
2c08214a AD |
78 | if (DB_TYPE == "pgsql") { |
79 | $login_thresh_qpart = "AND ttrss_users.last_login >= NOW() - INTERVAL '".DAEMON_UPDATE_LOGIN_LIMIT." days'"; | |
80 | } else { | |
81 | $login_thresh_qpart = "AND ttrss_users.last_login >= DATE_SUB(NOW(), INTERVAL ".DAEMON_UPDATE_LOGIN_LIMIT." DAY)"; | |
82 | } | |
83 | } else { | |
84 | $login_thresh_qpart = ""; | |
85 | } | |
86 | ||
537eca40 | 87 | // Test if the feed need a update (update interval exceeded). |
2c08214a AD |
88 | if (DB_TYPE == "pgsql") { |
89 | $update_limit_qpart = "AND (( | |
90 | ttrss_feeds.update_interval = 0 | |
ee0542ce | 91 | AND ttrss_user_prefs.value != '-1' |
2c08214a AD |
92 | AND ttrss_feeds.last_updated < NOW() - CAST((ttrss_user_prefs.value || ' minutes') AS INTERVAL) |
93 | ) OR ( | |
94 | ttrss_feeds.update_interval > 0 | |
95 | AND ttrss_feeds.last_updated < NOW() - CAST((ttrss_feeds.update_interval || ' minutes') AS INTERVAL) | |
f08426e3 AD |
96 | ) OR (ttrss_feeds.last_updated IS NULL |
97 | AND ttrss_user_prefs.value != '-1') | |
98 | OR (last_updated = '1970-01-01 00:00:00' | |
99 | AND ttrss_user_prefs.value != '-1'))"; | |
2c08214a AD |
100 | } else { |
101 | $update_limit_qpart = "AND (( | |
102 | ttrss_feeds.update_interval = 0 | |
ee0542ce | 103 | AND ttrss_user_prefs.value != '-1' |
2c08214a AD |
104 | AND ttrss_feeds.last_updated < DATE_SUB(NOW(), INTERVAL CONVERT(ttrss_user_prefs.value, SIGNED INTEGER) MINUTE) |
105 | ) OR ( | |
106 | ttrss_feeds.update_interval > 0 | |
107 | AND ttrss_feeds.last_updated < DATE_SUB(NOW(), INTERVAL ttrss_feeds.update_interval MINUTE) | |
f08426e3 AD |
108 | ) OR (ttrss_feeds.last_updated IS NULL |
109 | AND ttrss_user_prefs.value != '-1') | |
110 | OR (last_updated = '1970-01-01 00:00:00' | |
111 | AND ttrss_user_prefs.value != '-1'))"; | |
2c08214a AD |
112 | } |
113 | ||
114 | // Test if feed is currently being updated by another process. | |
115 | if (DB_TYPE == "pgsql") { | |
566417c4 | 116 | $updstart_thresh_qpart = "AND (ttrss_feeds.last_update_started IS NULL OR ttrss_feeds.last_update_started < NOW() - INTERVAL '10 minutes')"; |
2c08214a | 117 | } else { |
566417c4 | 118 | $updstart_thresh_qpart = "AND (ttrss_feeds.last_update_started IS NULL OR ttrss_feeds.last_update_started < DATE_SUB(NOW(), INTERVAL 10 MINUTE))"; |
2c08214a AD |
119 | } |
120 | ||
121 | // Test if there is a limit to number of updated feeds | |
122 | $query_limit = ""; | |
123 | if($limit) $query_limit = sprintf("LIMIT %d", $limit); | |
124 | ||
98070db0 TK |
125 | // Update the least recently updated feeds first |
126 | $query_order = "ORDER BY last_updated"; | |
127 | if (DB_TYPE == "pgsql") $query_order .= " NULLS FIRST"; | |
128 | ||
fce451a4 | 129 | $query = "SELECT DISTINCT ttrss_feeds.feed_url, ttrss_feeds.last_updated |
2c08214a AD |
130 | FROM |
131 | ttrss_feeds, ttrss_users, ttrss_user_prefs | |
f4ae0f05 | 132 | WHERE |
2c08214a | 133 | ttrss_feeds.owner_uid = ttrss_users.id |
f08426e3 | 134 | AND ttrss_user_prefs.profile IS NULL |
2c08214a AD |
135 | AND ttrss_users.id = ttrss_user_prefs.owner_uid |
136 | AND ttrss_user_prefs.pref_name = 'DEFAULT_UPDATE_INTERVAL' | |
137 | $login_thresh_qpart $update_limit_qpart | |
1c4421fc | 138 | $updstart_thresh_qpart |
98070db0 | 139 | $query_order $query_limit"; |
fce451a4 AD |
140 | |
141 | // We search for feed needing update. | |
142 | $result = db_query($query); | |
2c08214a | 143 | |
ca5ff2d9 | 144 | if($debug) _debug(sprintf("Scheduled %d feeds to update...", db_num_rows($result))); |
2c08214a AD |
145 | |
146 | // Here is a little cache magic in order to minimize risk of double feed updates. | |
147 | $feeds_to_update = array(); | |
148 | while ($line = db_fetch_assoc($result)) { | |
a42c55f0 | 149 | array_push($feeds_to_update, db_escape_string($line['feed_url'])); |
2c08214a AD |
150 | } |
151 | ||
152 | // We update the feed last update started date before anything else. | |
153 | // There is no lag due to feed contents downloads | |
154 | // It prevent an other process to update the same feed. | |
1c4421fc AD |
155 | |
156 | if(count($feeds_to_update) > 0) { | |
157 | $feeds_quoted = array(); | |
158 | ||
159 | foreach ($feeds_to_update as $feed) { | |
a42c55f0 | 160 | array_push($feeds_quoted, "'" . db_escape_string($feed) . "'"); |
1c4421fc AD |
161 | } |
162 | ||
a42c55f0 | 163 | db_query(sprintf("UPDATE ttrss_feeds SET last_update_started = NOW() |
1c4421fc | 164 | WHERE feed_url IN (%s)", implode(',', $feeds_quoted))); |
2c08214a AD |
165 | } |
166 | ||
8292d05b | 167 | $nf = 0; |
2d9c5684 | 168 | $bstarted = microtime(true); |
8292d05b | 169 | |
5cbd1fe8 AD |
170 | $batch_owners = array(); |
171 | ||
2c08214a | 172 | // For each feed, we call the feed update function. |
1c4421fc AD |
173 | foreach ($feeds_to_update as $feed) { |
174 | if($debug) _debug("Base feed: $feed"); | |
2c08214a | 175 | |
a42c55f0 | 176 | //update_rss_feed($line["id"], true); |
2c08214a | 177 | |
1c4421fc AD |
178 | // since we have the data cached, we can deal with other feeds with the same url |
179 | ||
15c762be | 180 | $tmp_result = db_query("SELECT DISTINCT ttrss_feeds.id,last_updated,ttrss_feeds.owner_uid |
ee0542ce AD |
181 | FROM ttrss_feeds, ttrss_users, ttrss_user_prefs WHERE |
182 | ttrss_user_prefs.owner_uid = ttrss_feeds.owner_uid AND | |
183 | ttrss_users.id = ttrss_user_prefs.owner_uid AND | |
184 | ttrss_user_prefs.pref_name = 'DEFAULT_UPDATE_INTERVAL' AND | |
f08426e3 | 185 | ttrss_user_prefs.profile IS NULL AND |
9e84bab4 AD |
186 | feed_url = '".db_escape_string($feed)."' |
187 | $update_limit_qpart | |
1c4421fc | 188 | $login_thresh_qpart |
5929a0c1 | 189 | ORDER BY ttrss_feeds.id $query_limit"); |
1c4421fc AD |
190 | |
191 | if (db_num_rows($tmp_result) > 0) { | |
192 | while ($tline = db_fetch_assoc($tmp_result)) { | |
15c762be | 193 | if($debug) _debug(" => " . $tline["last_updated"] . ", " . $tline["id"] . " " . $tline["owner_uid"]); |
f08426e3 | 194 | |
5cbd1fe8 AD |
195 | if (array_search($tline["owner_uid"], $batch_owners) === FALSE) |
196 | array_push($batch_owners, $tline["owner_uid"]); | |
197 | ||
2d9c5684 | 198 | $fstarted = microtime(true); |
7b55001e | 199 | update_rss_feed($tline["id"], true, false); |
4f71d743 | 200 | _debug_suppress(false); |
2d9c5684 AD |
201 | |
202 | _debug(sprintf(" %.4f (sec)", microtime(true) - $fstarted)); | |
203 | ||
8292d05b | 204 | ++$nf; |
1c4421fc AD |
205 | } |
206 | } | |
2c08214a AD |
207 | } |
208 | ||
2d9c5684 AD |
209 | if ($nf > 0) { |
210 | _debug(sprintf("Processed %d feeds in %.4f (sec), %.4f (sec/feed avg)", $nf, | |
211 | microtime(true) - $bstarted, (microtime(true) - $bstarted) / $nf)); | |
212 | } | |
213 | ||
5cbd1fe8 AD |
214 | foreach ($batch_owners as $owner_uid) { |
215 | _debug("Running housekeeping tasks for user $owner_uid..."); | |
216 | ||
217 | housekeeping_user($owner_uid); | |
218 | } | |
219 | ||
f1611683 AD |
220 | require_once "digest.php"; |
221 | ||
2c08214a | 222 | // Send feed digests by email if needed. |
a42c55f0 | 223 | send_headlines_digests($debug); |
2c08214a | 224 | |
8292d05b AD |
225 | return $nf; |
226 | ||
7b55001e | 227 | } |
2c08214a | 228 | |
6022776d AD |
229 | // this is used when subscribing |
230 | function set_basic_feed_info($feed) { | |
231 | ||
232 | $feed = db_escape_string($feed); | |
233 | ||
9bd422c2 | 234 | $result = db_query("SELECT feed_url,auth_pass,auth_login,auth_pass_encrypted |
6022776d AD |
235 | FROM ttrss_feeds WHERE id = '$feed'"); |
236 | ||
237 | $auth_pass_encrypted = sql_bool_to_bool(db_fetch_result($result, | |
238 | 0, "auth_pass_encrypted")); | |
239 | ||
240 | $auth_login = db_fetch_result($result, 0, "auth_login"); | |
