]>
Commit | Line | Data |
---|---|---|
2c08214a | 1 | <?php |
fce451a4 AD |
2 | define_default('DAEMON_UPDATE_LOGIN_LIMIT', 30); |
3 | define_default('DAEMON_FEED_LIMIT', 500); | |
e2cf81e2 | 4 | define_default('DAEMON_SLEEP_INTERVAL', 120); |
09e8bdfd | 5 | |
6322ac79 | 6 | function update_feedbrowser_cache() { |
79178062 | 7 | |
a42c55f0 | 8 | $result = db_query("SELECT feed_url, site_url, title, COUNT(id) AS subscribers |
79178062 AD |
9 | FROM ttrss_feeds WHERE (SELECT COUNT(id) = 0 FROM ttrss_feeds AS tf |
10 | WHERE tf.feed_url = ttrss_feeds.feed_url | |
11 | AND (private IS true OR auth_login != '' OR auth_pass != '' OR feed_url LIKE '%:%@%/%')) | |
12 | GROUP BY feed_url, site_url, title ORDER BY subscribers DESC LIMIT 1000"); | |
13 | ||
a42c55f0 | 14 | db_query("BEGIN"); |
79178062 | 15 | |
a42c55f0 | 16 | db_query("DELETE FROM ttrss_feedbrowser_cache"); |
79178062 AD |
17 | |
18 | $count = 0; | |
19 | ||
20 | while ($line = db_fetch_assoc($result)) { | |
a42c55f0 AD |
21 | $subscribers = db_escape_string($line["subscribers"]); |
22 | $feed_url = db_escape_string($line["feed_url"]); | |
23 | $title = db_escape_string($line["title"]); | |
24 | $site_url = db_escape_string($line["site_url"]); | |
79178062 | 25 | |
a42c55f0 | 26 | $tmp_result = db_query("SELECT subscribers FROM |
79178062 AD |
27 | ttrss_feedbrowser_cache WHERE feed_url = '$feed_url'"); |
28 | ||
29 | if (db_num_rows($tmp_result) == 0) { | |
30 | ||
a42c55f0 | 31 | db_query("INSERT INTO ttrss_feedbrowser_cache |
79178062 AD |
32 | (feed_url, site_url, title, subscribers) VALUES ('$feed_url', |
33 | '$site_url', '$title', '$subscribers')"); | |
34 | ||
35 | ++$count; | |
36 | ||
37 | } | |
38 | ||
39 | } | |
40 | ||
a42c55f0 | 41 | db_query("COMMIT"); |
79178062 AD |
42 | |
43 | return $count; | |
44 | ||
45 | } | |
46 | ||
47 | ||
2c08214a AD |
48 | /** |
49 | * Update a feed batch. | |
50 | * Used by daemons to update n feeds by run. | |
51 | * Only update feed needing a update, and not being processed | |
52 | * by another process. | |
53 | * | |
54 | * @param mixed $link Database link | |
55 | * @param integer $limit Maximum number of feeds in update batch. Default to DAEMON_FEED_LIMIT. | |
56 | * @param boolean $from_http Set to true if you call this function from http to disable cli specific code. | |
57 | * @param boolean $debug Set to false to disable debug output. Default to true. | |
58 | * @return void | |
59 | */ | |
a42c55f0 | 60 | function update_daemon_common($limit = DAEMON_FEED_LIMIT, $from_http = false, $debug = true) { |
2c08214a AD |
61 | // Process all other feeds using last_updated and interval parameters |
62 | ||
6322ac79 | 63 | $schema_version = get_schema_version(); |
857efe49 AD |
64 | |
65 | if ($schema_version != SCHEMA_VERSION) { | |
66 | die("Schema version is wrong, please upgrade the database.\n"); | |
67 | } | |
68 | ||
61c1812f AD |
69 | define('PREFS_NO_CACHE', true); |
70 | ||
2c08214a | 71 | // Test if the user has loggued in recently. If not, it does not update its feeds. |
09e8bdfd | 72 | if (!SINGLE_USER_MODE && DAEMON_UPDATE_LOGIN_LIMIT > 0) { |
2c08214a AD |
73 | if (DB_TYPE == "pgsql") { |
74 | $login_thresh_qpart = "AND ttrss_users.last_login >= NOW() - INTERVAL '".DAEMON_UPDATE_LOGIN_LIMIT." days'"; | |
75 | } else { | |
76 | $login_thresh_qpart = "AND ttrss_users.last_login >= DATE_SUB(NOW(), INTERVAL ".DAEMON_UPDATE_LOGIN_LIMIT." DAY)"; | |
77 | } | |
78 | } else { | |
79 | $login_thresh_qpart = ""; | |
80 | } | |
81 | ||
82 | // Test if the feed need a update (update interval exceded). | |
83 | if (DB_TYPE == "pgsql") { | |
84 | $update_limit_qpart = "AND (( | |
85 | ttrss_feeds.update_interval = 0 | |
ee0542ce | 86 | AND ttrss_user_prefs.value != '-1' |
2c08214a AD |
87 | AND ttrss_feeds.last_updated < NOW() - CAST((ttrss_user_prefs.value || ' minutes') AS INTERVAL) |
88 | ) OR ( | |
89 | ttrss_feeds.update_interval > 0 | |
90 | AND ttrss_feeds.last_updated < NOW() - CAST((ttrss_feeds.update_interval || ' minutes') AS INTERVAL) | |
941e48a4 AD |
91 | ) OR ttrss_feeds.last_updated IS NULL |
92 | OR last_updated = '1970-01-01 00:00:00')"; | |
2c08214a AD |
93 | } else { |
94 | $update_limit_qpart = "AND (( | |
95 | ttrss_feeds.update_interval = 0 | |
ee0542ce | 96 | AND ttrss_user_prefs.value != '-1' |
2c08214a AD |
97 | AND ttrss_feeds.last_updated < DATE_SUB(NOW(), INTERVAL CONVERT(ttrss_user_prefs.value, SIGNED INTEGER) MINUTE) |
98 | ) OR ( | |
99 | ttrss_feeds.update_interval > 0 | |
100 | AND ttrss_feeds.last_updated < DATE_SUB(NOW(), INTERVAL ttrss_feeds.update_interval MINUTE) | |
941e48a4 AD |
101 | ) OR ttrss_feeds.last_updated IS NULL |
102 | OR last_updated = '1970-01-01 00:00:00')"; | |
2c08214a AD |
103 | } |
104 | ||
105 | // Test if feed is currently being updated by another process. | |
106 | if (DB_TYPE == "pgsql") { | |
8cabc200 | 107 | $updstart_thresh_qpart = "AND (ttrss_feeds.last_update_started IS NULL OR ttrss_feeds.last_update_started < NOW() - INTERVAL '10 minutes')"; |
2c08214a | 108 | } else { |
8cabc200 | 109 | $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 |
110 | } |
111 | ||
112 | // Test if there is a limit to number of updated feeds | |
113 | $query_limit = ""; | |
114 | if($limit) $query_limit = sprintf("LIMIT %d", $limit); | |
