]>
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); |
a3d50184 | 5 | define_default('_MIN_CACHE_IMAGE_SIZE', 1024); |
09e8bdfd | 6 | |
eb16bd9f | 7 | function calculate_article_hash($article, $pluginhost) { |
af244f92 AD |
8 | $tmp = ""; |
9 | ||
10 | foreach ($article as $k => $v) { | |
11 | if ($k != "feed" && isset($v)) { | |
12 | $tmp .= sha1("$k:" . (is_array($v) ? implode(",", $v) : $v)); | |
13 | } | |
14 | } | |
15 | ||
eb16bd9f | 16 | return sha1(implode(",", $pluginhost->get_plugin_names()) . $tmp); |
b1840673 AD |
17 | } |
18 | ||
6322ac79 | 19 | function update_feedbrowser_cache() { |
79178062 | 20 | |
a42c55f0 | 21 | $result = db_query("SELECT feed_url, site_url, title, COUNT(id) AS subscribers |
79178062 AD |
22 | FROM ttrss_feeds WHERE (SELECT COUNT(id) = 0 FROM ttrss_feeds AS tf |
23 | WHERE tf.feed_url = ttrss_feeds.feed_url | |
24 | AND (private IS true OR auth_login != '' OR auth_pass != '' OR feed_url LIKE '%:%@%/%')) | |
25 | GROUP BY feed_url, site_url, title ORDER BY subscribers DESC LIMIT 1000"); | |
26 | ||
a42c55f0 | 27 | db_query("BEGIN"); |
79178062 | 28 | |
a42c55f0 | 29 | db_query("DELETE FROM ttrss_feedbrowser_cache"); |
79178062 AD |
30 | |
31 | $count = 0; | |
32 | ||
33 | while ($line = db_fetch_assoc($result)) { | |
a42c55f0 AD |
34 | $subscribers = db_escape_string($line["subscribers"]); |
35 | $feed_url = db_escape_string($line["feed_url"]); | |
36 | $title = db_escape_string($line["title"]); | |
37 | $site_url = db_escape_string($line["site_url"]); | |
79178062 | 38 | |
a42c55f0 | 39 | $tmp_result = db_query("SELECT subscribers FROM |
79178062 AD |
40 | ttrss_feedbrowser_cache WHERE feed_url = '$feed_url'"); |
41 | ||
42 | if (db_num_rows($tmp_result) == 0) { | |
43 | ||
a42c55f0 | 44 | db_query("INSERT INTO ttrss_feedbrowser_cache |
79178062 AD |
45 | (feed_url, site_url, title, subscribers) VALUES ('$feed_url', |
46 | '$site_url', '$title', '$subscribers')"); | |
47 | ||
48 | ++$count; | |
49 | ||
50 | } | |
51 | ||
52 | } | |
53 | ||
a42c55f0 | 54 | db_query("COMMIT"); |
79178062 AD |
55 | |
56 | return $count; | |
57 | ||
58 | } | |
59 | ||
60 | ||
2c08214a AD |
61 | /** |
62 | * Update a feed batch. | |
63 | * Used by daemons to update n feeds by run. | |
64 | * Only update feed needing a update, and not being processed | |
65 | * by another process. | |
66 | * | |
67 | * @param mixed $link Database link | |
68 | * @param integer $limit Maximum number of feeds in update batch. Default to DAEMON_FEED_LIMIT. | |
69 | * @param boolean $from_http Set to true if you call this function from http to disable cli specific code. | |
70 | * @param boolean $debug Set to false to disable debug output. Default to true. | |
71 | * @return void | |
72 | */ | |
a42c55f0 | 73 | function update_daemon_common($limit = DAEMON_FEED_LIMIT, $from_http = false, $debug = true) { |
2c08214a AD |
74 | // Process all other feeds using last_updated and interval parameters |
75 | ||
6322ac79 | 76 | $schema_version = get_schema_version(); |
857efe49 AD |
77 | |
78 | if ($schema_version != SCHEMA_VERSION) { | |
79 | die("Schema version is wrong, please upgrade the database.\n"); | |
80 | } | |
81 | ||
61c1812f AD |
82 | define('PREFS_NO_CACHE', true); |
83 | ||
2c08214a | 84 | // Test if the user has loggued in recently. If not, it does not update its feeds. |
09e8bdfd | 85 | if (!SINGLE_USER_MODE && DAEMON_UPDATE_LOGIN_LIMIT > 0) { |
2c08214a AD |
86 | if (DB_TYPE == "pgsql") { |
87 | $login_thresh_qpart = "AND ttrss_users.last_login >= NOW() - INTERVAL '".DAEMON_UPDATE_LOGIN_LIMIT." days'"; | |
88 | } else { | |
89 | $login_thresh_qpart = "AND ttrss_users.last_login >= DATE_SUB(NOW(), INTERVAL ".DAEMON_UPDATE_LOGIN_LIMIT." DAY)"; | |
90 | } | |
91 | } else { | |
92 | $login_thresh_qpart = ""; | |
93 | } | |
94 | ||
537eca40 | 95 | // Test if the feed need a update (update interval exceeded). |
2c08214a AD |
96 | if (DB_TYPE == "pgsql") { |
97 | $update_limit_qpart = "AND (( | |
98 | ttrss_feeds.update_interval = 0 | |
ee0542ce | 99 | AND ttrss_user_prefs.value != '-1' |
2c08214a AD |
100 | AND ttrss_feeds.last_updated < NOW() - CAST((ttrss_user_prefs.value || ' minutes') AS INTERVAL) |
101 | ) OR ( | |
102 | ttrss_feeds.update_interval > 0 | |
103 | AND ttrss_feeds.last_updated < NOW() - CAST((ttrss_feeds.update_interval || ' minutes') AS INTERVAL) | |
f08426e3 AD |
104 | ) OR (ttrss_feeds.last_updated IS NULL |
105 | AND ttrss_user_prefs.value != '-1') | |
106 | OR (last_updated = '1970-01-01 00:00:00' | |
107 | AND ttrss_user_prefs.value != '-1'))"; | |
2c08214a AD |
108 | } else { |
109 | $update_limit_qpart = "AND (( | |
110 | ttrss_feeds.update_interval = 0 | |
ee0542ce | 111 | AND ttrss_user_prefs.value != '-1' |
2c08214a AD |
112 | AND ttrss_feeds.last_updated < DATE_SUB(NOW(), INTERVAL CONVERT(ttrss_user_prefs.value, SIGNED INTEGER) MINUTE) |
113 | ) OR ( | |
114 | ttrss_feeds.update_interval > 0 | |
115 | AND ttrss_feeds.last_updated < DATE_SUB(NOW(), INTERVAL ttrss_feeds.update_interval MINUTE) | |
f08426e3 AD |
116 | ) OR (ttrss_feeds.last_updated IS NULL |
117 | AND ttrss_user_prefs.value != '-1') | |
118 | OR (last_updated = '1970-01-01 00:00:00' | |
119 | AND ttrss_user_prefs.value != '-1'))"; | |
2c08214a AD |
120 | } |
121 | ||
122 | // Test if feed is currently being updated by another process. | |
123 | if (DB_TYPE == "pgsql") { | |
566417c4 | 124 | $updstart_thresh_qpart = "AND (ttrss_feeds.last_update_started IS NULL OR ttrss_feeds.last_update_started < NOW() - INTERVAL '10 minutes')"; |
2c08214a | 125 | } else { |
566417c4 | 126 | $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 |
127 | } |
128 | ||
129 | // Test if there is a limit to number of updated feeds | |
130 | $query_limit = ""; | |
131 | if($limit) $query_limit = sprintf("LIMIT %d", $limit); | |
132 | ||
fce451a4 | 133 | $query = "SELECT DISTINCT ttrss_feeds.feed_url, ttrss_feeds.last_updated |
2c08214a AD |
134 | FROM |
135 | ttrss_feeds, ttrss_users, ttrss_user_prefs | |
f4ae0f05 | 136 | WHERE |
2c08214a | 137 | ttrss_feeds.owner_uid = ttrss_users.id |
f08426e3 | 138 | AND ttrss_user_prefs.profile IS NULL |
2c08214a AD |
139 | AND ttrss_users.id = ttrss_user_prefs.owner_uid |
140 | AND ttrss_user_prefs.pref_name = 'DEFAULT_UPDATE_INTERVAL' | |
141 | $login_thresh_qpart $update_limit_qpart | |
1c4421fc | 142 | $updstart_thresh_qpart |
fce451a4 AD |
143 | ORDER BY last_updated $query_limit"; |
144 | ||
145 | // We search for feed needing update. | |
146 | $result = db_query($query); | |
2c08214a | 147 | |
ca5ff2d9 | 148 | if($debug) _debug(sprintf("Scheduled %d feeds to update...", db_num_rows($result))); |
2c08214a AD |
149 | |
150 | // Here is a little cache magic in order to minimize risk of double feed updates. | |
151 | $feeds_to_update = array(); | |
152 | while ($line = db_fetch_assoc($result)) { | |
a42c55f0 | 153 | array_push($feeds_to_update, db_escape_string($line['feed_url'])); |
2c08214a AD |
154 | } |
155 | ||
156 | // We update the feed last update started date before anything else. | |
157 | // There is no lag due to feed contents downloads | |