241 | $auth_pass = db_fetch_result($result, 0, "auth_pass"); | |
242 | ||
17a8e61d | 243 | if ($auth_pass_encrypted && function_exists("mcrypt_decrypt")) { |
6022776d AD |
244 | require_once "crypt.php"; |
245 | $auth_pass = decrypt_string($auth_pass); | |
246 | } | |
247 | ||
248 | $fetch_url = db_fetch_result($result, 0, "feed_url"); | |
249 | ||
250 | $feed_data = fetch_file_contents($fetch_url, false, | |
251 | $auth_login, $auth_pass, false, | |
3d5d2890 | 252 | FEED_FETCH_TIMEOUT, |
6022776d AD |
253 | 0); |
254 | ||
255 | global $fetch_curl_used; | |
256 | ||
257 | if (!$fetch_curl_used) { | |
258 | $tmp = @gzdecode($feed_data); | |
259 | ||
260 | if ($tmp) $feed_data = $tmp; | |
261 | } | |
262 | ||
263 | $feed_data = trim($feed_data); | |
264 | ||
265 | $rss = new FeedParser($feed_data); | |
266 | $rss->init(); | |
267 | ||
268 | if (!$rss->error()) { | |
269 | ||
270 | $result = db_query("SELECT title, site_url FROM ttrss_feeds WHERE id = '$feed'"); | |
271 | ||
272 | $registered_title = db_fetch_result($result, 0, "title"); | |
273 | $orig_site_url = db_fetch_result($result, 0, "site_url"); | |
274 | ||
275 | $site_url = db_escape_string(mb_substr(rewrite_relative_url($fetch_url, $rss->get_link()), 0, 245)); | |
276 | $feed_title = db_escape_string(mb_substr($rss->get_title(), 0, 199)); | |
277 | ||
278 | if ($feed_title && (!$registered_title || $registered_title == "[Unknown]")) { | |
279 | db_query("UPDATE ttrss_feeds SET | |
280 | title = '$feed_title' WHERE id = '$feed'"); | |
281 | } | |
282 | ||
283 | if ($site_url && $orig_site_url != $site_url) { | |
284 | db_query("UPDATE ttrss_feeds SET | |
285 | site_url = '$site_url' WHERE id = '$feed'"); | |
286 | } | |
287 | } | |
288 | } | |
289 | ||
7b55001e AD |
290 | /** |
291 | * @SuppressWarnings(PHPMD.UnusedFormalParameter) | |
292 | */ | |
293 | function update_rss_feed($feed, $no_cache = false) { | |
2c08214a | 294 | |
2c08214a AD |
295 | $debug_enabled = defined('DAEMON_EXTENDED_DEBUG') || $_REQUEST['xdebug']; |
296 | ||
4f71d743 | 297 | _debug_suppress(!$debug_enabled); |
68cccafc | 298 | _debug("start", $debug_enabled); |
2c08214a | 299 | |
6bb96beb AD |
300 | $result = db_query("SELECT title FROM ttrss_feeds |
301 | WHERE id = '$feed'"); | |
bfe1eb4e AD |
302 | |
303 | if (db_num_rows($result) == 0) { | |
304 | _debug("feed $feed NOT FOUND/SKIPPED", $debug_enabled); | |
305 | user_error("Attempt to update unknown/invalid feed $feed", E_USER_WARNING); | |
306 | return false; | |
307 | } | |
308 | ||
6bb96beb AD |
309 | $title = db_fetch_result($result, 0, "title"); |
310 | ||
311 | // feed was batch-subscribed or something, we need to get basic info | |
312 | // this is not optimal currently as it fetches stuff separately TODO: optimize | |
313 | if ($title == "[Unknown]") { | |
314 | _debug("setting basic feed info for $feed..."); | |
315 | set_basic_feed_info($feed); | |
316 | } | |
317 | ||
a42c55f0 | 318 | $result = db_query("SELECT id,update_interval,auth_login, |
5ba1ddd4 | 319 | feed_url,auth_pass,cache_images, |
5321e775 | 320 | mark_unread_on_update, owner_uid, |
90e5f4f1 | 321 | pubsub_state, auth_pass_encrypted, |
0edf1d0d | 322 | feed_language |
87764a50 | 323 | FROM ttrss_feeds WHERE id = '$feed'"); |
2c08214a | 324 | |
2c08214a AD |
325 | $owner_uid = db_fetch_result($result, 0, "owner_uid"); |
326 | $mark_unread_on_update = sql_bool_to_bool(db_fetch_result($result, | |
327 | 0, "mark_unread_on_update")); | |
2c08214a | 328 | $pubsub_state = db_fetch_result($result, 0, "pubsub_state"); |
044cff2d AD |
329 | $auth_pass_encrypted = sql_bool_to_bool(db_fetch_result($result, |
330 | 0, "auth_pass_encrypted")); | |
2c08214a | 331 | |
a42c55f0 | 332 | db_query("UPDATE ttrss_feeds SET last_update_started = NOW() |
2c08214a AD |
333 | WHERE id = '$feed'"); |
334 | ||
335 | $auth_login = db_fetch_result($result, 0, "auth_login"); | |
336 | $auth_pass = db_fetch_result($result, 0, "auth_pass"); | |
337 | ||
17a8e61d | 338 | if ($auth_pass_encrypted && function_exists("mcrypt_decrypt")) { |
044cff2d AD |
339 | require_once "crypt.php"; |
340 | $auth_pass = decrypt_string($auth_pass); | |
341 | } | |
342 | ||
2c08214a AD |
343 | $cache_images = sql_bool_to_bool(db_fetch_result($result, 0, "cache_images")); |
344 | $fetch_url = db_fetch_result($result, 0, "feed_url"); | |
df659891 AD |
345 | $feed_language = db_escape_string(mb_strtolower(db_fetch_result($result, 0, "feed_language"))); |
346 | if (!$feed_language) $feed_language = 'english'; | |
2c08214a | 347 | |
a42c55f0 | 348 | $feed = db_escape_string($feed); |
2c08214a | 349 | |
f074ffe9 | 350 | $date_feed_processed = date('Y-m-d H:i'); |
2c08214a | 351 | |
865a3ed6 | 352 | $cache_filename = CACHE_DIR . "/simplepie/" . sha1($fetch_url) . ".xml"; |
f074ffe9 | 353 | |
ee65bef4 AD |
354 | $pluginhost = new PluginHost(); |
355 | $pluginhost->set_debug($debug_enabled); | |
356 | $user_plugins = get_pref("_ENABLED_PLUGINS", $owner_uid); | |
357 | ||
358 | $pluginhost->load(PLUGINS, PluginHost::KIND_ALL); | |
359 | $pluginhost->load($user_plugins, PluginHost::KIND_USER, $owner_uid); | |
360 | $pluginhost->load_data(); | |
361 | ||
7b55001e | 362 | $rss_hash = false; |
4f9cbdff | 363 | |
7b55001e AD |
364 | $force_refetch = isset($_REQUEST["force_refetch"]); |
365 | $feed_data = ""; | |
687a4f59 | 366 | |
7b55001e AD |
367 | foreach ($pluginhost->get_hooks(PluginHost::HOOK_FETCH_FEED) as $plugin) { |
368 | $feed_data = $plugin->hook_fetch_feed($feed_data, $fetch_url, $owner_uid, $feed, 0, $auth_login, $auth_pass); | |
369 | } | |
2c08214a | 370 | |
7b55001e AD |
371 | // try cache |
372 | if (!$feed_data && | |
373 | file_exists($cache_filename) && | |
374 | is_readable($cache_filename) && | |
375 | !$auth_login && !$auth_pass && | |
376 | filemtime($cache_filename) > time() - 30) { | |
be574731 | 377 | |
7b55001e | 378 | _debug("using local cache [$cache_filename].", $debug_enabled); |
52637d3b | 379 | |
7b55001e | 380 | @$feed_data = file_get_contents($cache_filename); |
f074ffe9 | 381 | |
7b55001e AD |
382 | if ($feed_data) { |
383 | $rss_hash = sha1($feed_data); | |
88edaa93 | 384 | } |
ee65bef4 | 385 | |
7b55001e AD |
386 | } else { |
387 | _debug("local cache will not be used for this feed", $debug_enabled); | |
388 | } | |
312742db | 389 | |
7b55001e AD |
390 | // fetch feed from source |
391 | if (!$feed_data) { | |
392 | _debug("fetching [$fetch_url]...", $debug_enabled); | |
312742db | 393 | |
7b55001e AD |
394 | if (ini_get("open_basedir") && function_exists("curl_init")) { |
395 | _debug("not using CURL due to open_basedir restrictions"); | |
396 | } | |
3f6f0857 | 397 | |
7b55001e AD |
398 | $feed_data = fetch_file_contents($fetch_url, false, |
399 | $auth_login, $auth_pass, false, | |
400 | $no_cache ? FEED_FETCH_NO_CACHE_TIMEOUT : FEED_FETCH_TIMEOUT, | |
401 | 0); | |
3f6f0857 | 402 | |
7b55001e | 403 | global $fetch_curl_used; |
3f6f0857 | 404 | |
7b55001e AD |
405 | if (!$fetch_curl_used) { |
406 | $tmp = @gzdecode($feed_data); | |
1367bc3f | 407 | |
7b55001e AD |
408 | if ($tmp) $feed_data = $tmp; |
409 | } | |
017401dd | 410 | |
7b55001e | 411 | $feed_data = trim($feed_data); |
fd687300 | 412 | |
7b55001e | 413 | _debug("fetch done.", $debug_enabled); |
95beaa14 | 414 | |
7b55001e AD |
415 | // cache vanilla feed data for re-use |
416 | if ($feed_data && !$auth_pass && !$auth_login && is_writable(CACHE_DIR . "/simplepie")) { | |
417 | $new_rss_hash = sha1($feed_data); | |
418 | ||
419 | if ($new_rss_hash != $rss_hash) { | |
420 | _debug("saving $cache_filename", $debug_enabled); | |
421 | @file_put_contents($cache_filename, $feed_data); | |
95beaa14 | 422 | } |
4f9cbdff | 423 | } |
7b55001e | 424 | } |