115 | ||
fce451a4 | 116 | $query = "SELECT DISTINCT ttrss_feeds.feed_url, ttrss_feeds.last_updated |
2c08214a AD |
117 | FROM |
118 | ttrss_feeds, ttrss_users, ttrss_user_prefs | |
119 | WHERE | |
120 | ttrss_feeds.owner_uid = ttrss_users.id | |
121 | AND ttrss_users.id = ttrss_user_prefs.owner_uid | |
122 | AND ttrss_user_prefs.pref_name = 'DEFAULT_UPDATE_INTERVAL' | |
123 | $login_thresh_qpart $update_limit_qpart | |
1c4421fc | 124 | $updstart_thresh_qpart |
fce451a4 AD |
125 | ORDER BY last_updated $query_limit"; |
126 | ||
127 | // We search for feed needing update. | |
128 | $result = db_query($query); | |
2c08214a | 129 | |
ca5ff2d9 | 130 | if($debug) _debug(sprintf("Scheduled %d feeds to update...", db_num_rows($result))); |
2c08214a AD |
131 | |
132 | // Here is a little cache magic in order to minimize risk of double feed updates. | |
133 | $feeds_to_update = array(); | |
134 | while ($line = db_fetch_assoc($result)) { | |
a42c55f0 | 135 | array_push($feeds_to_update, db_escape_string($line['feed_url'])); |
2c08214a AD |
136 | } |
137 | ||
138 | // We update the feed last update started date before anything else. | |
139 | // There is no lag due to feed contents downloads | |
140 | // It prevent an other process to update the same feed. | |
1c4421fc AD |
141 | |
142 | if(count($feeds_to_update) > 0) { | |
143 | $feeds_quoted = array(); | |
144 | ||
145 | foreach ($feeds_to_update as $feed) { | |
a42c55f0 | 146 | array_push($feeds_quoted, "'" . db_escape_string($feed) . "'"); |
1c4421fc AD |
147 | } |
148 | ||
a42c55f0 | 149 | db_query(sprintf("UPDATE ttrss_feeds SET last_update_started = NOW() |
1c4421fc | 150 | WHERE feed_url IN (%s)", implode(',', $feeds_quoted))); |
2c08214a AD |
151 | } |
152 | ||
8292d05b AD |
153 | $nf = 0; |
154 | ||
2c08214a | 155 | // For each feed, we call the feed update function. |
1c4421fc AD |
156 | foreach ($feeds_to_update as $feed) { |
157 | if($debug) _debug("Base feed: $feed"); | |
2c08214a | 158 | |
a42c55f0 | 159 | //update_rss_feed($line["id"], true); |
2c08214a | 160 | |
1c4421fc AD |
161 | // since we have the data cached, we can deal with other feeds with the same url |
162 | ||
a42c55f0 | 163 | $tmp_result = db_query("SELECT DISTINCT ttrss_feeds.id,last_updated |
ee0542ce AD |
164 | FROM ttrss_feeds, ttrss_users, ttrss_user_prefs WHERE |
165 | ttrss_user_prefs.owner_uid = ttrss_feeds.owner_uid AND | |
166 | ttrss_users.id = ttrss_user_prefs.owner_uid AND | |
167 | ttrss_user_prefs.pref_name = 'DEFAULT_UPDATE_INTERVAL' AND | |
a42c55f0 | 168 | feed_url = '".db_escape_string($feed)."' AND |
ee0542ce AD |
169 | (ttrss_feeds.update_interval > 0 OR |
170 | ttrss_user_prefs.value != '-1') | |
1c4421fc | 171 | $login_thresh_qpart |
5929a0c1 | 172 | ORDER BY ttrss_feeds.id $query_limit"); |
1c4421fc AD |
173 | |
174 | if (db_num_rows($tmp_result) > 0) { | |
175 | while ($tline = db_fetch_assoc($tmp_result)) { | |
32b05702 | 176 | if($debug) _debug(" => " . $tline["last_updated"] . ", " . $tline["id"]); |
a42c55f0 | 177 | update_rss_feed($tline["id"], true); |
8292d05b | 178 | ++$nf; |
1c4421fc AD |
179 | } |
180 | } | |
2c08214a AD |
181 | } |
182 | ||
f1611683 AD |
183 | require_once "digest.php"; |
184 | ||
2c08214a | 185 | // Send feed digests by email if needed. |
a42c55f0 | 186 | send_headlines_digests($debug); |
2c08214a | 187 | |
8292d05b AD |
188 | return $nf; |
189 | ||
2c08214a AD |
190 | } // function update_daemon_common |
191 | ||
87764a50 | 192 | // ignore_daemon is not used |
a42c55f0 | 193 | function update_rss_feed($feed, $ignore_daemon = false, $no_cache = false, |
2c08214a AD |
194 | $override_url = false) { |
195 | ||
2c08214a AD |
196 | $debug_enabled = defined('DAEMON_EXTENDED_DEBUG') || $_REQUEST['xdebug']; |
197 | ||
68cccafc | 198 | _debug("start", $debug_enabled); |
2c08214a | 199 | |
a42c55f0 | 200 | $result = db_query("SELECT id,update_interval,auth_login, |
0a3fd79b | 201 | feed_url,auth_pass,cache_images,last_updated, |
5321e775 | 202 | mark_unread_on_update, owner_uid, |
90e5f4f1 AD |
203 | pubsub_state, auth_pass_encrypted, |
204 | (SELECT max(date_entered) FROM | |
205 | ttrss_entries, ttrss_user_entries where ref_id = id AND feed_id = '$feed') AS last_article_timestamp | |
87764a50 | 206 | FROM ttrss_feeds WHERE id = '$feed'"); |
2c08214a AD |
207 | |
208 | if (db_num_rows($result) == 0) { | |
68cccafc | 209 | _debug("feed $feed NOT FOUND/SKIPPED", $debug_enabled); |
2c08214a AD |
210 | return false; |
211 | } | |
212 | ||
65c8d5e7 AD |
213 | // set last update to now so if anything *simplepie* crashes later we won't be |
214 | // continuously failing on the same feed | |
215 | db_query("UPDATE ttrss_feeds SET last_updated = NOW() WHERE id = '$feed'"); | |
216 | ||
2c08214a | 217 | $last_updated = db_fetch_result($result, 0, "last_updated"); |
90e5f4f1 | 218 | $last_article_timestamp = @strtotime(db_fetch_result($result, 0, "last_article_timestamp")); |
2c08214a AD |
219 | $owner_uid = db_fetch_result($result, 0, "owner_uid"); |
220 | $mark_unread_on_update = sql_bool_to_bool(db_fetch_result($result, | |
221 | 0, "mark_unread_on_update")); | |
2c08214a | 222 | $pubsub_state = db_fetch_result($result, 0, "pubsub_state"); |
044cff2d AD |
223 | $auth_pass_encrypted = sql_bool_to_bool(db_fetch_result($result, |
