]> git.wh0rd.org - tt-rss.git/blame - include/rssfuncs.php
show ugly minimum word count for auto categorizing in prefs ui
[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 $nf;
222
2c08214a
AD
223 } // function update_daemon_common
224
87764a50 225 // ignore_daemon is not used
88edaa93 226 function update_rss_feed($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");
a29fe121 674 $article_labels = get_article_labels($base_entry_id, $owner_uid);
b30abdad 675 } else {
b1840673
AD
676 $base_entry_id = false;
677 $entry_stored_hash = "";
a29fe121 678 $article_labels = array();
b30abdad
AD
679 }
680
455b1401 681 $article = array("owner_uid" => $owner_uid, // read only
b30abdad 682 "guid" => $entry_guid, // read only
59e83455 683 "guid_hashed" => $entry_guid_hashed, // read only
19b3992b
AD
684 "title" => $entry_title,
685 "content" => $entry_content,
686 "link" => $entry_link,
a29fe121 687 "labels" => $article_labels, // current limitation: can add labels to article, can't remove them
19b3992b 688 "tags" => $entry_tags,
e02555c1 689 "author" => $entry_author,
c9299c28 690 "force_catchup" => false, // ugly hack for the time being
6de3a1be 691 "score_modifier" => 0, // no previous value, plugin should recalculate score modifier based on content if needed
02b0348a 692 "language" => $entry_language, // read only
f73e03e0
AD
693 "feed" => array("id" => $feed,
694 "fetch_url" => $fetch_url,
695 "site_url" => $site_url)
696 );
cc85704f 697
b1840673 698 $entry_plugin_data = "";
eb16bd9f 699 $entry_current_hash = calculate_article_hash($article, $pluginhost);
b1840673
AD
700
701 _debug("article hash: $entry_current_hash [stored=$entry_stored_hash]", $debug_enabled);
702
522e8b35 703 if ($entry_current_hash == $entry_stored_hash && !isset($_REQUEST["force_rehash"])) {
b1840673
AD
704 _debug("stored article seems up to date [IID: $base_entry_id], updating timestamp only", $debug_enabled);
705
706 // we keep encountering the entry in feeds, so we need to
707 // update date_updated column so that we don't get horrible
708 // dupes when the entry gets purged and reinserted again e.g.
709 // in the case of SLOW SLOW OMG SLOW updating feeds
710
711 $base_entry_id = db_fetch_result($result, 0, "id");
712
713 db_query("UPDATE ttrss_entries SET date_updated = NOW()
714 WHERE id = '$base_entry_id'");
715
e44280a0 716 // if we allow duplicate posts, we have to continue to
da7e3e91
GP
717 // create the user entries for this feed
718 if (!get_pref("ALLOW_DUPLICATE_POSTS", $owner_uid, false)) {
719 continue;
720 }
b1840673
AD
721 }
722
723 _debug("hash differs, applying plugin filters:", $debug_enabled);
724
1ffe3391 725 foreach ($pluginhost->get_hooks(PluginHost::HOOK_ARTICLE_FILTER) as $plugin) {
b1840673
AD
726 _debug("... " . get_class($plugin), $debug_enabled);
727
728 $start = microtime(true);
19b3992b 729 $article = $plugin->hook_article_filter($article);
0084f0d1 730
b1840673
AD
731 _debug("=== " . sprintf("%.4f (sec)", microtime(true) - $start), $debug_enabled);
732
733 $entry_plugin_data .= mb_strtolower(get_class($plugin)) . ",";
cc85704f
AD
734 }
735
b1840673
AD
736 $entry_plugin_data = db_escape_string($entry_plugin_data);
737
738 _debug("plugin data: $entry_plugin_data", $debug_enabled);
739
35c12dc4
AD
740 // Workaround: 4-byte unicode requires utf8mb4 in MySQL. See https://tt-rss.org/forum/viewtopic.php?f=1&t=3377&p=20077#p20077
