]>
Commit | Line | Data |
---|---|---|
1 | <?php | |
2 | class RSSUtils { | |
3 | static function calculate_article_hash($article, $pluginhost) { | |
4 | $tmp = ""; | |
5 | ||
6 | foreach ($article as $k => $v) { | |
7 | if ($k != "feed" && isset($v)) { | |
8 | $x = strip_tags(is_array($v) ? implode(",", $v) : $v); | |
9 | ||
10 | //_debug("$k:" . sha1($x) . ":" . htmlspecialchars($x), true); | |
11 | ||
12 | $tmp .= sha1("$k:" . sha1($x)); | |
13 | } | |
14 | } | |
15 | ||
16 | return sha1(implode(",", $pluginhost->get_plugin_names()) . $tmp); | |
17 | } | |
18 | ||
19 | static function update_feedbrowser_cache() { | |
20 | ||
21 | $pdo = Db::pdo(); | |
22 | ||
23 | $sth = $pdo->query("SELECT feed_url, site_url, title, COUNT(id) AS subscribers | |
24 | FROM ttrss_feeds WHERE feed_url NOT IN (SELECT feed_url FROM ttrss_feeds | |
25 | WHERE private IS true OR auth_login != '' OR auth_pass != '' OR feed_url LIKE '%:%@%/%') | |
26 | GROUP BY feed_url, site_url, title ORDER BY subscribers DESC LIMIT 1000"); | |
27 | ||
28 | $pdo->beginTransaction(); | |
29 | ||
30 | $pdo->query("DELETE FROM ttrss_feedbrowser_cache"); | |
31 | ||
32 | $count = 0; | |
33 | ||
34 | while ($line = $sth->fetch()) { | |
35 | ||
36 | $subscribers = db_escape_string($line["subscribers"]); | |
37 | $feed_url = db_escape_string($line["feed_url"]); | |
38 | $title = db_escape_string($line["title"]); | |
39 | $site_url = db_escape_string($line["site_url"]); | |
40 | ||
41 | $tmph = $pdo->prepare("SELECT subscribers FROM | |
42 | ttrss_feedbrowser_cache WHERE feed_url = ?"); | |
43 | $tmph->execute([$feed_url]); | |
44 | ||
45 | if (!$tmph->fetch()) { | |
46 | ||
47 | $tmph = $pdo->prepare("INSERT INTO ttrss_feedbrowser_cache | |
48 | (feed_url, site_url, title, subscribers) | |
49 | VALUES | |
50 | (?, ?, ?, ?)"); | |
51 | ||
52 | $tmph->execute([$feed_url, $site_url, $title, $subscribers]); | |
53 | ||
54 | ++$count; | |
55 | ||
56 | } | |
57 | ||
58 | } | |
59 | ||
60 | $pdo->commit(); | |
61 | ||
62 | return $count; | |
63 | ||
64 | } | |
65 | ||
66 | static function update_daemon_common($limit = DAEMON_FEED_LIMIT, $debug = true) { | |
67 | $schema_version = get_schema_version(); | |
68 | ||
69 | if ($schema_version != SCHEMA_VERSION) { | |
70 | die("Schema version is wrong, please upgrade the database.\n"); | |
71 | } | |
72 | ||
73 | $pdo = Db::pdo(); | |
74 | ||
75 | if (!SINGLE_USER_MODE && DAEMON_UPDATE_LOGIN_LIMIT > 0) { | |
76 | if (DB_TYPE == "pgsql") { | |
77 | $login_thresh_qpart = "AND ttrss_users.last_login >= NOW() - INTERVAL '".DAEMON_UPDATE_LOGIN_LIMIT." days'"; | |
78 | } else { | |
79 | $login_thresh_qpart = "AND ttrss_users.last_login >= DATE_SUB(NOW(), INTERVAL ".DAEMON_UPDATE_LOGIN_LIMIT." DAY)"; | |
80 | } | |
81 | } else { | |
82 | $login_thresh_qpart = ""; | |
83 | } | |
84 | ||
85 | if (DB_TYPE == "pgsql") { | |
86 | $update_limit_qpart = "AND (( | |
87 | ttrss_feeds.update_interval = 0 | |
88 | AND ttrss_user_prefs.value != '-1' | |
89 | AND ttrss_feeds.last_updated < NOW() - CAST((ttrss_user_prefs.value || ' minutes') AS INTERVAL) | |
90 | ) OR ( | |
91 | ttrss_feeds.update_interval > 0 | |
92 | AND ttrss_feeds.last_updated < NOW() - CAST((ttrss_feeds.update_interval || ' minutes') AS INTERVAL) | |
93 | ) OR (ttrss_feeds.last_updated IS NULL | |
94 | AND ttrss_user_prefs.value != '-1') | |
95 | OR (last_updated = '1970-01-01 00:00:00' | |
96 | AND ttrss_user_prefs.value != '-1'))"; | |
97 | } else { | |
98 | $update_limit_qpart = "AND (( | |
99 | ttrss_feeds.update_interval = 0 | |
100 | AND ttrss_user_prefs.value != '-1' | |
101 | AND ttrss_feeds.last_updated < DATE_SUB(NOW(), INTERVAL CONVERT(ttrss_user_prefs.value, SIGNED INTEGER) MINUTE) | |
102 | ) OR ( | |
103 | ttrss_feeds.update_interval > 0 | |
104 | AND ttrss_feeds.last_updated < DATE_SUB(NOW(), INTERVAL ttrss_feeds.update_interval MINUTE) | |
105 | ) OR (ttrss_feeds.last_updated IS NULL | |
106 | AND ttrss_user_prefs.value != '-1') | |
107 | OR (last_updated = '1970-01-01 00:00:00' | |
108 | AND ttrss_user_prefs.value != '-1'))"; | |
109 | } | |
110 | ||
111 | // Test if feed is currently being updated by another process. | |
112 | if (DB_TYPE == "pgsql") { | |
113 | $updstart_thresh_qpart = "AND (ttrss_feeds.last_update_started IS NULL OR ttrss_feeds.last_update_started < NOW() - INTERVAL '10 minutes')"; | |
114 | } else { | |
115 | $updstart_thresh_qpart = "AND (ttrss_feeds.last_update_started IS NULL OR ttrss_feeds.last_update_started < DATE_SUB(NOW(), INTERVAL 10 MINUTE))"; | |
116 | } | |
117 | ||
118 | $query_limit = $limit ? sprintf("LIMIT %d", $limit) : ""; | |
119 | ||
120 | // Update the least recently updated feeds first | |
121 | $query_order = "ORDER BY last_updated"; | |
122 | if (DB_TYPE == "pgsql") $query_order .= " NULLS FIRST"; | |
123 | ||
124 | $query = "SELECT DISTINCT ttrss_feeds.feed_url, ttrss_feeds.last_updated | |
125 | FROM | |
126 | ttrss_feeds, ttrss_users, ttrss_user_prefs | |
127 | WHERE | |
128 | ttrss_feeds.owner_uid = ttrss_users.id | |
129 | AND ttrss_user_prefs.profile IS NULL | |
130 | AND ttrss_users.id = ttrss_user_prefs.owner_uid | |
131 | AND ttrss_user_prefs.pref_name = 'DEFAULT_UPDATE_INTERVAL' | |
132 | $login_thresh_qpart $update_limit_qpart | |
133 | $updstart_thresh_qpart | |
134 | $query_order $query_limit"; | |
135 | ||
136 | $res = $pdo->query($query); | |
137 | ||
138 | $feeds_to_update = array(); | |
139 | while ($line = $res->fetch()) { | |
140 | array_push($feeds_to_update, $line['feed_url']); | |
141 | } | |
142 | ||
143 | if ($debug) _debug(sprintf("Scheduled %d feeds to update...", count($feeds_to_update))); | |
144 | ||
145 | // Update last_update_started before actually starting the batch | |
146 | // in order to minimize collision risk for parallel daemon tasks | |
147 | if (count($feeds_to_update) > 0) { | |
148 | $feeds_qmarks = arr_qmarks($feeds_to_update); | |
149 | ||
150 | $tmph = $pdo->prepare("UPDATE ttrss_feeds SET last_update_started = NOW() | |
151 | WHERE feed_url IN ($feeds_qmarks)"); | |
152 | $tmph->execute($feeds_to_update); | |
153 | } | |
154 | ||
155 | $nf = 0; | |
156 | $bstarted = microtime(true); | |
157 | ||
158 | $batch_owners = array(); | |
159 | ||
160 | // since we have the data cached, we can deal with other feeds with the same url | |
161 | $usth = $pdo->prepare("SELECT DISTINCT ttrss_feeds.id,last_updated,ttrss_feeds.owner_uid | |
162 | FROM ttrss_feeds, ttrss_users, ttrss_user_prefs WHERE | |
163 | ttrss_user_prefs.owner_uid = ttrss_feeds.owner_uid AND | |
164 | ttrss_users.id = ttrss_user_prefs.owner_uid AND | |
165 | ttrss_user_prefs.pref_name = 'DEFAULT_UPDATE_INTERVAL' AND | |
166 | ttrss_user_prefs.profile IS NULL AND | |
167 | feed_url = ? | |
168 | $update_limit_qpart | |
169 | $login_thresh_qpart | |
170 | ORDER BY ttrss_feeds.id $query_limit"); | |
171 | ||
172 | foreach ($feeds_to_update as $feed) { | |
173 | if($debug) _debug("Base feed: $feed"); | |
174 | ||
175 | $usth->execute([$feed]); | |
176 | //update_rss_feed($line["id"], true); | |
177 | ||
178 | if ($tline = $usth->fetch()) { | |
179 | if ($debug) _debug(" => " . $tline["last_updated"] . ", " . $tline["id"] . " " . $tline["owner_uid"]); | |