158 | // It prevent an other process to update the same feed. | |
1c4421fc AD |
159 | |
160 | if(count($feeds_to_update) > 0) { | |
161 | $feeds_quoted = array(); | |
162 | ||
163 | foreach ($feeds_to_update as $feed) { | |
a42c55f0 | 164 | array_push($feeds_quoted, "'" . db_escape_string($feed) . "'"); |
1c4421fc AD |
165 | } |
166 | ||
a42c55f0 | 167 | db_query(sprintf("UPDATE ttrss_feeds SET last_update_started = NOW() |
1c4421fc | 168 | WHERE feed_url IN (%s)", implode(',', $feeds_quoted))); |
2c08214a AD |
169 | } |
170 | ||
8292d05b | 171 | $nf = 0; |
2d9c5684 | 172 | $bstarted = microtime(true); |
8292d05b | 173 | |
2c08214a | 174 | // For each feed, we call the feed update function. |
1c4421fc AD |
175 | foreach ($feeds_to_update as $feed) { |
176 | if($debug) _debug("Base feed: $feed"); | |
2c08214a | 177 | |
a42c55f0 | 178 | //update_rss_feed($line["id"], true); |
2c08214a | 179 | |
1c4421fc AD |
180 | // since we have the data cached, we can deal with other feeds with the same url |
181 | ||
15c762be | 182 | $tmp_result = db_query("SELECT DISTINCT ttrss_feeds.id,last_updated,ttrss_feeds.owner_uid |
ee0542ce AD |
183 | FROM ttrss_feeds, ttrss_users, ttrss_user_prefs WHERE |
184 | ttrss_user_prefs.owner_uid = ttrss_feeds.owner_uid AND | |
185 | ttrss_users.id = ttrss_user_prefs.owner_uid AND | |
186 | ttrss_user_prefs.pref_name = 'DEFAULT_UPDATE_INTERVAL' AND | |
f08426e3 | 187 | ttrss_user_prefs.profile IS NULL AND |
a42c55f0 | 188 | feed_url = '".db_escape_string($feed)."' AND |
ee0542ce AD |
189 | (ttrss_feeds.update_interval > 0 OR |
190 | ttrss_user_prefs.value != '-1') | |
1c4421fc | 191 | $login_thresh_qpart |
5929a0c1 | 192 | ORDER BY ttrss_feeds.id $query_limit"); |
1c4421fc AD |
193 | |
194 | if (db_num_rows($tmp_result) > 0) { | |
88edaa93 AD |
195 | $rss = false; |
196 | ||
1c4421fc | 197 | while ($tline = db_fetch_assoc($tmp_result)) { |
15c762be | 198 | if($debug) _debug(" => " . $tline["last_updated"] . ", " . $tline["id"] . " " . $tline["owner_uid"]); |
f08426e3 | 199 | |
2d9c5684 | 200 | $fstarted = microtime(true); |
f77c7203 | 201 | $rss = update_rss_feed($tline["id"], true, false); |
4f71d743 | 202 | _debug_suppress(false); |
2d9c5684 AD |
203 | |
204 | _debug(sprintf(" %.4f (sec)", microtime(true) - $fstarted)); | |
205 | ||
8292d05b | 206 | ++$nf; |
1c4421fc AD |
207 | } |
208 | } | |
2c08214a AD |
209 | } |
210 | ||
2d9c5684 AD |
211 | if ($nf > 0) { |
212 | _debug(sprintf("Processed %d feeds in %.4f (sec), %.4f (sec/feed avg)", $nf, | |
213 | microtime(true) - $bstarted, (microtime(true) - $bstarted) / $nf)); | |
214 | } | |
215 | ||
f1611683 AD |
216 | require_once "digest.php"; |
217 | ||
2c08214a | 218 | // Send feed digests by email if needed. |
a42c55f0 | 219 | send_headlines_digests($debug); |
2c08214a | 220 | |
8292d05b AD |
221 | return $nf; |
222 | ||
2c08214a AD |
223 | } // function update_daemon_common |
224 | ||
87764a50 | 225 | // ignore_daemon is not used |
88edaa93 | 226 | function update_rss_feed($feed, $ignore_daemon = false, $no_cache = false, $rss = false) { |
2c08214a | 227 | |
2c08214a AD |
228 | $debug_enabled = defined('DAEMON_EXTENDED_DEBUG') || $_REQUEST['xdebug']; |
229 | ||
4f71d743 | 230 | _debug_suppress(!$debug_enabled); |
68cccafc | 231 | _debug("start", $debug_enabled); |
2c08214a | 232 | |
a42c55f0 | 233 | $result = db_query("SELECT id,update_interval,auth_login, |
5ba1ddd4 | 234 | feed_url,auth_pass,cache_images, |
5321e775 | 235 | mark_unread_on_update, owner_uid, |
90e5f4f1 AD |
236 | pubsub_state, auth_pass_encrypted, |
237 | (SELECT max(date_entered) FROM | |
238 | ttrss_entries, ttrss_user_entries where ref_id = id AND feed_id = '$feed') AS last_article_timestamp | |
87764a50 | 239 | FROM ttrss_feeds WHERE id = '$feed'"); |
2c08214a AD |
240 | |
241 | if (db_num_rows($result) == 0) { | |
68cccafc | 242 | _debug("feed $feed NOT FOUND/SKIPPED", $debug_enabled); |
2c08214a AD |
243 | return false; |
244 | } | |
245 | ||
90e5f4f1 | 246 | $last_article_timestamp = @strtotime(db_fetch_result($result, 0, "last_article_timestamp")); |
5d0d3887 AD |
247 | |
248 | if (defined('_DISABLE_HTTP_304')) | |
249 | $last_article_timestamp = 0; | |
250 | ||
2c08214a AD |
251 | $owner_uid = db_fetch_result($result, 0, "owner_uid"); |
252 | $mark_unread_on_update = sql_bool_to_bool(db_fetch_result($result, | |
253 | 0, "mark_unread_on_update")); | |
2c08214a | 254 | $pubsub_state = db_fetch_result($result, 0, "pubsub_state"); |
044cff2d AD |
255 | $auth_pass_encrypted = sql_bool_to_bool(db_fetch_result($result, |
256 | 0, "auth_pass_encrypted")); | |
2c08214a | 257 | |
a42c55f0 | 258 | db_query("UPDATE ttrss_feeds SET last_update_started = NOW() |
2c08214a AD |
259 | WHERE id = '$feed'"); |
260 | ||
261 | $auth_login = db_fetch_result($result, 0, "auth_login"); | |
262 | $auth_pass = db_fetch_result($result, 0, "auth_pass"); | |
263 | ||
044cff2d AD |
264 | if ($auth_pass_encrypted) { |
265 | require_once "crypt.php"; | |
266 | $auth_pass = decrypt_string($auth_pass); | |
267 | } | |
268 | ||
2c08214a AD |
269 | $cache_images = sql_bool_to_bool(db_fetch_result($result, 0, "cache_images")); |
270 | $fetch_url = db_fetch_result($result, 0, "feed_url"); | |
271 | ||
a42c55f0 | 272 | $feed = db_escape_string($feed); |
2c08214a | 273 | |
f074ffe9 | 274 | $date_feed_processed = date('Y-m-d H:i'); |
2c08214a | 275 | |
865a3ed6 | 276 | $cache_filename = CACHE_DIR . "/simplepie/" . sha1($fetch_url) . ".xml"; |
f074ffe9 | 277 | |
ee65bef4 AD |
278 | $pluginhost = new PluginHost(); |
279 | $pluginhost->set_debug($debug_enabled); | |
280 | $user_plugins = get_pref("_ENABLED_PLUGINS", $owner_uid); | |
281 | ||
282 | $pluginhost->load(PLUGINS, PluginHost::KIND_ALL); | |
283 | $pluginhost->load($user_plugins, PluginHost::KIND_USER, $owner_uid); | |
284 | $pluginhost->load_data(); | |
285 | ||
88edaa93 AD |
286 | if ($rss && is_object($rss) && get_class($rss) == "FeedParser") { |
287 | _debug("using previously initialized parser object"); | |
288 | } else { | |
289 | $rss_hash = false; | |
4f9cbdff | 290 | |
88edaa93 | 291 | $force_refetch = isset($_REQUEST["force_refetch"]); |
687a4f59 | 292 | |
95beaa14 | 293 | foreach ($pluginhost->get_hooks(PluginHost::HOOK_FETCH_FEED) as $plugin) { |
01465325 | 294 | $feed_data = $plugin->hook_fetch_feed($feed_data, $fetch_url, $owner_uid, $feed, $last_article_timestamp, $auth_login, $auth_pass); |
95beaa14 RA |
295 | } |
296 | ||
297 | // try cache | |
298 | if (!$feed_data && | |
299 | file_exists($cache_filename) && | |
88edaa93 AD |
300 | is_readable($cache_filename) && |
301 | !$auth_login && !$auth_pass && | |
302 | filemtime($cache_filename) > time() - 30) { | |
2c08214a | 303 | |
15d0528a | 304 | _debug("using local cache [$cache_filename].", $debug_enabled); |
be574731 | 305 | |
88edaa93 | 306 | @$feed_data = file_get_contents($cache_filename); |
52637d3b | 307 | |
88edaa93 AD |
308 | if ($feed_data) { |
309 | $rss_hash = sha1($feed_data); | |
310 | } | |
f074ffe9 | 311 | |
88edaa93 AD |
312 | } else { |
313 | _debug("local cache will not be used for this feed", $debug_enabled); | |
314 | } | |
ee65bef4 | 315 | |