741 if (DB_TYPE == "mysql") {
742 foreach ($article as $k => $v) {
743 $article[$k] = preg_replace('/[\x{10000}-\x{10FFFF}]/u', "\xEF\xBF\xBD", $v);
744 }
745 }
746
19b3992b 747 $entry_tags = $article["tags"];
a42c55f0
AD
748 $entry_guid = db_escape_string($entry_guid);
749 $entry_title = db_escape_string($article["title"]);
750 $entry_author = db_escape_string($article["author"]);
751 $entry_link = db_escape_string($article["link"]);
f935d98e 752 $entry_content = $article["content"]; // escaped below
c9299c28 753 $entry_force_catchup = $article["force_catchup"];
a29fe121 754 $article_labels = $article["labels"];
6de3a1be 755 $entry_score_modifier = (int) $article["score_modifier"];
a29fe121
AD
756
757 if ($debug_enabled) {
758 _debug("article labels:", $debug_enabled);
759 print_r($article_labels);
760 }
c9299c28
AD
761
762 _debug("force catchup: $entry_force_catchup");
f935d98e 763
0a3fd79b 764 if ($cache_images && is_writable(CACHE_DIR . '/images'))
f0bd8e65 765 cache_images($entry_content, $site_url, $debug_enabled);
0a3fd79b 766
a42c55f0 767 $entry_content = db_escape_string($entry_content, false);
1f45c857 768
a42c55f0 769 db_query("BEGIN");
2c08214a 770
a42c55f0 771 $result = db_query("SELECT id FROM ttrss_entries
5e3d5480 772 WHERE (guid = '$entry_guid' OR guid = '$entry_guid_hashed')");
9e222305 773
2c08214a
AD
774 if (db_num_rows($result) == 0) {
775
68cccafc 776 _debug("base guid [$entry_guid] not found", $debug_enabled);
2c08214a
AD
777
778 // base post entry does not exist, create it
779
6322ac79 780 $result = db_query(
2c08214a
AD
781 "INSERT INTO ttrss_entries
782 (title,
783 guid,
784 link,
785 updated,
786 content,
787 content_hash,
788 no_orig_date,
789 date_updated,
790 date_entered,
791 comments,
792 num_comments,
b30abdad 793 plugin_data,
6b461797 794 lang,
2c08214a
AD
795 author)
796 VALUES
797 ('$entry_title',
5e3d5480 798 '$entry_guid_hashed',
2c08214a
AD
799 '$entry_link',
800 '$entry_timestamp_fmt',
801 '$entry_content',
b1840673 802 '$entry_current_hash',
5ba1ddd4 803 false,
2c08214a 804 NOW(),
be574731 805 '$date_feed_processed',
2c08214a
AD
806 '$entry_comments',
807 '$num_comments',
b30abdad 808 '$entry_plugin_data',
6b461797 809 '$entry_language',
2c08214a 810 '$entry_author')");
e8291805 811
2c08214a 812 } else {
2c08214a 813 $base_entry_id = db_fetch_result($result, 0, "id");
2c08214a
AD
814 }
815
816 // now it should exist, if not - bad luck then
817
b1840673 818 $result = db_query("SELECT id FROM ttrss_entries
5e3d5480 819 WHERE guid = '$entry_guid' OR guid = '$entry_guid_hashed'");
2c08214a
AD
820
821 $entry_ref_id = 0;
822 $entry_int_id = 0;
823
824 if (db_num_rows($result) == 1) {
825
68cccafc 826 _debug("base guid found, checking for user record", $debug_enabled);
2c08214a 827
2c08214a
AD
828 $ref_id = db_fetch_result($result, 0, "id");
829 $entry_ref_id = $ref_id;
830
5e3d5480
AD
831 /* $stored_guid = db_fetch_result($result, 0, "guid");
832 if ($stored_guid != $entry_guid_hashed) {
68cccafc 833 if ($debug_enabled) _debug("upgrading compat guid to hashed one", $debug_enabled);
5e3d5480 834
a42c55f0 835 db_query("UPDATE ttrss_entries SET guid = '$entry_guid_hashed' WHERE
5e3d5480
AD
836 id = '$ref_id'");
837 } */
838
2c08214a
AD
839 // check for user post link to main table
840
841 // do we allow duplicate posts with same GUID in different feeds?
