]> git.wh0rd.org - tt-rss.git/blame - include/rssfuncs.php
bayes, filter: do not categorize already processed articles
[tt-rss.git] / include / rssfuncs.php
CommitLineData
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 $nf;
222
2c08214a
AD
223 } // function update_daemon_common
224
6022776d
AD
225 // this is used when subscribing
226 function set_basic_feed_info($feed) {
227
228 $feed = db_escape_string($feed);
229
230 $result = db_query("SELECT feed_url,auth_pass,auth_pass_encrypted
231 FROM ttrss_feeds WHERE id = '$feed'");
232
233 $auth_pass_encrypted = sql_bool_to_bool(db_fetch_result($result,
234 0, "auth_pass_encrypted"));
235
236 $auth_login = db_fetch_result($result, 0, "auth_login");
237 $auth_pass = db_fetch_result($result, 0, "auth_pass");
238
239 if ($auth_pass_encrypted) {
240 require_once "crypt.php";
241 $auth_pass = decrypt_string($auth_pass);
242 }
243
244 $fetch_url = db_fetch_result($result, 0, "feed_url");
245
246 $feed_data = fetch_file_contents($fetch_url, false,
247 $auth_login, $auth_pass, false,
248 FEED_FETCH_TIMEOUT_TIMEOUT,
249 0);
250
251 global $fetch_curl_used;
252
253 if (!$fetch_curl_used) {
254 $tmp = @gzdecode($feed_data);
255
256 if ($tmp) $feed_data = $tmp;
257 }
258
259 $feed_data = trim($feed_data);
260
261 $rss = new FeedParser($feed_data);
262 $rss->init();
263
264 if (!$rss->error()) {
265
266 $result = db_query("SELECT title, site_url FROM ttrss_feeds WHERE id = '$feed'");
267
268 $registered_title = db_fetch_result($result, 0, "title");
269 $orig_site_url = db_fetch_result($result, 0, "site_url");
270
271 $site_url = db_escape_string(mb_substr(rewrite_relative_url($fetch_url, $rss->get_link()), 0, 245));
272 $feed_title = db_escape_string(mb_substr($rss->get_title(), 0, 199));
273
274 if ($feed_title && (!$registered_title || $registered_title == "[Unknown]")) {
275 db_query("UPDATE ttrss_feeds SET
276 title = '$feed_title' WHERE id = '$feed'");
277 }
278
279 if ($site_url && $orig_site_url != $site_url) {
280 db_query("UPDATE ttrss_feeds SET
281 site_url = '$site_url' WHERE id = '$feed'");
282 }
283 }
284 }
285
87764a50 286 // ignore_daemon is not used
88edaa93 287 function update_rss_feed($feed, $ignore_daemon = false, $no_cache = false, $rss = false) {
2c08214a 288
2c08214a
AD
289 $debug_enabled = defined('DAEMON_EXTENDED_DEBUG') || $_REQUEST['xdebug'];
290
4f71d743 291 _debug_suppress(!$debug_enabled);
68cccafc 292 _debug("start", $debug_enabled);
2c08214a 293
a42c55f0 294 $result = db_query("SELECT id,update_interval,auth_login,
5ba1ddd4 295 feed_url,auth_pass,cache_images,
5321e775 296 mark_unread_on_update, owner_uid,
90e5f4f1
AD
297 pubsub_state, auth_pass_encrypted,
298 (SELECT max(date_entered) FROM
299 ttrss_entries, ttrss_user_entries where ref_id = id AND feed_id = '$feed') AS last_article_timestamp
87764a50 300 FROM ttrss_feeds WHERE id = '$feed'");
2c08214a
AD
301
302 if (db_num_rows($result) == 0) {
68cccafc 303 _debug("feed $feed NOT FOUND/SKIPPED", $debug_enabled);
2c08214a
AD
304 return false;
305 }
306
90e5f4f1 307 $last_article_timestamp = @strtotime(db_fetch_result($result, 0, "last_article_timestamp"));
5d0d3887
AD
308
309 if (defined('_DISABLE_HTTP_304'))
310 $last_article_timestamp = 0;
311
2c08214a
AD
312 $owner_uid = db_fetch_result($result, 0, "owner_uid");
313 $mark_unread_on_update = sql_bool_to_bool(db_fetch_result($result,
314 0, "mark_unread_on_update"));
2c08214a 315 $pubsub_state = db_fetch_result($result, 0, "pubsub_state");
044cff2d
AD
316 $auth_pass_encrypted = sql_bool_to_bool(db_fetch_result($result,
317 0, "auth_pass_encrypted"));
2c08214a 318
a42c55f0 319 db_query("UPDATE ttrss_feeds SET last_update_started = NOW()
2c08214a
AD
320 WHERE id = '$feed'");
321
322 $auth_login = db_fetch_result($result, 0, "auth_login");
323 $auth_pass = db_fetch_result($result, 0, "auth_pass");
324
044cff2d
AD
325 if ($auth_pass_encrypted) {
326 require_once "crypt.php";
327 $auth_pass = decrypt_string($auth_pass);
328 }
329
2c08214a
AD
330 $cache_images = sql_bool_to_bool(db_fetch_result($result, 0, "cache_images"));
331 $fetch_url = db_fetch_result($result, 0, "feed_url");
332
a42c55f0 333 $feed = db_escape_string($feed);
2c08214a 334
f074ffe9 335 $date_feed_processed = date('Y-m-d H:i');
2c08214a 336
865a3ed6 337 $cache_filename = CACHE_DIR . "/simplepie/" . sha1($fetch_url) . ".xml";
f074ffe9 338
ee65bef4
AD
339 $pluginhost = new PluginHost();
340 $pluginhost->set_debug($debug_enabled);
341 $user_plugins = get_pref("_ENABLED_PLUGINS", $owner_uid);
342
343 $pluginhost->load(PLUGINS, PluginHost::KIND_ALL);
344 $pluginhost->load($user_plugins, PluginHost::KIND_USER, $owner_uid);
345 $pluginhost->load_data();
346
88edaa93
AD
347 if ($rss && is_object($rss) && get_class($rss) == "FeedParser") {
348 _debug("using previously initialized parser object");
349 } else {
350 $rss_hash = false;
4f9cbdff 351
88edaa93 352 $force_refetch = isset($_REQUEST["force_refetch"]);
687a4f59 353
95beaa14 354 foreach ($pluginhost->get_hooks(PluginHost::HOOK_FETCH_FEED) as $plugin) {
01465325 355 $feed_data = $plugin->hook_fetch_feed($feed_data, $fetch_url, $owner_uid, $feed, $last_article_timestamp, $auth_login, $auth_pass);
95beaa14
RA
356 }
357
358 // try cache
359 if (!$feed_data &&
360 file_exists($cache_filename) &&
88edaa93
AD
361 is_readable($cache_filename) &&
362 !$auth_login && !$auth_pass &&
363 filemtime($cache_filename) > time() - 30) {