224 | 0, "auth_pass_encrypted")); | |
2c08214a | 225 | |
a42c55f0 | 226 | db_query("UPDATE ttrss_feeds SET last_update_started = NOW() |
2c08214a AD |
227 | WHERE id = '$feed'"); |
228 | ||
229 | $auth_login = db_fetch_result($result, 0, "auth_login"); | |
230 | $auth_pass = db_fetch_result($result, 0, "auth_pass"); | |
231 | ||
044cff2d AD |
232 | if ($auth_pass_encrypted) { |
233 | require_once "crypt.php"; | |
234 | $auth_pass = decrypt_string($auth_pass); | |
235 | } | |
236 | ||
2c08214a AD |
237 | $cache_images = sql_bool_to_bool(db_fetch_result($result, 0, "cache_images")); |
238 | $fetch_url = db_fetch_result($result, 0, "feed_url"); | |
239 | ||
a42c55f0 | 240 | $feed = db_escape_string($feed); |
2c08214a | 241 | |
f074ffe9 | 242 | if ($override_url) $fetch_url = $override_url; |
2c08214a | 243 | |
f074ffe9 | 244 | $date_feed_processed = date('Y-m-d H:i'); |
2c08214a | 245 | |
865a3ed6 | 246 | $cache_filename = CACHE_DIR . "/simplepie/" . sha1($fetch_url) . ".xml"; |
f074ffe9 AD |
247 | |
248 | // Ignore cache if new feed or manual update. | |
23283f11 | 249 | $cache_age = ($no_cache || is_null($last_updated) || strpos($last_updated, '1970-01-01') === 0) ? 30 : get_feed_update_interval($feed) * 60; |
2c08214a | 250 | |
68cccafc AD |
251 | _debug("cache filename: $cache_filename exists: " . file_exists($cache_filename), $debug_enabled); |
252 | _debug("cache age: $cache_age; no cache: $no_cache", $debug_enabled); | |
2c08214a | 253 | |
f074ffe9 | 254 | $cached_feed_data_hash = false; |
2c08214a | 255 | |
4f9cbdff AD |
256 | $rss = false; |
257 | $rss_hash = false; | |
7a01dc77 | 258 | $cache_timestamp = file_exists($cache_filename) ? filemtime($cache_filename) : 0; |
4f9cbdff | 259 | |
687a4f59 AD |
260 | $force_refetch = isset($_REQUEST["force_refetch"]); |
261 | ||
f074ffe9 AD |
262 | if (file_exists($cache_filename) && |
263 | is_readable($cache_filename) && | |
264 | !$auth_login && !$auth_pass && | |
265 | filemtime($cache_filename) > time() - $cache_age) { | |
2c08214a | 266 | |
68cccafc | 267 | _debug("using local cache.", $debug_enabled); |
be574731 | 268 | |
90e5f4f1 | 269 | if ($cache_timestamp > $last_article_timestamp) { |
b8f316dc | 270 | @$feed_data = file_get_contents($cache_filename); |
f074ffe9 | 271 | |
b8f316dc AD |
272 | if ($feed_data) { |
273 | $rss_hash = sha1($feed_data); | |
17e74b21 | 274 | } |
687a4f59 | 275 | } else if (!$force_refetch) { |
68cccafc | 276 | _debug("local cache valid and older than last_updated, nothing to do.", $debug_enabled); |
17e74b21 | 277 | return; |
4f9cbdff | 278 | } |
81c20663 AD |
279 | } else { |
280 | _debug("local cache will not be used for this feed", $debug_enabled); | |
f074ffe9 | 281 | } |
017401dd | 282 | |
4f9cbdff | 283 | if (!$rss) { |
017401dd | 284 | |
4f9cbdff | 285 | if (!$feed_data) { |
90e5f4f1 AD |
286 | _debug("fetching [$fetch_url]...", $debug_enabled); |
287 | _debug("If-Modified-Since: ".gmdate('D, d M Y H:i:s \G\M\T', $last_article_timestamp), $debug_enabled); | |
017401dd | 288 | |
4f9cbdff | 289 | $feed_data = fetch_file_contents($fetch_url, false, |
5929a0c1 | 290 | $auth_login, $auth_pass, false, |
8401101d | 291 | $no_cache ? FEED_FETCH_NO_CACHE_TIMEOUT : FEED_FETCH_TIMEOUT, |
90e5f4f1 | 292 | $force_refetch ? 0 : $last_article_timestamp); |
96f98cb0 | 293 | |
3f6f0857 AD |
294 | global $fetch_curl_used; |
295 | ||
296 | if (!$fetch_curl_used) { | |
297 | $tmp = @gzdecode($feed_data); | |
298 | ||
299 | if ($tmp) $feed_data = $tmp; | |
300 | } | |
301 | ||
1367bc3f AD |
302 | $feed_data = trim($feed_data); |
303 | ||
68cccafc | 304 | _debug("fetch done.", $debug_enabled); |
017401dd | 305 | |
68cccafc AD |
306 | if ($feed_data) { |
307 | $error = verify_feed_xml($feed_data); | |
ebec81a6 | 308 | |
68cccafc AD |
309 | if ($error) { |
310 | _debug("error verifying XML, code: " . $error->code, $debug_enabled); | |
ebec81a6 | 311 | |
68cccafc AD |
312 | if ($error->code == 26) { |
313 | _debug("got error 26, trying to decode entities...", $debug_enabled); | |
ebec81a6 | 314 | |
68cccafc | 315 | $feed_data = html_entity_decode($feed_data, ENT_COMPAT, 'UTF-8'); |
ebec81a6 | 316 | |
68cccafc | 317 | $error = verify_feed_xml($feed_data); |
ebec81a6 | 318 | |
68cccafc AD |
319 | if ($error) $feed_data = ''; |
320 | } | |
ebec81a6 AD |
321 | } |
322 | } | |
4f9cbdff | 323 | } |
017401dd | 324 | |
4f9cbdff AD |
325 | if (!$feed_data) { |
326 | global $fetch_last_error; | |
7a01dc77 | 327 | global $fetch_last_error_code; |
f074ffe9 | 328 | |
68cccafc | 329 | _debug("unable to fetch: $fetch_last_error [$fetch_last_error_code]", $debug_enabled); |
f074ffe9 | 330 | |
7a01dc77 AD |
331 | $error_escaped = ''; |
332 | ||
333 | // If-Modified-Since | |
334 | if ($fetch_last_error_code != 304) { | |
a42c55f0 | 335 | $error_escaped = db_escape_string($fetch_last_error); |
7a01dc77 | 336 | } else { |
68cccafc | 337 | _debug("source claims data not modified, nothing to do.", $debug_enabled); |
7a01dc77 | 338 | } |
4f9cbdff | 339 | |
6322ac79 | 340 | db_query( |
4f9cbdff AD |
341 | "UPDATE ttrss_feeds SET last_error = '$error_escaped', |
342 | last_updated = NOW() WHERE id = '$feed'"); | |
343 | ||
344 | return; | |
f074ffe9 AD |
345 | } |
346 | } | |
347 | ||