a42c55f0 842 if (get_pref("ALLOW_DUPLICATE_POSTS", $owner_uid, false)) {
2c08214a
AD
843 $dupcheck_qpart = "AND (feed_id = '$feed' OR feed_id IS NULL)";
844 } else {
845 $dupcheck_qpart = "";
846 }
847
848 /* Collect article tags here so we could filter by them: */
849
850 $article_filters = get_article_filters($filters, $entry_title,
4021d61a 851 $entry_content, $entry_link, $entry_timestamp, $entry_author,
2c08214a
AD
852 $entry_tags);
853
854 if ($debug_enabled) {
68cccafc 855 _debug("article filters: ", $debug_enabled);
2c08214a
AD
856 if (count($article_filters) != 0) {
857 print_r($article_filters);
858 }
859 }
860
861 if (find_article_filter($article_filters, "filter")) {
a42c55f0 862 db_query("COMMIT"); // close transaction in progress
2c08214a
AD
863 continue;
864 }
865
6de3a1be 866 $score = calculate_article_score($article_filters) + $entry_score_modifier;
2c08214a 867
6de3a1be 868 _debug("initial score: $score [including plugin modifier: $entry_score_modifier]", $debug_enabled);
2c08214a
AD
869
870 $query = "SELECT ref_id, int_id FROM ttrss_user_entries WHERE
871 ref_id = '$ref_id' AND owner_uid = '$owner_uid'
872 $dupcheck_qpart";
873
874// if ($_REQUEST["xdebug"]) print "$query\n";
875
a42c55f0 876 $result = db_query($query);
2c08214a
AD
877
878 // okay it doesn't exist - create user entry
879 if (db_num_rows($result) == 0) {
880
68cccafc 881 _debug("user record not found, creating...", $debug_enabled);
2c08214a 882
c9299c28 883 if ($score >= -500 && !find_article_filter($article_filters, 'catchup') && !$entry_force_catchup) {
2c08214a
AD
884 $unread = 'true';
885 $last_read_qpart = 'NULL';
886 } else {
887 $unread = 'false';
888 $last_read_qpart = 'NOW()';
889 }
890
891 if (find_article_filter($article_filters, 'mark') || $score > 1000) {
892 $marked = 'true';
893 } else {
894 $marked = 'false';
895 }
896
897 if (find_article_filter($article_filters, 'publish')) {
898 $published = 'true';
899 } else {
900 $published = 'false';
901 }
902
2ea9bbfd
AD
903 // N-grams
904
c9299c28 905 /* if (DB_TYPE == "pgsql" and defined('_NGRAM_TITLE_DUPLICATE_THRESHOLD')) {
2ea9bbfd 906
a42c55f0 907 $result = db_query("SELECT COUNT(*) AS similar FROM
2ea9bbfd
AD
908 ttrss_entries,ttrss_user_entries
909 WHERE ref_id = id AND updated >= NOW() - INTERVAL '7 day'
910 AND similarity(title, '$entry_title') >= "._NGRAM_TITLE_DUPLICATE_THRESHOLD."
911 AND owner_uid = $owner_uid");
912
913 $ngram_similar = db_fetch_result($result, 0, "similar");
914
68cccafc 915 _debug("N-gram similar results: $ngram_similar", $debug_enabled);
2ea9bbfd
AD
916
917 if ($ngram_similar > 0) {
918 $unread = 'false';
919 }
c9299c28 920 } */
2ea9bbfd 921
7873d588
AD
922 $last_marked = ($marked == 'true') ? 'NOW()' : 'NULL';
923 $last_published = ($published == 'true') ? 'NOW()' : 'NULL';
924
6322ac79 925 $result = db_query(
2c08214a
AD
926 "INSERT INTO ttrss_user_entries
927 (ref_id, owner_uid, feed_id, unread, last_read, marked,
7873d588
AD
928 published, score, tag_cache, label_cache, uuid,
929 last_marked, last_published)
2c08214a 930 VALUES ('$ref_id', '$owner_uid', '$feed', $unread,
7873d588
AD
931 $last_read_qpart, $marked, $published, '$score', '', '',
932 '', $last_marked, $last_published)");
2c08214a
AD
933
934 if (PUBSUBHUBBUB_HUB && $published == 'true') {
935 $rss_link = get_self_url_prefix() .