017401dd | 425 | |
7b55001e AD |
426 | if (!$feed_data) { |
427 | global $fetch_last_error; | |
428 | global $fetch_last_error_code; | |
f074ffe9 | 429 | |
7b55001e | 430 | _debug("unable to fetch: $fetch_last_error [$fetch_last_error_code]", $debug_enabled); |
f074ffe9 | 431 | |
7b55001e | 432 | $error_escaped = ''; |
7a01dc77 | 433 | |
7b55001e AD |
434 | // If-Modified-Since |
435 | if ($fetch_last_error_code != 304) { | |
436 | $error_escaped = db_escape_string($fetch_last_error); | |
437 | } else { | |
438 | _debug("source claims data not modified, nothing to do.", $debug_enabled); | |
439 | } | |
4f9cbdff | 440 | |
7b55001e AD |
441 | db_query( |
442 | "UPDATE ttrss_feeds SET last_error = '$error_escaped', | |
443 | last_updated = NOW() WHERE id = '$feed'"); | |
4f9cbdff | 444 | |
7b55001e | 445 | return; |
f074ffe9 AD |
446 | } |
447 | ||
1ffe3391 | 448 | foreach ($pluginhost->get_hooks(PluginHost::HOOK_FEED_FETCHED) as $plugin) { |
6791af0c | 449 | $feed_data = $plugin->hook_feed_fetched($feed_data, $fetch_url, $owner_uid, $feed); |
017401dd AD |
450 | } |
451 | ||
07d3431e AD |
452 | $rss = new FeedParser($feed_data); |
453 | $rss->init(); | |
2c08214a | 454 | |
a42c55f0 | 455 | $feed = db_escape_string($feed); |
2c08214a | 456 | |
19b3992b | 457 | if (!$rss->error()) { |
2c08214a | 458 | |
d2a421e3 | 459 | // We use local pluginhost here because we need to load different per-user feed plugins |
1ffe3391 | 460 | $pluginhost->run_hooks(PluginHost::HOOK_FEED_PARSED, "hook_feed_parsed", $rss); |
4412b877 | 461 | |
df659891 | 462 | _debug("language: $feed_language", $debug_enabled); |
68cccafc | 463 | _debug("processing feed data...", $debug_enabled); |
2c08214a | 464 | |
a42c55f0 | 465 | // db_query("BEGIN"); |
2c08214a | 466 | |
382268c6 AD |
467 | if (DB_TYPE == "pgsql") { |
468 | $favicon_interval_qpart = "favicon_last_checked < NOW() - INTERVAL '12 hour'"; | |
469 | } else { | |
470 | $favicon_interval_qpart = "favicon_last_checked < DATE_SUB(NOW(), INTERVAL 12 HOUR)"; | |
471 | } | |
472 | ||
fe4535e6 | 473 | $result = db_query("SELECT owner_uid,favicon_avg_color, |
382268c6 AD |
474 | (favicon_last_checked IS NULL OR $favicon_interval_qpart) AS |
475 | favicon_needs_check | |
2c08214a AD |
476 | FROM ttrss_feeds WHERE id = '$feed'"); |
477 | ||
382268c6 AD |
478 | $favicon_needs_check = sql_bool_to_bool(db_fetch_result($result, 0, |
479 | "favicon_needs_check")); | |
36490f11 | 480 | $favicon_avg_color = db_fetch_result($result, 0, "favicon_avg_color"); |
2c08214a AD |
481 | |
482 | $owner_uid = db_fetch_result($result, 0, "owner_uid"); | |
483 | ||
a42c55f0 | 484 | $site_url = db_escape_string(mb_substr(rewrite_relative_url($fetch_url, $rss->get_link()), 0, 245)); |
2c08214a | 485 | |
cd07592c AD |
486 | _debug("site_url: $site_url", $debug_enabled); |
487 | _debug("feed_title: " . $rss->get_title(), $debug_enabled); | |
488 | ||
687a4f59 | 489 | if ($favicon_needs_check || $force_refetch) { |
36490f11 AD |
490 | |
491 | /* terrible hack: if we crash on floicon shit here, we won't check | |
560cbd8c | 492 | * the icon avgcolor again (unless the icon got updated) */ |
36490f11 | 493 | |
560cbd8c AD |
494 | $favicon_file = ICONS_DIR . "/$feed.ico"; |
495 | $favicon_modified = @filemtime($favicon_file); | |
496 | ||
68cccafc | 497 | _debug("checking favicon...", $debug_enabled); |
687a4f59 | 498 | |
6f7798b6 | 499 | check_feed_favicon($site_url, $feed); |
560cbd8c AD |
500 | $favicon_modified_new = @filemtime($favicon_file); |
501 | ||
502 | if ($favicon_modified_new > $favicon_modified) | |
503 | $favicon_avg_color = ''; | |
687a4f59 | 504 | |
6ee0d4b0 | 505 | if (file_exists($favicon_file) && function_exists("imagecreatefromstring") && $favicon_avg_color == '') { |
bc7a144d | 506 | require_once "colors.php"; |
687a4f59 | 507 | |
aafd55ba AD |
508 | db_query("UPDATE ttrss_feeds SET favicon_avg_color = 'fail' WHERE |
509 | id = '$feed'"); | |
510 | ||
6322ac79 | 511 | $favicon_color = db_escape_string( |
bc7a144d | 512 | calculate_avg_color($favicon_file)); |
63c323f7 | 513 | |
bc7a144d | 514 | $favicon_colorstring = ",favicon_avg_color = '".$favicon_color."'"; |
36490f11 | 515 | } else if ($favicon_avg_color == 'fail') { |
84ceb6bd | 516 | _debug("floicon failed on this file, not trying to recalculate avg color", $debug_enabled); |
6ac722d5 | 517 | } |
687a4f59 | 518 | |
a42c55f0 | 519 | db_query("UPDATE ttrss_feeds SET favicon_last_checked = NOW() |
bc7a144d | 520 | $favicon_colorstring |
f2798eb6 AD |
521 | WHERE id = '$feed'"); |
522 | } | |
2c08214a | 523 | |
68cccafc | 524 | _debug("loading filters & labels...", $debug_enabled); |
2c08214a | 525 | |
a42c55f0 | 526 | $filters = load_filters($feed, $owner_uid); |
2c08214a | 527 | |
68cccafc | 528 | _debug("" . count($filters) . " filters loaded.", $debug_enabled); |
2c08214a | 529 | |
19b3992b | 530 | $items = $rss->get_items(); |
2c08214a | 531 | |
19b3992b | 532 | if (!is_array($items)) { |
68cccafc | 533 | _debug("no articles found.", $debug_enabled); |
2c08214a | 534 | |
a42c55f0 | 535 | db_query("UPDATE ttrss_feeds |
2c08214a AD |
536 | SET last_updated = NOW(), last_error = '' WHERE id = '$feed'"); |
537 | ||
538 | return; // no articles | |
539 | } | |
540 | ||
541 | if ($pubsub_state != 2 && PUBSUBHUBBUB_ENABLED) { | |
542 | ||
68cccafc | 543 | _debug("checking for PUSH hub...", $debug_enabled); |
2c08214a AD |
544 | |
545 | $feed_hub_url = false; | |
2c08214a | 546 | |
19b3992b | 547 | $links = $rss->get_links('hub'); |
2c08214a | 548 | |
19b3992b AD |
549 | if ($links && is_array($links)) { |
550 | foreach ($links as $l) { | |
551 | $feed_hub_url = $l; | |
552 | break; | |
2c08214a AD |
553 | } |
554 | } | |
555 | ||
68cccafc | 556 | _debug("feed hub url: $feed_hub_url", $debug_enabled); |
2c08214a | 557 | |
d8179cb9 AD |
558 | $feed_self_url = $fetch_url; |
559 | ||
560 | $links = $rss->get_links('self'); | |
561 | ||
562 | if ($links && is_array($links)) { | |
563 | foreach ($links as $l) { | |
564 | $feed_self_url = $l; | |
565 | break; | |
566 | } | |
567 | } | |
568 | ||
569 | _debug("feed self url = $feed_self_url"); | |
570 | ||
571 | if ($feed_hub_url && $feed_self_url && function_exists('curl_init') && | |
2c08214a AD |
572 | !ini_get("open_basedir")) { |
573 | ||
becd215a | 574 | require_once 'lib/pubsubhubbub/Subscriber.php'; |
2c08214a AD |
575 | |
576 | $callback_url = get_self_url_prefix() . | |
577 | "/public.php?op=pubsub&id=$feed"; | |
578 | ||
cb3f8773 | 579 | $s = new Pubsubhubbub\Subscriber\Subscriber($feed_hub_url, $callback_url); |
2c08214a | 580 | |
d8179cb9 | 581 | $rc = $s->subscribe($feed_self_url); |
2c08214a | 582 | |
5ba1ddd4 | 583 | _debug("feed hub url found, subscribe request sent. [rc=$rc]", $debug_enabled); |
2c08214a | 584 | |
a42c55f0 | 585 | db_query("UPDATE ttrss_feeds SET pubsub_state = 1 |
2c08214a AD |
586 | WHERE id = '$feed'"); |
587 | } | |
588 | } | |
589 | ||
68cccafc | 590 | _debug("processing articles...", $debug_enabled); |
2c08214a | 591 | |
6c9f3d4a AD |
592 | $tstart = time(); |
593 | ||
19b3992b | 594 | foreach ($items as $item) { |
5d56d100 | 595 | if ($_REQUEST['xdebug'] == 3) { |
2c08214a AD |
596 | print_r($item); |
597 | } | |
598 | ||
6c9f3d4a AD |
599 | if (ini_get("max_execution_time") > 0 && time() - $tstart >= ini_get("max_execution_time") * 0.7) { |
600 | _debug("looks like there's too many articles to process at once, breaking out", $debug_enabled); | |
601 | break; | |
602 | } | |
603 | ||
19b3992b AD |
604 | $entry_guid = $item->get_id(); |
605 | if (!$entry_guid) $entry_guid = $item->get_link(); | |
606 | if (!$entry_guid) $entry_guid = make_guid_from_title($item->get_title()); | |