180 | ||
181 | if (array_search($tline["owner_uid"], $batch_owners) === FALSE) | |
182 | array_push($batch_owners, $tline["owner_uid"]); | |
183 | ||
184 | $fstarted = microtime(true); | |
185 | RSSUtils::update_rss_feed($tline["id"], true, false); | |
186 | _debug_suppress(false); | |
187 | ||
188 | _debug(sprintf(" %.4f (sec)", microtime(true) - $fstarted)); | |
189 | ||
190 | ++$nf; | |
191 | } | |
192 | } | |
193 | ||
194 | if ($nf > 0) { | |
195 | _debug(sprintf("Processed %d feeds in %.4f (sec), %.4f (sec/feed avg)", $nf, | |
196 | microtime(true) - $bstarted, (microtime(true) - $bstarted) / $nf)); | |
197 | } | |
198 | ||
199 | foreach ($batch_owners as $owner_uid) { | |
200 | _debug("Running housekeeping tasks for user $owner_uid..."); | |
201 | ||
202 | RSSUtils::housekeeping_user($owner_uid); | |
203 | } | |
204 | ||
205 | // Send feed digests by email if needed. | |
206 | Digest::send_headlines_digests($debug); | |
207 | ||
208 | return $nf; | |
209 | } | |
210 | ||
211 | // this is used when subscribing | |
212 | static function set_basic_feed_info($feed) { | |
213 | ||
214 | $feed = db_escape_string($feed); | |
215 | ||
216 | $result = db_query("SELECT owner_uid,feed_url,auth_pass,auth_login,auth_pass_encrypted | |
217 | FROM ttrss_feeds WHERE id = '$feed'"); | |
218 | ||
219 | $owner_uid = db_fetch_result($result, 0, "owner_uid"); | |
220 | ||
221 | $auth_pass_encrypted = sql_bool_to_bool(db_fetch_result($result, | |
222 | 0, "auth_pass_encrypted")); | |
223 | ||
224 | $auth_login = db_fetch_result($result, 0, "auth_login"); | |
225 | $auth_pass = db_fetch_result($result, 0, "auth_pass"); | |
226 | ||
227 | if ($auth_pass_encrypted && function_exists("mcrypt_decrypt")) { | |
228 | require_once "crypt.php"; | |
229 | $auth_pass = decrypt_string($auth_pass); | |
230 | } | |
231 | ||
232 | $fetch_url = db_fetch_result($result, 0, "feed_url"); | |
233 | ||
234 | $pluginhost = new PluginHost(); | |
235 | $user_plugins = get_pref("_ENABLED_PLUGINS", $owner_uid); | |
236 | ||
237 | $pluginhost->load(PLUGINS, PluginHost::KIND_ALL); | |
238 | $pluginhost->load($user_plugins, PluginHost::KIND_USER, $owner_uid); | |
239 | $pluginhost->load_data(); | |
240 | ||
241 | $basic_info = array(); | |
242 | foreach ($pluginhost->get_hooks(PluginHost::HOOK_FEED_BASIC_INFO) as $plugin) { | |
243 | $basic_info = $plugin->hook_feed_basic_info($basic_info, $fetch_url, $owner_uid, $feed, $auth_login, $auth_pass); | |
244 | } | |
245 | ||
246 | if (!$basic_info) { | |
247 | $feed_data = fetch_file_contents($fetch_url, false, | |
248 | $auth_login, $auth_pass, false, | |
249 | FEED_FETCH_TIMEOUT, | |
250 | 0); | |
251 | ||
252 | global $fetch_curl_used; | |
253 | ||
254 | if (!$fetch_curl_used) { | |
255 | $tmp = @gzdecode($feed_data); | |
256 | ||
257 | if ($tmp) $feed_data = $tmp; | |
258 | } | |
259 | ||
260 | $feed_data = trim($feed_data); | |
261 | ||
262 | $rss = new FeedParser($feed_data); | |
263 | $rss->init(); | |
264 | ||
265 | if (!$rss->error()) { | |
266 | $basic_info = array( | |
267 | 'title' => db_escape_string(mb_substr($rss->get_title(), 0, 199)), | |
268 | 'site_url' => db_escape_string(mb_substr(rewrite_relative_url($fetch_url, $rss->get_link()), 0, 245)) | |
269 | ); | |
270 | } | |
271 | } | |
272 | ||
273 | if ($basic_info && is_array($basic_info)) { | |
274 | $result = db_query("SELECT title, site_url FROM ttrss_feeds WHERE id = '$feed'"); | |
275 | ||
276 | $registered_title = db_fetch_result($result, 0, "title"); | |
277 | $orig_site_url = db_fetch_result($result, 0, "site_url"); | |
278 | ||
279 | if ($basic_info['title'] && (!$registered_title || $registered_title == "[Unknown]")) { | |
280 | db_query("UPDATE ttrss_feeds SET | |
281 | title = '${basic_info['title']}' WHERE id = '$feed'"); | |
282 | } | |
283 | ||
284 | if ($basic_info['site_url'] && $orig_site_url != $basic_info['site_url']) { | |
285 | db_query("UPDATE ttrss_feeds SET | |
286 | site_url = '${basic_info['site_url']}' WHERE id = '$feed'"); | |
287 | } | |
288 | } | |
289 | } | |
290 | ||
291 | /** | |
292 | * @SuppressWarnings(PHPMD.UnusedFormalParameter) | |
293 | */ | |
294 | static function update_rss_feed($feed, $no_cache = false) { | |
295 | ||
296 | $debug_enabled = defined('DAEMON_EXTENDED_DEBUG') || $_REQUEST['xdebug']; | |
297 | ||
298 | _debug_suppress(!$debug_enabled); | |
299 | _debug("start", $debug_enabled); | |
300 | ||
301 | $result = db_query("SELECT title FROM ttrss_feeds | |
302 | WHERE id = '$feed'"); | |
303 | ||
304 | if (db_num_rows($result) == 0) { | |
305 | _debug("feed $feed NOT FOUND/SKIPPED", $debug_enabled); | |
306 | user_error("Attempt to update unknown/invalid feed $feed", E_USER_WARNING); | |
307 | return false; | |
308 | } | |
309 | ||
310 | $title = db_fetch_result($result, 0, "title"); | |
311 | ||
312 | // feed was batch-subscribed or something, we need to get basic info | |
313 | // this is not optimal currently as it fetches stuff separately TODO: optimize | |
314 | if ($title == "[Unknown]") { | |
315 | _debug("setting basic feed info for $feed..."); | |
316 | RSSUtils::set_basic_feed_info($feed); | |
317 | } | |
318 | ||
319 | $result = db_query("SELECT id,update_interval,auth_login, | |
320 | feed_url,auth_pass,cache_images, | |
321 | mark_unread_on_update, owner_uid, | |
322 | auth_pass_encrypted, feed_language, | |
323 | last_modified, | |
324 | ".SUBSTRING_FOR_DATE."(last_unconditional, 1, 19) AS last_unconditional | |
325 | FROM ttrss_feeds WHERE id = '$feed'"); | |
326 | ||
327 | $owner_uid = db_fetch_result($result, 0, "owner_uid"); | |
328 | $mark_unread_on_update = sql_bool_to_bool(db_fetch_result($result, | |
329 | 0, "mark_unread_on_update")); | |
330 | $auth_pass_encrypted = sql_bool_to_bool(db_fetch_result($result, | |
331 | 0, "auth_pass_encrypted")); | |
332 | ||
333 | db_query("UPDATE ttrss_feeds SET last_update_started = NOW() | |
334 | WHERE id = '$feed'"); | |
335 | ||
336 | $auth_login = db_fetch_result($result, 0, "auth_login"); | |
337 | $auth_pass = db_fetch_result($result, 0, "auth_pass"); | |
338 | ||
339 | if ($auth_pass_encrypted && function_exists("mcrypt_decrypt")) { | |
340 | require_once "crypt.php"; | |
341 | $auth_pass = decrypt_string($auth_pass); | |
342 | } | |
343 | ||
344 | $stored_last_modified = db_fetch_result($result, 0, "last_modified"); | |
345 | $last_unconditional = db_fetch_result($result, 0, "last_unconditional"); | |
346 | $cache_images = sql_bool_to_bool(db_fetch_result($result, 0, "cache_images")); | |
347 | $fetch_url = db_fetch_result($result, 0, "feed_url"); | |
348 | $feed_language = db_escape_string(mb_strtolower(db_fetch_result($result, 0, "feed_language"))); | |
349 | if (!$feed_language) $feed_language = 'english'; | |
350 | ||
351 | $feed = db_escape_string($feed); | |
352 | ||
353 | $date_feed_processed = date('Y-m-d H:i'); | |
354 | ||
355 | $cache_filename = CACHE_DIR . "/simplepie/" . sha1($fetch_url) . ".xml"; | |
356 | ||
357 | $pluginhost = new PluginHost(); | |
358 | $pluginhost->set_debug($debug_enabled); | |
359 | $user_plugins = get_pref("_ENABLED_PLUGINS", $owner_uid); | |
360 | ||
361 | $pluginhost->load(PLUGINS, PluginHost::KIND_ALL); | |