936 "/public.php?op=rss&id=-2&key=" .
a42c55f0 937 get_feed_access_key(-2, false, $owner_uid);
2c08214a
AD
938
939 $p = new Publisher(PUBSUBHUBBUB_HUB);
940
5ba1ddd4 941 /* $pubsub_result = */ $p->publish_update($rss_link);
2c08214a
AD
942 }
943
6322ac79 944 $result = db_query(
2c08214a
AD
945 "SELECT int_id FROM ttrss_user_entries WHERE
946 ref_id = '$ref_id' AND owner_uid = '$owner_uid' AND
947 feed_id = '$feed' LIMIT 1");
948
949 if (db_num_rows($result) == 1) {
950 $entry_int_id = db_fetch_result($result, 0, "int_id");
951 }
952 } else {
68cccafc 953 _debug("user record FOUND", $debug_enabled);
2c08214a
AD
954
955 $entry_ref_id = db_fetch_result($result, 0, "ref_id");
956 $entry_int_id = db_fetch_result($result, 0, "int_id");
957 }
958
68cccafc 959 _debug("RID: $entry_ref_id, IID: $entry_int_id", $debug_enabled);
2c08214a 960
b1840673
AD
961 db_query("UPDATE ttrss_entries
962 SET title = '$entry_title',
963 content = '$entry_content',
964 content_hash = '$entry_current_hash',
965 updated = '$entry_timestamp_fmt',
966 num_comments = '$num_comments',
967 plugin_data = '$entry_plugin_data',
968 author = '$entry_author',
969 lang = '$entry_language'
970 WHERE id = '$ref_id'");
971
59e83455
AD
972 // update aux data
973 db_query("UPDATE ttrss_user_entries
974 SET score = '$score' WHERE ref_id = '$ref_id'");
975
b1840673
AD
976 if ($mark_unread_on_update) {
977 db_query("UPDATE ttrss_user_entries
978 SET last_read = null, unread = true WHERE ref_id = '$ref_id'");
2c08214a
AD
979 }
980 }
981
a42c55f0 982 db_query("COMMIT");
2c08214a 983
a29fe121
AD
984 _debug("assigning labels [other]...", $debug_enabled);
985
986 foreach ($article_labels as $label) {
987 label_add_article($entry_ref_id, $label[1], $owner_uid);
988 }
989
990 _debug("assigning labels [filters]...", $debug_enabled);
2c08214a 991
a42c55f0 992 assign_article_to_label_filters($entry_ref_id, $article_filters,
b24504b1 993 $owner_uid, $article_labels);
2c08214a 994
68cccafc 995 _debug("looking for enclosures...", $debug_enabled);
2c08214a
AD
996
997 // enclosures
998
999 $enclosures = array();
1000
19b3992b 1001 $encs = $item->get_enclosures();
2c08214a 1002
19b3992b
AD
1003 if (is_array($encs)) {
1004 foreach ($encs as $e) {
1005 $e_item = array(
523bd90b 1006 $e->link, $e->type, $e->length, $e->title, $e->width, $e->height);
2c08214a 1007 array_push($enclosures, $e_item);
2c08214a
AD
1008 }
1009 }
1010
2c08214a 1011 if ($debug_enabled) {
68cccafc 1012 _debug("article enclosures:", $debug_enabled);
2c08214a
AD
1013 print_r($enclosures);
1014 }
1015
a42c55f0 1016 db_query("BEGIN");
2c08214a 1017
5c54e683
AD
1018// debugging
1019// db_query("DELETE FROM ttrss_enclosures WHERE post_id = '$entry_ref_id'");
1020
2c08214a 1021 foreach ($enclosures as $enc) {
a42c55f0
AD
1022 $enc_url = db_escape_string($enc[0]);
1023 $enc_type = db_escape_string($enc[1]);
1024 $enc_dur = db_escape_string($enc[2]);
5c54e683 1025 $enc_title = db_escape_string($enc[3]);
523bd90b
FE
1026 $enc_width = intval($enc[4]);
1027 $enc_height = intval($enc[5]);
2c08214a 1028