2c08214a AD |
607 | if (!$entry_guid) continue; |
608 | ||
3a4c8973 AD |
609 | $entry_guid = "$owner_uid,$entry_guid"; |
610 | ||
a42c55f0 | 611 | $entry_guid_hashed = db_escape_string('SHA1:' . sha1($entry_guid)); |
5e3d5480 | 612 | |
68cccafc | 613 | _debug("guid $entry_guid / $entry_guid_hashed", $debug_enabled); |
5e3d5480 | 614 | |
2c08214a AD |
615 | $entry_timestamp = ""; |
616 | ||
04d2f9c8 AD |
617 | $entry_timestamp = $item->get_date(); |
618 | ||
619 | _debug("orig date: " . $item->get_date(), $debug_enabled); | |
2c08214a | 620 | |
30123fe6 | 621 | if ($entry_timestamp == -1 || !$entry_timestamp || $entry_timestamp > time()) { |
2c08214a | 622 | $entry_timestamp = time(); |
2c08214a AD |
623 | } |
624 | ||
625 | $entry_timestamp_fmt = strftime("%Y/%m/%d %H:%M:%S", $entry_timestamp); | |
626 | ||
68cccafc | 627 | _debug("date $entry_timestamp [$entry_timestamp_fmt]", $debug_enabled); |
2c08214a | 628 | |
99429e57 AD |
629 | // $entry_title = html_entity_decode($item->get_title(), ENT_COMPAT, 'UTF-8'); |
630 | // $entry_title = decode_numeric_entities($entry_title); | |
631 | $entry_title = $item->get_title(); | |
1b35d30c | 632 | |
5d56d100 | 633 | $entry_link = rewrite_relative_url($site_url, $item->get_link()); |
2c08214a | 634 | |
68cccafc AD |
635 | _debug("title $entry_title", $debug_enabled); |
636 | _debug("link $entry_link", $debug_enabled); | |
2c08214a AD |
637 | |
638 | if (!$entry_title) $entry_title = date("Y-m-d H:i:s", $entry_timestamp);; | |
639 | ||
19b3992b AD |
640 | $entry_content = $item->get_content(); |
641 | if (!$entry_content) $entry_content = $item->get_description(); | |
2c08214a AD |
642 | |
643 | if ($_REQUEST["xdebug"] == 2) { | |
9ec10352 | 644 | print "content: "; |
0bc503ff | 645 | print htmlspecialchars($entry_content); |
3c696512 | 646 | print "\n"; |
2c08214a AD |
647 | } |
648 | ||
12ff230b AD |
649 | $entry_comments = db_escape_string(mb_substr($item->get_comments_url(), 0, 245)); |
650 | $num_comments = (int) $item->get_comments_count(); | |
2c08214a | 651 | |
12ff230b | 652 | $entry_author = $item->get_author(); // escaped later |
a42c55f0 | 653 | $entry_guid = db_escape_string(mb_substr($entry_guid, 0, 245)); |
2c08214a | 654 | |
68cccafc AD |
655 | _debug("author $entry_author", $debug_enabled); |
656 | _debug("num_comments: $num_comments", $debug_enabled); | |
ee78f81c | 657 | _debug("looking for tags...", $debug_enabled); |
2c08214a AD |
658 | |
659 | // parse <category> entries into tags | |
660 | ||
661 | $additional_tags = array(); | |
662 | ||
19b3992b | 663 | $additional_tags_src = $item->get_categories(); |
2c08214a | 664 | |
19b3992b AD |
665 | if (is_array($additional_tags_src)) { |
666 | foreach ($additional_tags_src as $tobj) { | |
cd07592c | 667 | array_push($additional_tags, $tobj); |
2c08214a | 668 | } |
19b3992b | 669 | } |
2c08214a | 670 | |
fa6fbd36 | 671 | $entry_tags = array_unique($additional_tags); |
2c08214a AD |
672 | |
673 | for ($i = 0; $i < count($entry_tags); $i++) | |
674 | $entry_tags[$i] = mb_strtolower($entry_tags[$i], 'utf-8'); | |
675 | ||
ee78f81c AD |
676 | _debug("tags found: " . join(",", $entry_tags), $debug_enabled); |
677 | ||
68cccafc | 678 | _debug("done collecting data.", $debug_enabled); |
2c08214a | 679 | |
3318d324 | 680 | $result = db_query("SELECT id, content_hash, lang FROM ttrss_entries |
b1840673 | 681 | WHERE guid = '".db_escape_string($entry_guid)."' OR guid = '$entry_guid_hashed'"); |
b30abdad AD |
682 | |
683 | if (db_num_rows($result) != 0) { | |
b1840673 AD |
684 | $base_entry_id = db_fetch_result($result, 0, "id"); |
685 | $entry_stored_hash = db_fetch_result($result, 0, "content_hash"); | |
a29fe121 | 686 | $article_labels = get_article_labels($base_entry_id, $owner_uid); |
3318d324 | 687 | $entry_language = db_fetch_result($result, 0, "lang"); |
a8ac7661 | 688 | |
040aa540 | 689 | $existing_tags = get_article_tags($base_entry_id, $owner_uid); |
a8ac7661 AD |
690 | $entry_tags = array_unique(array_merge($entry_tags, $existing_tags)); |
691 | ||
b30abdad | 692 | } else { |
b1840673 AD |
693 | $base_entry_id = false; |
694 | $entry_stored_hash = ""; | |
a29fe121 | 695 | $article_labels = array(); |
3318d324 | 696 | $entry_language = ""; |
b30abdad AD |
697 | } |
698 | ||
455b1401 | 699 | $article = array("owner_uid" => $owner_uid, // read only |
b30abdad | 700 | "guid" => $entry_guid, // read only |
59e83455 | 701 | "guid_hashed" => $entry_guid_hashed, // read only |
19b3992b AD |
702 | "title" => $entry_title, |
703 | "content" => $entry_content, | |
704 | "link" => $entry_link, | |
a29fe121 | 705 | "labels" => $article_labels, // current limitation: can add labels to article, can't remove them |
19b3992b | 706 | "tags" => $entry_tags, |
e02555c1 | 707 | "author" => $entry_author, |
c9299c28 | 708 | "force_catchup" => false, // ugly hack for the time being |
6de3a1be | 709 | "score_modifier" => 0, // no previous value, plugin should recalculate score modifier based on content if needed |
3318d324 | 710 | "language" => $entry_language, |
f73e03e0 AD |
711 | "feed" => array("id" => $feed, |
712 | "fetch_url" => $fetch_url, | |
babfadbf J |
713 | "site_url" => $site_url, |
714 | "cache_images" => $cache_images) | |
f73e03e0 | 715 | ); |
cc85704f | 716 | |
b1840673 | 717 | $entry_plugin_data = ""; |
eb16bd9f | 718 | $entry_current_hash = calculate_article_hash($article, $pluginhost); |
b1840673 AD |
719 | |
720 | _debug("article hash: $entry_current_hash [stored=$entry_stored_hash]", $debug_enabled); | |
721 | ||
522e8b35 | 722 | if ($entry_current_hash == $entry_stored_hash && !isset($_REQUEST["force_rehash"])) { |
b1840673 AD |
723 | _debug("stored article seems up to date [IID: $base_entry_id], updating timestamp only", $debug_enabled); |
724 | ||
725 | // we keep encountering the entry in feeds, so we need to | |
726 | // update date_updated column so that we don't get horrible | |
727 | // dupes when the entry gets purged and reinserted again e.g. | |
728 | // in the case of SLOW SLOW OMG SLOW updating feeds | |
729 | ||
730 | $base_entry_id = db_fetch_result($result, 0, "id"); | |
731 | ||
732 | db_query("UPDATE ttrss_entries SET date_updated = NOW() | |
733 | WHERE id = '$base_entry_id'"); | |
734 | ||
5bdcb8fd | 735 | continue; |
b1840673 AD |
736 | } |
737 | ||
738 | _debug("hash differs, applying plugin filters:", $debug_enabled); | |
739 | ||
1ffe3391 | 740 | foreach ($pluginhost->get_hooks(PluginHost::HOOK_ARTICLE_FILTER) as $plugin) { |
b1840673 AD |
741 | _debug("... " . get_class($plugin), $debug_enabled); |
742 | ||
743 | $start = microtime(true); | |
19b3992b | 744 | $article = $plugin->hook_article_filter($article); |
0084f0d1 | 745 | |
b1840673 AD |
746 | _debug("=== " . sprintf("%.4f (sec)", microtime(true) - $start), $debug_enabled); |
747 | ||
748 | $entry_plugin_data .= mb_strtolower(get_class($plugin)) . ","; | |
cc85704f AD |
749 | } |
750 | ||
0bc503ff AD |
751 | if ($_REQUEST["xdebug"] == 2) { |
752 | print "processed content: "; | |
753 | print htmlspecialchars($article["content"]); | |
754 | print "\n"; | |
755 | } | |
756 | ||
b1840673 AD |
757 | $entry_plugin_data = db_escape_string($entry_plugin_data); |
758 | ||
759 | _debug("plugin data: $entry_plugin_data", $debug_enabled); | |
760 | ||
35c12dc4 AD |
761 | // Workaround: 4-byte unicode requires utf8mb4 in MySQL. See https://tt-rss.org/forum/viewtopic.php?f=1&t=3377&p=20077#p20077 |
762 | if (DB_TYPE == "mysql") { | |
763 | foreach ($article as $k => $v) { | |
35c37354 AD |
764 | |
765 | // i guess we'll have to take the risk of 4byte unicode labels & tags here | |
dae16f72 | 766 | if (is_string($article[$k])) { |
35c37354 AD |