362 | $pluginhost->load($user_plugins, PluginHost::KIND_USER, $owner_uid); | |
363 | $pluginhost->load_data(); | |
364 | ||
365 | $rss_hash = false; | |
366 | ||
367 | $force_refetch = isset($_REQUEST["force_refetch"]); | |
368 | $feed_data = ""; | |
369 | ||
370 | foreach ($pluginhost->get_hooks(PluginHost::HOOK_FETCH_FEED) as $plugin) { | |
371 | $feed_data = $plugin->hook_fetch_feed($feed_data, $fetch_url, $owner_uid, $feed, 0, $auth_login, $auth_pass); | |
372 | } | |
373 | ||
374 | // try cache | |
375 | if (!$feed_data && | |
376 | file_exists($cache_filename) && | |
377 | is_readable($cache_filename) && | |
378 | !$auth_login && !$auth_pass && | |
379 | filemtime($cache_filename) > time() - 30) { | |
380 | ||
381 | _debug("using local cache [$cache_filename].", $debug_enabled); | |
382 | ||
383 | @$feed_data = file_get_contents($cache_filename); | |
384 | ||
385 | if ($feed_data) { | |
386 | $rss_hash = sha1($feed_data); | |
387 | } | |
388 | ||
389 | } else { | |
390 | _debug("local cache will not be used for this feed", $debug_enabled); | |
391 | } | |
392 | ||
393 | global $fetch_last_modified; | |
394 | ||
395 | // fetch feed from source | |
396 | if (!$feed_data) { | |
397 | _debug("last unconditional update request: $last_unconditional"); | |
398 | ||
399 | if (ini_get("open_basedir") && function_exists("curl_init")) { | |
400 | _debug("not using CURL due to open_basedir restrictions"); | |
401 | } | |
402 | ||
403 | if (time() - strtotime($last_unconditional) > MAX_CONDITIONAL_INTERVAL) { | |
404 | _debug("maximum allowed interval for conditional requests exceeded, forcing refetch"); | |
405 | ||
406 | $force_refetch = true; | |
407 | } else { | |
408 | _debug("stored last modified for conditional request: $stored_last_modified", $debug_enabled); | |
409 | } | |
410 | ||
411 | _debug("fetching [$fetch_url] (force_refetch: $force_refetch)...", $debug_enabled); | |
412 | ||
413 | $feed_data = fetch_file_contents([ | |
414 | "url" => $fetch_url, | |
415 | "login" => $auth_login, | |
416 | "pass" => $auth_pass, | |
417 | "timeout" => $no_cache ? FEED_FETCH_NO_CACHE_TIMEOUT : FEED_FETCH_TIMEOUT, | |
418 | "last_modified" => $force_refetch ? "" : $stored_last_modified | |
419 | ]); | |
420 | ||
421 | global $fetch_curl_used; | |
422 | ||
423 | if (!$fetch_curl_used) { | |
424 | $tmp = @gzdecode($feed_data); | |
425 | ||
426 | if ($tmp) $feed_data = $tmp; | |
427 | } | |
428 | ||
429 | $feed_data = trim($feed_data); | |
430 | ||
431 | _debug("fetch done.", $debug_enabled); | |
432 | _debug("source last modified: " . $fetch_last_modified, $debug_enabled); | |
433 | ||
434 | if ($feed_data && $fetch_last_modified != $stored_last_modified) { | |
435 | $last_modified_escaped = db_escape_string(substr($fetch_last_modified, 0, 245)); | |
436 | ||
437 | db_query("UPDATE ttrss_feeds SET last_modified = '$last_modified_escaped' WHERE id = '$feed'"); | |
438 | ||
439 | } | |
440 | ||
441 | // cache vanilla feed data for re-use | |
442 | if ($feed_data && !$auth_pass && !$auth_login && is_writable(CACHE_DIR . "/simplepie")) { | |
443 | $new_rss_hash = sha1($feed_data); | |
444 | ||
445 | if ($new_rss_hash != $rss_hash) { | |
446 | _debug("saving $cache_filename", $debug_enabled); | |
447 | @file_put_contents($cache_filename, $feed_data); | |
448 | } | |
449 | } | |
450 | } | |
451 | ||
452 | if (!$feed_data) { | |
453 | global $fetch_last_error; | |
454 | global $fetch_last_error_code; | |
455 | ||
456 | _debug("unable to fetch: $fetch_last_error [$fetch_last_error_code]", $debug_enabled); | |
457 | ||
458 | $error_escaped = ''; | |
459 | ||
460 | // If-Modified-Since | |
461 | if ($fetch_last_error_code != 304) { | |
462 | $error_escaped = db_escape_string($fetch_last_error); | |
463 | } else { | |
464 | _debug("source claims data not modified, nothing to do.", $debug_enabled); | |
465 | } | |
466 | ||
467 | db_query( | |
468 | "UPDATE ttrss_feeds SET last_error = '$error_escaped', | |
469 | last_updated = NOW() WHERE id = '$feed'"); | |
470 | ||
471 | return; | |
472 | } | |
473 | ||
474 | foreach ($pluginhost->get_hooks(PluginHost::HOOK_FEED_FETCHED) as $plugin) { | |
475 | $feed_data = $plugin->hook_feed_fetched($feed_data, $fetch_url, $owner_uid, $feed); | |
476 | } | |
477 | ||
478 | $rss = new FeedParser($feed_data); | |
479 | $rss->init(); | |
480 | ||
481 | $feed = db_escape_string($feed); | |
482 | ||
483 | if (!$rss->error()) { | |
484 | ||
485 | // We use local pluginhost here because we need to load different per-user feed plugins | |
486 | $pluginhost->run_hooks(PluginHost::HOOK_FEED_PARSED, "hook_feed_parsed", $rss); | |
487 | ||
488 | _debug("language: $feed_language", $debug_enabled); | |
489 | _debug("processing feed data...", $debug_enabled); | |
490 | ||
491 | // db_query("BEGIN"); | |
492 | ||
493 | if (DB_TYPE == "pgsql") { | |
494 | $favicon_interval_qpart = "favicon_last_checked < NOW() - INTERVAL '12 hour'"; | |
495 | } else { | |
496 | $favicon_interval_qpart = "favicon_last_checked < DATE_SUB(NOW(), INTERVAL 12 HOUR)"; | |
497 | } | |
498 | ||
499 | $result = db_query("SELECT owner_uid,favicon_avg_color, | |
500 | (favicon_last_checked IS NULL OR $favicon_interval_qpart) AS | |
501 | favicon_needs_check | |
502 | FROM ttrss_feeds WHERE id = '$feed'"); | |
503 | ||
504 | $favicon_needs_check = sql_bool_to_bool(db_fetch_result($result, 0, | |
505 | "favicon_needs_check")); | |
506 | $favicon_avg_color = db_fetch_result($result, 0, "favicon_avg_color"); | |
507 | ||
508 | $owner_uid = db_fetch_result($result, 0, "owner_uid"); | |
509 | ||
510 | $site_url = db_escape_string(mb_substr(rewrite_relative_url($fetch_url, $rss->get_link()), 0, 245)); | |
511 | ||
512 | _debug("site_url: $site_url", $debug_enabled); | |
513 | _debug("feed_title: " . $rss->get_title(), $debug_enabled); | |
514 | ||
515 | if ($favicon_needs_check || $force_refetch) { | |
516 | ||
517 | /* terrible hack: if we crash on floicon shit here, we won't check | |
518 | * the icon avgcolor again (unless the icon got updated) */ | |
519 | ||
520 | $favicon_file = ICONS_DIR . "/$feed.ico"; | |
521 | $favicon_modified = @filemtime($favicon_file); | |
522 | ||
523 | _debug("checking favicon...", $debug_enabled); | |
524 | ||
525 | RSSUtils::check_feed_favicon($site_url, $feed); | |
526 | $favicon_modified_new = @filemtime($favicon_file); | |
527 | ||
528 | if ($favicon_modified_new > $favicon_modified) | |
529 | $favicon_avg_color = ''; | |
530 | ||
531 | if (file_exists($favicon_file) && function_exists("imagecreatefromstring") && $favicon_avg_color == '') { | |
532 | require_once "colors.php"; | |
533 | ||
534 | db_query("UPDATE ttrss_feeds SET favicon_avg_color = 'fail' WHERE | |
535 | id = '$feed'"); | |
536 | ||
537 | $favicon_color = db_escape_string( | |
538 | calculate_avg_color($favicon_file)); | |
539 | ||
540 | $favicon_colorstring = ",favicon_avg_color = '".$favicon_color."'"; | |
541 | } else if ($favicon_avg_color == 'fail') { | |
542 | _debug("floicon failed on this file, not trying to recalculate avg color", $debug_enabled); | |
543 | } | |
544 | ||
545 | db_query("UPDATE ttrss_feeds SET favicon_last_checked = NOW() | |
546 | $favicon_colorstring | |
547 | WHERE id = '$feed'"); | |