a42c55f0 1029 $result = db_query("SELECT id FROM ttrss_enclosures
2c08214a
AD
1030 WHERE content_url = '$enc_url' AND post_id = '$entry_ref_id'");
1031
1032 if (db_num_rows($result) == 0) {
a42c55f0 1033 db_query("INSERT INTO ttrss_enclosures
523bd90b
FE
1034 (content_url, content_type, title, duration, post_id, width, height) VALUES
1035 ('$enc_url', '$enc_type', '$enc_title', '$enc_dur', '$entry_ref_id', $enc_width, $enc_height)");
2c08214a
AD
1036 }
1037 }
1038
a42c55f0 1039 db_query("COMMIT");
2c08214a
AD
1040
1041 // check for manual tags (we have to do it here since they're loaded from filters)
1042
1043 foreach ($article_filters as $f) {
6aff7845 1044 if ($f["type"] == "tag") {
2c08214a 1045
6aff7845 1046 $manual_tags = trim_array(explode(",", $f["param"]));
2c08214a
AD
1047
1048 foreach ($manual_tags as $tag) {
1049 if (tag_is_valid($tag)) {
1050 array_push($entry_tags, $tag);
1051 }
1052 }
1053 }
1054 }
1055
1056 // Skip boring tags
1057
6322ac79 1058 $boring_tags = trim_array(explode(",", mb_strtolower(get_pref(
2c08214a
AD
1059 'BLACKLISTED_TAGS', $owner_uid, ''), 'utf-8')));
1060
1061 $filtered_tags = array();
1062 $tags_to_cache = array();
1063
1064 if ($entry_tags && is_array($entry_tags)) {
1065 foreach ($entry_tags as $tag) {
1066 if (array_search($tag, $boring_tags) === false) {
1067 array_push($filtered_tags, $tag);
1068 }
1069 }
1070 }
1071
1072 $filtered_tags = array_unique($filtered_tags);
1073
1074 if ($debug_enabled) {
68cccafc 1075 _debug("filtered article tags:", $debug_enabled);
2c08214a
AD
1076 print_r($filtered_tags);
1077 }
1078
1079 // Save article tags in the database
1080
1081 if (count($filtered_tags) > 0) {
1082
a42c55f0 1083 db_query("BEGIN");
2c08214a
AD
1084
1085 foreach ($filtered_tags as $tag) {
1086
1087 $tag = sanitize_tag($tag);
a42c55f0 1088 $tag = db_escape_string($tag);
2c08214a
AD
1089
1090 if (!tag_is_valid($tag)) continue;
1091
a42c55f0 1092 $result = db_query("SELECT id FROM ttrss_tags
2c08214a
AD
1093 WHERE tag_name = '$tag' AND post_int_id = '$entry_int_id' AND
1094 owner_uid = '$owner_uid' LIMIT 1");
1095
1096 if ($result && db_num_rows($result) == 0) {
1097
a42c55f0 1098 db_query("INSERT INTO ttrss_tags
2c08214a
AD
1099 (owner_uid,tag_name,post_int_id)
1100 VALUES ('$owner_uid','$tag', '$entry_int_id')");
1101 }
1102
1103 array_push($tags_to_cache, $tag);
1104 }
1105
1106 /* update the cache */
1107
1108 $tags_to_cache = array_unique($tags_to_cache);
1109
a42c55f0 1110 $tags_str = db_escape_string(join(",", $tags_to_cache));
2c08214a 1111
a42c55f0 1112 db_query("UPDATE ttrss_user_entries
2c08214a
AD
1113 SET tag_cache = '$tags_str' WHERE ref_id = '$entry_ref_id'
1114 AND owner_uid = $owner_uid");
1115
a42c55f0 1116 db_query("COMMIT");
2c08214a
AD
1117 }
1118
68cccafc 1119 _debug("article processed", $debug_enabled);
2c08214a
AD
1120 }
1121
68cccafc 1122 _debug("purging feed...", $debug_enabled);
2c08214a 1123
a42c55f0 1124 purge_feed($feed, 0, $debug_enabled);
2c08214a 1125
a42c55f0 1126 db_query("UPDATE ttrss_feeds
f2798eb6 1127 SET last_updated = NOW(), last_error = '' WHERE id = '$feed'");