2c08214a 364
15d0528a 365 _debug("using local cache [$cache_filename].", $debug_enabled);
be574731 366
88edaa93 367 @$feed_data = file_get_contents($cache_filename);
52637d3b 368
88edaa93
AD
369 if ($feed_data) {
370 $rss_hash = sha1($feed_data);
371 }
f074ffe9 372
88edaa93
AD
373 } else {
374 _debug("local cache will not be used for this feed", $debug_enabled);
375 }
ee65bef4 376
95beaa14 377 // fetch feed from source
fd687300
AD
378 if (!$feed_data) {
379 _debug("fetching [$fetch_url]...", $debug_enabled);
380 _debug("If-Modified-Since: ".gmdate('D, d M Y H:i:s \G\M\T', $last_article_timestamp), $debug_enabled);
96f98cb0 381
fd687300
AD
382 $feed_data = fetch_file_contents($fetch_url, false,
383 $auth_login, $auth_pass, false,
384 $no_cache ? FEED_FETCH_NO_CACHE_TIMEOUT : FEED_FETCH_TIMEOUT,
385 $force_refetch ? 0 : $last_article_timestamp);
3f6f0857 386
fd687300 387 global $fetch_curl_used;
3f6f0857 388
fd687300
AD
389 if (!$fetch_curl_used) {
390 $tmp = @gzdecode($feed_data);
3f6f0857 391
fd687300
AD
392 if ($tmp) $feed_data = $tmp;
393 }
1367bc3f 394
fd687300 395 $feed_data = trim($feed_data);
017401dd 396
fd687300
AD
397 _debug("fetch done.", $debug_enabled);
398
95beaa14
RA
399 // cache vanilla feed data for re-use
400 if ($feed_data && !$auth_pass && !$auth_login && is_writable(CACHE_DIR . "/simplepie")) {
401 $new_rss_hash = sha1($feed_data);
402
403 if ($new_rss_hash != $rss_hash) {
404 _debug("saving $cache_filename", $debug_enabled);
405 @file_put_contents($cache_filename, $feed_data);
406 }
407 }
4f9cbdff 408 }
017401dd 409
fd687300
AD
410 if (!$feed_data) {
411 global $fetch_last_error;
412 global $fetch_last_error_code;
f074ffe9 413
fd687300 414 _debug("unable to fetch: $fetch_last_error [$fetch_last_error_code]", $debug_enabled);
f074ffe9 415
fd687300 416 $error_escaped = '';
7a01dc77 417
fd687300
AD
418 // If-Modified-Since
419 if ($fetch_last_error_code != 304) {
420 $error_escaped = db_escape_string($fetch_last_error);
421 } else {
422 _debug("source claims data not modified, nothing to do.", $debug_enabled);
423 }
4f9cbdff 424
fd687300
AD
425 db_query(
426 "UPDATE ttrss_feeds SET last_error = '$error_escaped',
427 last_updated = NOW() WHERE id = '$feed'");
4f9cbdff 428
fd687300
AD
429 return;
430 }
f074ffe9
AD
431 }
432
1ffe3391 433 foreach ($pluginhost->get_hooks(PluginHost::HOOK_FEED_FETCHED) as $plugin) {
6791af0c 434 $feed_data = $plugin->hook_feed_fetched($feed_data, $fetch_url, $owner_uid, $feed);
017401dd
AD
435 }
436
5ef84097
AD
437 // set last update to now so if anything *simplepie* crashes later we won't be
438 // continuously failing on the same feed
5de40104 439 //db_query("UPDATE ttrss_feeds SET last_updated = NOW() WHERE id = '$feed'");
5ef84097 440
4f9cbdff 441 if (!$rss) {
cd07592c
AD
442 $rss = new FeedParser($feed_data);
443 $rss->init();
5de51df7
AD
444 }
445
5ddd2705
AD
446 if (DETECT_ARTICLE_LANGUAGE) {
447 require_once "lib/languagedetect/LanguageDetect.php";
6b461797 448
5ddd2705
AD
449 $lang = new Text_LanguageDetect();
450 $lang->setNameMode(2);
451 }
6b461797 452
2c08214a
AD
453// print_r($rss);
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
68cccafc 462 _debug("processing feed data...", $debug_enabled);
2c08214a 463
a42c55f0 464// db_query("BEGIN");
2c08214a 465
382268c6
AD
466 if (DB_TYPE == "pgsql") {
467 $favicon_interval_qpart = "favicon_last_checked < NOW() - INTERVAL '12 hour'";
468 } else {
469 $favicon_interval_qpart = "favicon_last_checked < DATE_SUB(NOW(), INTERVAL 12 HOUR)";
470 }
471
fe4535e6 472 $result = db_query("SELECT owner_uid,favicon_avg_color,
382268c6
AD
473 (favicon_last_checked IS NULL OR $favicon_interval_qpart) AS
474 favicon_needs_check
2c08214a
AD
475 FROM ttrss_feeds WHERE id = '$feed'");
476
382268c6
AD
477 $favicon_needs_check = sql_bool_to_bool(db_fetch_result($result, 0,
478 "favicon_needs_check"));
36490f11 479 $favicon_avg_color = db_fetch_result($result, 0, "favicon_avg_color");
2c08214a
AD
480
481 $owner_uid = db_fetch_result($result, 0, "owner_uid");
482
a42c55f0 483 $site_url = db_escape_string(mb_substr(rewrite_relative_url($fetch_url, $rss->get_link()), 0, 245));
2c08214a 484
cd07592c
AD
485 _debug("site_url: $site_url", $debug_enabled);
486 _debug("feed_title: " . $rss->get_title(), $debug_enabled);
487
687a4f59 488 if ($favicon_needs_check || $force_refetch) {
36490f11
AD
489
490 /* terrible hack: if we crash on floicon shit here, we won't check
560cbd8c 491 * the icon avgcolor again (unless the icon got updated) */
36490f11 492
560cbd8c
AD
493 $favicon_file = ICONS_DIR . "/$feed.ico";
494 $favicon_modified = @filemtime($favicon_file);
495
68cccafc 496 _debug("checking favicon...", $debug_enabled);
687a4f59 497
6f7798b6 498 check_feed_favicon($site_url, $feed);
560cbd8c
AD
499 $favicon_modified_new = @filemtime($favicon_file);
500
501 if ($favicon_modified_new > $favicon_modified)
502 $favicon_avg_color = '';
687a4f59 503
6ee0d4b0 504 if (file_exists($favicon_file) && function_exists("imagecreatefromstring") && $favicon_avg_color == '') {
bc7a144d 505 require_once "colors.php";
687a4f59 506
aafd55ba
AD
507 db_query("UPDATE ttrss_feeds SET favicon_avg_color = 'fail' WHERE
508 id = '$feed'");
509
6322ac79 510 $favicon_color = db_escape_string(
bc7a144d 511 calculate_avg_color($favicon_file));
63c323f7 512
bc7a144d 513 $favicon_colorstring = ",favicon_avg_color = '".$favicon_color."'";