95beaa14 | 316 | // fetch feed from source |
fd687300 AD |
317 | if (!$feed_data) { |
318 | _debug("fetching [$fetch_url]...", $debug_enabled); | |
319 | _debug("If-Modified-Since: ".gmdate('D, d M Y H:i:s \G\M\T', $last_article_timestamp), $debug_enabled); | |
96f98cb0 | 320 | |
fd687300 AD |
321 | $feed_data = fetch_file_contents($fetch_url, false, |
322 | $auth_login, $auth_pass, false, | |
323 | $no_cache ? FEED_FETCH_NO_CACHE_TIMEOUT : FEED_FETCH_TIMEOUT, | |
324 | $force_refetch ? 0 : $last_article_timestamp); | |
3f6f0857 | 325 | |
fd687300 | 326 | global $fetch_curl_used; |
3f6f0857 | 327 | |
fd687300 AD |
328 | if (!$fetch_curl_used) { |
329 | $tmp = @gzdecode($feed_data); | |
3f6f0857 | 330 | |
fd687300 AD |
331 | if ($tmp) $feed_data = $tmp; |
332 | } | |
1367bc3f | 333 | |
fd687300 | 334 | $feed_data = trim($feed_data); |
017401dd | 335 | |
fd687300 AD |
336 | _debug("fetch done.", $debug_enabled); |
337 | ||
d1f3fa97 | 338 | /* if ($feed_data) { |
fd687300 | 339 | $error = verify_feed_xml($feed_data); |
ebec81a6 | 340 | |
fd687300 AD |
341 | if ($error) { |
342 | _debug("error verifying XML, code: " . $error->code, $debug_enabled); | |
ebec81a6 | 343 | |
fd687300 AD |
344 | if ($error->code == 26) { |
345 | _debug("got error 26, trying to decode entities...", $debug_enabled); | |
ebec81a6 | 346 | |
fd687300 | 347 | $feed_data = html_entity_decode($feed_data, ENT_COMPAT, 'UTF-8'); |
ebec81a6 | 348 | |
fd687300 | 349 | $error = verify_feed_xml($feed_data); |
ebec81a6 | 350 | |
fd687300 AD |
351 | if ($error) $feed_data = ''; |
352 | } | |
ebec81a6 | 353 | } |
d1f3fa97 | 354 | } */ |
95beaa14 RA |
355 | |
356 | // cache vanilla feed data for re-use | |
357 | if ($feed_data && !$auth_pass && !$auth_login && is_writable(CACHE_DIR . "/simplepie")) { | |
358 | $new_rss_hash = sha1($feed_data); | |
359 | ||
360 | if ($new_rss_hash != $rss_hash) { | |
361 | _debug("saving $cache_filename", $debug_enabled); | |
362 | @file_put_contents($cache_filename, $feed_data); | |
363 | } | |
364 | } | |
4f9cbdff | 365 | } |
017401dd | 366 | |
fd687300 AD |
367 | if (!$feed_data) { |
368 | global $fetch_last_error; | |
369 | global $fetch_last_error_code; | |
f074ffe9 | 370 | |
fd687300 | 371 | _debug("unable to fetch: $fetch_last_error [$fetch_last_error_code]", $debug_enabled); |
f074ffe9 | 372 | |
fd687300 | 373 | $error_escaped = ''; |
7a01dc77 | 374 | |
fd687300 AD |
375 | // If-Modified-Since |
376 | if ($fetch_last_error_code != 304) { | |
377 | $error_escaped = db_escape_string($fetch_last_error); | |
378 | } else { | |
379 | _debug("source claims data not modified, nothing to do.", $debug_enabled); | |
380 | } | |
4f9cbdff | 381 | |
fd687300 AD |
382 | db_query( |
383 | "UPDATE ttrss_feeds SET last_error = '$error_escaped', | |
384 | last_updated = NOW() WHERE id = '$feed'"); | |
4f9cbdff | 385 | |
fd687300 AD |
386 | return; |
387 | } | |
f074ffe9 AD |
388 | } |
389 | ||
1ffe3391 | 390 | foreach ($pluginhost->get_hooks(PluginHost::HOOK_FEED_FETCHED) as $plugin) { |
6791af0c | 391 | $feed_data = $plugin->hook_feed_fetched($feed_data, $fetch_url, $owner_uid, $feed); |
017401dd AD |
392 | } |
393 | ||
5ef84097 AD |
394 | // set last update to now so if anything *simplepie* crashes later we won't be |
395 | // continuously failing on the same feed | |
5de40104 | 396 | //db_query("UPDATE ttrss_feeds SET last_updated = NOW() WHERE id = '$feed'"); |
5ef84097 | 397 | |
4f9cbdff | 398 | if (!$rss) { |
cd07592c AD |
399 | $rss = new FeedParser($feed_data); |
400 | $rss->init(); | |
5de51df7 AD |
401 | } |
402 | ||
5ddd2705 AD |
403 | if (DETECT_ARTICLE_LANGUAGE) { |
404 | require_once "lib/languagedetect/LanguageDetect.php"; | |
6b461797 | 405 | |
5ddd2705 AD |
406 | $lang = new Text_LanguageDetect(); |
407 | $lang->setNameMode(2); | |
408 | } | |
6b461797 | 409 | |
2c08214a AD |
410 | // print_r($rss); |
411 | ||
a42c55f0 | 412 | $feed = db_escape_string($feed); |
2c08214a | 413 | |
19b3992b | 414 | if (!$rss->error()) { |
2c08214a | 415 | |
d2a421e3 | 416 | // We use local pluginhost here because we need to load different per-user feed plugins |
1ffe3391 | 417 | $pluginhost->run_hooks(PluginHost::HOOK_FEED_PARSED, "hook_feed_parsed", $rss); |
4412b877 | 418 | |
68cccafc | 419 | _debug("processing feed data...", $debug_enabled); |
2c08214a | 420 | |
a42c55f0 | 421 | // db_query("BEGIN"); |
2c08214a | 422 | |
382268c6 AD |
423 | if (DB_TYPE == "pgsql") { |
424 | $favicon_interval_qpart = "favicon_last_checked < NOW() - INTERVAL '12 hour'"; | |
425 | } else { | |
426 | $favicon_interval_qpart = "favicon_last_checked < DATE_SUB(NOW(), INTERVAL 12 HOUR)"; | |
427 | } | |
428 | ||
36490f11 | 429 | $result = db_query("SELECT title,site_url,owner_uid,favicon_avg_color, |
382268c6 AD |
430 | (favicon_last_checked IS NULL OR $favicon_interval_qpart) AS |
431 | favicon_needs_check | |
2c08214a AD |
432 | FROM ttrss_feeds WHERE id = '$feed'"); |
433 | ||
434 | $registered_title = db_fetch_result($result, 0, "title"); | |
2c08214a | 435 | $orig_site_url = db_fetch_result($result, 0, "site_url"); |
382268c6 AD |
436 | $favicon_needs_check = sql_bool_to_bool(db_fetch_result($result, 0, |
437 | "favicon_needs_check")); | |
36490f11 | 438 | $favicon_avg_color = db_fetch_result($result, 0, "favicon_avg_color"); |
2c08214a AD |
439 | |
440 | $owner_uid = db_fetch_result($result, 0, "owner_uid"); | |
441 | ||
a42c55f0 | 442 | $site_url = db_escape_string(mb_substr(rewrite_relative_url($fetch_url, $rss->get_link()), 0, 245)); |
2c08214a | 443 | |
cd07592c AD |
444 | _debug("site_url: $site_url", $debug_enabled); |
445 | _debug("feed_title: " . $rss->get_title(), $debug_enabled); | |
446 | ||
687a4f59 | 447 | if ($favicon_needs_check || $force_refetch) { |
36490f11 AD |
448 | |
449 | /* terrible hack: if we crash on floicon shit here, we won't check | |
560cbd8c | 450 | * the icon avgcolor again (unless the icon got updated) */ |
36490f11 | 451 | |
560cbd8c AD |
452 | $favicon_file = ICONS_DIR . "/$feed.ico"; |
453 | $favicon_modified = @filemtime($favicon_file); | |
454 | ||
68cccafc | 455 | _debug("checking favicon...", $debug_enabled); |
687a4f59 | 456 | |
6f7798b6 | 457 | check_feed_favicon($site_url, $feed); |
560cbd8c AD |
458 | $favicon_modified_new = @filemtime($favicon_file); |
459 | ||
460 | if ($favicon_modified_new > $favicon_modified) | |
461 | $favicon_avg_color = ''; | |
687a4f59 | 462 | |
6ee0d4b0 | 463 | if (file_exists($favicon_file) && function_exists("imagecreatefromstring") && $favicon_avg_color == '') { |
bc7a144d | 464 | require_once "colors.php"; |
687a4f59 | 465 | |
aafd55ba AD |
466 | db_query("UPDATE ttrss_feeds SET favicon_avg_color = 'fail' WHERE |
467 | id = '$feed'"); | |
468 | ||
6322ac79 | 469 | $favicon_color = db_escape_string( |
bc7a144d | 470 | calculate_avg_color($favicon_file)); |
63c323f7 | 471 | |
bc7a144d | 472 | $favicon_colorstring = ",favicon_avg_color = '".$favicon_color."'"; |
36490f11 | 473 | } else if ($favicon_avg_color == 'fail') { |
84ceb6bd | 474 | _debug("floicon failed on this file, not trying to recalculate avg color", $debug_enabled); |
6ac722d5 | 475 | } |