767 | $article[$k] = preg_replace('/[\x{10000}-\x{10FFFF}]/u', "\xEF\xBF\xBD", $v); |
768 | } | |
35c12dc4 AD |
769 | } |
770 | } | |
771 | ||
b8774453 AD |
772 | /* Collect article tags here so we could filter by them: */ |
773 | ||
557d86fe AD |
774 | $matched_rules = array(); |
775 | ||
b8774453 | 776 | $article_filters = get_article_filters($filters, $article["title"], |
7b55001e | 777 | $article["content"], $article["link"], $article["author"], |
557d86fe | 778 | $article["tags"], $matched_rules); |
b8774453 AD |
779 | |
780 | if ($debug_enabled) { | |
557d86fe AD |
781 | _debug("matched filter rules: ", $debug_enabled); |
782 | ||
783 | if (count($matched_rules) != 0) { | |
784 | print_r($matched_rules); | |
785 | } | |
786 | ||
787 | _debug("filter actions: ", $debug_enabled); | |
788 | ||
b8774453 AD |
789 | if (count($article_filters) != 0) { |
790 | print_r($article_filters); | |
791 | } | |
792 | } | |
793 | ||
794 | $plugin_filter_names = find_article_filters($article_filters, "plugin"); | |
795 | $plugin_filter_actions = $pluginhost->get_filter_actions(); | |
796 | ||
797 | if (count($plugin_filter_names) > 0) { | |
798 | _debug("applying plugin filter actions...", $debug_enabled); | |
799 | ||
800 | foreach ($plugin_filter_names as $pfn) { | |
801 | list($pfclass,$pfaction) = explode(":", $pfn["param"]); | |
802 | ||
803 | if (isset($plugin_filter_actions[$pfclass])) { | |
804 | $plugin = $pluginhost->get_plugin($pfclass); | |
805 | ||
806 | _debug("... $pfclass: $pfaction", $debug_enabled); | |
807 | ||
808 | if ($plugin) { | |
809 | $start = microtime(true); | |
810 | $article = $plugin->hook_article_filter_action($article, $pfaction); | |
811 | ||
812 | _debug("=== " . sprintf("%.4f (sec)", microtime(true) - $start), $debug_enabled); | |
813 | } else { | |
814 | _debug("??? $pfclass: plugin object not found."); | |
815 | } | |
816 | } else { | |
817 | _debug("??? $pfclass: filter plugin not registered."); | |
818 | } | |
819 | } | |
820 | } | |
821 | ||
19b3992b | 822 | $entry_tags = $article["tags"]; |
a42c55f0 AD |
823 | $entry_guid = db_escape_string($entry_guid); |
824 | $entry_title = db_escape_string($article["title"]); | |
12ff230b | 825 | $entry_author = db_escape_string(mb_substr($article["author"], 0, 245)); |
a42c55f0 | 826 | $entry_link = db_escape_string($article["link"]); |
f935d98e | 827 | $entry_content = $article["content"]; // escaped below |
c9299c28 | 828 | $entry_force_catchup = $article["force_catchup"]; |
a29fe121 | 829 | $article_labels = $article["labels"]; |
6de3a1be | 830 | $entry_score_modifier = (int) $article["score_modifier"]; |
3318d324 | 831 | $entry_language = db_escape_string($article["language"]); |
a29fe121 AD |
832 | |
833 | if ($debug_enabled) { | |
834 | _debug("article labels:", $debug_enabled); | |
557d86fe AD |
835 | |
836 | if (count($article_labels) != 0) { | |
837 | print_r($article_labels); | |
838 | } | |
a29fe121 | 839 | } |
c9299c28 AD |
840 | |
841 | _debug("force catchup: $entry_force_catchup"); | |
f935d98e | 842 | |
0a3fd79b | 843 | if ($cache_images && is_writable(CACHE_DIR . '/images')) |
41bead9b | 844 | cache_media($entry_content, $site_url, $debug_enabled); |
0a3fd79b | 845 | |
a42c55f0 | 846 | $entry_content = db_escape_string($entry_content, false); |
1f45c857 | 847 | |
07877caf | 848 | //db_query("BEGIN"); |
2c08214a | 849 | |
a42c55f0 | 850 | $result = db_query("SELECT id FROM ttrss_entries |
5e3d5480 | 851 | WHERE (guid = '$entry_guid' OR guid = '$entry_guid_hashed')"); |
9e222305 | 852 | |
2c08214a AD |
853 | if (db_num_rows($result) == 0) { |
854 | ||
07d3431e | 855 | _debug("base guid [$entry_guid or $entry_guid_hashed] not found, creating...", $debug_enabled); |
2c08214a AD |
856 | |
857 | // base post entry does not exist, create it | |
858 | ||
07d3431e | 859 | db_query( |
2c08214a AD |
860 | "INSERT INTO ttrss_entries |
861 | (title, | |
862 | guid, | |
863 | link, | |
864 | updated, | |
865 | content, | |
866 | content_hash, | |
867 | no_orig_date, | |
868 | date_updated, | |
869 | date_entered, | |
870 | comments, | |
871 | num_comments, | |
b30abdad | 872 | plugin_data, |
6b461797 | 873 | lang, |
2c08214a AD |
874 | author) |
875 | VALUES | |
876 | ('$entry_title', | |
5e3d5480 | 877 | '$entry_guid_hashed', |
2c08214a AD |
878 | '$entry_link', |
879 | '$entry_timestamp_fmt', | |
880 | '$entry_content', | |
b1840673 | 881 | '$entry_current_hash', |
5ba1ddd4 | 882 | false, |
2c08214a | 883 | NOW(), |
be574731 | 884 | '$date_feed_processed', |
2c08214a AD |
885 | '$entry_comments', |
886 | '$num_comments', | |
b30abdad | 887 | '$entry_plugin_data', |
6b461797 | 888 | '$entry_language', |
2c08214a | 889 | '$entry_author')"); |
e8291805 | 890 | |
2c08214a AD |
891 | } |
892 | ||
893 | // now it should exist, if not - bad luck then | |
894 | ||
b1840673 | 895 | $result = db_query("SELECT id FROM ttrss_entries |
5e3d5480 | 896 | WHERE guid = '$entry_guid' OR guid = '$entry_guid_hashed'"); |
2c08214a AD |
897 | |
898 | $entry_ref_id = 0; | |
899 | $entry_int_id = 0; | |
900 | ||
901 | if (db_num_rows($result) == 1) { | |
902 | ||
68cccafc | 903 | _debug("base guid found, checking for user record", $debug_enabled); |
2c08214a | 904 | |
2c08214a AD |
905 | $ref_id = db_fetch_result($result, 0, "id"); |
906 | $entry_ref_id = $ref_id; | |
907 | ||
5e3d5480 AD |
908 | /* $stored_guid = db_fetch_result($result, 0, "guid"); |
909 | if ($stored_guid != $entry_guid_hashed) { | |
68cccafc | 910 | if ($debug_enabled) _debug("upgrading compat guid to hashed one", $debug_enabled); |
5e3d5480 | 911 | |
a42c55f0 | 912 | db_query("UPDATE ttrss_entries SET guid = '$entry_guid_hashed' WHERE |
5e3d5480 AD |
913 | id = '$ref_id'"); |
914 | } */ | |
915 | ||
2c08214a | 916 | if (find_article_filter($article_filters, "filter")) { |
07877caf | 917 | //db_query("COMMIT"); // close transaction in progress |
2c08214a AD |
918 | continue; |
919 | } | |
920 | ||
6de3a1be | 921 | $score = calculate_article_score($article_filters) + $entry_score_modifier; |
2c08214a | 922 | |
6de3a1be | 923 | _debug("initial score: $score [including plugin modifier: $entry_score_modifier]", $debug_enabled); |
2c08214a | 924 | |
4f186b1f AD |
925 | // check for user post link to main table |
926 | ||
2c08214a | 927 | $query = "SELECT ref_id, int_id FROM ttrss_user_entries WHERE |
5bdcb8fd | 928 | ref_id = '$ref_id' AND owner_uid = '$owner_uid'"; |
2c08214a AD |
929 | |
930 | // if ($_REQUEST["xdebug"]) print "$query\n"; | |
931 | ||
a42c55f0 | 932 | $result = db_query($query); |
2c08214a AD |
933 | |
934 | // okay it doesn't exist - create user entry | |
935 | if (db_num_rows($result) == 0) { | |
936 | ||
68cccafc | 937 | _debug("user record not found, creating...", $debug_enabled); |
2c08214a | 938 | |
c9299c28 | 939 | if ($score >= -500 && !find_article_filter($article_filters, 'catchup') && !$entry_force_catchup) { |
2c08214a AD |
940 | $unread = 'true'; |
941 | $last_read_qpart = 'NULL'; | |
942 | } else { | |
943 | $unread = 'false'; | |
944 | $last_read_qpart = 'NOW()'; | |
945 | } | |
946 | ||
947 | if (find_article_filter($article_filters, 'mark') || $score > 1000) { | |
948 | $marked = 'true'; | |
949 | } else { | |
950 | $marked = 'false'; | |
951 | } | |
952 | ||
953 | if (find_article_filter($article_filters, 'publish')) { | |
954 | $published = 'true'; | |
955 | } else { | |
956 | $published = 'false'; | |
957 | } | |
958 | ||
7873d588 AD |
959 | $last_marked = ($marked == 'true') ? 'NOW()' : 'NULL'; |
960 | $last_published = ($published == 'true') ? 'NOW()' : 'NULL'; | |
961 | ||
6322ac79 | 962 | $result = db_query( |
2c08214a AD |
963 | "INSERT INTO ttrss_user_entries |