36490f11 514 } else if ($favicon_avg_color == 'fail') {
84ceb6bd 515 _debug("floicon failed on this file, not trying to recalculate avg color", $debug_enabled);
6ac722d5 516 }
687a4f59 517
a42c55f0 518 db_query("UPDATE ttrss_feeds SET favicon_last_checked = NOW()
bc7a144d 519 $favicon_colorstring
f2798eb6
AD
520 WHERE id = '$feed'");
521 }
2c08214a 522
68cccafc 523 _debug("loading filters & labels...", $debug_enabled);
2c08214a 524
a42c55f0 525 $filters = load_filters($feed, $owner_uid);
2c08214a 526
68cccafc 527 _debug("" . count($filters) . " filters loaded.", $debug_enabled);
2c08214a 528
19b3992b 529 $items = $rss->get_items();
2c08214a 530
19b3992b 531 if (!is_array($items)) {
68cccafc 532 _debug("no articles found.", $debug_enabled);
2c08214a 533
a42c55f0 534 db_query("UPDATE ttrss_feeds
2c08214a
AD
535 SET last_updated = NOW(), last_error = '' WHERE id = '$feed'");
536
537 return; // no articles
538 }
539
540 if ($pubsub_state != 2 && PUBSUBHUBBUB_ENABLED) {
541
68cccafc 542 _debug("checking for PUSH hub...", $debug_enabled);
2c08214a
AD
543
544 $feed_hub_url = false;
2c08214a 545
19b3992b 546 $links = $rss->get_links('hub');
2c08214a 547
19b3992b
AD
548 if ($links && is_array($links)) {
549 foreach ($links as $l) {
550 $feed_hub_url = $l;
551 break;
2c08214a
AD
552 }
553 }
554
68cccafc 555 _debug("feed hub url: $feed_hub_url", $debug_enabled);
2c08214a 556
d8179cb9
AD
557 $feed_self_url = $fetch_url;
558
559 $links = $rss->get_links('self');
560
561 if ($links && is_array($links)) {
562 foreach ($links as $l) {
563 $feed_self_url = $l;
564 break;
565 }
566 }
567
568 _debug("feed self url = $feed_self_url");
569
570 if ($feed_hub_url && $feed_self_url && function_exists('curl_init') &&
2c08214a
AD
571 !ini_get("open_basedir")) {
572
573 require_once 'lib/pubsubhubbub/subscriber.php';
574
575 $callback_url = get_self_url_prefix() .
576 "/public.php?op=pubsub&id=$feed";
577
578 $s = new Subscriber($feed_hub_url, $callback_url);
579
d8179cb9 580 $rc = $s->subscribe($feed_self_url);
2c08214a 581
5ba1ddd4 582 _debug("feed hub url found, subscribe request sent. [rc=$rc]", $debug_enabled);
2c08214a 583
a42c55f0 584 db_query("UPDATE ttrss_feeds SET pubsub_state = 1
2c08214a
AD
585 WHERE id = '$feed'");
586 }
587 }
588
68cccafc 589 _debug("processing articles...", $debug_enabled);
2c08214a 590
19b3992b 591 foreach ($items as $item) {
5d56d100 592 if ($_REQUEST['xdebug'] == 3) {
2c08214a
AD
593 print_r($item);
594 }
595
19b3992b
AD
596 $entry_guid = $item->get_id();
597 if (!$entry_guid) $entry_guid = $item->get_link();
598 if (!$entry_guid) $entry_guid = make_guid_from_title($item->get_title());
2c08214a
AD
599 if (!$entry_guid) continue;
600
3a4c8973
AD
601 $entry_guid = "$owner_uid,$entry_guid";
602
a42c55f0 603 $entry_guid_hashed = db_escape_string('SHA1:' . sha1($entry_guid));
5e3d5480 604
68cccafc 605 _debug("guid $entry_guid / $entry_guid_hashed", $debug_enabled);
5e3d5480 606
2c08214a
AD
607 $entry_timestamp = "";
608
04d2f9c8
AD
609 $entry_timestamp = $item->get_date();
610
611 _debug("orig date: " . $item->get_date(), $debug_enabled);
2c08214a 612
30123fe6 613 if ($entry_timestamp == -1 || !$entry_timestamp || $entry_timestamp > time()) {
2c08214a 614 $entry_timestamp = time();
2c08214a
AD
615 }
616
617 $entry_timestamp_fmt = strftime("%Y/%m/%d %H:%M:%S", $entry_timestamp);
618
68cccafc 619 _debug("date $entry_timestamp [$entry_timestamp_fmt]", $debug_enabled);
2c08214a 620
99429e57
AD
621// $entry_title = html_entity_decode($item->get_title(), ENT_COMPAT, 'UTF-8');
622// $entry_title = decode_numeric_entities($entry_title);
623 $entry_title = $item->get_title();
1b35d30c 624
5d56d100 625 $entry_link = rewrite_relative_url($site_url, $item->get_link());
2c08214a 626
68cccafc
AD
627 _debug("title $entry_title", $debug_enabled);
628 _debug("link $entry_link", $debug_enabled);
2c08214a
AD
629
630 if (!$entry_title) $entry_title = date("Y-m-d H:i:s", $entry_timestamp);;
631
19b3992b
AD
632 $entry_content = $item->get_content();
633 if (!$entry_content) $entry_content = $item->get_description();
2c08214a
AD
634
635 if ($_REQUEST["xdebug"] == 2) {
9ec10352 636 print "content: ";
487f0750 637 print $entry_content;
3c696512 638 print "\n";
2c08214a
AD
639 }
640
5ddd2705
AD
641 $entry_language = "";
642
643 if (DETECT_ARTICLE_LANGUAGE) {
644 $entry_language = $lang->detect($entry_title . " " . $entry_content, 1);
6b461797 645
5ddd2705 646 if (count($entry_language) > 0) {
5d2e74a9
JP
647 $possible = array_keys($entry_language);
648 $entry_language = $possible[0];
6b461797 649
5ddd2705 650 _debug("detected language: $entry_language", $debug_enabled);
e95bb179
AD
651 } else {
652 $entry_language = "";
5ddd2705 653 }
6b461797
AD
654 }
655
cd07592c
AD
656 $entry_comments = $item->get_comments_url();
657 $entry_author = $item->get_author();
2c08214a 658
a42c55f0 659 $entry_guid = db_escape_string(mb_substr($entry_guid, 0, 245));
2c08214a 660
ce2b219c
AD
661 $entry_comments = db_escape_string(mb_substr(trim($entry_comments), 0, 245));
662 $entry_author = db_escape_string(mb_substr(trim($entry_author), 0, 245));
2c08214a 663
d4992d6b 664 $num_comments = (int) $item->get_comments_count();
2c08214a 665
68cccafc
AD
666 _debug("author $entry_author", $debug_enabled);
667 _debug("num_comments: $num_comments", $debug_enabled);
ee78f81c 668 _debug("looking for tags...", $debug_enabled);