687a4f59 | 476 | |
a42c55f0 | 477 | db_query("UPDATE ttrss_feeds SET favicon_last_checked = NOW() |
bc7a144d | 478 | $favicon_colorstring |
f2798eb6 AD |
479 | WHERE id = '$feed'"); |
480 | } | |
2c08214a AD |
481 | |
482 | if (!$registered_title || $registered_title == "[Unknown]") { | |
483 | ||
e44280a0 | 484 | $feed_title = db_escape_string(mb_substr($rss->get_title(), 0, 199)); |
2c08214a | 485 | |
852d4ac8 AD |
486 | if ($feed_title) { |
487 | _debug("registering title: $feed_title", $debug_enabled); | |
2c08214a | 488 | |
852d4ac8 AD |
489 | db_query("UPDATE ttrss_feeds SET |
490 | title = '$feed_title' WHERE id = '$feed'"); | |
491 | } | |
2c08214a AD |
492 | } |
493 | ||
0cf81637 | 494 | if ($site_url && $orig_site_url != $site_url) { |
a42c55f0 | 495 | db_query("UPDATE ttrss_feeds SET |
2c08214a AD |
496 | site_url = '$site_url' WHERE id = '$feed'"); |
497 | } | |
498 | ||
68cccafc | 499 | _debug("loading filters & labels...", $debug_enabled); |
2c08214a | 500 | |
a42c55f0 AD |
501 | $filters = load_filters($feed, $owner_uid); |
502 | $labels = get_all_labels($owner_uid); | |
2c08214a | 503 | |
68cccafc | 504 | _debug("" . count($filters) . " filters loaded.", $debug_enabled); |
2c08214a | 505 | |
19b3992b | 506 | $items = $rss->get_items(); |
2c08214a | 507 | |
19b3992b | 508 | if (!is_array($items)) { |
68cccafc | 509 | _debug("no articles found.", $debug_enabled); |
2c08214a | 510 | |
a42c55f0 | 511 | db_query("UPDATE ttrss_feeds |
2c08214a AD |
512 | SET last_updated = NOW(), last_error = '' WHERE id = '$feed'"); |
513 | ||
514 | return; // no articles | |
515 | } | |
516 | ||
517 | if ($pubsub_state != 2 && PUBSUBHUBBUB_ENABLED) { | |
518 | ||
68cccafc | 519 | _debug("checking for PUSH hub...", $debug_enabled); |
2c08214a AD |
520 | |
521 | $feed_hub_url = false; | |
2c08214a | 522 | |
19b3992b | 523 | $links = $rss->get_links('hub'); |
2c08214a | 524 | |
19b3992b AD |
525 | if ($links && is_array($links)) { |
526 | foreach ($links as $l) { | |
527 | $feed_hub_url = $l; | |
528 | break; | |
2c08214a AD |
529 | } |
530 | } | |
531 | ||
68cccafc | 532 | _debug("feed hub url: $feed_hub_url", $debug_enabled); |
2c08214a | 533 | |
d8179cb9 AD |
534 | $feed_self_url = $fetch_url; |
535 | ||
536 | $links = $rss->get_links('self'); | |
537 | ||
538 | if ($links && is_array($links)) { | |
539 | foreach ($links as $l) { | |
540 | $feed_self_url = $l; | |
541 | break; | |
542 | } | |
543 | } | |
544 | ||
545 | _debug("feed self url = $feed_self_url"); | |
546 | ||
547 | if ($feed_hub_url && $feed_self_url && function_exists('curl_init') && | |
2c08214a AD |
548 | !ini_get("open_basedir")) { |
549 | ||
550 | require_once 'lib/pubsubhubbub/subscriber.php'; | |
551 | ||
552 | $callback_url = get_self_url_prefix() . | |
553 | "/public.php?op=pubsub&id=$feed"; | |
554 | ||
555 | $s = new Subscriber($feed_hub_url, $callback_url); | |
556 | ||
d8179cb9 | 557 | $rc = $s->subscribe($feed_self_url); |
2c08214a | 558 | |
5ba1ddd4 | 559 | _debug("feed hub url found, subscribe request sent. [rc=$rc]", $debug_enabled); |
2c08214a | 560 | |
a42c55f0 | 561 | db_query("UPDATE ttrss_feeds SET pubsub_state = 1 |
2c08214a AD |
562 | WHERE id = '$feed'"); |
563 | } | |
564 | } | |
565 | ||
68cccafc | 566 | _debug("processing articles...", $debug_enabled); |
2c08214a | 567 | |
19b3992b | 568 | foreach ($items as $item) { |
5d56d100 | 569 | if ($_REQUEST['xdebug'] == 3) { |
2c08214a AD |
570 | print_r($item); |
571 | } | |
572 | ||
19b3992b AD |
573 | $entry_guid = $item->get_id(); |
574 | if (!$entry_guid) $entry_guid = $item->get_link(); | |
575 | if (!$entry_guid) $entry_guid = make_guid_from_title($item->get_title()); | |
2c08214a AD |
576 | if (!$entry_guid) continue; |
577 | ||
3a4c8973 AD |
578 | $entry_guid = "$owner_uid,$entry_guid"; |
579 | ||
a42c55f0 | 580 | $entry_guid_hashed = db_escape_string('SHA1:' . sha1($entry_guid)); |
5e3d5480 | 581 | |
68cccafc | 582 | _debug("guid $entry_guid / $entry_guid_hashed", $debug_enabled); |
5e3d5480 | 583 | |
2c08214a AD |
584 | $entry_timestamp = ""; |
585 | ||
04d2f9c8 AD |
586 | $entry_timestamp = $item->get_date(); |
587 | ||
588 | _debug("orig date: " . $item->get_date(), $debug_enabled); | |
2c08214a | 589 | |
30123fe6 | 590 | if ($entry_timestamp == -1 || !$entry_timestamp || $entry_timestamp > time()) { |
2c08214a | 591 | $entry_timestamp = time(); |
2c08214a AD |
592 | } |
593 | ||
594 | $entry_timestamp_fmt = strftime("%Y/%m/%d %H:%M:%S", $entry_timestamp); | |
595 | ||
68cccafc | 596 | _debug("date $entry_timestamp [$entry_timestamp_fmt]", $debug_enabled); |
2c08214a | 597 | |
99429e57 AD |
598 | // $entry_title = html_entity_decode($item->get_title(), ENT_COMPAT, 'UTF-8'); |
599 | // $entry_title = decode_numeric_entities($entry_title); | |
600 | $entry_title = $item->get_title(); | |
1b35d30c | 601 | |
5d56d100 | 602 | $entry_link = rewrite_relative_url($site_url, $item->get_link()); |
2c08214a | 603 | |
68cccafc AD |
604 | _debug("title $entry_title", $debug_enabled); |
605 | _debug("link $entry_link", $debug_enabled); | |
2c08214a AD |
606 | |
607 | if (!$entry_title) $entry_title = date("Y-m-d H:i:s", $entry_timestamp);; | |
608 | ||
19b3992b AD |
609 | $entry_content = $item->get_content(); |
610 | if (!$entry_content) $entry_content = $item->get_description(); | |
2c08214a AD |
611 | |
612 | if ($_REQUEST["xdebug"] == 2) { | |
9ec10352 | 613 | print "content: "; |
487f0750 | 614 | print $entry_content; |
3c696512 | 615 | print "\n"; |
2c08214a AD |
616 | } |
617 | ||
5ddd2705 AD |
618 | $entry_language = ""; |
619 | ||
620 | if (DETECT_ARTICLE_LANGUAGE) { | |
621 | $entry_language = $lang->detect($entry_title . " " . $entry_content, 1); | |
6b461797 | 622 | |
5ddd2705 | 623 | if (count($entry_language) > 0) { |
5d2e74a9 JP |
624 | $possible = array_keys($entry_language); |
625 | $entry_language = $possible[0]; | |
6b461797 | 626 | |
5ddd2705 | 627 | _debug("detected language: $entry_language", $debug_enabled); |
e95bb179 AD |
628 | } else { |
629 | $entry_language = ""; | |
5ddd2705 | 630 | } |
6b461797 AD |
631 | } |
632 | ||
cd07592c AD |
633 | $entry_comments = $item->get_comments_url(); |
634 | $entry_author = $item->get_author(); | |
2c08214a | 635 | |
a42c55f0 | 636 | $entry_guid = db_escape_string(mb_substr($entry_guid, 0, 245)); |
2c08214a | 637 | |
ce2b219c AD |
638 | $entry_comments = db_escape_string(mb_substr(trim($entry_comments), 0, 245)); |
639 | $entry_author = db_escape_string(mb_substr(trim($entry_author), 0, 245)); | |
2c08214a | 640 | |
d4992d6b | 641 | $num_comments = (int) $item->get_comments_count(); |
2c08214a | 642 | |
68cccafc AD |
643 | _debug("author $entry_author", $debug_enabled); |
644 | _debug("num_comments: $num_comments", $debug_enabled); | |
ee78f81c | 645 | _debug("looking for tags...", $debug_enabled); |
2c08214a AD |
646 | |
647 | // parse <category> entries into tags | |
648 | ||
649 | $additional_tags = array(); | |
650 | ||
19b3992b | 651 | $additional_tags_src = $item->get_categories(); |
2c08214a | 652 | |
19b3992b AD |
653 | if (is_array($additional_tags_src)) { |
654 | foreach ($additional_tags_src as $tobj) { | |