1ffe3391 | 348 | $pluginhost = new PluginHost(); |
68cccafc | 349 | $pluginhost->set_debug($debug_enabled, $debug_enabled); |
a42c55f0 | 350 | $user_plugins = get_pref("_ENABLED_PLUGINS", $owner_uid); |
ab457a9c | 351 | |
1ffe3391 AD |
352 | $pluginhost->load(PLUGINS, PluginHost::KIND_ALL); |
353 | $pluginhost->load($user_plugins, PluginHost::KIND_USER, $owner_uid); | |
ab457a9c | 354 | $pluginhost->load_data(); |
017401dd | 355 | |
1ffe3391 | 356 | foreach ($pluginhost->get_hooks(PluginHost::HOOK_FEED_FETCHED) as $plugin) { |
017401dd AD |
357 | $feed_data = $plugin->hook_feed_fetched($feed_data); |
358 | } | |
359 | ||
4f9cbdff | 360 | if (!$rss) { |
cd07592c AD |
361 | $rss = new FeedParser($feed_data); |
362 | $rss->init(); | |
5de51df7 AD |
363 | } |
364 | ||
2c08214a AD |
365 | // print_r($rss); |
366 | ||
a42c55f0 | 367 | $feed = db_escape_string($feed); |
2c08214a | 368 | |
19b3992b | 369 | if (!$rss->error()) { |
2c08214a | 370 | |
4f9cbdff AD |
371 | // cache data for later |
372 | if (!$auth_pass && !$auth_login && is_writable(CACHE_DIR . "/simplepie")) { | |
4f9cbdff AD |
373 | $new_rss_hash = sha1($rss_data); |
374 | ||
b8f316dc | 375 | if ($new_rss_hash != $rss_hash && count($rss->get_items()) > 0 ) { |
68cccafc | 376 | _debug("saving $cache_filename", $debug_enabled); |
431e2785 | 377 | @file_put_contents($cache_filename, $feed_data); |
4f9cbdff AD |
378 | } |
379 | } | |
380 | ||
d2a421e3 | 381 | // We use local pluginhost here because we need to load different per-user feed plugins |
1ffe3391 | 382 | $pluginhost->run_hooks(PluginHost::HOOK_FEED_PARSED, "hook_feed_parsed", $rss); |
4412b877 | 383 | |
68cccafc | 384 | _debug("processing feed data...", $debug_enabled); |
2c08214a | 385 | |
a42c55f0 | 386 | // db_query("BEGIN"); |
2c08214a | 387 | |
382268c6 AD |
388 | if (DB_TYPE == "pgsql") { |
389 | $favicon_interval_qpart = "favicon_last_checked < NOW() - INTERVAL '12 hour'"; | |
390 | } else { | |
391 | $favicon_interval_qpart = "favicon_last_checked < DATE_SUB(NOW(), INTERVAL 12 HOUR)"; | |
392 | } | |
393 | ||
36490f11 | 394 | $result = db_query("SELECT title,site_url,owner_uid,favicon_avg_color, |
382268c6 AD |
395 | (favicon_last_checked IS NULL OR $favicon_interval_qpart) AS |
396 | favicon_needs_check | |
2c08214a AD |
397 | FROM ttrss_feeds WHERE id = '$feed'"); |
398 | ||
399 | $registered_title = db_fetch_result($result, 0, "title"); | |
2c08214a | 400 | $orig_site_url = db_fetch_result($result, 0, "site_url"); |
382268c6 AD |
401 | $favicon_needs_check = sql_bool_to_bool(db_fetch_result($result, 0, |
402 | "favicon_needs_check")); | |
36490f11 | 403 | $favicon_avg_color = db_fetch_result($result, 0, "favicon_avg_color"); |
2c08214a AD |
404 | |
405 | $owner_uid = db_fetch_result($result, 0, "owner_uid"); | |
406 | ||
a42c55f0 | 407 | $site_url = db_escape_string(mb_substr(rewrite_relative_url($fetch_url, $rss->get_link()), 0, 245)); |
2c08214a | 408 | |
cd07592c AD |
409 | _debug("site_url: $site_url", $debug_enabled); |
410 | _debug("feed_title: " . $rss->get_title(), $debug_enabled); | |
411 | ||
687a4f59 | 412 | if ($favicon_needs_check || $force_refetch) { |
36490f11 AD |
413 | |
414 | /* terrible hack: if we crash on floicon shit here, we won't check | |
560cbd8c | 415 | * the icon avgcolor again (unless the icon got updated) */ |
36490f11 | 416 | |
560cbd8c AD |
417 | $favicon_file = ICONS_DIR . "/$feed.ico"; |
418 | $favicon_modified = @filemtime($favicon_file); | |
419 | ||
68cccafc | 420 | _debug("checking favicon...", $debug_enabled); |
687a4f59 AD |
421 | |
422 | check_feed_favicon($site_url, $feed, $link); | |
560cbd8c AD |
423 | $favicon_modified_new = @filemtime($favicon_file); |
424 | ||
425 | if ($favicon_modified_new > $favicon_modified) | |
426 | $favicon_avg_color = ''; | |
687a4f59 | 427 | |
6ee0d4b0 | 428 | if (file_exists($favicon_file) && function_exists("imagecreatefromstring") && $favicon_avg_color == '') { |
bc7a144d | 429 | require_once "colors.php"; |
687a4f59 | 430 | |
aafd55ba AD |
431 | db_query("UPDATE ttrss_feeds SET favicon_avg_color = 'fail' WHERE |
432 | id = '$feed'"); | |
433 | ||
6322ac79 | 434 | $favicon_color = db_escape_string( |
bc7a144d | 435 | calculate_avg_color($favicon_file)); |
63c323f7 | 436 | |
bc7a144d | 437 | $favicon_colorstring = ",favicon_avg_color = '".$favicon_color."'"; |
36490f11 | 438 | } else if ($favicon_avg_color == 'fail') { |
84ceb6bd | 439 | _debug("floicon failed on this file, not trying to recalculate avg color", $debug_enabled); |
6ac722d5 | 440 | } |
687a4f59 | 441 | |
a42c55f0 | 442 | db_query("UPDATE ttrss_feeds SET favicon_last_checked = NOW() |
bc7a144d | 443 | $favicon_colorstring |
f2798eb6 AD |
444 | WHERE id = '$feed'"); |
445 | } | |
2c08214a AD |
446 | |
447 | if (!$registered_title || $registered_title == "[Unknown]") { | |
448 | ||
a42c55f0 | 449 | $feed_title = db_escape_string($rss->get_title()); |
2c08214a | 450 | |
852d4ac8 AD |
451 | if ($feed_title) { |
452 | _debug("registering title: $feed_title", $debug_enabled); | |
2c08214a | 453 | |
852d4ac8 AD |
454 | db_query("UPDATE ttrss_feeds SET |
455 | title = '$feed_title' WHERE id = '$feed'"); | |
456 | } | |
2c08214a AD |
457 | } |
458 | ||
0cf81637 | 459 | if ($site_url && $orig_site_url != $site_url) { |