2c08214a
AD
669
670 // parse <category> entries into tags
671
672 $additional_tags = array();
673
19b3992b 674 $additional_tags_src = $item->get_categories();
2c08214a 675
19b3992b
AD
676 if (is_array($additional_tags_src)) {
677 foreach ($additional_tags_src as $tobj) {
cd07592c 678 array_push($additional_tags, $tobj);
2c08214a 679 }
19b3992b 680 }
2c08214a 681
fa6fbd36 682 $entry_tags = array_unique($additional_tags);
2c08214a
AD
683
684 for ($i = 0; $i < count($entry_tags); $i++)
685 $entry_tags[$i] = mb_strtolower($entry_tags[$i], 'utf-8');
686
ee78f81c
AD
687 _debug("tags found: " . join(",", $entry_tags), $debug_enabled);
688
68cccafc 689 _debug("done collecting data.", $debug_enabled);
2c08214a 690
b1840673
AD
691 $result = db_query("SELECT id, content_hash FROM ttrss_entries
692 WHERE guid = '".db_escape_string($entry_guid)."' OR guid = '$entry_guid_hashed'");
b30abdad
AD
693
694 if (db_num_rows($result) != 0) {
b1840673
AD
695 $base_entry_id = db_fetch_result($result, 0, "id");
696 $entry_stored_hash = db_fetch_result($result, 0, "content_hash");
a29fe121 697 $article_labels = get_article_labels($base_entry_id, $owner_uid);
b30abdad 698 } else {
b1840673
AD
699 $base_entry_id = false;
700 $entry_stored_hash = "";
a29fe121 701 $article_labels = array();
b30abdad
AD
702 }
703
455b1401 704 $article = array("owner_uid" => $owner_uid, // read only
b30abdad 705 "guid" => $entry_guid, // read only
59e83455 706 "guid_hashed" => $entry_guid_hashed, // read only
19b3992b
AD
707 "title" => $entry_title,
708 "content" => $entry_content,
709 "link" => $entry_link,
a29fe121 710 "labels" => $article_labels, // current limitation: can add labels to article, can't remove them
19b3992b 711 "tags" => $entry_tags,
e02555c1 712 "author" => $entry_author,
c9299c28 713 "force_catchup" => false, // ugly hack for the time being
6de3a1be 714 "score_modifier" => 0, // no previous value, plugin should recalculate score modifier based on content if needed
02b0348a 715 "language" => $entry_language, // read only
f73e03e0
AD
716 "feed" => array("id" => $feed,
717 "fetch_url" => $fetch_url,
718 "site_url" => $site_url)
719 );
cc85704f 720
b1840673 721 $entry_plugin_data = "";
eb16bd9f 722 $entry_current_hash = calculate_article_hash($article, $pluginhost);
b1840673
AD
723
724 _debug("article hash: $entry_current_hash [stored=$entry_stored_hash]", $debug_enabled);
725
522e8b35 726 if ($entry_current_hash == $entry_stored_hash && !isset($_REQUEST["force_rehash"])) {
b1840673
AD
727 _debug("stored article seems up to date [IID: $base_entry_id], updating timestamp only", $debug_enabled);
728
729 // we keep encountering the entry in feeds, so we need to
730 // update date_updated column so that we don't get horrible
731 // dupes when the entry gets purged and reinserted again e.g.
732 // in the case of SLOW SLOW OMG SLOW updating feeds
733
734 $base_entry_id = db_fetch_result($result, 0, "id");
735
736 db_query("UPDATE ttrss_entries SET date_updated = NOW()
737 WHERE id = '$base_entry_id'");
738
e44280a0 739 // if we allow duplicate posts, we have to continue to
da7e3e91
GP
740 // create the user entries for this feed
741 if (!get_pref("ALLOW_DUPLICATE_POSTS", $owner_uid, false)) {
742 continue;
743 }
b1840673
AD
744 }
745
746 _debug("hash differs, applying plugin filters:", $debug_enabled);
747
1ffe3391 748 foreach ($pluginhost->get_hooks(PluginHost::HOOK_ARTICLE_FILTER) as $plugin) {
b1840673
AD
749 _debug("... " . get_class($plugin), $debug_enabled);
750
751 $start = microtime(true);
19b3992b 752 $article = $plugin->hook_article_filter($article);
0084f0d1 753
b1840673
AD
754 _debug("=== " . sprintf("%.4f (sec)", microtime(true) - $start), $debug_enabled);
755
756 $entry_plugin_data .= mb_strtolower(get_class($plugin)) . ",";
cc85704f
AD
757 }
758
b1840673
AD
759 $entry_plugin_data = db_escape_string($entry_plugin_data);
760
761 _debug("plugin data: $entry_plugin_data", $debug_enabled);
762
35c12dc4
AD
763 // Workaround: 4-byte unicode requires utf8mb4 in MySQL. See https://tt-rss.org/forum/viewtopic.php?f=1&t=3377&p=20077#p20077
764 if (DB_TYPE == "mysql") {
765 foreach ($article as $k => $v) {
766 $article[$k] = preg_replace('/[\x{10000}-\x{10FFFF}]/u', "\xEF\xBF\xBD", $v);
767 }
768 }
769
19b3992b 770 $entry_tags = $article["tags"];
a42c55f0
AD
771 $entry_guid = db_escape_string($entry_guid);
772 $entry_title = db_escape_string($article["title"]);
773 $entry_author = db_escape_string($article["author"]);
774 $entry_link = db_escape_string($article["link"]);
f935d98e 775 $entry_content = $article["content"]; // escaped below
c9299c28 776 $entry_force_catchup = $article["force_catchup"];
a29fe121 777 $article_labels = $article["labels"];
6de3a1be 778 $entry_score_modifier = (int) $article["score_modifier"];
a29fe121
AD
779
780 if ($debug_enabled) {
781 _debug("article labels:", $debug_enabled);
782 print_r($article_labels);
783 }
c9299c28
AD
784
785 _debug("force catchup: $entry_force_catchup");
f935d98e 786
0a3fd79b 787 if ($cache_images && is_writable(CACHE_DIR . '/images'))
f0bd8e65 788 cache_images($entry_content, $site_url, $debug_enabled);
0a3fd79b 789
a42c55f0 790 $entry_content = db_escape_string($entry_content, false);
1f45c857 791
a42c55f0 792 db_query("BEGIN");
2c08214a 793
a42c55f0 794 $result = db_query("SELECT id FROM ttrss_entries
5e3d5480 795 WHERE (guid = '$entry_guid' OR guid = '$entry_guid_hashed')");