cd07592c | 655 | array_push($additional_tags, $tobj); |
2c08214a | 656 | } |
19b3992b | 657 | } |
2c08214a | 658 | |
fa6fbd36 | 659 | $entry_tags = array_unique($additional_tags); |
2c08214a AD |
660 | |
661 | for ($i = 0; $i < count($entry_tags); $i++) | |
662 | $entry_tags[$i] = mb_strtolower($entry_tags[$i], 'utf-8'); | |
663 | ||
ee78f81c AD |
664 | _debug("tags found: " . join(",", $entry_tags), $debug_enabled); |
665 | ||
68cccafc | 666 | _debug("done collecting data.", $debug_enabled); |
2c08214a | 667 | |
b1840673 AD |
668 | $result = db_query("SELECT id, content_hash FROM ttrss_entries |
669 | WHERE guid = '".db_escape_string($entry_guid)."' OR guid = '$entry_guid_hashed'"); | |
b30abdad AD |
670 | |
671 | if (db_num_rows($result) != 0) { | |
b1840673 AD |
672 | $base_entry_id = db_fetch_result($result, 0, "id"); |
673 | $entry_stored_hash = db_fetch_result($result, 0, "content_hash"); | |
b30abdad | 674 | } else { |
b1840673 AD |
675 | $base_entry_id = false; |
676 | $entry_stored_hash = ""; | |
b30abdad AD |
677 | } |
678 | ||
455b1401 | 679 | $article = array("owner_uid" => $owner_uid, // read only |
b30abdad | 680 | "guid" => $entry_guid, // read only |
19b3992b AD |
681 | "title" => $entry_title, |
682 | "content" => $entry_content, | |
683 | "link" => $entry_link, | |
684 | "tags" => $entry_tags, | |
e02555c1 | 685 | "author" => $entry_author, |
c9299c28 | 686 | "force_catchup" => false, // ugly hack for the time being |
02b0348a | 687 | "language" => $entry_language, // read only |
f73e03e0 AD |
688 | "feed" => array("id" => $feed, |
689 | "fetch_url" => $fetch_url, | |
690 | "site_url" => $site_url) | |
691 | ); | |
cc85704f | 692 | |
b1840673 | 693 | $entry_plugin_data = ""; |
eb16bd9f | 694 | $entry_current_hash = calculate_article_hash($article, $pluginhost); |
b1840673 AD |
695 | |
696 | _debug("article hash: $entry_current_hash [stored=$entry_stored_hash]", $debug_enabled); | |
697 | ||
522e8b35 | 698 | if ($entry_current_hash == $entry_stored_hash && !isset($_REQUEST["force_rehash"])) { |
b1840673 AD |
699 | _debug("stored article seems up to date [IID: $base_entry_id], updating timestamp only", $debug_enabled); |
700 | ||
701 | // we keep encountering the entry in feeds, so we need to | |
702 | // update date_updated column so that we don't get horrible | |
703 | // dupes when the entry gets purged and reinserted again e.g. | |
704 | // in the case of SLOW SLOW OMG SLOW updating feeds | |
705 | ||
706 | $base_entry_id = db_fetch_result($result, 0, "id"); | |
707 | ||
708 | db_query("UPDATE ttrss_entries SET date_updated = NOW() | |
709 | WHERE id = '$base_entry_id'"); | |
710 | ||
e44280a0 | 711 | // if we allow duplicate posts, we have to continue to |
da7e3e91 GP |
712 | // create the user entries for this feed |
713 | if (!get_pref("ALLOW_DUPLICATE_POSTS", $owner_uid, false)) { | |
714 | continue; | |
715 | } | |
b1840673 AD |
716 | } |
717 | ||
718 | _debug("hash differs, applying plugin filters:", $debug_enabled); | |
719 | ||
1ffe3391 | 720 | foreach ($pluginhost->get_hooks(PluginHost::HOOK_ARTICLE_FILTER) as $plugin) { |
b1840673 AD |
721 | _debug("... " . get_class($plugin), $debug_enabled); |
722 | ||
723 | $start = microtime(true); | |
19b3992b | 724 | $article = $plugin->hook_article_filter($article); |
0084f0d1 | 725 | |
b1840673 AD |
726 | _debug("=== " . sprintf("%.4f (sec)", microtime(true) - $start), $debug_enabled); |
727 | ||
728 | $entry_plugin_data .= mb_strtolower(get_class($plugin)) . ","; | |
cc85704f AD |
729 | } |
730 | ||
b1840673 AD |
731 | $entry_plugin_data = db_escape_string($entry_plugin_data); |
732 | ||
733 | _debug("plugin data: $entry_plugin_data", $debug_enabled); | |
734 | ||
19b3992b | 735 | $entry_tags = $article["tags"]; |
a42c55f0 AD |
736 | $entry_guid = db_escape_string($entry_guid); |
737 | $entry_title = db_escape_string($article["title"]); | |
738 | $entry_author = db_escape_string($article["author"]); | |
739 | $entry_link = db_escape_string($article["link"]); | |
f935d98e | 740 | $entry_content = $article["content"]; // escaped below |
c9299c28 AD |
741 | $entry_force_catchup = $article["force_catchup"]; |
742 | ||
743 | _debug("force catchup: $entry_force_catchup"); | |
f935d98e | 744 | |
0a3fd79b | 745 | if ($cache_images && is_writable(CACHE_DIR . '/images')) |
f0bd8e65 | 746 | cache_images($entry_content, $site_url, $debug_enabled); |
0a3fd79b | 747 | |
a42c55f0 | 748 | $entry_content = db_escape_string($entry_content, false); |
1f45c857 | 749 | |
a42c55f0 | 750 | db_query("BEGIN"); |
2c08214a | 751 | |
a42c55f0 | 752 | $result = db_query("SELECT id FROM ttrss_entries |
5e3d5480 | 753 | WHERE (guid = '$entry_guid' OR guid = '$entry_guid_hashed')"); |
9e222305 | 754 | |
2c08214a AD |
755 | if (db_num_rows($result) == 0) { |
756 | ||
68cccafc | 757 | _debug("base guid [$entry_guid] not found", $debug_enabled); |
2c08214a AD |
758 | |
759 | // base post entry does not exist, create it | |
760 | ||
6322ac79 | 761 | $result = db_query( |
2c08214a AD |
762 | "INSERT INTO ttrss_entries |
763 | (title, | |
764 | guid, | |
765 | link, | |
766 | updated, | |
767 | content, | |
768 | content_hash, | |
769 | no_orig_date, | |
770 | date_updated, | |
771 | date_entered, | |
772 | comments, | |
773 | num_comments, | |
b30abdad | 774 | plugin_data, |
6b461797 | 775 | lang, |
2c08214a AD |
776 | author) |
777 | VALUES | |
778 | ('$entry_title', | |
5e3d5480 | 779 | '$entry_guid_hashed', |
2c08214a AD |
780 | '$entry_link', |
781 | '$entry_timestamp_fmt', | |
782 | '$entry_content', | |
b1840673 | 783 | '$entry_current_hash', |
5ba1ddd4 | 784 | false, |
2c08214a | 785 | NOW(), |
be574731 | 786 | '$date_feed_processed', |
2c08214a AD |
787 | '$entry_comments', |
788 | '$num_comments', | |
b30abdad | 789 | '$entry_plugin_data', |
6b461797 | 790 | '$entry_language', |
2c08214a | 791 | '$entry_author')"); |
e8291805 AD |
792 | |
793 | $article_labels = array(); | |
794 | ||
2c08214a | 795 | } else { |
2c08214a AD |
796 | $base_entry_id = db_fetch_result($result, 0, "id"); |
797 | ||
a42c55f0 | 798 | $article_labels = get_article_labels($base_entry_id, $owner_uid); |
2c08214a AD |
799 | } |
800 | ||
801 | // now it should exist, if not - bad luck then | |
802 | ||
b1840673 | 803 | $result = db_query("SELECT id FROM ttrss_entries |
5e3d5480 | 804 | WHERE guid = '$entry_guid' OR guid = '$entry_guid_hashed'"); |
2c08214a AD |
805 | |
806 | $entry_ref_id = 0; | |
807 | $entry_int_id = 0; | |
808 | ||
809 | if (db_num_rows($result) == 1) { | |
810 | ||
68cccafc | 811 | _debug("base guid found, checking for user record", $debug_enabled); |
2c08214a | 812 | |
2c08214a AD |
813 | $ref_id = db_fetch_result($result, 0, "id"); |
814 | $entry_ref_id = $ref_id; | |
815 | ||
5e3d5480 AD |
816 | /* $stored_guid = db_fetch_result($result, 0, "guid"); |
817 | if ($stored_guid != $entry_guid_hashed) { | |
68cccafc | 818 | if ($debug_enabled) _debug("upgrading compat guid to hashed one", $debug_enabled); |
5e3d5480 | 819 | |
a42c55f0 | 820 | db_query("UPDATE ttrss_entries SET guid = '$entry_guid_hashed' WHERE |
5e3d5480 AD |