964 | (ref_id, owner_uid, feed_id, unread, last_read, marked, | |
7873d588 AD |
965 | published, score, tag_cache, label_cache, uuid, |
966 | last_marked, last_published) | |
2c08214a | 967 | VALUES ('$ref_id', '$owner_uid', '$feed', $unread, |
7873d588 AD |
968 | $last_read_qpart, $marked, $published, '$score', '', '', |
969 | '', $last_marked, $last_published)"); | |
2c08214a AD |
970 | |
971 | if (PUBSUBHUBBUB_HUB && $published == 'true') { | |
972 | $rss_link = get_self_url_prefix() . | |
973 | "/public.php?op=rss&id=-2&key=" . | |
a42c55f0 | 974 | get_feed_access_key(-2, false, $owner_uid); |
2c08214a | 975 | |
cb3f8773 | 976 | $p = new pubsubhubbub\publisher\Publisher(PUBSUBHUBBUB_HUB); |
2c08214a | 977 | |
5ba1ddd4 | 978 | /* $pubsub_result = */ $p->publish_update($rss_link); |
2c08214a AD |
979 | } |
980 | ||
6322ac79 | 981 | $result = db_query( |
2c08214a AD |
982 | "SELECT int_id FROM ttrss_user_entries WHERE |
983 | ref_id = '$ref_id' AND owner_uid = '$owner_uid' AND | |
984 | feed_id = '$feed' LIMIT 1"); | |
985 | ||
986 | if (db_num_rows($result) == 1) { | |
987 | $entry_int_id = db_fetch_result($result, 0, "int_id"); | |
988 | } | |
989 | } else { | |
68cccafc | 990 | _debug("user record FOUND", $debug_enabled); |
2c08214a AD |
991 | |
992 | $entry_ref_id = db_fetch_result($result, 0, "ref_id"); | |
993 | $entry_int_id = db_fetch_result($result, 0, "int_id"); | |
994 | } | |
995 | ||
68cccafc | 996 | _debug("RID: $entry_ref_id, IID: $entry_int_id", $debug_enabled); |
2c08214a | 997 | |
e854442e | 998 | if (DB_TYPE == "pgsql") { |
978989bb | 999 | $tsvector_combined = db_escape_string(mb_substr($entry_title . ' ' . strip_tags(str_replace('<', ' <', $entry_content)), |
e854442e AD |
1000 | 0, 1000000)); |
1001 | ||
df659891 | 1002 | $tsvector_qpart = "tsvector_combined = to_tsvector('$feed_language', '$tsvector_combined'),"; |
e854442e AD |
1003 | |
1004 | } else { | |
1005 | $tsvector_qpart = ""; | |
1006 | } | |
1007 | ||
b1840673 AD |
1008 | db_query("UPDATE ttrss_entries |
1009 | SET title = '$entry_title', | |
1010 | content = '$entry_content', | |
1011 | content_hash = '$entry_current_hash', | |
1012 | updated = '$entry_timestamp_fmt', | |
e854442e | 1013 | $tsvector_qpart |
b1840673 AD |
1014 | num_comments = '$num_comments', |
1015 | plugin_data = '$entry_plugin_data', | |
1016 | author = '$entry_author', | |
1017 | lang = '$entry_language' | |
1018 | WHERE id = '$ref_id'"); | |
1019 | ||
59e83455 AD |
1020 | // update aux data |
1021 | db_query("UPDATE ttrss_user_entries | |
1022 | SET score = '$score' WHERE ref_id = '$ref_id'"); | |
1023 | ||
b1840673 | 1024 | if ($mark_unread_on_update) { |
24e6ff5d AD |
1025 | _debug("article updated, marking unread as requested.", $debug_enabled); |
1026 | ||
b1840673 AD |
1027 | db_query("UPDATE ttrss_user_entries |
1028 | SET last_read = null, unread = true WHERE ref_id = '$ref_id'"); | |
2c08214a AD |
1029 | } |
1030 | } | |
1031 | ||
07877caf | 1032 | //db_query("COMMIT"); |
2c08214a | 1033 | |
a29fe121 AD |
1034 | _debug("assigning labels [other]...", $debug_enabled); |
1035 | ||
1036 | foreach ($article_labels as $label) { | |
1037 | label_add_article($entry_ref_id, $label[1], $owner_uid); | |
1038 | } | |
1039 | ||
1040 | _debug("assigning labels [filters]...", $debug_enabled); | |
2c08214a | 1041 | |
a42c55f0 | 1042 | assign_article_to_label_filters($entry_ref_id, $article_filters, |
b24504b1 | 1043 | $owner_uid, $article_labels); |
2c08214a | 1044 | |
68cccafc | 1045 | _debug("looking for enclosures...", $debug_enabled); |
2c08214a AD |
1046 | |
1047 | // enclosures | |
1048 | ||
1049 | $enclosures = array(); | |
1050 | ||
19b3992b | 1051 | $encs = $item->get_enclosures(); |
2c08214a | 1052 | |
19b3992b AD |
1053 | if (is_array($encs)) { |
1054 | foreach ($encs as $e) { | |
1055 | $e_item = array( | |
86e53429 AD |
1056 | rewrite_relative_url($site_url, $e->link), |
1057 | $e->type, $e->length, $e->title, $e->width, $e->height); | |
2c08214a | 1058 | array_push($enclosures, $e_item); |
2c08214a AD |
1059 | } |
1060 | } | |
1061 | ||
388d4dfa AD |
1062 | if ($cache_images && is_writable(CACHE_DIR . '/images')) |
1063 | cache_enclosures($enclosures, $site_url, $debug_enabled); | |
1064 | ||
2c08214a | 1065 | if ($debug_enabled) { |
68cccafc | 1066 | _debug("article enclosures:", $debug_enabled); |
2c08214a AD |
1067 | print_r($enclosures); |
1068 | } | |
1069 | ||
dab229cd | 1070 | //db_query("BEGIN"); |
2c08214a | 1071 | |
5c54e683 AD |
1072 | // debugging |
1073 | // db_query("DELETE FROM ttrss_enclosures WHERE post_id = '$entry_ref_id'"); | |
1074 | ||
2c08214a | 1075 | foreach ($enclosures as $enc) { |
a42c55f0 AD |
1076 | $enc_url = db_escape_string($enc[0]); |
1077 | $enc_type = db_escape_string($enc[1]); | |
1078 | $enc_dur = db_escape_string($enc[2]); | |
5c54e683 | 1079 | $enc_title = db_escape_string($enc[3]); |
523bd90b FE |
1080 | $enc_width = intval($enc[4]); |
1081 | $enc_height = intval($enc[5]); | |
2c08214a | 1082 | |
a42c55f0 | 1083 | $result = db_query("SELECT id FROM ttrss_enclosures |
2c08214a AD |
1084 | WHERE content_url = '$enc_url' AND post_id = '$entry_ref_id'"); |
1085 | ||
1086 | if (db_num_rows($result) == 0) { | |
a42c55f0 | 1087 | db_query("INSERT INTO ttrss_enclosures |
523bd90b FE |
1088 | (content_url, content_type, title, duration, post_id, width, height) VALUES |
1089 | ('$enc_url', '$enc_type', '$enc_title', '$enc_dur', '$entry_ref_id', $enc_width, $enc_height)"); | |
2c08214a AD |
1090 | } |
1091 | } | |
1092 | ||
dab229cd | 1093 | //db_query("COMMIT"); |
2c08214a AD |
1094 | |
1095 | // check for manual tags (we have to do it here since they're loaded from filters) | |
1096 | ||
1097 | foreach ($article_filters as $f) { | |
6aff7845 | 1098 | if ($f["type"] == "tag") { |
2c08214a | 1099 | |
6aff7845 | 1100 | $manual_tags = trim_array(explode(",", $f["param"])); |
2c08214a AD |
1101 | |
1102 | foreach ($manual_tags as $tag) { | |
1103 | if (tag_is_valid($tag)) { | |
1104 | array_push($entry_tags, $tag); | |
1105 | } | |
1106 | } | |
1107 | } | |
1108 | } | |
1109 | ||
1110 | // Skip boring tags | |
1111 | ||
6322ac79 | 1112 | $boring_tags = trim_array(explode(",", mb_strtolower(get_pref( |
2c08214a AD |
1113 | 'BLACKLISTED_TAGS', $owner_uid, ''), 'utf-8'))); |
1114 | ||
1115 | $filtered_tags = array(); | |
1116 | $tags_to_cache = array(); | |
1117 | ||
1118 | if ($entry_tags && is_array($entry_tags)) { | |
1119 | foreach ($entry_tags as $tag) { | |
1120 | if (array_search($tag, $boring_tags) === false) { | |
1121 | array_push($filtered_tags, $tag); | |
1122 | } | |
1123 | } | |
1124 | } | |
1125 | ||
1126 | $filtered_tags = array_unique($filtered_tags); | |
1127 | ||
1128 | if ($debug_enabled) { | |
68cccafc | 1129 | _debug("filtered article tags:", $debug_enabled); |
2c08214a AD |
1130 | print_r($filtered_tags); |
1131 | } | |
1132 | ||
1133 | // Save article tags in the database | |
1134 | ||
1135 | if (count($filtered_tags) > 0) { | |
1136 | ||
dab229cd | 1137 | //db_query("BEGIN"); |
2c08214a AD |
1138 | |
1139 | foreach ($filtered_tags as $tag) { | |
1140 | ||
1141 | $tag = sanitize_tag($tag); | |
a42c55f0 | 1142 | $tag = db_escape_string($tag); |
2c08214a AD |
1143 | |
1144 | if (!tag_is_valid($tag)) continue; | |
1145 | ||
a42c55f0 | 1146 | $result = db_query("SELECT id FROM ttrss_tags |
2c08214a AD |
1147 | WHERE tag_name = '$tag' AND post_int_id = '$entry_int_id' AND |
1148 | owner_uid = '$owner_uid' LIMIT 1"); | |
1149 | ||
1150 | if ($result && db_num_rows($result) == 0) { | |
1151 | ||
a42c55f0 | 1152 | db_query("INSERT INTO ttrss_tags |
2c08214a AD |
1153 | (owner_uid,tag_name,post_int_id) |
1154 | VALUES ('$owner_uid','$tag', '$entry_int_id')"); | |