9e222305 796
2c08214a
AD
797 if (db_num_rows($result) == 0) {
798
68cccafc 799 _debug("base guid [$entry_guid] not found", $debug_enabled);
2c08214a
AD
800
801 // base post entry does not exist, create it
802
6322ac79 803 $result = db_query(
2c08214a
AD
804 "INSERT INTO ttrss_entries
805 (title,
806 guid,
807 link,
808 updated,
809 content,
810 content_hash,
811 no_orig_date,
812 date_updated,
813 date_entered,
814 comments,
815 num_comments,
b30abdad 816 plugin_data,
6b461797 817 lang,
2c08214a
AD
818 author)
819 VALUES
820 ('$entry_title',
5e3d5480 821 '$entry_guid_hashed',
2c08214a
AD
822 '$entry_link',
823 '$entry_timestamp_fmt',
824 '$entry_content',
b1840673 825 '$entry_current_hash',
5ba1ddd4 826 false,
2c08214a 827 NOW(),
be574731 828 '$date_feed_processed',
2c08214a
AD
829 '$entry_comments',
830 '$num_comments',
b30abdad 831 '$entry_plugin_data',
6b461797 832 '$entry_language',
2c08214a 833 '$entry_author')");
e8291805 834
2c08214a 835 } else {
2c08214a 836 $base_entry_id = db_fetch_result($result, 0, "id");
2c08214a
AD
837 }
838
839 // now it should exist, if not - bad luck then
840
b1840673 841 $result = db_query("SELECT id FROM ttrss_entries
5e3d5480 842 WHERE guid = '$entry_guid' OR guid = '$entry_guid_hashed'");
2c08214a
AD
843
844 $entry_ref_id = 0;
845 $entry_int_id = 0;
846
847 if (db_num_rows($result) == 1) {
848
68cccafc 849 _debug("base guid found, checking for user record", $debug_enabled);
2c08214a 850
2c08214a
AD
851 $ref_id = db_fetch_result($result, 0, "id");
852 $entry_ref_id = $ref_id;
853
5e3d5480
AD
854 /* $stored_guid = db_fetch_result($result, 0, "guid");
855 if ($stored_guid != $entry_guid_hashed) {
68cccafc 856 if ($debug_enabled) _debug("upgrading compat guid to hashed one", $debug_enabled);
5e3d5480 857
a42c55f0 858 db_query("UPDATE ttrss_entries SET guid = '$entry_guid_hashed' WHERE
5e3d5480
AD
859 id = '$ref_id'");
860 } */
861
2c08214a
AD
862 // check for user post link to main table
863
864 // do we allow duplicate posts with same GUID in different feeds?
a42c55f0 865 if (get_pref("ALLOW_DUPLICATE_POSTS", $owner_uid, false)) {
2c08214a
AD
866 $dupcheck_qpart = "AND (feed_id = '$feed' OR feed_id IS NULL)";
867 } else {
868 $dupcheck_qpart = "";
869 }
870
871 /* Collect article tags here so we could filter by them: */
872
873 $article_filters = get_article_filters($filters, $entry_title,
4021d61a 874 $entry_content, $entry_link, $entry_timestamp, $entry_author,
2c08214a
AD
875 $entry_tags);
876
877 if ($debug_enabled) {
68cccafc 878 _debug("article filters: ", $debug_enabled);
2c08214a
AD
879 if (count($article_filters) != 0) {
880 print_r($article_filters);
881 }
882 }
883
884 if (find_article_filter($article_filters, "filter")) {
a42c55f0 885 db_query("COMMIT"); // close transaction in progress
2c08214a
AD
886 continue;
887 }
888
6de3a1be 889 $score = calculate_article_score($article_filters) + $entry_score_modifier;
2c08214a 890
6de3a1be 891 _debug("initial score: $score [including plugin modifier: $entry_score_modifier]", $debug_enabled);
2c08214a
AD
892
893 $query = "SELECT ref_id, int_id FROM ttrss_user_entries WHERE
894 ref_id = '$ref_id' AND owner_uid = '$owner_uid'
895 $dupcheck_qpart";
896
897// if ($_REQUEST["xdebug"]) print "$query\n";
898
a42c55f0 899 $result = db_query($query);
2c08214a
AD
900
901 // okay it doesn't exist - create user entry
902 if (db_num_rows($result) == 0) {
903
68cccafc 904 _debug("user record not found, creating...", $debug_enabled);
2c08214a 905
c9299c28 906 if ($score >= -500 && !find_article_filter($article_filters, 'catchup') && !$entry_force_catchup) {
2c08214a
AD
907 $unread = 'true';
908 $last_read_qpart = 'NULL';
909 } else {
910 $unread = 'false';
911 $last_read_qpart = 'NOW()';
912 }
913
914 if (find_article_filter($article_filters, 'mark') || $score > 1000) {
915 $marked = 'true';
916 } else {
917 $marked = 'false';
918 }
919
920 if (find_article_filter($article_filters, 'publish')) {
921 $published = 'true';
922 } else {
923 $published = 'false';
924 }
925
2ea9bbfd
AD
926 // N-grams
927
c9299c28 928 /* if (DB_TYPE == "pgsql" and defined('_NGRAM_TITLE_DUPLICATE_THRESHOLD')) {
2ea9bbfd 929
a42c55f0 930 $result = db_query("SELECT COUNT(*) AS similar FROM
2ea9bbfd
AD
931 ttrss_entries,ttrss_user_entries
932 WHERE ref_id = id AND updated >= NOW() - INTERVAL '7 day'
933 AND similarity(title, '$entry_title') >= "._NGRAM_TITLE_DUPLICATE_THRESHOLD."
934 AND owner_uid = $owner_uid");
935
936 $ngram_similar = db_fetch_result($result, 0, "similar");
937
68cccafc 938 _debug("N-gram similar results: $ngram_similar", $debug_enabled);
2ea9bbfd
AD
939
940 if ($ngram_similar > 0) {
941 $unread = 'false';
942 }
c9299c28 943 } */
2ea9bbfd 944
7873d588
AD
945 $last_marked = ($marked == 'true') ? 'NOW()' : 'NULL';
946 $last_published = ($published == 'true') ? 'NOW()' : 'NULL';
947
6322ac79 948 $result = db_query(
2c08214a
AD
949 "INSERT INTO ttrss_user_entries
950 (ref_id, owner_uid, feed_id, unread, last_read, marked,
7873d588
AD
951 published, score, tag_cache, label_cache, uuid,
952 last_marked, last_published)
2c08214a 953 VALUES ('$ref_id', '$owner_uid', '$feed', $unread,
7873d588
AD
954 $last_read_qpart, $marked, $published, '$score', '', '',
955 '', $last_marked, $last_published)");
2c08214a
AD
956
957 if (PUBSUBHUBBUB_HUB && $published == 'true') {
958 $rss_link = get_self_url_prefix() .