821 | id = '$ref_id'"); |
822 | } */ | |
823 | ||
2c08214a AD |
824 | // check for user post link to main table |
825 | ||
826 | // do we allow duplicate posts with same GUID in different feeds? | |
a42c55f0 | 827 | if (get_pref("ALLOW_DUPLICATE_POSTS", $owner_uid, false)) { |
2c08214a AD |
828 | $dupcheck_qpart = "AND (feed_id = '$feed' OR feed_id IS NULL)"; |
829 | } else { | |
830 | $dupcheck_qpart = ""; | |
831 | } | |
832 | ||
833 | /* Collect article tags here so we could filter by them: */ | |
834 | ||
835 | $article_filters = get_article_filters($filters, $entry_title, | |
4021d61a | 836 | $entry_content, $entry_link, $entry_timestamp, $entry_author, |
2c08214a AD |
837 | $entry_tags); |
838 | ||
839 | if ($debug_enabled) { | |
68cccafc | 840 | _debug("article filters: ", $debug_enabled); |
2c08214a AD |
841 | if (count($article_filters) != 0) { |
842 | print_r($article_filters); | |
843 | } | |
844 | } | |
845 | ||
846 | if (find_article_filter($article_filters, "filter")) { | |
a42c55f0 | 847 | db_query("COMMIT"); // close transaction in progress |
2c08214a AD |
848 | continue; |
849 | } | |
850 | ||
851 | $score = calculate_article_score($article_filters); | |
852 | ||
68cccafc | 853 | _debug("initial score: $score", $debug_enabled); |
2c08214a AD |
854 | |
855 | $query = "SELECT ref_id, int_id FROM ttrss_user_entries WHERE | |
856 | ref_id = '$ref_id' AND owner_uid = '$owner_uid' | |
857 | $dupcheck_qpart"; | |
858 | ||
859 | // if ($_REQUEST["xdebug"]) print "$query\n"; | |
860 | ||
a42c55f0 | 861 | $result = db_query($query); |
2c08214a AD |
862 | |
863 | // okay it doesn't exist - create user entry | |
864 | if (db_num_rows($result) == 0) { | |
865 | ||
68cccafc | 866 | _debug("user record not found, creating...", $debug_enabled); |
2c08214a | 867 | |
c9299c28 | 868 | if ($score >= -500 && !find_article_filter($article_filters, 'catchup') && !$entry_force_catchup) { |
2c08214a AD |
869 | $unread = 'true'; |
870 | $last_read_qpart = 'NULL'; | |
871 | } else { | |
872 | $unread = 'false'; | |
873 | $last_read_qpart = 'NOW()'; | |
874 | } | |
875 | ||
876 | if (find_article_filter($article_filters, 'mark') || $score > 1000) { | |
877 | $marked = 'true'; | |
878 | } else { | |
879 | $marked = 'false'; | |
880 | } | |
881 | ||
882 | if (find_article_filter($article_filters, 'publish')) { | |
883 | $published = 'true'; | |
884 | } else { | |
885 | $published = 'false'; | |
886 | } | |
887 | ||
2ea9bbfd AD |
888 | // N-grams |
889 | ||
c9299c28 | 890 | /* if (DB_TYPE == "pgsql" and defined('_NGRAM_TITLE_DUPLICATE_THRESHOLD')) { |
2ea9bbfd | 891 | |
a42c55f0 | 892 | $result = db_query("SELECT COUNT(*) AS similar FROM |
2ea9bbfd AD |
893 | ttrss_entries,ttrss_user_entries |
894 | WHERE ref_id = id AND updated >= NOW() - INTERVAL '7 day' | |
895 | AND similarity(title, '$entry_title') >= "._NGRAM_TITLE_DUPLICATE_THRESHOLD." | |
896 | AND owner_uid = $owner_uid"); | |
897 | ||
898 | $ngram_similar = db_fetch_result($result, 0, "similar"); | |
899 | ||
68cccafc | 900 | _debug("N-gram similar results: $ngram_similar", $debug_enabled); |
2ea9bbfd AD |
901 | |
902 | if ($ngram_similar > 0) { | |
903 | $unread = 'false'; | |
904 | } | |
c9299c28 | 905 | } */ |
2ea9bbfd | 906 | |
7873d588 AD |
907 | $last_marked = ($marked == 'true') ? 'NOW()' : 'NULL'; |
908 | $last_published = ($published == 'true') ? 'NOW()' : 'NULL'; | |
909 | ||
6322ac79 | 910 | $result = db_query( |
2c08214a AD |
911 | "INSERT INTO ttrss_user_entries |
912 | (ref_id, owner_uid, feed_id, unread, last_read, marked, | |
7873d588 AD |
913 | published, score, tag_cache, label_cache, uuid, |
914 | last_marked, last_published) | |
2c08214a | 915 | VALUES ('$ref_id', '$owner_uid', '$feed', $unread, |
7873d588 AD |
916 | $last_read_qpart, $marked, $published, '$score', '', '', |
917 | '', $last_marked, $last_published)"); | |
2c08214a AD |
918 | |
919 | if (PUBSUBHUBBUB_HUB && $published == 'true') { | |
920 | $rss_link = get_self_url_prefix() . | |
921 | "/public.php?op=rss&id=-2&key=" . | |
a42c55f0 | 922 | get_feed_access_key(-2, false, $owner_uid); |
2c08214a AD |
923 | |
924 | $p = new Publisher(PUBSUBHUBBUB_HUB); | |
925 | ||
5ba1ddd4 | 926 | /* $pubsub_result = */ $p->publish_update($rss_link); |
2c08214a AD |
927 | } |
928 | ||
6322ac79 | 929 | $result = db_query( |
2c08214a AD |
930 | "SELECT int_id FROM ttrss_user_entries WHERE |
931 | ref_id = '$ref_id' AND owner_uid = '$owner_uid' AND | |
932 | feed_id = '$feed' LIMIT 1"); | |
933 | ||
934 | if (db_num_rows($result) == 1) { | |
935 | $entry_int_id = db_fetch_result($result, 0, "int_id"); | |
936 | } | |
937 | } else { | |
68cccafc | 938 | _debug("user record FOUND", $debug_enabled); |
2c08214a AD |
939 | |
940 | $entry_ref_id = db_fetch_result($result, 0, "ref_id"); | |
941 | $entry_int_id = db_fetch_result($result, 0, "int_id"); | |
942 | } | |
943 | ||
68cccafc | 944 | _debug("RID: $entry_ref_id, IID: $entry_int_id", $debug_enabled); |
2c08214a | 945 | |
b1840673 AD |
946 | db_query("UPDATE ttrss_entries |
947 | SET title = '$entry_title', | |
948 | content = '$entry_content', | |
949 | content_hash = '$entry_current_hash', | |
950 | updated = '$entry_timestamp_fmt', | |
951 | num_comments = '$num_comments', | |
952 | plugin_data = '$entry_plugin_data', | |
953 | author = '$entry_author', | |
954 | lang = '$entry_language' | |
955 | WHERE id = '$ref_id'"); | |
956 | ||
957 | if ($mark_unread_on_update) { | |
958 | db_query("UPDATE ttrss_user_entries | |
959 | SET last_read = null, unread = true WHERE ref_id = '$ref_id'"); | |
2c08214a AD |
960 | } |
961 | } | |
962 | ||
a42c55f0 | 963 | db_query("COMMIT"); |
2c08214a | 964 | |
68cccafc | 965 | _debug("assigning labels...", $debug_enabled); |
2c08214a | 966 | |
a42c55f0 | 967 | assign_article_to_label_filters($entry_ref_id, $article_filters, |
b24504b1 | 968 | $owner_uid, $article_labels); |
2c08214a | 969 | |
68cccafc | 970 | _debug("looking for enclosures...", $debug_enabled); |
2c08214a AD |
971 | |
972 | // enclosures | |
973 | ||
974 | $enclosures = array(); | |
975 | ||
19b3992b | 976 | $encs = $item->get_enclosures(); |
2c08214a | 977 | |
19b3992b AD |
978 | if (is_array($encs)) { |
979 | foreach ($encs as $e) { | |
980 | $e_item = array( | |
523bd90b | 981 | $e->link, $e->type, $e->length, $e->title, $e->width, $e->height); |
2c08214a | 982 | array_push($enclosures, $e_item); |
2c08214a AD |
983 | } |
984 | } | |
985 | ||
2c08214a | 986 | if ($debug_enabled) { |
68cccafc | 987 | _debug("article enclosures:", $debug_enabled); |
2c08214a AD |
988 | print_r($enclosures); |
989 | } | |
990 | ||
a42c55f0 | 991 | db_query("BEGIN"); |
2c08214a | 992 | |
5c54e683 AD |
993 | // debugging |
994 | // db_query("DELETE FROM ttrss_enclosures WHERE post_id = '$entry_ref_id'"); | |
995 | ||
2c08214a | 996 | foreach ($enclosures as $enc) { |
a42c55f0 AD |
997 | $enc_url = db_escape_string($enc[0]); |
998 | $enc_type = db_escape_string($enc[1]); | |
999 | $enc_dur = db_escape_string($enc[2]); | |
5c54e683 | 1000 | $enc_title = db_escape_string($enc[3]); |
523bd90b FE |
1001 | $enc_width = intval($enc[4]); |
1002 | $enc_height = intval($enc[5]); | |