548 | } | |
549 | ||
550 | _debug("loading filters & labels...", $debug_enabled); | |
551 | ||
552 | $filters = load_filters($feed, $owner_uid); | |
553 | ||
554 | if ($debug_enabled) { | |
555 | print_r($filters); | |
556 | } | |
557 | ||
558 | _debug("" . count($filters) . " filters loaded.", $debug_enabled); | |
559 | ||
560 | $items = $rss->get_items(); | |
561 | ||
562 | if (!is_array($items)) { | |
563 | _debug("no articles found.", $debug_enabled); | |
564 | ||
565 | db_query("UPDATE ttrss_feeds | |
566 | SET last_updated = NOW(), last_unconditional = NOW(), last_error = '' WHERE id = '$feed'"); | |
567 | ||
568 | return; // no articles | |
569 | } | |
570 | ||
571 | _debug("processing articles...", $debug_enabled); | |
572 | ||
573 | $tstart = time(); | |
574 | ||
575 | foreach ($items as $item) { | |
576 | if ($_REQUEST['xdebug'] == 3) { | |
577 | print_r($item); | |
578 | } | |
579 | ||
580 | if (ini_get("max_execution_time") > 0 && time() - $tstart >= ini_get("max_execution_time") * 0.7) { | |
581 | _debug("looks like there's too many articles to process at once, breaking out", $debug_enabled); | |
582 | break; | |
583 | } | |
584 | ||
585 | $entry_guid = $item->get_id(); | |
586 | if (!$entry_guid) $entry_guid = $item->get_link(); | |
587 | if (!$entry_guid) $entry_guid = RSSUtils::make_guid_from_title($item->get_title()); | |
588 | if (!$entry_guid) continue; | |
589 | ||
590 | $entry_guid = "$owner_uid,$entry_guid"; | |
591 | ||
592 | $entry_guid_hashed = db_escape_string('SHA1:' . sha1($entry_guid)); | |
593 | ||
594 | _debug("guid $entry_guid / $entry_guid_hashed", $debug_enabled); | |
595 | ||
596 | $entry_timestamp = ""; | |
597 | ||
598 | $entry_timestamp = $item->get_date(); | |
599 | ||
600 | _debug("orig date: " . $item->get_date(), $debug_enabled); | |
601 | ||
602 | if ($entry_timestamp == -1 || !$entry_timestamp || $entry_timestamp > time()) { | |
603 | $entry_timestamp = time(); | |
604 | } | |
605 | ||
606 | $entry_timestamp_fmt = strftime("%Y/%m/%d %H:%M:%S", $entry_timestamp); | |
607 | ||
608 | _debug("date $entry_timestamp [$entry_timestamp_fmt]", $debug_enabled); | |
609 | ||
610 | // $entry_title = html_entity_decode($item->get_title(), ENT_COMPAT, 'UTF-8'); | |
611 | // $entry_title = decode_numeric_entities($entry_title); | |
612 | $entry_title = $item->get_title(); | |
613 | ||
614 | $entry_link = rewrite_relative_url($site_url, $item->get_link()); | |
615 | ||
616 | _debug("title $entry_title", $debug_enabled); | |
617 | _debug("link $entry_link", $debug_enabled); | |
618 | ||
619 | if (!$entry_title) $entry_title = date("Y-m-d H:i:s", $entry_timestamp);; | |
620 | ||
621 | $entry_content = $item->get_content(); | |
622 | if (!$entry_content) $entry_content = $item->get_description(); | |
623 | ||
624 | if ($_REQUEST["xdebug"] == 2) { | |
625 | print "content: "; | |
626 | print htmlspecialchars($entry_content); | |
627 | print "\n"; | |
628 | } | |
629 | ||
630 | $entry_comments = db_escape_string(mb_substr($item->get_comments_url(), 0, 245)); | |
631 | $num_comments = (int) $item->get_comments_count(); | |
632 | ||
633 | $entry_author = $item->get_author(); // escaped later | |
634 | $entry_guid = db_escape_string(mb_substr($entry_guid, 0, 245)); | |
635 | ||
636 | _debug("author $entry_author", $debug_enabled); | |
637 | _debug("num_comments: $num_comments", $debug_enabled); | |
638 | _debug("looking for tags...", $debug_enabled); | |
639 | ||
640 | // parse <category> entries into tags | |
641 | ||
642 | $additional_tags = array(); | |
643 | ||
644 | $additional_tags_src = $item->get_categories(); | |
645 | ||
646 | if (is_array($additional_tags_src)) { | |
647 | foreach ($additional_tags_src as $tobj) { | |
648 | array_push($additional_tags, $tobj); | |
649 | } | |
650 | } | |
651 | ||
652 | $entry_tags = array_unique($additional_tags); | |
653 | ||
654 | for ($i = 0; $i < count($entry_tags); $i++) | |
655 | $entry_tags[$i] = mb_strtolower($entry_tags[$i], 'utf-8'); | |
656 | ||
657 | _debug("tags found: " . join(",", $entry_tags), $debug_enabled); | |
658 | ||
659 | _debug("done collecting data.", $debug_enabled); | |
660 | ||
661 | $result = db_query("SELECT id, content_hash, lang FROM ttrss_entries | |
662 | WHERE guid = '".db_escape_string($entry_guid)."' OR guid = '$entry_guid_hashed'"); | |
663 | ||
664 | if (db_num_rows($result) != 0) { | |
665 | $base_entry_id = db_fetch_result($result, 0, "id"); | |
666 | $entry_stored_hash = db_fetch_result($result, 0, "content_hash"); | |
667 | $article_labels = Article::get_article_labels($base_entry_id, $owner_uid); | |
668 | $entry_language = db_fetch_result($result, 0, "lang"); | |
669 | ||
670 | $existing_tags = Article::get_article_tags($base_entry_id, $owner_uid); | |
671 | $entry_tags = array_unique(array_merge($entry_tags, $existing_tags)); | |
672 | ||
673 | } else { | |
674 | $base_entry_id = false; | |
675 | $entry_stored_hash = ""; | |
676 | $article_labels = array(); | |
677 | $entry_language = ""; | |
678 | } | |
679 | ||
680 | $article = array("owner_uid" => $owner_uid, // read only | |
681 | "guid" => $entry_guid, // read only | |
682 | "guid_hashed" => $entry_guid_hashed, // read only | |
683 | "title" => $entry_title, | |
684 | "content" => $entry_content, | |
685 | "link" => $entry_link, | |
686 | "labels" => $article_labels, // current limitation: can add labels to article, can't remove them | |
687 | "tags" => $entry_tags, | |
688 | "author" => $entry_author, | |
689 | "force_catchup" => false, // ugly hack for the time being | |
690 | "score_modifier" => 0, // no previous value, plugin should recalculate score modifier based on content if needed | |
691 | "language" => $entry_language, | |
692 | "num_comments" => $num_comments, // read only | |
693 | "feed" => array("id" => $feed, | |
694 | "fetch_url" => $fetch_url, | |
695 | "site_url" => $site_url, | |
696 | "cache_images" => $cache_images) | |
697 | ); | |
698 | ||
699 | $entry_plugin_data = ""; | |
700 | $entry_current_hash = RSSUtils::calculate_article_hash($article, $pluginhost); | |
701 | ||
702 | _debug("article hash: $entry_current_hash [stored=$entry_stored_hash]", $debug_enabled); | |
703 | ||
704 | if ($entry_current_hash == $entry_stored_hash && !isset($_REQUEST["force_rehash"])) { | |
705 | _debug("stored article seems up to date [IID: $base_entry_id], updating timestamp only", $debug_enabled); | |
706 | ||
707 | // we keep encountering the entry in feeds, so we need to | |
708 | // update date_updated column so that we don't get horrible | |
709 | // dupes when the entry gets purged and reinserted again e.g. | |
710 | // in the case of SLOW SLOW OMG SLOW updating feeds | |
711 | ||
712 | $base_entry_id = db_fetch_result($result, 0, "id"); | |
713 | ||
714 | db_query("UPDATE ttrss_entries SET date_updated = NOW() | |
715 | WHERE id = '$base_entry_id'"); | |
716 | ||
717 | continue; | |
718 | } | |
719 | ||
720 | _debug("hash differs, applying plugin filters:", $debug_enabled); | |
721 | ||
722 | foreach ($pluginhost->get_hooks(PluginHost::HOOK_ARTICLE_FILTER) as $plugin) { | |
723 | _debug("... " . get_class($plugin), $debug_enabled); | |
724 | ||
725 | $start = microtime(true); | |