a42c55f0 | 460 | db_query("UPDATE ttrss_feeds SET |
2c08214a AD |
461 | site_url = '$site_url' WHERE id = '$feed'"); |
462 | } | |
463 | ||
68cccafc | 464 | _debug("loading filters & labels...", $debug_enabled); |
2c08214a | 465 | |
a42c55f0 AD |
466 | $filters = load_filters($feed, $owner_uid); |
467 | $labels = get_all_labels($owner_uid); | |
2c08214a | 468 | |
68cccafc | 469 | _debug("" . count($filters) . " filters loaded.", $debug_enabled); |
2c08214a | 470 | |
19b3992b | 471 | $items = $rss->get_items(); |
2c08214a | 472 | |
19b3992b | 473 | if (!is_array($items)) { |
68cccafc | 474 | _debug("no articles found.", $debug_enabled); |
2c08214a | 475 | |
a42c55f0 | 476 | db_query("UPDATE ttrss_feeds |
2c08214a AD |
477 | SET last_updated = NOW(), last_error = '' WHERE id = '$feed'"); |
478 | ||
479 | return; // no articles | |
480 | } | |
481 | ||
482 | if ($pubsub_state != 2 && PUBSUBHUBBUB_ENABLED) { | |
483 | ||
68cccafc | 484 | _debug("checking for PUSH hub...", $debug_enabled); |
2c08214a AD |
485 | |
486 | $feed_hub_url = false; | |
2c08214a | 487 | |
19b3992b | 488 | $links = $rss->get_links('hub'); |
2c08214a | 489 | |
19b3992b AD |
490 | if ($links && is_array($links)) { |
491 | foreach ($links as $l) { | |
492 | $feed_hub_url = $l; | |
493 | break; | |
2c08214a AD |
494 | } |
495 | } | |
496 | ||
68cccafc | 497 | _debug("feed hub url: $feed_hub_url", $debug_enabled); |
2c08214a AD |
498 | |
499 | if ($feed_hub_url && function_exists('curl_init') && | |
500 | !ini_get("open_basedir")) { | |
501 | ||
502 | require_once 'lib/pubsubhubbub/subscriber.php'; | |
503 | ||
504 | $callback_url = get_self_url_prefix() . | |
505 | "/public.php?op=pubsub&id=$feed"; | |
506 | ||
507 | $s = new Subscriber($feed_hub_url, $callback_url); | |
508 | ||
509 | $rc = $s->subscribe($fetch_url); | |
510 | ||
68cccafc | 511 | _debug("feed hub url found, subscribe request sent.", $debug_enabled); |
2c08214a | 512 | |
a42c55f0 | 513 | db_query("UPDATE ttrss_feeds SET pubsub_state = 1 |
2c08214a AD |
514 | WHERE id = '$feed'"); |
515 | } | |
516 | } | |
517 | ||
68cccafc | 518 | _debug("processing articles...", $debug_enabled); |
2c08214a | 519 | |
19b3992b | 520 | foreach ($items as $item) { |
5d56d100 | 521 | if ($_REQUEST['xdebug'] == 3) { |
2c08214a AD |
522 | print_r($item); |
523 | } | |
524 | ||
19b3992b AD |
525 | $entry_guid = $item->get_id(); |
526 | if (!$entry_guid) $entry_guid = $item->get_link(); | |
527 | if (!$entry_guid) $entry_guid = make_guid_from_title($item->get_title()); | |
2c08214a | 528 | |
cd07592c AD |
529 | _debug("f_guid $entry_guid", $debug_enabled); |
530 | ||
2c08214a AD |
531 | if (!$entry_guid) continue; |
532 | ||
3a4c8973 AD |
533 | $entry_guid = "$owner_uid,$entry_guid"; |
534 | ||
a42c55f0 | 535 | $entry_guid_hashed = db_escape_string('SHA1:' . sha1($entry_guid)); |
5e3d5480 | 536 | |
68cccafc | 537 | _debug("guid $entry_guid / $entry_guid_hashed", $debug_enabled); |
5e3d5480 | 538 | |
2c08214a AD |
539 | $entry_timestamp = ""; |
540 | ||
04d2f9c8 AD |
541 | $entry_timestamp = $item->get_date(); |
542 | ||
543 | _debug("orig date: " . $item->get_date(), $debug_enabled); | |
2c08214a | 544 | |
30123fe6 | 545 | if ($entry_timestamp == -1 || !$entry_timestamp || $entry_timestamp > time()) { |
2c08214a AD |
546 | $entry_timestamp = time(); |
547 | $no_orig_date = 'true'; | |
548 | } else { | |
549 | $no_orig_date = 'false'; | |
550 | } | |
551 | ||
552 | $entry_timestamp_fmt = strftime("%Y/%m/%d %H:%M:%S", $entry_timestamp); | |
553 | ||
68cccafc | 554 | _debug("date $entry_timestamp [$entry_timestamp_fmt]", $debug_enabled); |
2c08214a | 555 | |
99429e57 AD |
556 | // $entry_title = html_entity_decode($item->get_title(), ENT_COMPAT, 'UTF-8'); |
557 | // $entry_title = decode_numeric_entities($entry_title); | |
558 | $entry_title = $item->get_title(); | |
1b35d30c | 559 | |
5d56d100 | 560 | $entry_link = rewrite_relative_url($site_url, $item->get_link()); |
2c08214a | 561 | |
68cccafc AD |
562 | _debug("title $entry_title", $debug_enabled); |
563 | _debug("link $entry_link", $debug_enabled); | |
2c08214a AD |
564 | |
565 | if (!$entry_title) $entry_title = date("Y-m-d H:i:s", $entry_timestamp);; | |
566 | ||
19b3992b AD |
567 | $entry_content = $item->get_content(); |
568 | if (!$entry_content) $entry_content = $item->get_description(); | |
2c08214a AD |
569 | |
570 | if ($_REQUEST["xdebug"] == 2) { | |
9ec10352 | 571 | print "content: "; |
487f0750 | 572 | print $entry_content; |
3c696512 | 573 | print "\n"; |
2c08214a AD |
574 | } |
575 | ||
cd07592c AD |
576 | $entry_comments = $item->get_comments_url(); |
577 | $entry_author = $item->get_author(); | |
2c08214a | 578 | |
a42c55f0 | 579 | $entry_guid = db_escape_string(mb_substr($entry_guid, 0, 245)); |
2c08214a | 580 | |
ce2b219c AD |
581 | $entry_comments = db_escape_string(mb_substr(trim($entry_comments), 0, 245)); |
582 | $entry_author = db_escape_string(mb_substr(trim($entry_author), 0, 245)); | |
2c08214a | 583 | |
d4992d6b | 584 | $num_comments = (int) $item->get_comments_count(); |
2c08214a | 585 | |
68cccafc AD |
586 | _debug("author $entry_author", $debug_enabled); |
587 | _debug("num_comments: $num_comments", $debug_enabled); | |
ee78f81c | 588 | _debug("looking for tags...", $debug_enabled); |
2c08214a AD |
589 | |