2c08214a 1128
a42c55f0 1129// db_query("COMMIT");
2c08214a
AD
1130
1131 } else {
1132
a42c55f0 1133 $error_msg = db_escape_string(mb_substr($rss->error(), 0, 245));
2c08214a 1134
4ad04ee2
AD
1135 _debug("fetch error: $error_msg", $debug_enabled);
1136
1137 if (count($rss->errors()) > 1) {
1138 foreach ($rss->errors() as $error) {
1139 _debug("+ $error");
1140 }
1141 }
2c08214a 1142
6322ac79 1143 db_query(
2c08214a 1144 "UPDATE ttrss_feeds SET last_error = '$error_msg',
88edaa93 1145 last_updated = NOW() WHERE id = '$feed'");
2c08214a 1146
88edaa93
AD
1147 unset($rss);
1148 }
2c08214a 1149
68cccafc 1150 _debug("done", $debug_enabled);
88edaa93
AD
1151
1152 return $rss;
2c08214a
AD
1153 }
1154
3c696512 1155 function cache_images($html, $site_url, $debug) {
3c696512
AD
1156 libxml_use_internal_errors(true);
1157
1158 $charset_hack = '<head>
1159 <meta http-equiv="Content-Type" content="text/html; charset=utf-8"/>
1160 </head>';
1161
1162 $doc = new DOMDocument();
1163 $doc->loadHTML($charset_hack . $html);
1164 $xpath = new DOMXPath($doc);
1165
1166 $entries = $xpath->query('(//img[@src])');
1167
1168 foreach ($entries as $entry) {
1169 if ($entry->hasAttribute('src')) {
1170 $src = rewrite_relative_url($site_url, $entry->getAttribute('src'));
1171
1172 $local_filename = CACHE_DIR . "/images/" . sha1($src) . ".png";
1173
1174 if ($debug) _debug("cache_images: downloading: $src to $local_filename");
1175
1176 if (!file_exists($local_filename)) {
1177 $file_content = fetch_file_contents($src);
1178
a3d50184 1179 if ($file_content && strlen($file_content) > _MIN_CACHE_IMAGE_SIZE) {
3c696512
AD
1180 file_put_contents($local_filename, $file_content);
1181 }
1182 }
3c696512
AD
1183 }
1184 }
3c696512
AD
1185 }
1186
a42c55f0 1187 function expire_error_log($debug) {
e2261e17
AD
1188 if ($debug) _debug("Removing old error log entries...");
1189
1190 if (DB_TYPE == "pgsql") {
a42c55f0 1191 db_query("DELETE FROM ttrss_error_log
e2261e17
AD
1192 WHERE created_at < NOW() - INTERVAL '7 days'");
1193 } else {
a42c55f0 1194 db_query("DELETE FROM ttrss_error_log
e2261e17
AD
1195 WHERE created_at < DATE_SUB(NOW(), INTERVAL 7 DAY)");
1196 }
1197
1198 }
1199
2a91b6ff 1200 function expire_lock_files($debug) {
65465085 1201 //if ($debug) _debug("Removing old lock files...");
2a91b6ff
AD
1202
1203 $num_deleted = 0;
1204
1205 if (is_writable(LOCK_DIRECTORY)) {
1206 $files = glob(LOCK_DIRECTORY . "/*.lock");
1207
1208 if ($files) {
1209 foreach ($files as $file) {
11344971 1210 if (!file_is_locked(basename($file)) && time() - filemtime($file) > 86400*2) {
2a91b6ff
AD
1211 unlink($file);
1212 ++$num_deleted;
1213 }
1214 }
1215 }
1216 }
1217
65465085 1218 if ($debug) _debug("Removed $num_deleted old lock files.");
2a91b6ff
AD
1219 }
1220
3c696512 1221 function expire_cached_files($debug) {
3306daec 1222 foreach (array("simplepie", "images", "export", "upload") as $dir) {
3c696512 1223 $cache_dir = CACHE_DIR . "/$dir";
2c08214a 1224
65465085 1225// if ($debug) _debug("Expiring $cache_dir");
2c08214a 1226
3c696512
AD
1227 $num_deleted = 0;