726 | $article = $plugin->hook_article_filter($article); | |
727 | ||
728 | _debug("=== " . sprintf("%.4f (sec)", microtime(true) - $start), $debug_enabled); | |
729 | ||
730 | $entry_plugin_data .= mb_strtolower(get_class($plugin)) . ","; | |
731 | } | |
732 | ||
733 | if ($_REQUEST["xdebug"] == 2) { | |
734 | print "processed content: "; | |
735 | print htmlspecialchars($article["content"]); | |
736 | print "\n"; | |
737 | } | |
738 | ||
739 | $entry_plugin_data = db_escape_string($entry_plugin_data); | |
740 | ||
741 | _debug("plugin data: $entry_plugin_data", $debug_enabled); | |
742 | ||
743 | // Workaround: 4-byte unicode requires utf8mb4 in MySQL. See https://tt-rss.org/forum/viewtopic.php?f=1&t=3377&p=20077#p20077 | |
744 | if (DB_TYPE == "mysql") { | |
745 | foreach ($article as $k => $v) { | |
746 | ||
747 | // i guess we'll have to take the risk of 4byte unicode labels & tags here | |
748 | if (is_string($article[$k])) { | |
749 | $article[$k] = preg_replace('/[\x{10000}-\x{10FFFF}]/u', "\xEF\xBF\xBD", $v); | |
750 | } | |
751 | } | |
752 | } | |
753 | ||
754 | /* Collect article tags here so we could filter by them: */ | |
755 | ||
756 | $matched_rules = array(); | |
757 | ||
758 | $article_filters = RSSUtils::get_article_filters($filters, $article["title"], | |
759 | $article["content"], $article["link"], $article["author"], | |
760 | $article["tags"], $matched_rules); | |
761 | ||
762 | if ($debug_enabled) { | |
763 | _debug("matched filter rules: ", $debug_enabled); | |
764 | ||
765 | if (count($matched_rules) != 0) { | |
766 | print_r($matched_rules); | |
767 | } | |
768 | ||
769 | _debug("filter actions: ", $debug_enabled); | |
770 | ||
771 | if (count($article_filters) != 0) { | |
772 | print_r($article_filters); | |
773 | } | |
774 | } | |
775 | ||
776 | $plugin_filter_names = RSSUtils::find_article_filters($article_filters, "plugin"); | |
777 | $plugin_filter_actions = $pluginhost->get_filter_actions(); | |
778 | ||
779 | if (count($plugin_filter_names) > 0) { | |
780 | _debug("applying plugin filter actions...", $debug_enabled); | |
781 | ||
782 | foreach ($plugin_filter_names as $pfn) { | |
783 | list($pfclass,$pfaction) = explode(":", $pfn["param"]); | |
784 | ||
785 | if (isset($plugin_filter_actions[$pfclass])) { | |
786 | $plugin = $pluginhost->get_plugin($pfclass); | |
787 | ||
788 | _debug("... $pfclass: $pfaction", $debug_enabled); | |
789 | ||
790 | if ($plugin) { | |
791 | $start = microtime(true); | |
792 | $article = $plugin->hook_article_filter_action($article, $pfaction); | |
793 | ||
794 | _debug("=== " . sprintf("%.4f (sec)", microtime(true) - $start), $debug_enabled); | |
795 | } else { | |
796 | _debug("??? $pfclass: plugin object not found."); | |
797 | } | |
798 | } else { | |
799 | _debug("??? $pfclass: filter plugin not registered."); | |
800 | } | |
801 | } | |
802 | } | |
803 | ||
804 | $entry_tags = $article["tags"]; | |
805 | $entry_guid = db_escape_string($entry_guid); | |
806 | $entry_title = db_escape_string($article["title"]); | |
807 | $entry_author = db_escape_string(mb_substr($article["author"], 0, 245)); | |
808 | $entry_link = db_escape_string($article["link"]); | |
809 | $entry_content = $article["content"]; // escaped below | |
810 | $entry_force_catchup = $article["force_catchup"]; | |
811 | $article_labels = $article["labels"]; | |
812 | $entry_score_modifier = (int) $article["score_modifier"]; | |
813 | $entry_language = db_escape_string($article["language"]); | |
814 | ||
815 | if ($debug_enabled) { | |
816 | _debug("article labels:", $debug_enabled); | |
817 | ||
818 | if (count($article_labels) != 0) { | |
819 | print_r($article_labels); | |
820 | } | |
821 | } | |
822 | ||
823 | _debug("force catchup: $entry_force_catchup"); | |
824 | ||
825 | if ($cache_images && is_writable(CACHE_DIR . '/images')) | |
826 | RSSUtils::cache_media($entry_content, $site_url, $debug_enabled); | |
827 | ||
828 | $entry_content = db_escape_string($entry_content, false); | |
829 | ||
830 | //db_query("BEGIN"); | |
831 | ||
832 | $result = db_query("SELECT id FROM ttrss_entries | |
833 | WHERE (guid = '$entry_guid' OR guid = '$entry_guid_hashed')"); | |
834 | ||
835 | if (db_num_rows($result) == 0) { | |
836 | ||
837 | _debug("base guid [$entry_guid or $entry_guid_hashed] not found, creating...", $debug_enabled); | |
838 | ||
839 | // base post entry does not exist, create it | |
840 | ||
841 | db_query( | |
842 | "INSERT INTO ttrss_entries | |
843 | (title, | |
844 | guid, | |
845 | link, | |
846 | updated, | |
847 | content, | |
848 | content_hash, | |
849 | no_orig_date, | |
850 | date_updated, | |
851 | date_entered, | |
852 | comments, | |
853 | num_comments, | |
854 | plugin_data, | |
855 | lang, | |
856 | author) | |
857 | VALUES | |
858 | ('$entry_title', | |
859 | '$entry_guid_hashed', | |
860 | '$entry_link', | |
861 | '$entry_timestamp_fmt', | |
862 | '$entry_content', | |
863 | '$entry_current_hash', | |
864 | false, | |
865 | NOW(), | |
866 | '$date_feed_processed', | |
867 | '$entry_comments', | |
868 | '$num_comments', | |
869 | '$entry_plugin_data', | |
870 | '$entry_language', | |
871 | '$entry_author')"); | |
872 | ||
873 | } | |
874 | ||
875 | // now it should exist, if not - bad luck then | |
876 | ||
877 | $result = db_query("SELECT id FROM ttrss_entries | |
878 | WHERE guid = '$entry_guid' OR guid = '$entry_guid_hashed'"); | |
879 | ||
880 | $entry_ref_id = 0; | |
881 | $entry_int_id = 0; | |
882 | ||
883 | if (db_num_rows($result) == 1) { | |
884 | ||
885 | _debug("base guid found, checking for user record", $debug_enabled); | |
886 | ||
887 | $ref_id = db_fetch_result($result, 0, "id"); | |
888 | $entry_ref_id = $ref_id; | |
889 | ||
890 | /* $stored_guid = db_fetch_result($result, 0, "guid"); | |
891 | if ($stored_guid != $entry_guid_hashed) { | |
892 | if ($debug_enabled) _debug("upgrading compat guid to hashed one", $debug_enabled); | |
893 | ||
894 | db_query("UPDATE ttrss_entries SET guid = '$entry_guid_hashed' WHERE | |
895 | id = '$ref_id'"); | |
896 | } */ | |
897 | ||
898 | if (RSSUtils::find_article_filter($article_filters, "filter")) { | |
899 | //db_query("COMMIT"); // close transaction in progress | |
900 | continue; | |
901 | } | |
902 | ||
903 | $score = RSSUtils::calculate_article_score($article_filters) + $entry_score_modifier; | |
904 | ||
905 | _debug("initial score: $score [including plugin modifier: $entry_score_modifier]", $debug_enabled); | |
906 | ||
907 | // check for user post link to main table | |
908 | ||
909 | $query = "SELECT ref_id, int_id FROM ttrss_user_entries WHERE | |
910 | ref_id = '$ref_id' AND owner_uid = '$owner_uid'"; | |
911 | ||
912 | // if ($_REQUEST["xdebug"]) print "$query\n"; | |
913 | ||
914 | $result = db_query($query); | |
915 | ||
916 | // okay it doesn't exist - create user entry | |
917 | if (db_num_rows($result) == 0) { | |
918 | ||
919 | _debug("user record not found, creating...", $debug_enabled); | |
920 | ||
921 | if ($score >= -500 && !RSSUtils::find_article_filter($article_filters, 'catchup') && !$entry_force_catchup) { | |
922 | $unread = 'true'; | |
923 | $last_read_qpart = 'NULL'; | |
924 | } else { | |
925 | $unread = 'false'; | |