590 | // parse <category> entries into tags | |
591 | ||
592 | $additional_tags = array(); | |
593 | ||
19b3992b | 594 | $additional_tags_src = $item->get_categories(); |
2c08214a | 595 | |
19b3992b AD |
596 | if (is_array($additional_tags_src)) { |
597 | foreach ($additional_tags_src as $tobj) { | |
cd07592c | 598 | array_push($additional_tags, $tobj); |
2c08214a | 599 | } |
19b3992b | 600 | } |
2c08214a | 601 | |
fa6fbd36 | 602 | $entry_tags = array_unique($additional_tags); |
2c08214a AD |
603 | |
604 | for ($i = 0; $i < count($entry_tags); $i++) | |
605 | $entry_tags[$i] = mb_strtolower($entry_tags[$i], 'utf-8'); | |
606 | ||
ee78f81c AD |
607 | _debug("tags found: " . join(",", $entry_tags), $debug_enabled); |
608 | ||
68cccafc | 609 | _debug("done collecting data.", $debug_enabled); |
2c08214a | 610 | |
cc85704f | 611 | // TODO: less memory-hungry implementation |
19c73507 | 612 | |
68cccafc | 613 | _debug("applying plugin filters..", $debug_enabled); |
cc85704f | 614 | |
e02555c1 | 615 | // FIXME not sure if owner_uid is a good idea here, we may have a base entry without user entry (?) |
a42c55f0 AD |
616 | $result = db_query("SELECT plugin_data,title,content,link,tag_cache,author FROM ttrss_entries, ttrss_user_entries |
617 | WHERE ref_id = id AND (guid = '".db_escape_string($entry_guid)."' OR guid = '$entry_guid_hashed') AND owner_uid = $owner_uid"); | |
b30abdad AD |
618 | |
619 | if (db_num_rows($result) != 0) { | |
620 | $entry_plugin_data = db_fetch_result($result, 0, "plugin_data"); | |
e02555c1 AD |
621 | $stored_article = array("title" => db_fetch_result($result, 0, "title"), |
622 | "content" => db_fetch_result($result, 0, "content"), | |
623 | "link" => db_fetch_result($result, 0, "link"), | |
624 | "tags" => explode(",", db_fetch_result($result, 0, "tag_cache")), | |
625 | "author" => db_fetch_result($result, 0, "author")); | |
b30abdad AD |
626 | } else { |
627 | $entry_plugin_data = ""; | |
e02555c1 | 628 | $stored_article = array(); |
b30abdad AD |
629 | } |
630 | ||
455b1401 | 631 | $article = array("owner_uid" => $owner_uid, // read only |
b30abdad | 632 | "guid" => $entry_guid, // read only |
19b3992b AD |
633 | "title" => $entry_title, |
634 | "content" => $entry_content, | |
635 | "link" => $entry_link, | |
636 | "tags" => $entry_tags, | |
b30abdad | 637 | "plugin_data" => $entry_plugin_data, |
e02555c1 AD |
638 | "author" => $entry_author, |
639 | "stored" => $stored_article); | |
cc85704f | 640 | |
1ffe3391 | 641 | foreach ($pluginhost->get_hooks(PluginHost::HOOK_ARTICLE_FILTER) as $plugin) { |
19b3992b | 642 | $article = $plugin->hook_article_filter($article); |
cc85704f AD |
643 | } |
644 | ||
19b3992b | 645 | $entry_tags = $article["tags"]; |
a42c55f0 AD |
646 | $entry_guid = db_escape_string($entry_guid); |
647 | $entry_title = db_escape_string($article["title"]); | |
648 | $entry_author = db_escape_string($article["author"]); | |
649 | $entry_link = db_escape_string($article["link"]); | |
650 | $entry_plugin_data = db_escape_string($article["plugin_data"]); | |
f935d98e AD |
651 | $entry_content = $article["content"]; // escaped below |
652 | ||
b30abdad | 653 | |
68cccafc | 654 | _debug("plugin data: $entry_plugin_data", $debug_enabled); |
2bbd6994 | 655 | |
0a3fd79b | 656 | if ($cache_images && is_writable(CACHE_DIR . '/images')) |
f0bd8e65 | 657 | cache_images($entry_content, $site_url, $debug_enabled); |
0a3fd79b | 658 | |
a42c55f0 | 659 | $entry_content = db_escape_string($entry_content, false); |
1f45c857 | 660 | |
9e222305 | 661 | $content_hash = "SHA1:" . sha1($entry_content); |
cc85704f | 662 | |
a42c55f0 | 663 | db_query("BEGIN"); |
2c08214a | 664 | |
a42c55f0 | 665 | $result = db_query("SELECT id FROM ttrss_entries |
5e3d5480 | 666 | WHERE (guid = '$entry_guid' OR guid = '$entry_guid_hashed')"); |
9e222305 | 667 | |
2c08214a AD |
668 | if (db_num_rows($result) == 0) { |
669 | ||
68cccafc | 670 | _debug("base guid [$entry_guid] not found", $debug_enabled); |
2c08214a AD |
671 | |
672 | // base post entry does not exist, create it | |
673 | ||
6322ac79 | 674 | $result = db_query( |
2c08214a AD |
675 | "INSERT INTO ttrss_entries |
676 | (title, | |
677 | guid, | |
678 | link, | |
679 | updated, | |
680 | content, | |
681 | content_hash, | |
87764a50 | 682 | cached_content, |
2c08214a AD |
683 | no_orig_date, |
684 | date_updated, | |
685 | date_entered, | |
686 | comments, | |
687 | num_comments, | |
b30abdad | 688 | plugin_data, |
2c08214a AD |
689 | author) |
690 | VALUES | |
691 | ('$entry_title', | |
5e3d5480 | 692 | '$entry_guid_hashed', |
2c08214a AD |
693 | '$entry_link', |
694 | '$entry_timestamp_fmt', | |
695 | '$entry_content', | |
696 | '$content_hash', | |
0a3fd79b | 697 | '', |
2c08214a AD |
698 | $no_orig_date, |
699 | NOW(), | |
be574731 | 700 | '$date_feed_processed', |
2c08214a AD |
701 | '$entry_comments', |
702 | '$num_comments', | |
b30abdad | 703 | '$entry_plugin_data', |
2c08214a | 704 | '$entry_author')"); |
e8291805 AD |
705 | |
706 | $article_labels = array(); | |
707 | ||
2c08214a AD |
708 | } else { |
709 | // we keep encountering the entry in feeds, so we need to | |
710 | // update date_updated column so that we don't get horrible | |
711 | // dupes when the entry gets purged and reinserted again e.g. | |