1228
1229 if (is_writable($cache_dir)) {
1230 $files = glob("$cache_dir/*");
1231
2a91b6ff 1232 if ($files) {
2ab20c31
AD
1233 foreach ($files as $file) {
1234 if (time() - filemtime($file) > 86400*7) {
1235 unlink($file);
3c696512 1236
2ab20c31
AD
1237 ++$num_deleted;
1238 }
3c696512
AD
1239 }
1240 }
2a91b6ff 1241 }
3c696512 1242
65465085 1243 if ($debug) _debug("$cache_dir: removed $num_deleted files.");
3c696512
AD
1244 }
1245 }
2c08214a 1246
a3e0bdcf
AD
1247 /**
1248 * Source: http://www.php.net/manual/en/function.parse-url.php#104527
1249 * Returns the url query as associative array
1250 *
1251 * @param string query
1252 * @return array params
1253 */
1254 function convertUrlQuery($query) {
1255 $queryParts = explode('&', $query);
1256
1257 $params = array();
1258
1259 foreach ($queryParts as $param) {
1260 $item = explode('=', $param);
1261 $params[$item[0]] = $item[1];
1262 }
1263
1264 return $params;
1265 }
92c14e9d 1266
48cb2536 1267 function get_article_filters($filters, $title, $content, $link, $timestamp, $author, $tags) {
92c14e9d
AD
1268 $matches = array();
1269
1270 foreach ($filters as $filter) {
1271 $match_any_rule = $filter["match_any_rule"];
a3a896a1 1272 $inverse = $filter["inverse"];
92c14e9d
AD
1273 $filter_match = false;
1274
1275 foreach ($filter["rules"] as $rule) {
1276 $match = false;
ffa1bd7b 1277 $reg_exp = str_replace('/', '\/', $rule["reg_exp"]);
a3a896a1 1278 $rule_inverse = $rule["inverse"];
92c14e9d
AD
1279
1280 if (!$reg_exp)
1281 continue;
1282
1283 switch ($rule["type"]) {
1284 case "title":
1285 $match = @preg_match("/$reg_exp/i", $title);
1286 break;
1287 case "content":
d03ae73e
AD
1288 // we don't need to deal with multiline regexps
1289 $content = preg_replace("/[\r\n\t]/", "", $content);
1290
92c14e9d
AD
1291 $match = @preg_match("/$reg_exp/i", $content);
1292 break;
1293 case "both":
d03ae73e
AD
1294 // we don't need to deal with multiline regexps
1295 $content = preg_replace("/[\r\n\t]/", "", $content);
1296
72d1d067 1297 $match = (@preg_match("/$reg_exp/i", $title) || @preg_match("/$reg_exp/i", $content));
92c14e9d
AD
1298 break;
1299 case "link":
1300 $match = @preg_match("/$reg_exp/i", $link);
1301 break;
1302 case "author":
1303 $match = @preg_match("/$reg_exp/i", $author);
1304 break;
1305 case "tag":
7b80b5e1
DK
1306 foreach ($tags as $tag) {
1307 if (@preg_match("/$reg_exp/i", $tag)) {
1308 $match = true;
1309 break;
1310 }
1311 }
92c14e9d
AD
1312 break;
1313 }
1314
a3a896a1
AD
1315 if ($rule_inverse) $match = !$match;
1316
92c14e9d
AD
1317 if ($match_any_rule) {
1318 if ($match) {
1319 $filter_match = true;
1320 break;
1321 }
1322 } else {
1323 $filter_match = $match;
1324 if (!$match) {
1325 break;
1326 }
1327 }
1328 }
1329
a3a896a1
AD
1330 if ($inverse) $filter_match = !$filter_match;
1331
92c14e9d
AD
1332 if ($filter_match) {
1333 foreach ($filter["actions"] AS $action) {
1334 array_push($matches, $action);
5e736e45
AD
1335
1336 // if Stop action encountered, perform no further processing
1337 if ($action["type"] == "stop") return $matches;
92c14e9d
AD
1338 }