926 | $last_read_qpart = 'NOW()'; | |
927 | } | |
928 | ||
929 | if (RSSUtils::find_article_filter($article_filters, 'mark') || $score > 1000) { | |
930 | $marked = 'true'; | |
931 | } else { | |
932 | $marked = 'false'; | |
933 | } | |
934 | ||
935 | if (RSSUtils::find_article_filter($article_filters, 'publish')) { | |
936 | $published = 'true'; | |
937 | } else { | |
938 | $published = 'false'; | |
939 | } | |
940 | ||
941 | $last_marked = ($marked == 'true') ? 'NOW()' : 'NULL'; | |
942 | $last_published = ($published == 'true') ? 'NOW()' : 'NULL'; | |
943 | ||
944 | $result = db_query( | |
945 | "INSERT INTO ttrss_user_entries | |
946 | (ref_id, owner_uid, feed_id, unread, last_read, marked, | |
947 | published, score, tag_cache, label_cache, uuid, | |
948 | last_marked, last_published) | |
949 | VALUES ('$ref_id', '$owner_uid', '$feed', $unread, | |
950 | $last_read_qpart, $marked, $published, '$score', '', '', | |
951 | '', $last_marked, $last_published)"); | |
952 | ||
953 | $result = db_query( | |
954 | "SELECT int_id FROM ttrss_user_entries WHERE | |
955 | ref_id = '$ref_id' AND owner_uid = '$owner_uid' AND | |
956 | feed_id = '$feed' LIMIT 1"); | |
957 | ||
958 | if (db_num_rows($result) == 1) { | |
959 | $entry_int_id = db_fetch_result($result, 0, "int_id"); | |
960 | } | |
961 | } else { | |
962 | _debug("user record FOUND", $debug_enabled); | |
963 | ||
964 | $entry_ref_id = db_fetch_result($result, 0, "ref_id"); | |
965 | $entry_int_id = db_fetch_result($result, 0, "int_id"); | |
966 | } | |
967 | ||
968 | _debug("RID: $entry_ref_id, IID: $entry_int_id", $debug_enabled); | |
969 | ||
970 | if (DB_TYPE == "pgsql") { | |
971 | $tsvector_combined = db_escape_string(mb_substr($entry_title . ' ' . strip_tags(str_replace('<', ' <', $entry_content)), | |
972 | 0, 1000000)); | |
973 | ||
974 | $tsvector_qpart = "tsvector_combined = to_tsvector('$feed_language', '$tsvector_combined'),"; | |
975 | ||
976 | } else { | |
977 | $tsvector_qpart = ""; | |
978 | } | |
979 | ||
980 | db_query("UPDATE ttrss_entries | |
981 | SET title = '$entry_title', | |
982 | content = '$entry_content', | |
983 | content_hash = '$entry_current_hash', | |
984 | updated = '$entry_timestamp_fmt', | |
985 | $tsvector_qpart | |
986 | num_comments = '$num_comments', | |
987 | plugin_data = '$entry_plugin_data', | |
988 | author = '$entry_author', | |
989 | lang = '$entry_language' | |
990 | WHERE id = '$ref_id'"); | |
991 | ||
992 | // update aux data | |
993 | db_query("UPDATE ttrss_user_entries | |
994 | SET score = '$score' WHERE ref_id = '$ref_id'"); | |
995 | ||
996 | if ($mark_unread_on_update) { | |
997 | _debug("article updated, marking unread as requested.", $debug_enabled); | |
998 | ||
999 | db_query("UPDATE ttrss_user_entries | |
1000 | SET last_read = null, unread = true WHERE ref_id = '$ref_id'"); | |
1001 | } | |
1002 | } | |
1003 | ||
1004 | //db_query("COMMIT"); | |
1005 | ||
1006 | _debug("assigning labels [other]...", $debug_enabled); | |
1007 | ||
1008 | foreach ($article_labels as $label) { | |
1009 | Labels::add_article($entry_ref_id, $label[1], $owner_uid); | |
1010 | } | |
1011 | ||
1012 | _debug("assigning labels [filters]...", $debug_enabled); | |
1013 | ||
1014 | RSSUtils::assign_article_to_label_filters($entry_ref_id, $article_filters, | |
1015 | $owner_uid, $article_labels); | |
1016 | ||
1017 | _debug("looking for enclosures...", $debug_enabled); | |
1018 | ||
1019 | // enclosures | |
1020 | ||
1021 | $enclosures = array(); | |
1022 | ||
1023 | $encs = $item->get_enclosures(); | |
1024 | ||
1025 | if (is_array($encs)) { | |
1026 | foreach ($encs as $e) { | |
1027 | $e_item = array( | |
1028 | rewrite_relative_url($site_url, $e->link), | |
1029 | $e->type, $e->length, $e->title, $e->width, $e->height); | |
1030 | array_push($enclosures, $e_item); | |
1031 | } | |
1032 | } | |
1033 | ||
1034 | if ($cache_images && is_writable(CACHE_DIR . '/images')) | |
1035 | RSSUtils::cache_enclosures($enclosures, $site_url, $debug_enabled); | |
1036 | ||
1037 | if ($debug_enabled) { | |
1038 | _debug("article enclosures:", $debug_enabled); | |
1039 | print_r($enclosures); | |
1040 | } | |
1041 | ||
1042 | //db_query("BEGIN"); | |
1043 | ||
1044 | // debugging | |
1045 | // db_query("DELETE FROM ttrss_enclosures WHERE post_id = '$entry_ref_id'"); | |
1046 | ||
1047 | foreach ($enclosures as $enc) { | |
1048 | $enc_url = db_escape_string($enc[0]); | |
1049 | $enc_type = db_escape_string($enc[1]); | |
1050 | $enc_dur = db_escape_string($enc[2]); | |
1051 | $enc_title = db_escape_string($enc[3]); | |
1052 | $enc_width = intval($enc[4]); | |
1053 | $enc_height = intval($enc[5]); | |
1054 | ||
1055 | $result = db_query("SELECT id FROM ttrss_enclosures | |
1056 | WHERE content_url = '$enc_url' AND post_id = '$entry_ref_id'"); | |
1057 | ||
1058 | if (db_num_rows($result) == 0) { | |
1059 | db_query("INSERT INTO ttrss_enclosures | |
1060 | (content_url, content_type, title, duration, post_id, width, height) VALUES | |
1061 | ('$enc_url', '$enc_type', '$enc_title', '$enc_dur', '$entry_ref_id', $enc_width, $enc_height)"); | |
1062 | } | |
1063 | } | |
1064 | ||
1065 | //db_query("COMMIT"); | |
1066 | ||
1067 | // check for manual tags (we have to do it here since they're loaded from filters) | |
1068 | ||
1069 | foreach ($article_filters as $f) { | |
1070 | if ($f["type"] == "tag") { | |
1071 | ||
1072 | $manual_tags = trim_array(explode(",", $f["param"])); | |
1073 | ||
1074 | foreach ($manual_tags as $tag) { | |
1075 | if (tag_is_valid($tag)) { | |
1076 | array_push($entry_tags, $tag); | |
1077 | } | |
1078 | } | |
1079 | } | |
1080 | } | |
1081 | ||
1082 | // Skip boring tags | |
1083 | ||
1084 | $boring_tags = trim_array(explode(",", mb_strtolower(get_pref( | |
1085 | 'BLACKLISTED_TAGS', $owner_uid, ''), 'utf-8'))); | |
1086 | ||
1087 | $filtered_tags = array(); | |
1088 | $tags_to_cache = array(); | |
1089 | ||
1090 | if ($entry_tags && is_array($entry_tags)) { | |
1091 | foreach ($entry_tags as $tag) { | |
1092 | if (array_search($tag, $boring_tags) === false) { | |
1093 | array_push($filtered_tags, $tag); | |
1094 | } | |
1095 | } | |
1096 | } | |
1097 | ||
1098 | $filtered_tags = array_unique($filtered_tags); | |
1099 | ||
1100 | if ($debug_enabled) { | |
1101 | _debug("filtered article tags:", $debug_enabled); | |
1102 | print_r($filtered_tags); | |
1103 | } | |
1104 | ||
1105 | // Save article tags in the database | |
1106 | ||
1107 | if (count($filtered_tags) > 0) { | |
1108 | ||
1109 | //db_query("BEGIN"); | |
1110 | ||
1111 | foreach ($filtered_tags as $tag) { | |
1112 | ||
1113 | $tag = sanitize_tag($tag); | |
1114 | $tag = db_escape_string($tag); | |
1115 | ||
1116 | if (!tag_is_valid($tag)) continue; | |
1117 | ||
1118 | $result = db_query("SELECT id FROM ttrss_tags | |
1119 | WHERE tag_name = '$tag' AND post_int_id = '$entry_int_id' AND | |
1120 | owner_uid = '$owner_uid' LIMIT 1"); | |
1121 | ||
1122 | if ($result && db_num_rows($result) == 0) { | |
1123 | ||
1124 | db_query("INSERT INTO ttrss_tags | |
1125 | (owner_uid,tag_name,post_int_id) | |
1126 | VALUES ('$owner_uid','$tag', '$entry_int_id')"); | |
1127 | } | |
1128 | ||