2c08214a | 1003 | |
a42c55f0 | 1004 | $result = db_query("SELECT id FROM ttrss_enclosures |
2c08214a AD |
1005 | WHERE content_url = '$enc_url' AND post_id = '$entry_ref_id'"); |
1006 | ||
1007 | if (db_num_rows($result) == 0) { | |
a42c55f0 | 1008 | db_query("INSERT INTO ttrss_enclosures |
523bd90b FE |
1009 | (content_url, content_type, title, duration, post_id, width, height) VALUES |
1010 | ('$enc_url', '$enc_type', '$enc_title', '$enc_dur', '$entry_ref_id', $enc_width, $enc_height)"); | |
2c08214a AD |
1011 | } |
1012 | } | |
1013 | ||
a42c55f0 | 1014 | db_query("COMMIT"); |
2c08214a AD |
1015 | |
1016 | // check for manual tags (we have to do it here since they're loaded from filters) | |
1017 | ||
1018 | foreach ($article_filters as $f) { | |
6aff7845 | 1019 | if ($f["type"] == "tag") { |
2c08214a | 1020 | |
6aff7845 | 1021 | $manual_tags = trim_array(explode(",", $f["param"])); |
2c08214a AD |
1022 | |
1023 | foreach ($manual_tags as $tag) { | |
1024 | if (tag_is_valid($tag)) { | |
1025 | array_push($entry_tags, $tag); | |
1026 | } | |
1027 | } | |
1028 | } | |
1029 | } | |
1030 | ||
1031 | // Skip boring tags | |
1032 | ||
6322ac79 | 1033 | $boring_tags = trim_array(explode(",", mb_strtolower(get_pref( |
2c08214a AD |
1034 | 'BLACKLISTED_TAGS', $owner_uid, ''), 'utf-8'))); |
1035 | ||
1036 | $filtered_tags = array(); | |
1037 | $tags_to_cache = array(); | |
1038 | ||
1039 | if ($entry_tags && is_array($entry_tags)) { | |
1040 | foreach ($entry_tags as $tag) { | |
1041 | if (array_search($tag, $boring_tags) === false) { | |
1042 | array_push($filtered_tags, $tag); | |
1043 | } | |
1044 | } | |
1045 | } | |
1046 | ||
1047 | $filtered_tags = array_unique($filtered_tags); | |
1048 | ||
1049 | if ($debug_enabled) { | |
68cccafc | 1050 | _debug("filtered article tags:", $debug_enabled); |
2c08214a AD |
1051 | print_r($filtered_tags); |
1052 | } | |
1053 | ||
1054 | // Save article tags in the database | |
1055 | ||
1056 | if (count($filtered_tags) > 0) { | |
1057 | ||
a42c55f0 | 1058 | db_query("BEGIN"); |
2c08214a AD |
1059 | |
1060 | foreach ($filtered_tags as $tag) { | |
1061 | ||
1062 | $tag = sanitize_tag($tag); | |
a42c55f0 | 1063 | $tag = db_escape_string($tag); |
2c08214a AD |
1064 | |
1065 | if (!tag_is_valid($tag)) continue; | |
1066 | ||
a42c55f0 | 1067 | $result = db_query("SELECT id FROM ttrss_tags |
2c08214a AD |
1068 | WHERE tag_name = '$tag' AND post_int_id = '$entry_int_id' AND |
1069 | owner_uid = '$owner_uid' LIMIT 1"); | |
1070 | ||
1071 | if ($result && db_num_rows($result) == 0) { | |
1072 | ||
a42c55f0 | 1073 | db_query("INSERT INTO ttrss_tags |
2c08214a AD |
1074 | (owner_uid,tag_name,post_int_id) |
1075 | VALUES ('$owner_uid','$tag', '$entry_int_id')"); | |
1076 | } | |
1077 | ||
1078 | array_push($tags_to_cache, $tag); | |
1079 | } | |
1080 | ||
1081 | /* update the cache */ | |
1082 | ||
1083 | $tags_to_cache = array_unique($tags_to_cache); | |
1084 | ||
a42c55f0 | 1085 | $tags_str = db_escape_string(join(",", $tags_to_cache)); |
2c08214a | 1086 | |
a42c55f0 | 1087 | db_query("UPDATE ttrss_user_entries |
2c08214a AD |
1088 | SET tag_cache = '$tags_str' WHERE ref_id = '$entry_ref_id' |
1089 | AND owner_uid = $owner_uid"); | |
1090 | ||
a42c55f0 | 1091 | db_query("COMMIT"); |
2c08214a AD |
1092 | } |
1093 | ||
68cccafc | 1094 | _debug("article processed", $debug_enabled); |
2c08214a AD |
1095 | } |
1096 | ||
68cccafc | 1097 | _debug("purging feed...", $debug_enabled); |
2c08214a | 1098 | |
a42c55f0 | 1099 | purge_feed($feed, 0, $debug_enabled); |
2c08214a | 1100 | |
a42c55f0 | 1101 | db_query("UPDATE ttrss_feeds |
f2798eb6 | 1102 | SET last_updated = NOW(), last_error = '' WHERE id = '$feed'"); |
2c08214a | 1103 | |
a42c55f0 | 1104 | // db_query("COMMIT"); |
2c08214a AD |
1105 | |
1106 | } else { | |
1107 | ||
a42c55f0 | 1108 | $error_msg = db_escape_string(mb_substr($rss->error(), 0, 245)); |
2c08214a | 1109 | |
4ad04ee2 AD |
1110 | _debug("fetch error: $error_msg", $debug_enabled); |
1111 | ||
1112 | if (count($rss->errors()) > 1) { | |
1113 | foreach ($rss->errors() as $error) { | |
1114 | _debug("+ $error"); | |
1115 | } | |
1116 | } | |
2c08214a | 1117 | |
6322ac79 | 1118 | db_query( |
2c08214a | 1119 | "UPDATE ttrss_feeds SET last_error = '$error_msg', |
88edaa93 | 1120 | last_updated = NOW() WHERE id = '$feed'"); |
2c08214a | 1121 | |
88edaa93 AD |
1122 | unset($rss); |
1123 | } | |
2c08214a | 1124 | |
68cccafc | 1125 | _debug("done", $debug_enabled); |
88edaa93 AD |
1126 | |
1127 | return $rss; | |
2c08214a AD |
1128 | } |
1129 | ||
3c696512 | 1130 | function cache_images($html, $site_url, $debug) { |
3c696512 AD |
1131 | libxml_use_internal_errors(true); |
1132 | ||
1133 | $charset_hack = '<head> | |
1134 | <meta http-equiv="Content-Type" content="text/html; charset=utf-8"/> | |
1135 | </head>'; | |
1136 | ||
1137 | $doc = new DOMDocument(); | |
1138 | $doc->loadHTML($charset_hack . $html); | |
1139 | $xpath = new DOMXPath($doc); | |
1140 | ||
1141 | $entries = $xpath->query('(//img[@src])'); | |
1142 | ||
1143 | foreach ($entries as $entry) { | |
1144 | if ($entry->hasAttribute('src')) { | |
1145 | $src = rewrite_relative_url($site_url, $entry->getAttribute('src')); | |
1146 | ||
1147 | $local_filename = CACHE_DIR . "/images/" . sha1($src) . ".png"; | |
1148 | ||
1149 | if ($debug) _debug("cache_images: downloading: $src to $local_filename"); | |
1150 | ||
1151 | if (!file_exists($local_filename)) { | |
1152 | $file_content = fetch_file_contents($src); | |
1153 | ||
a3d50184 | 1154 | if ($file_content && strlen($file_content) > _MIN_CACHE_IMAGE_SIZE) { |
3c696512 AD |
1155 | file_put_contents($local_filename, $file_content); |
1156 | } | |
1157 | } | |
1158 | ||
2fc4d981 | 1159 | /* if (file_exists($local_filename)) { |
3c696512 | 1160 | $entry->setAttribute('src', SELF_URL_PATH . '/image.php?url=' . |
487f0750 | 1161 | base64_encode($src)); |
2fc4d981 | 1162 | } */ |
3c696512 AD |
1163 | } |
1164 | } | |
1165 | ||
2fc4d981 AD |
1166 | //$node = $doc->getElementsByTagName('body')->item(0); |
1167 | //return $doc->saveXML($node); | |
3c696512 AD |
1168 | } |
1169 | ||
a42c55f0 | 1170 | function expire_error_log($debug) { |
e2261e17 AD |
1171 | if ($debug) _debug("Removing old error log entries..."); |
1172 | ||
1173 | if (DB_TYPE == "pgsql") { | |
a42c55f0 | 1174 | db_query("DELETE FROM ttrss_error_log |
e2261e17 AD |
1175 | WHERE created_at < NOW() - INTERVAL '7 days'"); |
1176 | } else { | |
a42c55f0 | 1177 | db_query("DELETE FROM ttrss_error_log |
e2261e17 AD |
1178 | WHERE created_at < DATE_SUB(NOW(), INTERVAL 7 DAY)"); |
1179 | } | |
1180 | ||
1181 | } | |
1182 | ||
2a91b6ff | 1183 | function expire_lock_files($debug) { |
65465085 | 1184 | //if ($debug) _debug("Removing old lock files..."); |
2a91b6ff AD |
1185 | |
1186 | $num_deleted = 0; | |
1187 | ||
1188 | if (is_writable(LOCK_DIRECTORY)) { | |
1189 | $files = glob(LOCK_DIRECTORY . "/*.lock"); | |
1190 | ||
1191 | if ($files) { | |
1192 | foreach ($files as $file) { | |
11344971 | 1193 | if (!file_is_locked(basename($file)) && time() - filemtime($file) > 86400*2) { |