1339 }
1340 }
1341
1342 return $matches;
1343 }
1344
1345 function find_article_filter($filters, $filter_name) {
1346 foreach ($filters as $f) {
1347 if ($f["type"] == $filter_name) {
1348 return $f;
1349 };
1350 }
1351 return false;
1352 }
1353
1354 function find_article_filters($filters, $filter_name) {
1355 $results = array();
1356
1357 foreach ($filters as $f) {
1358 if ($f["type"] == $filter_name) {
1359 array_push($results, $f);
1360 };
1361 }
1362 return $results;
1363 }
1364
1365 function calculate_article_score($filters) {
1366 $score = 0;
1367
1368 foreach ($filters as $f) {
1369 if ($f["type"] == "score") {
1370 $score += $f["param"];
1371 };
1372 }
1373 return $score;
1374 }
1375
b24504b1
AD
1376 function labels_contains_caption($labels, $caption) {
1377 foreach ($labels as $label) {
1378 if ($label[1] == $caption) {
1379 return true;
1380 }
1381 }
1382
1383 return false;
1384 }
1385
a42c55f0 1386 function assign_article_to_label_filters($id, $filters, $owner_uid, $article_labels) {
92c14e9d
AD
1387 foreach ($filters as $f) {
1388 if ($f["type"] == "label") {
b24504b1 1389 if (!labels_contains_caption($article_labels, $f["param"])) {
a42c55f0 1390 label_add_article($id, $f["param"], $owner_uid);
b24504b1
AD
1391 }
1392 }
92c14e9d
AD
1393 }
1394 }
87764a50 1395
87d7e850
AD
1396 function make_guid_from_title($title) {
1397 return preg_replace("/[ \"\',.:;]/", "-",
1398 mb_strtolower(strip_tags($title), 'utf-8'));
1399 }
1400
d1f3fa97 1401 /* function verify_feed_xml($feed_data) {
ebec81a6
AD
1402 libxml_use_internal_errors(true);
1403 $doc = new DOMDocument();
1404 $doc->loadXML($feed_data);
1405 $error = libxml_get_last_error();
1406 libxml_clear_errors();
1407 return $error;
d1f3fa97 1408 } */
87d7e850 1409
168cf351
AD
1410 function cleanup_counters_cache($debug) {
1411 $result = db_query("DELETE FROM ttrss_counters_cache
1412 WHERE feed_id > 0 AND
1413 (SELECT COUNT(id) FROM ttrss_feeds WHERE
1414 id = feed_id AND
1415 ttrss_counters_cache.owner_uid = ttrss_feeds.owner_uid) = 0");
1416 $frows = db_affected_rows($result);
1417
1418 $result = db_query("DELETE FROM ttrss_cat_counters_cache
1419 WHERE feed_id > 0 AND
1420 (SELECT COUNT(id) FROM ttrss_feed_categories WHERE
1421 id = feed_id AND
1422 ttrss_cat_counters_cache.owner_uid = ttrss_feed_categories.owner_uid) = 0");
1423 $crows = db_affected_rows($result);
1424
1425 _debug("Removed $frows (feeds) $crows (cats) orphaned counter cache entries.");
1426 }
1427
e2cf81e2
AD
1428 function housekeeping_common($debug) {
1429 expire_cached_files($debug);
1430 expire_lock_files($debug);
1431 expire_error_log($debug);
1432
1433 $count = update_feedbrowser_cache();
1434 _debug("Feedbrowser updated, $count feeds processed.");
1435
1436 purge_orphans( true);
168cf351 1437 cleanup_counters_cache($debug);
e2cf81e2
AD
1438 $rc = cleanup_tags( 14, 50000);
1439
1440 _debug("Cleaned $rc cached tags.");
8e470220 1441
00f22824 1442 PluginHost::getInstance()->run_hooks(PluginHost::HOOK_HOUSE_KEEPING, "hook_house_keeping", "");
8e470220 1443
e2cf81e2 1444 }
2c08214a 1445?>