959 "/public.php?op=rss&id=-2&key=" .
a42c55f0 960 get_feed_access_key(-2, false, $owner_uid);
2c08214a
AD
961
962 $p = new Publisher(PUBSUBHUBBUB_HUB);
963
5ba1ddd4 964 /* $pubsub_result = */ $p->publish_update($rss_link);
2c08214a
AD
965 }
966
6322ac79 967 $result = db_query(
2c08214a
AD
968 "SELECT int_id FROM ttrss_user_entries WHERE
969 ref_id = '$ref_id' AND owner_uid = '$owner_uid' AND
970 feed_id = '$feed' LIMIT 1");
971
972 if (db_num_rows($result) == 1) {
973 $entry_int_id = db_fetch_result($result, 0, "int_id");
974 }
975 } else {
68cccafc 976 _debug("user record FOUND", $debug_enabled);
2c08214a
AD
977
978 $entry_ref_id = db_fetch_result($result, 0, "ref_id");
979 $entry_int_id = db_fetch_result($result, 0, "int_id");
980 }
981
68cccafc 982 _debug("RID: $entry_ref_id, IID: $entry_int_id", $debug_enabled);
2c08214a 983
b1840673
AD
984 db_query("UPDATE ttrss_entries
985 SET title = '$entry_title',
986 content = '$entry_content',
987 content_hash = '$entry_current_hash',
988 updated = '$entry_timestamp_fmt',
989 num_comments = '$num_comments',
990 plugin_data = '$entry_plugin_data',
991 author = '$entry_author',
992 lang = '$entry_language'
993 WHERE id = '$ref_id'");
994
59e83455
AD
995 // update aux data
996 db_query("UPDATE ttrss_user_entries
997 SET score = '$score' WHERE ref_id = '$ref_id'");
998
b1840673
AD
999 if ($mark_unread_on_update) {
1000 db_query("UPDATE ttrss_user_entries
1001 SET last_read = null, unread = true WHERE ref_id = '$ref_id'");
2c08214a
AD
1002 }
1003 }
1004
a42c55f0 1005 db_query("COMMIT");
2c08214a 1006
a29fe121
AD
1007 _debug("assigning labels [other]...", $debug_enabled);
1008
1009 foreach ($article_labels as $label) {
1010 label_add_article($entry_ref_id, $label[1], $owner_uid);
1011 }
1012
1013 _debug("assigning labels [filters]...", $debug_enabled);
2c08214a 1014
a42c55f0 1015 assign_article_to_label_filters($entry_ref_id, $article_filters,
b24504b1 1016 $owner_uid, $article_labels);
2c08214a 1017
68cccafc 1018 _debug("looking for enclosures...", $debug_enabled);
2c08214a
AD
1019
1020 // enclosures
1021
1022 $enclosures = array();
1023
19b3992b 1024 $encs = $item->get_enclosures();
2c08214a 1025
19b3992b
AD
1026 if (is_array($encs)) {
1027 foreach ($encs as $e) {
1028 $e_item = array(
523bd90b 1029 $e->link, $e->type, $e->length, $e->title, $e->width, $e->height);
2c08214a 1030 array_push($enclosures, $e_item);
2c08214a
AD
1031 }
1032 }
1033
2c08214a 1034 if ($debug_enabled) {
68cccafc 1035 _debug("article enclosures:", $debug_enabled);
2c08214a
AD
1036 print_r($enclosures);
1037 }
1038
a42c55f0 1039 db_query("BEGIN");
2c08214a 1040
5c54e683
AD
1041// debugging
1042// db_query("DELETE FROM ttrss_enclosures WHERE post_id = '$entry_ref_id'");
1043
2c08214a 1044 foreach ($enclosures as $enc) {
a42c55f0
AD
1045 $enc_url = db_escape_string($enc[0]);
1046 $enc_type = db_escape_string($enc[1]);
1047 $enc_dur = db_escape_string($enc[2]);
5c54e683 1048 $enc_title = db_escape_string($enc[3]);
523bd90b
FE
1049 $enc_width = intval($enc[4]);
1050 $enc_height = intval($enc[5]);
2c08214a 1051
a42c55f0 1052 $result = db_query("SELECT id FROM ttrss_enclosures
2c08214a
AD
1053 WHERE content_url = '$enc_url' AND post_id = '$entry_ref_id'");
1054
1055 if (db_num_rows($result) == 0) {
a42c55f0 1056 db_query("INSERT INTO ttrss_enclosures
523bd90b
FE
1057 (content_url, content_type, title, duration, post_id, width, height) VALUES
1058 ('$enc_url', '$enc_type', '$enc_title', '$enc_dur', '$entry_ref_id', $enc_width, $enc_height)");
2c08214a
AD
1059 }
1060 }
1061
a42c55f0 1062 db_query("COMMIT");
2c08214a
AD
1063
1064 // check for manual tags (we have to do it here since they're loaded from filters)
1065
1066 foreach ($article_filters as $f) {
6aff7845 1067 if ($f["type"] == "tag") {
2c08214a 1068
6aff7845 1069 $manual_tags = trim_array(explode(",", $f["param"]));
2c08214a
AD
1070
1071 foreach ($manual_tags as $tag) {
1072 if (tag_is_valid($tag)) {
1073 array_push($entry_tags, $tag);
1074 }
1075 }
1076 }
1077 }
1078
1079 // Skip boring tags
1080
6322ac79 1081 $boring_tags = trim_array(explode(",", mb_strtolower(get_pref(
2c08214a
AD
1082 'BLACKLISTED_TAGS', $owner_uid, ''), 'utf-8')));
1083
1084 $filtered_tags = array();
1085 $tags_to_cache = array();
1086
1087 if ($entry_tags && is_array($entry_tags)) {
1088 foreach ($entry_tags as $tag) {
1089 if (array_search($tag, $boring_tags) === false) {
1090 array_push($filtered_tags, $tag);
1091 }
1092 }
1093 }
1094
1095 $filtered_tags = array_unique($filtered_tags);
1096
1097 if ($debug_enabled) {
68cccafc 1098 _debug("filtered article tags:", $debug_enabled);
2c08214a
AD
1099 print_r($filtered_tags);
1100 }
1101
1102 // Save article tags in the database
1103
1104 if (count($filtered_tags) > 0) {
1105
a42c55f0 1106 db_query("BEGIN");
2c08214a
AD
1107
1108 foreach ($filtered_tags as $tag) {
1109
1110 $tag = sanitize_tag($tag);
a42c55f0 1111 $tag = db_escape_string($tag);
2c08214a
AD
1112
1113 if (!tag_is_valid($tag)) continue;
1114
a42c55f0 1115 $result = db_query("SELECT id FROM ttrss_tags
2c08214a
AD
1116 WHERE tag_name = '$tag' AND post_int_id = '$entry_int_id' AND