2a91b6ff AD |
1194 | unlink($file); |
1195 | ++$num_deleted; | |
1196 | } | |
1197 | } | |
1198 | } | |
1199 | } | |
1200 | ||
65465085 | 1201 | if ($debug) _debug("Removed $num_deleted old lock files."); |
2a91b6ff AD |
1202 | } |
1203 | ||
3c696512 | 1204 | function expire_cached_files($debug) { |
3306daec | 1205 | foreach (array("simplepie", "images", "export", "upload") as $dir) { |
3c696512 | 1206 | $cache_dir = CACHE_DIR . "/$dir"; |
2c08214a | 1207 | |
65465085 | 1208 | // if ($debug) _debug("Expiring $cache_dir"); |
2c08214a | 1209 | |
3c696512 AD |
1210 | $num_deleted = 0; |
1211 | ||
1212 | if (is_writable($cache_dir)) { | |
1213 | $files = glob("$cache_dir/*"); | |
1214 | ||
2a91b6ff | 1215 | if ($files) { |
2ab20c31 AD |
1216 | foreach ($files as $file) { |
1217 | if (time() - filemtime($file) > 86400*7) { | |
1218 | unlink($file); | |
3c696512 | 1219 | |
2ab20c31 AD |
1220 | ++$num_deleted; |
1221 | } | |
3c696512 AD |
1222 | } |
1223 | } | |
2a91b6ff | 1224 | } |
3c696512 | 1225 | |
65465085 | 1226 | if ($debug) _debug("$cache_dir: removed $num_deleted files."); |
3c696512 AD |
1227 | } |
1228 | } | |
2c08214a | 1229 | |
a3e0bdcf AD |
1230 | /** |
1231 | * Source: http://www.php.net/manual/en/function.parse-url.php#104527 | |
1232 | * Returns the url query as associative array | |
1233 | * | |
1234 | * @param string query | |
1235 | * @return array params | |
1236 | */ | |
1237 | function convertUrlQuery($query) { | |
1238 | $queryParts = explode('&', $query); | |
1239 | ||
1240 | $params = array(); | |
1241 | ||
1242 | foreach ($queryParts as $param) { | |
1243 | $item = explode('=', $param); | |
1244 | $params[$item[0]] = $item[1]; | |
1245 | } | |
1246 | ||
1247 | return $params; | |
1248 | } | |
92c14e9d | 1249 | |
48cb2536 | 1250 | function get_article_filters($filters, $title, $content, $link, $timestamp, $author, $tags) { |
92c14e9d AD |
1251 | $matches = array(); |
1252 | ||
1253 | foreach ($filters as $filter) { | |
1254 | $match_any_rule = $filter["match_any_rule"]; | |
a3a896a1 | 1255 | $inverse = $filter["inverse"]; |
92c14e9d AD |
1256 | $filter_match = false; |
1257 | ||
1258 | foreach ($filter["rules"] as $rule) { | |
1259 | $match = false; | |
ffa1bd7b | 1260 | $reg_exp = str_replace('/', '\/', $rule["reg_exp"]); |
a3a896a1 | 1261 | $rule_inverse = $rule["inverse"]; |
92c14e9d AD |
1262 | |
1263 | if (!$reg_exp) | |
1264 | continue; | |
1265 | ||
1266 | switch ($rule["type"]) { | |
1267 | case "title": | |
1268 | $match = @preg_match("/$reg_exp/i", $title); | |
1269 | break; | |
1270 | case "content": | |
d03ae73e AD |
1271 | // we don't need to deal with multiline regexps |
1272 | $content = preg_replace("/[\r\n\t]/", "", $content); | |
1273 | ||
92c14e9d AD |
1274 | $match = @preg_match("/$reg_exp/i", $content); |
1275 | break; | |
1276 | case "both": | |
d03ae73e AD |
1277 | // we don't need to deal with multiline regexps |
1278 | $content = preg_replace("/[\r\n\t]/", "", $content); | |
1279 | ||
72d1d067 | 1280 | $match = (@preg_match("/$reg_exp/i", $title) || @preg_match("/$reg_exp/i", $content)); |
92c14e9d AD |
1281 | break; |
1282 | case "link": | |
1283 | $match = @preg_match("/$reg_exp/i", $link); | |
1284 | break; | |
1285 | case "author": | |
1286 | $match = @preg_match("/$reg_exp/i", $author); | |
1287 | break; | |
1288 | case "tag": | |
7b80b5e1 DK |
1289 | foreach ($tags as $tag) { |
1290 | if (@preg_match("/$reg_exp/i", $tag)) { | |
1291 | $match = true; | |
1292 | break; | |
1293 | } | |
1294 | } | |
92c14e9d AD |
1295 | break; |
1296 | } | |
1297 | ||
a3a896a1 AD |
1298 | if ($rule_inverse) $match = !$match; |
1299 | ||
92c14e9d AD |
1300 | if ($match_any_rule) { |
1301 | if ($match) { | |
1302 | $filter_match = true; | |
1303 | break; | |
1304 | } | |
1305 | } else { | |
1306 | $filter_match = $match; | |
1307 | if (!$match) { | |
1308 | break; | |
1309 | } | |
1310 | } | |
1311 | } | |
1312 | ||
a3a896a1 AD |
1313 | if ($inverse) $filter_match = !$filter_match; |
1314 | ||
92c14e9d AD |
1315 | if ($filter_match) { |
1316 | foreach ($filter["actions"] AS $action) { | |
1317 | array_push($matches, $action); | |
5e736e45 AD |
1318 | |
1319 | // if Stop action encountered, perform no further processing | |
1320 | if ($action["type"] == "stop") return $matches; | |
92c14e9d AD |
1321 | } |
1322 | } | |
1323 | } | |
1324 | ||
1325 | return $matches; | |
1326 | } | |
1327 | ||
1328 | function find_article_filter($filters, $filter_name) { | |
1329 | foreach ($filters as $f) { | |
1330 | if ($f["type"] == $filter_name) { | |
1331 | return $f; | |
1332 | }; | |
1333 | } | |
1334 | return false; | |
1335 | } | |
1336 | ||
1337 | function find_article_filters($filters, $filter_name) { | |
1338 | $results = array(); | |
1339 | ||
1340 | foreach ($filters as $f) { | |
1341 | if ($f["type"] == $filter_name) { | |
1342 | array_push($results, $f); | |
1343 | }; | |
1344 | } | |
1345 | return $results; | |
1346 | } | |
1347 | ||
1348 | function calculate_article_score($filters) { | |
1349 | $score = 0; | |
1350 | ||
1351 | foreach ($filters as $f) { | |
1352 | if ($f["type"] == "score") { | |
1353 | $score += $f["param"]; | |
1354 | }; | |
1355 | } | |
1356 | return $score; | |
1357 | } | |
1358 | ||
b24504b1 AD |
1359 | function labels_contains_caption($labels, $caption) { |
1360 | foreach ($labels as $label) { | |
1361 | if ($label[1] == $caption) { | |
1362 | return true; | |
1363 | } | |
1364 | } | |
1365 | ||
1366 | return false; | |
1367 | } | |
1368 | ||
a42c55f0 | 1369 | function assign_article_to_label_filters($id, $filters, $owner_uid, $article_labels) { |
92c14e9d AD |
1370 | foreach ($filters as $f) { |
1371 | if ($f["type"] == "label") { | |
b24504b1 | 1372 | if (!labels_contains_caption($article_labels, $f["param"])) { |
a42c55f0 | 1373 | label_add_article($id, $f["param"], $owner_uid); |
b24504b1 AD |
1374 | } |
1375 | } | |
92c14e9d AD |
1376 | } |
1377 | } | |
87764a50 | 1378 | |
87d7e850 AD |
1379 | function make_guid_from_title($title) { |
1380 | return preg_replace("/[ \"\',.:;]/", "-", | |
1381 | mb_strtolower(strip_tags($title), 'utf-8')); | |
1382 | } | |
1383 | ||
d1f3fa97 | 1384 | /* function verify_feed_xml($feed_data) { |
ebec81a6 AD |
1385 | libxml_use_internal_errors(true); |
1386 | $doc = new DOMDocument(); | |
1387 | $doc->loadXML($feed_data); | |
1388 | $error = libxml_get_last_error(); | |
1389 | libxml_clear_errors(); | |
1390 | return $error; | |
d1f3fa97 | 1391 | } */ |
87d7e850 | 1392 | |
e2cf81e2 AD |
1393 | function housekeeping_common($debug) { |
1394 | expire_cached_files($debug); | |
1395 | expire_lock_files($debug); | |
1396 | expire_error_log($debug); | |
1397 | ||
1398 | $count = update_feedbrowser_cache(); | |
1399 | _debug("Feedbrowser updated, $count feeds processed."); | |
1400 | ||
1401 | purge_orphans( true); | |
1402 | $rc = cleanup_tags( 14, 50000); | |
1403 | ||
1404 | _debug("Cleaned $rc cached tags."); | |
8e470220 | 1405 | |
00f22824 | 1406 | PluginHost::getInstance()->run_hooks(PluginHost::HOOK_HOUSE_KEEPING, "hook_house_keeping", ""); |
8e470220 | 1407 | |
e2cf81e2 | 1408 | } |
2c08214a | 1409 | ?> |