1155 | } | |
1156 | ||
1157 | array_push($tags_to_cache, $tag); | |
1158 | } | |
1159 | ||
1160 | /* update the cache */ | |
1161 | ||
1162 | $tags_to_cache = array_unique($tags_to_cache); | |
1163 | ||
a42c55f0 | 1164 | $tags_str = db_escape_string(join(",", $tags_to_cache)); |
2c08214a | 1165 | |
a42c55f0 | 1166 | db_query("UPDATE ttrss_user_entries |
2c08214a AD |
1167 | SET tag_cache = '$tags_str' WHERE ref_id = '$entry_ref_id' |
1168 | AND owner_uid = $owner_uid"); | |
1169 | ||
dab229cd | 1170 | //db_query("COMMIT"); |
2c08214a AD |
1171 | } |
1172 | ||
68cccafc | 1173 | _debug("article processed", $debug_enabled); |
2c08214a AD |
1174 | } |
1175 | ||
68cccafc | 1176 | _debug("purging feed...", $debug_enabled); |
2c08214a | 1177 | |
a42c55f0 | 1178 | purge_feed($feed, 0, $debug_enabled); |
2c08214a | 1179 | |
a42c55f0 | 1180 | db_query("UPDATE ttrss_feeds |
f2798eb6 | 1181 | SET last_updated = NOW(), last_error = '' WHERE id = '$feed'"); |
2c08214a | 1182 | |
a42c55f0 | 1183 | // db_query("COMMIT"); |
2c08214a AD |
1184 | |
1185 | } else { | |
1186 | ||
a42c55f0 | 1187 | $error_msg = db_escape_string(mb_substr($rss->error(), 0, 245)); |
2c08214a | 1188 | |
4ad04ee2 AD |
1189 | _debug("fetch error: $error_msg", $debug_enabled); |
1190 | ||
1191 | if (count($rss->errors()) > 1) { | |
1192 | foreach ($rss->errors() as $error) { | |
1193 | _debug("+ $error"); | |
1194 | } | |
1195 | } | |
2c08214a | 1196 | |
6322ac79 | 1197 | db_query( |
2c08214a | 1198 | "UPDATE ttrss_feeds SET last_error = '$error_msg', |
88edaa93 | 1199 | last_updated = NOW() WHERE id = '$feed'"); |
2c08214a | 1200 | |
88edaa93 | 1201 | unset($rss); |
7b55001e | 1202 | return; |
88edaa93 | 1203 | } |
2c08214a | 1204 | |
68cccafc | 1205 | _debug("done", $debug_enabled); |
88edaa93 | 1206 | |
7b55001e | 1207 | return true; |
2c08214a AD |
1208 | } |
1209 | ||
388d4dfa AD |
1210 | function cache_enclosures($enclosures, $site_url, $debug) { |
1211 | foreach ($enclosures as $enc) { | |
1212 | ||
1213 | if (preg_match("/(image|audio|video)/", $enc[1])) { | |
1214 | ||
1215 | $src = rewrite_relative_url($site_url, $enc[0]); | |
1216 | ||
1217 | $local_filename = CACHE_DIR . "/images/" . sha1($src); | |
1218 | ||
1219 | if ($debug) _debug("cache_enclosures: downloading: $src to $local_filename"); | |
1220 | ||
1221 | if (!file_exists($local_filename)) { | |
1222 | $file_content = fetch_file_contents($src); | |
1223 | ||
1224 | if ($file_content && strlen($file_content) > _MIN_CACHE_FILE_SIZE) { | |
1225 | file_put_contents($local_filename, $file_content); | |
1226 | } | |
1227 | } else { | |
1228 | touch($local_filename); | |
1229 | } | |
1230 | } | |
1231 | } | |
1232 | } | |
1233 | ||
41bead9b | 1234 | function cache_media($html, $site_url, $debug) { |
3c696512 AD |
1235 | libxml_use_internal_errors(true); |
1236 | ||
1237 | $charset_hack = '<head> | |
1238 | <meta http-equiv="Content-Type" content="text/html; charset=utf-8"/> | |
1239 | </head>'; | |
1240 | ||
1241 | $doc = new DOMDocument(); | |
1242 | $doc->loadHTML($charset_hack . $html); | |
1243 | $xpath = new DOMXPath($doc); | |
1244 | ||
388d4dfa | 1245 | $entries = $xpath->query('(//img[@src])|(//video/source[@src])|(//audio/source[@src])'); |
3c696512 AD |
1246 | |
1247 | foreach ($entries as $entry) { | |
5edd605a | 1248 | if ($entry->hasAttribute('src') && strpos($entry->getAttribute('src'), "data:") !== 0) { |
3c696512 AD |
1249 | $src = rewrite_relative_url($site_url, $entry->getAttribute('src')); |
1250 | ||
41bead9b | 1251 | $local_filename = CACHE_DIR . "/images/" . sha1($src); |
3c696512 | 1252 | |
41bead9b | 1253 | if ($debug) _debug("cache_media: downloading: $src to $local_filename"); |
3c696512 AD |
1254 | |
1255 | if (!file_exists($local_filename)) { | |
1256 | $file_content = fetch_file_contents($src); | |
1257 | ||
41bead9b | 1258 | if ($file_content && strlen($file_content) > _MIN_CACHE_FILE_SIZE) { |
3c696512 AD |
1259 | file_put_contents($local_filename, $file_content); |
1260 | } | |
4a27966e J |
1261 | } else { |
1262 | touch($local_filename); | |
3c696512 | 1263 | } |
3c696512 AD |
1264 | } |
1265 | } | |
3c696512 AD |
1266 | } |
1267 | ||
a42c55f0 | 1268 | function expire_error_log($debug) { |
e2261e17 AD |
1269 | if ($debug) _debug("Removing old error log entries..."); |
1270 | ||
1271 | if (DB_TYPE == "pgsql") { | |
a42c55f0 | 1272 | db_query("DELETE FROM ttrss_error_log |
e2261e17 AD |
1273 | WHERE created_at < NOW() - INTERVAL '7 days'"); |
1274 | } else { | |
a42c55f0 | 1275 | db_query("DELETE FROM ttrss_error_log |
e2261e17 AD |
1276 | WHERE created_at < DATE_SUB(NOW(), INTERVAL 7 DAY)"); |
1277 | } | |
1278 | ||
1279 | } | |
1280 | ||
2a91b6ff | 1281 | function expire_lock_files($debug) { |
65465085 | 1282 | //if ($debug) _debug("Removing old lock files..."); |
2a91b6ff AD |
1283 | |
1284 | $num_deleted = 0; | |
1285 | ||
1286 | if (is_writable(LOCK_DIRECTORY)) { | |
1287 | $files = glob(LOCK_DIRECTORY . "/*.lock"); | |
1288 | ||
1289 | if ($files) { | |
1290 | foreach ($files as $file) { | |
11344971 | 1291 | if (!file_is_locked(basename($file)) && time() - filemtime($file) > 86400*2) { |
2a91b6ff AD |
1292 | unlink($file); |
1293 | ++$num_deleted; | |
1294 | } | |
1295 | } | |
1296 | } | |
1297 | } | |
1298 | ||
65465085 | 1299 | if ($debug) _debug("Removed $num_deleted old lock files."); |
2a91b6ff AD |
1300 | } |
1301 | ||
3c696512 | 1302 | function expire_cached_files($debug) { |
3306daec | 1303 | foreach (array("simplepie", "images", "export", "upload") as $dir) { |
3c696512 | 1304 | $cache_dir = CACHE_DIR . "/$dir"; |
2c08214a | 1305 | |
65465085 | 1306 | // if ($debug) _debug("Expiring $cache_dir"); |
2c08214a | 1307 | |
3c696512 AD |
1308 | $num_deleted = 0; |
1309 | ||
1310 | if (is_writable($cache_dir)) { | |
1311 | $files = glob("$cache_dir/*"); | |
1312 | ||
2a91b6ff | 1313 | if ($files) { |
2ab20c31 AD |
1314 | foreach ($files as $file) { |
1315 | if (time() - filemtime($file) > 86400*7) { | |
1316 | unlink($file); | |
3c696512 | 1317 | |
2ab20c31 AD |
1318 | ++$num_deleted; |
1319 | } | |
3c696512 AD |
1320 | } |
1321 | } | |
2a91b6ff | 1322 | } |
3c696512 | 1323 | |
65465085 | 1324 | if ($debug) _debug("$cache_dir: removed $num_deleted files."); |
3c696512 AD |
1325 | } |
1326 | } | |
2c08214a | 1327 | |
a3e0bdcf AD |
1328 | /** |
1329 | * Source: http://www.php.net/manual/en/function.parse-url.php#104527 | |
1330 | * Returns the url query as associative array | |
1331 | * | |
1332 | * @param string query | |
1333 | * @return array params | |
1334 | */ | |
1335 | function convertUrlQuery($query) { | |
1336 | $queryParts = explode('&', $query); | |
1337 | ||
1338 | $params = array(); | |
1339 | ||
1340 | foreach ($queryParts as $param) { | |
1341 | $item = explode('=', $param); | |
1342 | $params[$item[0]] = $item[1]; | |
1343 | } | |
1344 | ||
1345 | return $params; | |
1346 | } | |
92c14e9d | 1347 | |
7b55001e | 1348 | function get_article_filters($filters, $title, $content, $link, $author, $tags, &$matched_rules = false) { |
92c14e9d AD |
1349 | $matches = array(); |
1350 | ||
1351 | foreach ($filters as $filter) { | |
1352 | $match_any_rule = $filter["match_any_rule"]; | |
a3a896a1 | 1353 | $inverse = $filter["inverse"]; |
92c14e9d AD |
1354 | $filter_match = false; |
1355 | ||
1356 | foreach ($filter["rules"] as $rule) { | |
1357 | $match = false; | |
ffa1bd7b | 1358 | $reg_exp = str_replace('/', '\/', $rule["reg_exp"]); |
a3a896a1 | 1359 | $rule_inverse = $rule["inverse"]; |
92c14e9d AD |
1360 | |
1361 | if (!$reg_exp) | |
1362 | continue; | |
1363 | ||
1364 | switch ($rule["type"]) { | |
1365 | case "title": | |
e3dceca6 | 1366 | $match = @preg_match("/$reg_exp/iu", $title); |