1129 | array_push($tags_to_cache, $tag); | |
1130 | } | |
1131 | ||
1132 | /* update the cache */ | |
1133 | ||
1134 | $tags_to_cache = array_unique($tags_to_cache); | |
1135 | ||
1136 | $tags_str = db_escape_string(join(",", $tags_to_cache)); | |
1137 | ||
1138 | db_query("UPDATE ttrss_user_entries | |
1139 | SET tag_cache = '$tags_str' WHERE ref_id = '$entry_ref_id' | |
1140 | AND owner_uid = $owner_uid"); | |
1141 | ||
1142 | //db_query("COMMIT"); | |
1143 | } | |
1144 | ||
1145 | _debug("article processed", $debug_enabled); | |
1146 | } | |
1147 | ||
1148 | _debug("purging feed...", $debug_enabled); | |
1149 | ||
1150 | purge_feed($feed, 0, $debug_enabled); | |
1151 | ||
1152 | db_query("UPDATE ttrss_feeds | |
1153 | SET last_updated = NOW(), last_unconditional = NOW(), last_error = '' WHERE id = '$feed'"); | |
1154 | ||
1155 | // db_query("COMMIT"); | |
1156 | ||
1157 | } else { | |
1158 | ||
1159 | $error_msg = db_escape_string(mb_substr($rss->error(), 0, 245)); | |
1160 | ||
1161 | _debug("fetch error: $error_msg", $debug_enabled); | |
1162 | ||
1163 | if (count($rss->errors()) > 1) { | |
1164 | foreach ($rss->errors() as $error) { | |
1165 | _debug("+ $error"); | |
1166 | } | |
1167 | } | |
1168 | ||
1169 | db_query( | |
1170 | "UPDATE ttrss_feeds SET last_error = '$error_msg', | |
1171 | last_updated = NOW(), last_unconditional = NOW() WHERE id = '$feed'"); | |
1172 | ||
1173 | unset($rss); | |
1174 | return; | |
1175 | } | |
1176 | ||
1177 | _debug("done", $debug_enabled); | |
1178 | ||
1179 | return true; | |
1180 | } | |
1181 | ||
1182 | static function cache_enclosures($enclosures, $site_url, $debug) { | |
1183 | foreach ($enclosures as $enc) { | |
1184 | ||
1185 | if (preg_match("/(image|audio|video)/", $enc[1])) { | |
1186 | ||
1187 | $src = rewrite_relative_url($site_url, $enc[0]); | |
1188 | ||
1189 | $local_filename = CACHE_DIR . "/images/" . sha1($src); | |
1190 | ||
1191 | if ($debug) _debug("cache_enclosures: downloading: $src to $local_filename"); | |
1192 | ||
1193 | if (!file_exists($local_filename)) { | |
1194 | $file_content = fetch_file_contents($src); | |
1195 | ||
1196 | if ($file_content && strlen($file_content) > MIN_CACHE_FILE_SIZE) { | |
1197 | file_put_contents($local_filename, $file_content); | |
1198 | } | |
1199 | } else { | |
1200 | touch($local_filename); | |
1201 | } | |
1202 | } | |
1203 | } | |
1204 | } | |
1205 | ||
1206 | static function cache_media($html, $site_url, $debug) { | |
1207 | libxml_use_internal_errors(true); | |
1208 | ||
1209 | $charset_hack = '<head> | |
1210 | <meta http-equiv="Content-Type" content="text/html; charset=utf-8"/> | |
1211 | </head>'; | |
1212 | ||
1213 | $doc = new DOMDocument(); | |
1214 | $doc->loadHTML($charset_hack . $html); | |
1215 | $xpath = new DOMXPath($doc); | |
1216 | ||
1217 | $entries = $xpath->query('(//img[@src])|(//video/source[@src])|(//audio/source[@src])'); | |
1218 | ||
1219 | foreach ($entries as $entry) { | |
1220 | if ($entry->hasAttribute('src') && strpos($entry->getAttribute('src'), "data:") !== 0) { | |
1221 | $src = rewrite_relative_url($site_url, $entry->getAttribute('src')); | |
1222 | ||
1223 | $local_filename = CACHE_DIR . "/images/" . sha1($src); | |
1224 | ||
1225 | if ($debug) _debug("cache_media: downloading: $src to $local_filename"); | |
1226 | ||
1227 | if (!file_exists($local_filename)) { | |
1228 | $file_content = fetch_file_contents($src); | |
1229 | ||
1230 | if ($file_content && strlen($file_content) > MIN_CACHE_FILE_SIZE) { | |
1231 | file_put_contents($local_filename, $file_content); | |
1232 | } | |
1233 | } else { | |
1234 | touch($local_filename); | |
1235 | } | |
1236 | } | |
1237 | } | |
1238 | } | |
1239 | ||
1240 | static function expire_error_log($debug) { | |
1241 | if ($debug) _debug("Removing old error log entries..."); | |
1242 | ||
1243 | if (DB_TYPE == "pgsql") { | |
1244 | db_query("DELETE FROM ttrss_error_log | |
1245 | WHERE created_at < NOW() - INTERVAL '7 days'"); | |
1246 | } else { | |
1247 | db_query("DELETE FROM ttrss_error_log | |
1248 | WHERE created_at < DATE_SUB(NOW(), INTERVAL 7 DAY)"); | |
1249 | } | |
1250 | ||
1251 | } | |
1252 | ||
1253 | static function expire_lock_files($debug) { | |
1254 | //if ($debug) _debug("Removing old lock files..."); | |
1255 | ||
1256 | $num_deleted = 0; | |
1257 | ||
1258 | if (is_writable(LOCK_DIRECTORY)) { | |
1259 | $files = glob(LOCK_DIRECTORY . "/*.lock"); | |
1260 | ||
1261 | if ($files) { | |
1262 | foreach ($files as $file) { | |
1263 | if (!file_is_locked(basename($file)) && time() - filemtime($file) > 86400*2) { | |
1264 | unlink($file); | |
1265 | ++$num_deleted; | |
1266 | } | |
1267 | } | |
1268 | } | |
1269 | } | |
1270 | ||
1271 | if ($debug) _debug("Removed $num_deleted old lock files."); | |
1272 | } | |
1273 | ||
1274 | static function expire_cached_files($debug) { | |
1275 | foreach (array("simplepie", "images", "export", "upload") as $dir) { | |
1276 | $cache_dir = CACHE_DIR . "/$dir"; | |
1277 | ||
1278 | // if ($debug) _debug("Expiring $cache_dir"); | |
1279 | ||
1280 | $num_deleted = 0; | |
1281 | ||
1282 | if (is_writable($cache_dir)) { | |
1283 | $files = glob("$cache_dir/*"); | |
1284 | ||
1285 | if ($files) { | |
1286 | foreach ($files as $file) { | |
1287 | if (time() - filemtime($file) > 86400*CACHE_MAX_DAYS) { | |
1288 | unlink($file); | |
1289 | ||
1290 | ++$num_deleted; | |
1291 | } | |
1292 | } | |
1293 | } | |
1294 | } | |
1295 | ||
1296 | if ($debug) _debug("$cache_dir: removed $num_deleted files."); | |
1297 | } | |
1298 | } | |
1299 | ||
1300 | /** | |
1301 | * Source: http://www.php.net/manual/en/function.parse-url.php#104527 | |
1302 | * Returns the url query as associative array | |
1303 | * | |
1304 | * @param string query | |
1305 | * @return array params | |
1306 | */ | |
1307 | static function convertUrlQuery($query) { | |
1308 | $queryParts = explode('&', $query); | |
1309 | ||
1310 | $params = array(); | |
1311 | ||
1312 | foreach ($queryParts as $param) { | |
1313 | $item = explode('=', $param); | |
1314 | $params[$item[0]] = $item[1]; | |
1315 | } | |
1316 | ||
1317 | return $params; | |
1318 | } | |
1319 | ||
1320 | static function get_article_filters($filters, $title, $content, $link, $author, $tags, &$matched_rules = false) { | |
1321 | $matches = array(); | |
1322 | ||
1323 | foreach ($filters as $filter) { | |
1324 | $match_any_rule = $filter["match_any_rule"]; | |
1325 | $inverse = $filter["inverse"]; | |
1326 | $filter_match = false; | |
1327 | ||
1328 | foreach ($filter["rules"] as $rule) { | |
1329 | $match = false; | |
1330 | $reg_exp = str_replace('/', '\/', $rule["reg_exp"]); | |
1331 | $rule_inverse = $rule["inverse"]; | |
1332 | ||
1333 | if (!$reg_exp) | |
1334 | continue; | |
1335 | ||
1336 | switch ($rule["type"]) { | |
1337 | case "title": | |
1338 | $match = @preg_match("/$reg_exp/iu", $title); | |
1339 | break; | |
1340 | case "content": | |
1341 | // we don't need to deal with multiline regexps | |
1342 | $content = preg_replace("/[\r\n\t]/", "", $content); | |
1343 | ||
1344 | $match = @preg_match("/$reg_exp/iu", $content); | |
1345 | break; | |
1346 | case "both": | |
1347 | // we don't need to deal with multiline regexps | |