712 | // in the case of SLOW SLOW OMG SLOW updating feeds | |
713 | ||
714 | $base_entry_id = db_fetch_result($result, 0, "id"); | |
715 | ||
a42c55f0 | 716 | db_query("UPDATE ttrss_entries SET date_updated = NOW() |
2c08214a | 717 | WHERE id = '$base_entry_id'"); |
e8291805 | 718 | |
a42c55f0 | 719 | $article_labels = get_article_labels($base_entry_id, $owner_uid); |
2c08214a AD |
720 | } |
721 | ||
722 | // now it should exist, if not - bad luck then | |
723 | ||
a42c55f0 | 724 | $result = db_query("SELECT |
5e3d5480 | 725 | id,content_hash,no_orig_date,title,plugin_data,guid, |
2c08214a AD |
726 | ".SUBSTRING_FOR_DATE."(date_updated,1,19) as date_updated, |
727 | ".SUBSTRING_FOR_DATE."(updated,1,19) as updated, | |
0a3fd79b | 728 | num_comments |
2c08214a AD |
729 | FROM |
730 | ttrss_entries | |
5e3d5480 | 731 | WHERE guid = '$entry_guid' OR guid = '$entry_guid_hashed'"); |
2c08214a AD |
732 | |
733 | $entry_ref_id = 0; | |
734 | $entry_int_id = 0; | |
735 | ||
736 | if (db_num_rows($result) == 1) { | |
737 | ||
68cccafc | 738 | _debug("base guid found, checking for user record", $debug_enabled); |
2c08214a AD |
739 | |
740 | // this will be used below in update handler | |
741 | $orig_content_hash = db_fetch_result($result, 0, "content_hash"); | |
742 | $orig_title = db_fetch_result($result, 0, "title"); | |
743 | $orig_num_comments = db_fetch_result($result, 0, "num_comments"); | |
744 | $orig_date_updated = strtotime(db_fetch_result($result, | |
745 | 0, "date_updated")); | |
b30abdad | 746 | $orig_plugin_data = db_fetch_result($result, 0, "plugin_data"); |
2c08214a AD |
747 | |
748 | $ref_id = db_fetch_result($result, 0, "id"); | |
749 | $entry_ref_id = $ref_id; | |
750 | ||
5e3d5480 AD |
751 | /* $stored_guid = db_fetch_result($result, 0, "guid"); |
752 | if ($stored_guid != $entry_guid_hashed) { | |
68cccafc | 753 | if ($debug_enabled) _debug("upgrading compat guid to hashed one", $debug_enabled); |
5e3d5480 | 754 | |
a42c55f0 | 755 | db_query("UPDATE ttrss_entries SET guid = '$entry_guid_hashed' WHERE |
5e3d5480 AD |
756 | id = '$ref_id'"); |
757 | } */ | |
758 | ||
2c08214a AD |
759 | // check for user post link to main table |
760 | ||
761 | // do we allow duplicate posts with same GUID in different feeds? | |
a42c55f0 | 762 | if (get_pref("ALLOW_DUPLICATE_POSTS", $owner_uid, false)) { |
2c08214a AD |
763 | $dupcheck_qpart = "AND (feed_id = '$feed' OR feed_id IS NULL)"; |
764 | } else { | |
765 | $dupcheck_qpart = ""; | |
766 | } | |
767 | ||
768 | /* Collect article tags here so we could filter by them: */ | |
769 | ||
770 | $article_filters = get_article_filters($filters, $entry_title, | |
4021d61a | 771 | $entry_content, $entry_link, $entry_timestamp, $entry_author, |
2c08214a AD |
772 | $entry_tags); |
773 | ||
774 | if ($debug_enabled) { | |
68cccafc | 775 | _debug("article filters: ", $debug_enabled); |
2c08214a AD |
776 | if (count($article_filters) != 0) { |
777 | print_r($article_filters); | |
778 | } | |
779 | } | |
780 | ||
781 | if (find_article_filter($article_filters, "filter")) { | |
a42c55f0 | 782 | db_query("COMMIT"); // close transaction in progress |
2c08214a AD |
783 | continue; |
784 | } | |
785 | ||
786 | $score = calculate_article_score($article_filters); | |
787 | ||
68cccafc | 788 | _debug("initial score: $score", $debug_enabled); |
2c08214a AD |
789 | |
790 | $query = "SELECT ref_id, int_id FROM ttrss_user_entries WHERE | |
791 | ref_id = '$ref_id' AND owner_uid = '$owner_uid' | |
792 | $dupcheck_qpart"; | |
793 | ||
794 | // if ($_REQUEST["xdebug"]) print "$query\n"; | |
795 | ||
a42c55f0 | 796 | $result = db_query($query); |
2c08214a AD |
797 | |
798 | // okay it doesn't exist - create user entry | |
799 | if (db_num_rows($result) == 0) { | |
800 | ||
68cccafc | 801 | _debug("user record not found, creating...", $debug_enabled); |
2c08214a AD |
802 | |
803 | if ($score >= -500 && !find_article_filter($article_filters, 'catchup')) { | |
804 | $unread = 'true'; | |
805 | $last_read_qpart = 'NULL'; | |
806 | } else { | |
807 | $unread = 'false'; | |
808 | $last_read_qpart = 'NOW()'; | |
809 | } | |
810 | ||
811 | if (find_article_filter($article_filters, 'mark') || $score > 1000) { | |
812 | $marked = 'true'; | |
813 | } else { | |
814 | $marked = 'false'; | |
815 | } | |
816 | ||
817 | if (find_article_filter($article_filters, 'publish')) { | |
818 | $published = 'true'; | |
819 | } else { | |
820 | $published = 'false'; | |
821 | } | |
822 | ||
2ea9bbfd AD |
823 | // N-grams |
824 | ||
825 | if (DB_TYPE == "pgsql" and defined('_NGRAM_TITLE_DUPLICATE_THRESHOLD')) { | |
826 | ||
a42c55f0 | 827 | $result = db_query("SELECT COUNT(*) AS similar FROM |
2ea9bbfd AD |
828 | ttrss_entries,ttrss_user_entries |
829 | WHERE ref_id = id AND updated >= NOW() - INTERVAL '7 day' | |
830 | AND similarity(title, '$entry_title') >= "._NGRAM_TITLE_DUPLICATE_THRESHOLD." | |
831 | AND owner_uid = $owner_uid"); | |
832 | ||
833 | $ngram_similar = db_fetch_result($result, 0, "similar"); | |
834 | ||
68cccafc | 835 | _debug("N-gram similar results: $ngram_similar", $debug_enabled); |
2ea9bbfd AD |
836 | |
837 | if ($ngram_similar > 0) { | |
838 | $unread = 'false'; | |
839 | } | |
840 | } | |
841 | ||
7873d588 AD |
842 | $last_marked = ($marked == 'true') ? 'NOW()' : 'NULL'; |