1117 owner_uid = '$owner_uid' LIMIT 1");
1118
1119 if ($result && db_num_rows($result) == 0) {
1120
a42c55f0 1121 db_query("INSERT INTO ttrss_tags
2c08214a
AD
1122 (owner_uid,tag_name,post_int_id)
1123 VALUES ('$owner_uid','$tag', '$entry_int_id')");
1124 }
1125
1126 array_push($tags_to_cache, $tag);
1127 }
1128
1129 /* update the cache */
1130
1131 $tags_to_cache = array_unique($tags_to_cache);
1132
a42c55f0 1133 $tags_str = db_escape_string(join(",", $tags_to_cache));
2c08214a 1134
a42c55f0 1135 db_query("UPDATE ttrss_user_entries
2c08214a
AD
1136 SET tag_cache = '$tags_str' WHERE ref_id = '$entry_ref_id'
1137 AND owner_uid = $owner_uid");
1138
a42c55f0 1139 db_query("COMMIT");
2c08214a
AD
1140 }
1141
68cccafc 1142 _debug("article processed", $debug_enabled);
2c08214a
AD
1143 }
1144
68cccafc 1145 _debug("purging feed...", $debug_enabled);
2c08214a 1146
a42c55f0 1147 purge_feed($feed, 0, $debug_enabled);
2c08214a 1148
a42c55f0 1149 db_query("UPDATE ttrss_feeds
f2798eb6 1150 SET last_updated = NOW(), last_error = '' WHERE id = '$feed'");
2c08214a 1151
a42c55f0 1152// db_query("COMMIT");
2c08214a
AD
1153
1154 } else {
1155
a42c55f0 1156 $error_msg = db_escape_string(mb_substr($rss->error(), 0, 245));
2c08214a 1157
4ad04ee2
AD
1158 _debug("fetch error: $error_msg", $debug_enabled);
1159
1160 if (count($rss->errors()) > 1) {
1161 foreach ($rss->errors() as $error) {
1162 _debug("+ $error");
1163 }
1164 }
2c08214a 1165
6322ac79 1166 db_query(
2c08214a 1167 "UPDATE ttrss_feeds SET last_error = '$error_msg',
88edaa93 1168 last_updated = NOW() WHERE id = '$feed'");
2c08214a 1169
88edaa93
AD
1170 unset($rss);
1171 }
2c08214a 1172
68cccafc 1173 _debug("done", $debug_enabled);
88edaa93
AD
1174
1175 return $rss;
2c08214a
AD
1176 }
1177
3c696512 1178 function cache_images($html, $site_url, $debug) {
3c696512
AD
1179 libxml_use_internal_errors(true);
1180
1181 $charset_hack = '<head>
1182 <meta http-equiv="Content-Type" content="text/html; charset=utf-8"/>
1183 </head>';
1184
1185 $doc = new DOMDocument();
1186 $doc->loadHTML($charset_hack . $html);
1187 $xpath = new DOMXPath($doc);
1188
1189 $entries = $xpath->query('(//img[@src])');
1190
1191 foreach ($entries as $entry) {
1192 if ($entry->hasAttribute('src')) {
1193 $src = rewrite_relative_url($site_url, $entry->getAttribute('src'));
1194
1195 $local_filename = CACHE_DIR . "/images/" . sha1($src) . ".png";
1196
1197 if ($debug) _debug("cache_images: downloading: $src to $local_filename");
1198
1199 if (!file_exists($local_filename)) {
1200 $file_content = fetch_file_contents($src);
1201
a3d50184 1202 if ($file_content && strlen($file_content) > _MIN_CACHE_IMAGE_SIZE) {
3c696512
AD
1203 file_put_contents($local_filename, $file_content);
1204 }
1205 }
3c696512
AD
1206 }
1207 }
3c696512
AD
1208 }
1209
a42c55f0 1210 function expire_error_log($debug) {
e2261e17
AD
1211 if ($debug) _debug("Removing old error log entries...");
1212
1213 if (DB_TYPE == "pgsql") {
a42c55f0 1214 db_query("DELETE FROM ttrss_error_log
e2261e17
AD
1215 WHERE created_at < NOW() - INTERVAL '7 days'");
1216 } else {
a42c55f0 1217 db_query("DELETE FROM ttrss_error_log
e2261e17
AD
1218 WHERE created_at < DATE_SUB(NOW(), INTERVAL 7 DAY)");
1219 }
1220
1221 }
1222
2a91b6ff 1223 function expire_lock_files($debug) {
65465085 1224 //if ($debug) _debug("Removing old lock files...");
2a91b6ff
AD
1225
1226 $num_deleted = 0;
1227
1228 if (is_writable(LOCK_DIRECTORY)) {
1229 $files = glob(LOCK_DIRECTORY . "/*.lock");
1230
1231 if ($files) {
1232 foreach ($files as $file) {
11344971 1233 if (!file_is_locked(basename($file)) && time() - filemtime($file) > 86400*2) {
2a91b6ff
AD
1234 unlink($file);
1235 ++$num_deleted;
1236 }
1237 }
1238 }
1239 }
1240
65465085 1241 if ($debug) _debug("Removed $num_deleted old lock files.");
2a91b6ff
AD
1242 }
1243
3c696512 1244 function expire_cached_files($debug) {
3306daec 1245 foreach (array("simplepie", "images", "export", "upload") as $dir) {
3c696512 1246 $cache_dir = CACHE_DIR . "/$dir";
2c08214a 1247
65465085 1248// if ($debug) _debug("Expiring $cache_dir");
2c08214a 1249
3c696512
AD
1250 $num_deleted = 0;
1251
1252 if (is_writable($cache_dir)) {
1253 $files = glob("$cache_dir/*");
1254
2a91b6ff 1255 if ($files) {
2ab20c31
AD
1256 foreach ($files as $file) {
1257 if (time() - filemtime($file) > 86400*7) {
1258 unlink($file);
3c696512 1259
2ab20c31
AD
1260 ++$num_deleted;
1261 }
3c696512
AD
1262 }
1263 }
2a91b6ff 1264 }
3c696512 1265
65465085 1266 if ($debug) _debug("$cache_dir: removed $num_deleted files.");
3c696512
AD
1267 }
1268 }
2c08214a 1269
a3e0bdcf
AD
1270 /**
1271 * Source: http://www.php.net/manual/en/function.parse-url.php#104527
1272 * Returns the url query as associative array
1273 *
1274 * @param string query
1275 * @return array params
1276 */
1277 function convertUrlQuery($query) {
1278 $queryParts = explode('&', $query);
1279
1280 $params = array();
1281
1282 foreach ($queryParts as $param) {
1283 $item = explode('=', $param);
1284 $params[$item[0]] = $item[1];
1285 }
1286
1287 return $params;
1288 }
92c14e9d 1289
48cb2536 1290 function get_article_filters($filters, $title, $content, $link, $timestamp, $author, $tags) {