1348 | $content = preg_replace("/[\r\n\t]/", "", $content); | |
1349 | ||
1350 | $match = (@preg_match("/$reg_exp/iu", $title) || @preg_match("/$reg_exp/iu", $content)); | |
1351 | break; | |
1352 | case "link": | |
1353 | $match = @preg_match("/$reg_exp/iu", $link); | |
1354 | break; | |
1355 | case "author": | |
1356 | $match = @preg_match("/$reg_exp/iu", $author); | |
1357 | break; | |
1358 | case "tag": | |
1359 | foreach ($tags as $tag) { | |
1360 | if (@preg_match("/$reg_exp/iu", $tag)) { | |
1361 | $match = true; | |
1362 | break; | |
1363 | } | |
1364 | } | |
1365 | break; | |
1366 | } | |
1367 | ||
1368 | if ($rule_inverse) $match = !$match; | |
1369 | ||
1370 | if ($match_any_rule) { | |
1371 | if ($match) { | |
1372 | $filter_match = true; | |
1373 | break; | |
1374 | } | |
1375 | } else { | |
1376 | $filter_match = $match; | |
1377 | if (!$match) { | |
1378 | break; | |
1379 | } | |
1380 | } | |
1381 | } | |
1382 | ||
1383 | if ($inverse) $filter_match = !$filter_match; | |
1384 | ||
1385 | if ($filter_match) { | |
1386 | if (is_array($matched_rules)) array_push($matched_rules, $rule); | |
1387 | ||
1388 | foreach ($filter["actions"] AS $action) { | |
1389 | array_push($matches, $action); | |
1390 | ||
1391 | // if Stop action encountered, perform no further processing | |
1392 | if (isset($action["type"]) && $action["type"] == "stop") return $matches; | |
1393 | } | |
1394 | } | |
1395 | } | |
1396 | ||
1397 | return $matches; | |
1398 | } | |
1399 | ||
1400 | static function find_article_filter($filters, $filter_name) { | |
1401 | foreach ($filters as $f) { | |
1402 | if ($f["type"] == $filter_name) { | |
1403 | return $f; | |
1404 | }; | |
1405 | } | |
1406 | return false; | |
1407 | } | |
1408 | ||
1409 | static function find_article_filters($filters, $filter_name) { | |
1410 | $results = array(); | |
1411 | ||
1412 | foreach ($filters as $f) { | |
1413 | if ($f["type"] == $filter_name) { | |
1414 | array_push($results, $f); | |
1415 | }; | |
1416 | } | |
1417 | return $results; | |
1418 | } | |
1419 | ||
1420 | static function calculate_article_score($filters) { | |
1421 | $score = 0; | |
1422 | ||
1423 | foreach ($filters as $f) { | |
1424 | if ($f["type"] == "score") { | |
1425 | $score += $f["param"]; | |
1426 | }; | |
1427 | } | |
1428 | return $score; | |
1429 | } | |
1430 | ||
1431 | static function labels_contains_caption($labels, $caption) { | |
1432 | foreach ($labels as $label) { | |
1433 | if ($label[1] == $caption) { | |
1434 | return true; | |
1435 | } | |
1436 | } | |
1437 | ||
1438 | return false; | |
1439 | } | |
1440 | ||
1441 | static function assign_article_to_label_filters($id, $filters, $owner_uid, $article_labels) { | |
1442 | foreach ($filters as $f) { | |
1443 | if ($f["type"] == "label") { | |
1444 | if (!RSSUtils::labels_contains_caption($article_labels, $f["param"])) { | |
1445 | Labels::add_article($id, $f["param"], $owner_uid); | |
1446 | } | |
1447 | } | |
1448 | } | |
1449 | } | |
1450 | ||
1451 | static function make_guid_from_title($title) { | |
1452 | return preg_replace("/[ \"\',.:;]/", "-", | |
1453 | mb_strtolower(strip_tags($title), 'utf-8')); | |
1454 | } | |
1455 | ||
1456 | static function cleanup_counters_cache($debug) { | |
1457 | $result = db_query("DELETE FROM ttrss_counters_cache | |
1458 | WHERE feed_id > 0 AND | |
1459 | (SELECT COUNT(id) FROM ttrss_feeds WHERE | |
1460 | id = feed_id AND | |
1461 | ttrss_counters_cache.owner_uid = ttrss_feeds.owner_uid) = 0"); | |
1462 | $frows = db_affected_rows($result); | |
1463 | ||
1464 | $result = db_query("DELETE FROM ttrss_cat_counters_cache | |
1465 | WHERE feed_id > 0 AND | |
1466 | (SELECT COUNT(id) FROM ttrss_feed_categories WHERE | |
1467 | id = feed_id AND | |
1468 | ttrss_cat_counters_cache.owner_uid = ttrss_feed_categories.owner_uid) = 0"); | |
1469 | $crows = db_affected_rows($result); | |
1470 | ||
1471 | if ($debug) _debug("Removed $frows (feeds) $crows (cats) orphaned counter cache entries."); | |
1472 | } | |
1473 | ||
1474 | static function housekeeping_user($owner_uid) { | |
1475 | $tmph = new PluginHost(); | |
1476 | ||
1477 | load_user_plugins($owner_uid, $tmph); | |
1478 | ||
1479 | $tmph->run_hooks(PluginHost::HOOK_HOUSE_KEEPING, "hook_house_keeping", ""); | |
1480 | } | |
1481 | ||
1482 | static function housekeeping_common($debug) { | |
1483 | RSSUtils::expire_cached_files($debug); | |
1484 | RSSUtils::expire_lock_files($debug); | |
1485 | RSSUtils::expire_error_log($debug); | |
1486 | ||
1487 | $count = RSSUtils::update_feedbrowser_cache(); | |
1488 | _debug("Feedbrowser updated, $count feeds processed."); | |
1489 | ||
1490 | Article::purge_orphans( true); | |
1491 | RSSUtils::cleanup_counters_cache($debug); | |
1492 | ||
1493 | //$rc = cleanup_tags( 14, 50000); | |
1494 | //_debug("Cleaned $rc cached tags."); | |
1495 | ||
1496 | PluginHost::getInstance()->run_hooks(PluginHost::HOOK_HOUSE_KEEPING, "hook_house_keeping", ""); | |
1497 | } | |
1498 | ||
1499 | static function check_feed_favicon($site_url, $feed) { | |
1500 | # print "FAVICON [$site_url]: $favicon_url\n"; | |
1501 | ||
1502 | $icon_file = ICONS_DIR . "/$feed.ico"; | |
1503 | ||
1504 | if (!file_exists($icon_file)) { | |
1505 | $favicon_url = get_favicon_url($site_url); | |
1506 | ||
1507 | if ($favicon_url) { | |
1508 | // Limiting to "image" type misses those served with text/plain | |
1509 | $contents = fetch_file_contents($favicon_url); // , "image"); | |
1510 | ||
1511 | if ($contents) { | |
1512 | // Crude image type matching. | |
1513 | // Patterns gleaned from the file(1) source code. | |
1514 | if (preg_match('/^\x00\x00\x01\x00/', $contents)) { | |
1515 | // 0 string \000\000\001\000 MS Windows icon resource | |
1516 | //error_log("check_feed_favicon: favicon_url=$favicon_url isa MS Windows icon resource"); | |
1517 | } | |
1518 | elseif (preg_match('/^GIF8/', $contents)) { | |
1519 | // 0 string GIF8 GIF image data | |
1520 | //error_log("check_feed_favicon: favicon_url=$favicon_url isa GIF image"); | |
1521 | } | |
1522 | elseif (preg_match('/^\x89PNG\x0d\x0a\x1a\x0a/', $contents)) { | |
1523 | // 0 string \x89PNG\x0d\x0a\x1a\x0a PNG image data | |
1524 | //error_log("check_feed_favicon: favicon_url=$favicon_url isa PNG image"); | |
1525 | } | |
1526 | elseif (preg_match('/^\xff\xd8/', $contents)) { | |
1527 | // 0 beshort 0xffd8 JPEG image data | |
1528 | //error_log("check_feed_favicon: favicon_url=$favicon_url isa JPG image"); | |
1529 | } | |
1530 | elseif (preg_match('/^BM/', $contents)) { | |
1531 | // 0 string BM PC bitmap (OS2, Windows BMP files) | |
1532 | //error_log("check_feed_favicon, favicon_url=$favicon_url isa BMP image"); | |
1533 | } | |
1534 | else { | |
1535 | //error_log("check_feed_favicon: favicon_url=$favicon_url isa UNKNOWN type"); | |
1536 | $contents = ""; | |
1537 | } | |
1538 | } | |
1539 | ||
1540 | if ($contents) { | |
1541 | $fp = @fopen($icon_file, "w"); | |
1542 | ||
1543 | if ($fp) { | |
1544 | fwrite($fp, $contents); | |
1545 | fclose($fp); | |
1546 | chmod($icon_file, 0644); | |
1547 | } | |
1548 | } | |
1549 | } | |
1550 | return $icon_file; | |
1551 | } | |
1552 | } | |
1553 | ||
1554 | ||
1555 | ||
1556 | } |