92c14e9d AD |
1367 | break; |
1368 | case "content": | |
d03ae73e AD |
1369 | // we don't need to deal with multiline regexps |
1370 | $content = preg_replace("/[\r\n\t]/", "", $content); | |
1371 | ||
e3dceca6 | 1372 | $match = @preg_match("/$reg_exp/iu", $content); |
92c14e9d AD |
1373 | break; |
1374 | case "both": | |
d03ae73e AD |
1375 | // we don't need to deal with multiline regexps |
1376 | $content = preg_replace("/[\r\n\t]/", "", $content); | |
1377 | ||
e3dceca6 | 1378 | $match = (@preg_match("/$reg_exp/iu", $title) || @preg_match("/$reg_exp/iu", $content)); |
92c14e9d AD |
1379 | break; |
1380 | case "link": | |
e3dceca6 | 1381 | $match = @preg_match("/$reg_exp/iu", $link); |
92c14e9d AD |
1382 | break; |
1383 | case "author": | |
e3dceca6 | 1384 | $match = @preg_match("/$reg_exp/iu", $author); |
92c14e9d AD |
1385 | break; |
1386 | case "tag": | |
7b80b5e1 | 1387 | foreach ($tags as $tag) { |
e3dceca6 | 1388 | if (@preg_match("/$reg_exp/iu", $tag)) { |
7b80b5e1 DK |
1389 | $match = true; |
1390 | break; | |
1391 | } | |
1392 | } | |
92c14e9d AD |
1393 | break; |
1394 | } | |
1395 | ||
a3a896a1 AD |
1396 | if ($rule_inverse) $match = !$match; |
1397 | ||
92c14e9d AD |
1398 | if ($match_any_rule) { |
1399 | if ($match) { | |
1400 | $filter_match = true; | |
1401 | break; | |
1402 | } | |
1403 | } else { | |
1404 | $filter_match = $match; | |
1405 | if (!$match) { | |
1406 | break; | |
1407 | } | |
1408 | } | |
1409 | } | |
1410 | ||
a3a896a1 AD |
1411 | if ($inverse) $filter_match = !$filter_match; |
1412 | ||
92c14e9d | 1413 | if ($filter_match) { |
557d86fe AD |
1414 | if (is_array($matched_rules)) array_push($matched_rules, $rule); |
1415 | ||
92c14e9d AD |
1416 | foreach ($filter["actions"] AS $action) { |
1417 | array_push($matches, $action); | |
5e736e45 AD |
1418 | |
1419 | // if Stop action encountered, perform no further processing | |
fd3e5e8d | 1420 | if (isset($action["type"]) && $action["type"] == "stop") return $matches; |
92c14e9d AD |
1421 | } |
1422 | } | |
1423 | } | |
1424 | ||
1425 | return $matches; | |
1426 | } | |
1427 | ||
1428 | function find_article_filter($filters, $filter_name) { | |
1429 | foreach ($filters as $f) { | |
1430 | if ($f["type"] == $filter_name) { | |
1431 | return $f; | |
1432 | }; | |
1433 | } | |
1434 | return false; | |
1435 | } | |
1436 | ||
1437 | function find_article_filters($filters, $filter_name) { | |
1438 | $results = array(); | |
1439 | ||
1440 | foreach ($filters as $f) { | |
1441 | if ($f["type"] == $filter_name) { | |
1442 | array_push($results, $f); | |
1443 | }; | |
1444 | } | |
1445 | return $results; | |
1446 | } | |
1447 | ||
1448 | function calculate_article_score($filters) { | |
1449 | $score = 0; | |
1450 | ||
1451 | foreach ($filters as $f) { | |
1452 | if ($f["type"] == "score") { | |
1453 | $score += $f["param"]; | |
1454 | }; | |
1455 | } | |
1456 | return $score; | |
1457 | } | |
1458 | ||
b24504b1 AD |
1459 | function labels_contains_caption($labels, $caption) { |
1460 | foreach ($labels as $label) { | |
1461 | if ($label[1] == $caption) { | |
1462 | return true; | |
1463 | } | |
1464 | } | |
1465 | ||
1466 | return false; | |
1467 | } | |
1468 | ||
a42c55f0 | 1469 | function assign_article_to_label_filters($id, $filters, $owner_uid, $article_labels) { |
92c14e9d AD |
1470 | foreach ($filters as $f) { |
1471 | if ($f["type"] == "label") { | |
b24504b1 | 1472 | if (!labels_contains_caption($article_labels, $f["param"])) { |
a42c55f0 | 1473 | label_add_article($id, $f["param"], $owner_uid); |
b24504b1 AD |
1474 | } |
1475 | } | |
92c14e9d AD |
1476 | } |
1477 | } | |
87764a50 | 1478 | |
87d7e850 AD |
1479 | function make_guid_from_title($title) { |
1480 | return preg_replace("/[ \"\',.:;]/", "-", | |
1481 | mb_strtolower(strip_tags($title), 'utf-8')); | |
1482 | } | |
1483 | ||
168cf351 AD |
1484 | function cleanup_counters_cache($debug) { |
1485 | $result = db_query("DELETE FROM ttrss_counters_cache | |
1486 | WHERE feed_id > 0 AND | |
1487 | (SELECT COUNT(id) FROM ttrss_feeds WHERE | |
1488 | id = feed_id AND | |
1489 | ttrss_counters_cache.owner_uid = ttrss_feeds.owner_uid) = 0"); | |
1490 | $frows = db_affected_rows($result); | |
1491 | ||
1492 | $result = db_query("DELETE FROM ttrss_cat_counters_cache | |
1493 | WHERE feed_id > 0 AND | |
1494 | (SELECT COUNT(id) FROM ttrss_feed_categories WHERE | |
1495 | id = feed_id AND | |
1496 | ttrss_cat_counters_cache.owner_uid = ttrss_feed_categories.owner_uid) = 0"); | |
1497 | $crows = db_affected_rows($result); | |
1498 | ||
7b55001e | 1499 | if ($debug) _debug("Removed $frows (feeds) $crows (cats) orphaned counter cache entries."); |
168cf351 AD |
1500 | } |
1501 | ||
5cbd1fe8 AD |
1502 | function housekeeping_user($owner_uid) { |
1503 | $tmph = new PluginHost(); | |
1504 | ||
1505 | load_user_plugins($owner_uid, $tmph); | |
1506 | ||
1507 | $tmph->run_hooks(PluginHost::HOOK_HOUSE_KEEPING, "hook_house_keeping", ""); | |
1508 | } | |
1509 | ||
e2cf81e2 AD |
1510 | function housekeeping_common($debug) { |
1511 | expire_cached_files($debug); | |
1512 | expire_lock_files($debug); | |
1513 | expire_error_log($debug); | |
1514 | ||
1515 | $count = update_feedbrowser_cache(); | |
1516 | _debug("Feedbrowser updated, $count feeds processed."); | |
1517 | ||
a230bf88 | 1518 | Article::purge_orphans( true); |
168cf351 | 1519 | cleanup_counters_cache($debug); |
e2cf81e2 | 1520 | |
9b736a20 AD |
1521 | //$rc = cleanup_tags( 14, 50000); |
1522 | //_debug("Cleaned $rc cached tags."); | |
8e470220 | 1523 | |
00f22824 | 1524 | PluginHost::getInstance()->run_hooks(PluginHost::HOOK_HOUSE_KEEPING, "hook_house_keeping", ""); |
e2cf81e2 | 1525 | } |
ea79a0e0 | 1526 | |
a230bf88 AD |
1527 | function check_feed_favicon($site_url, $feed) { |
1528 | # print "FAVICON [$site_url]: $favicon_url\n"; | |
1529 | ||
1530 | $icon_file = ICONS_DIR . "/$feed.ico"; | |
1531 | ||
1532 | if (!file_exists($icon_file)) { | |
1533 | $favicon_url = get_favicon_url($site_url); | |
1534 | ||
1535 | if ($favicon_url) { | |
1536 | // Limiting to "image" type misses those served with text/plain | |
1537 | $contents = fetch_file_contents($favicon_url); // , "image"); | |
1538 | ||
1539 | if ($contents) { | |
1540 | // Crude image type matching. | |
1541 | // Patterns gleaned from the file(1) source code. | |
1542 | if (preg_match('/^\x00\x00\x01\x00/', $contents)) { | |
1543 | // 0 string \000\000\001\000 MS Windows icon resource | |
1544 | //error_log("check_feed_favicon: favicon_url=$favicon_url isa MS Windows icon resource"); | |
1545 | } | |
1546 | elseif (preg_match('/^GIF8/', $contents)) { | |
1547 | // 0 string GIF8 GIF image data | |
1548 | //error_log("check_feed_favicon: favicon_url=$favicon_url isa GIF image"); | |
1549 | } | |
1550 | elseif (preg_match('/^\x89PNG\x0d\x0a\x1a\x0a/', $contents)) { | |
1551 | // 0 string \x89PNG\x0d\x0a\x1a\x0a PNG image data | |
1552 | //error_log("check_feed_favicon: favicon_url=$favicon_url isa PNG image"); | |
1553 | } | |
1554 | elseif (preg_match('/^\xff\xd8/', $contents)) { | |
1555 | // 0 beshort 0xffd8 JPEG image data | |
1556 | //error_log("check_feed_favicon: favicon_url=$favicon_url isa JPG image"); | |
1557 | } | |
1558 | else { | |
1559 | //error_log("check_feed_favicon: favicon_url=$favicon_url isa UNKNOWN type"); | |
1560 | $contents = ""; | |
1561 | } | |
1562 | } | |
1563 | ||
1564 | if ($contents) { | |
1565 | $fp = @fopen($icon_file, "w"); | |
1566 | ||
1567 | if ($fp) { | |
1568 | fwrite($fp, $contents); | |
1569 | fclose($fp); | |
1570 | chmod($icon_file, 0644); | |
1571 | } | |
1572 | } | |
1573 | } | |
1574 | return $icon_file; | |
1575 | } | |
1576 | } |