92c14e9d
AD
1291 $matches = array();
1292
1293 foreach ($filters as $filter) {
1294 $match_any_rule = $filter["match_any_rule"];
a3a896a1 1295 $inverse = $filter["inverse"];
92c14e9d
AD
1296 $filter_match = false;
1297
1298 foreach ($filter["rules"] as $rule) {
1299 $match = false;
ffa1bd7b 1300 $reg_exp = str_replace('/', '\/', $rule["reg_exp"]);
a3a896a1 1301 $rule_inverse = $rule["inverse"];
92c14e9d
AD
1302
1303 if (!$reg_exp)
1304 continue;
1305
1306 switch ($rule["type"]) {
1307 case "title":
1308 $match = @preg_match("/$reg_exp/i", $title);
1309 break;
1310 case "content":
d03ae73e
AD
1311 // we don't need to deal with multiline regexps
1312 $content = preg_replace("/[\r\n\t]/", "", $content);
1313
92c14e9d
AD
1314 $match = @preg_match("/$reg_exp/i", $content);
1315 break;
1316 case "both":
d03ae73e
AD
1317 // we don't need to deal with multiline regexps
1318 $content = preg_replace("/[\r\n\t]/", "", $content);
1319
72d1d067 1320 $match = (@preg_match("/$reg_exp/i", $title) || @preg_match("/$reg_exp/i", $content));
92c14e9d
AD
1321 break;
1322 case "link":
1323 $match = @preg_match("/$reg_exp/i", $link);
1324 break;
1325 case "author":
1326 $match = @preg_match("/$reg_exp/i", $author);
1327 break;
1328 case "tag":
7b80b5e1
DK
1329 foreach ($tags as $tag) {
1330 if (@preg_match("/$reg_exp/i", $tag)) {
1331 $match = true;
1332 break;
1333 }
1334 }
92c14e9d
AD
1335 break;
1336 }
1337
a3a896a1
AD
1338 if ($rule_inverse) $match = !$match;
1339
92c14e9d
AD
1340 if ($match_any_rule) {
1341 if ($match) {
1342 $filter_match = true;
1343 break;
1344 }
1345 } else {
1346 $filter_match = $match;
1347 if (!$match) {
1348 break;
1349 }
1350 }
1351 }
1352
a3a896a1
AD
1353 if ($inverse) $filter_match = !$filter_match;
1354
92c14e9d
AD
1355 if ($filter_match) {
1356 foreach ($filter["actions"] AS $action) {
1357 array_push($matches, $action);
5e736e45
AD
1358
1359 // if Stop action encountered, perform no further processing
1360 if ($action["type"] == "stop") return $matches;
92c14e9d
AD
1361 }
1362 }
1363 }
1364
1365 return $matches;
1366 }
1367
1368 function find_article_filter($filters, $filter_name) {
1369 foreach ($filters as $f) {
1370 if ($f["type"] == $filter_name) {
1371 return $f;
1372 };
1373 }
1374 return false;
1375 }
1376
1377 function find_article_filters($filters, $filter_name) {
1378 $results = array();
1379
1380 foreach ($filters as $f) {
1381 if ($f["type"] == $filter_name) {
1382 array_push($results, $f);
1383 };
1384 }
1385 return $results;
1386 }
1387
1388 function calculate_article_score($filters) {
1389 $score = 0;
1390
1391 foreach ($filters as $f) {
1392 if ($f["type"] == "score") {
1393 $score += $f["param"];
1394 };
1395 }
1396 return $score;
1397 }
1398
b24504b1
AD
1399 function labels_contains_caption($labels, $caption) {
1400 foreach ($labels as $label) {
1401 if ($label[1] == $caption) {
1402 return true;
1403 }
1404 }
1405
1406 return false;
1407 }
1408
a42c55f0 1409 function assign_article_to_label_filters($id, $filters, $owner_uid, $article_labels) {
92c14e9d
AD
1410 foreach ($filters as $f) {
1411 if ($f["type"] == "label") {
b24504b1 1412 if (!labels_contains_caption($article_labels, $f["param"])) {
a42c55f0 1413 label_add_article($id, $f["param"], $owner_uid);
b24504b1
AD
1414 }
1415 }
92c14e9d
AD
1416 }
1417 }
87764a50 1418
87d7e850
AD
1419 function make_guid_from_title($title) {
1420 return preg_replace("/[ \"\',.:;]/", "-",
1421 mb_strtolower(strip_tags($title), 'utf-8'));
1422 }
1423
d1f3fa97 1424 /* function verify_feed_xml($feed_data) {
ebec81a6
AD
1425 libxml_use_internal_errors(true);
1426 $doc = new DOMDocument();
1427 $doc->loadXML($feed_data);
1428 $error = libxml_get_last_error();
1429 libxml_clear_errors();
1430 return $error;
d1f3fa97 1431 } */
87d7e850 1432
168cf351
AD
1433 function cleanup_counters_cache($debug) {
1434 $result = db_query("DELETE FROM ttrss_counters_cache
1435 WHERE feed_id > 0 AND
1436 (SELECT COUNT(id) FROM ttrss_feeds WHERE
1437 id = feed_id AND
1438 ttrss_counters_cache.owner_uid = ttrss_feeds.owner_uid) = 0");
1439 $frows = db_affected_rows($result);
1440
1441 $result = db_query("DELETE FROM ttrss_cat_counters_cache
1442 WHERE feed_id > 0 AND
1443 (SELECT COUNT(id) FROM ttrss_feed_categories WHERE
1444 id = feed_id AND
1445 ttrss_cat_counters_cache.owner_uid = ttrss_feed_categories.owner_uid) = 0");
1446 $crows = db_affected_rows($result);
1447
1448 _debug("Removed $frows (feeds) $crows (cats) orphaned counter cache entries.");
1449 }
1450
e2cf81e2
AD
1451 function housekeeping_common($debug) {
1452 expire_cached_files($debug);
1453 expire_lock_files($debug);
1454 expire_error_log($debug);
1455
1456 $count = update_feedbrowser_cache();
1457 _debug("Feedbrowser updated, $count feeds processed.");
1458
1459 purge_orphans( true);
168cf351 1460 cleanup_counters_cache($debug);
e2cf81e2
AD
1461 $rc = cleanup_tags( 14, 50000);
1462
1463 _debug("Cleaned $rc cached tags.");
8e470220 1464
00f22824 1465 PluginHost::getInstance()->run_hooks(PluginHost::HOOK_HOUSE_KEEPING, "hook_house_keeping", "");
8e470220 1466
e2cf81e2 1467 }
2c08214a 1468?>