843 | $last_published = ($published == 'true') ? 'NOW()' : 'NULL'; | |
844 | ||
6322ac79 | 845 | $result = db_query( |
2c08214a AD |
846 | "INSERT INTO ttrss_user_entries |
847 | (ref_id, owner_uid, feed_id, unread, last_read, marked, | |
7873d588 AD |
848 | published, score, tag_cache, label_cache, uuid, |
849 | last_marked, last_published) | |
2c08214a | 850 | VALUES ('$ref_id', '$owner_uid', '$feed', $unread, |
7873d588 AD |
851 | $last_read_qpart, $marked, $published, '$score', '', '', |
852 | '', $last_marked, $last_published)"); | |
2c08214a AD |
853 | |
854 | if (PUBSUBHUBBUB_HUB && $published == 'true') { | |
855 | $rss_link = get_self_url_prefix() . | |
856 | "/public.php?op=rss&id=-2&key=" . | |
a42c55f0 | 857 | get_feed_access_key(-2, false, $owner_uid); |
2c08214a AD |
858 | |
859 | $p = new Publisher(PUBSUBHUBBUB_HUB); | |
860 | ||
861 | $pubsub_result = $p->publish_update($rss_link); | |
862 | } | |
863 | ||
6322ac79 | 864 | $result = db_query( |
2c08214a AD |
865 | "SELECT int_id FROM ttrss_user_entries WHERE |
866 | ref_id = '$ref_id' AND owner_uid = '$owner_uid' AND | |
867 | feed_id = '$feed' LIMIT 1"); | |
868 | ||
869 | if (db_num_rows($result) == 1) { | |
870 | $entry_int_id = db_fetch_result($result, 0, "int_id"); | |
871 | } | |
872 | } else { | |
68cccafc | 873 | _debug("user record FOUND", $debug_enabled); |
2c08214a AD |
874 | |
875 | $entry_ref_id = db_fetch_result($result, 0, "ref_id"); | |
876 | $entry_int_id = db_fetch_result($result, 0, "int_id"); | |
877 | } | |
878 | ||
68cccafc | 879 | _debug("RID: $entry_ref_id, IID: $entry_int_id", $debug_enabled); |
2c08214a AD |
880 | |
881 | $post_needs_update = false; | |
882 | $update_insignificant = false; | |
883 | ||
884 | if ($orig_num_comments != $num_comments) { | |
885 | $post_needs_update = true; | |
886 | $update_insignificant = true; | |
887 | } | |
888 | ||
b30abdad AD |
889 | if ($entry_plugin_data != $orig_plugin_data) { |
890 | $post_needs_update = true; | |
891 | $update_insignificant = true; | |
892 | } | |
893 | ||
2c08214a AD |
894 | if ($content_hash != $orig_content_hash) { |
895 | $post_needs_update = true; | |
896 | $update_insignificant = false; | |
897 | } | |
898 | ||
a42c55f0 | 899 | if (db_escape_string($orig_title) != $entry_title) { |
2c08214a AD |
900 | $post_needs_update = true; |
901 | $update_insignificant = false; | |
902 | } | |
903 | ||
904 | // if post needs update, update it and mark all user entries | |
905 | // linking to this post as updated | |
906 | if ($post_needs_update) { | |
907 | ||
908 | if (defined('DAEMON_EXTENDED_DEBUG')) { | |
68cccafc | 909 | _debug("post $entry_guid_hashed needs update...", $debug_enabled); |
2c08214a AD |
910 | } |
911 | ||
912 | // print "<!-- post $orig_title needs update : $post_needs_update -->"; | |
913 | ||
a42c55f0 | 914 | db_query("UPDATE ttrss_entries |
2c08214a AD |
915 | SET title = '$entry_title', content = '$entry_content', |
916 | content_hash = '$content_hash', | |
917 | updated = '$entry_timestamp_fmt', | |
b30abdad AD |
918 | num_comments = '$num_comments', |
919 | plugin_data = '$entry_plugin_data' | |
2c08214a AD |
920 | WHERE id = '$ref_id'"); |
921 | ||
922 | if (!$update_insignificant) { | |
923 | if ($mark_unread_on_update) { | |
a42c55f0 | 924 | db_query("UPDATE ttrss_user_entries |
2c08214a | 925 | SET last_read = null, unread = true WHERE ref_id = '$ref_id'"); |
2c08214a AD |
926 | } |
927 | } | |
928 | } | |
929 | } | |
930 | ||
a42c55f0 | 931 | db_query("COMMIT"); |
2c08214a | 932 | |
68cccafc | 933 | _debug("assigning labels...", $debug_enabled); |
2c08214a | 934 | |
a42c55f0 | 935 | assign_article_to_label_filters($entry_ref_id, $article_filters, |
b24504b1 | 936 | $owner_uid, $article_labels); |
2c08214a | 937 | |
68cccafc | 938 | _debug("looking for enclosures...", $debug_enabled); |
2c08214a AD |
939 | |
940 | // enclosures | |
941 | ||
942 | $enclosures = array(); | |
943 | ||
19b3992b | 944 | $encs = $item->get_enclosures(); |
2c08214a | 945 | |
19b3992b AD |
946 | if (is_array($encs)) { |
947 | foreach ($encs as $e) { | |
948 | $e_item = array( | |
949 | $e->link, $e->type, $e->length); | |
2c08214a | 950 | array_push($enclosures, $e_item); |
2c08214a AD |
951 | } |
952 | } | |
953 | ||
2c08214a | 954 | if ($debug_enabled) { |
68cccafc | 955 | _debug("article enclosures:", $debug_enabled); |
2c08214a AD |
956 | print_r($enclosures); |
957 | } | |
958 | ||
a42c55f0 | 959 | db_query("BEGIN"); |
2c08214a AD |
960 | |
961 | foreach ($enclosures as $enc) { | |
a42c55f0 AD |
962 | $enc_url = db_escape_string($enc[0]); |
963 | $enc_type = db_escape_string($enc[1]); | |
964 | $enc_dur = db_escape_string($enc[2]); | |
2c08214a | 965 | |
a42c55f0 | 966 | $result = db_query("SELECT id FROM ttrss_enclosures |
2c08214a AD |
967 | WHERE content_url = '$enc_url' AND post_id = '$entry_ref_id'"); |
968 | ||
969 | if (db_num_rows($result) == 0) { | |
a42c55f0 | 970 | db_query("INSERT INTO ttrss_enclosures |
2c08214a AD |
971 | (content_url, content_type, title, duration, post_id) VALUES |
972 | ('$enc_url', '$enc_type', '', '$enc_dur', '$entry_ref_id')"); | |
973 | } | |
974 | } | |
975 | ||
a42c55f0 | 976 | db_query("COMMIT"); |
2c08214a AD |
977 | |
978 | // check for manual tags (we have to do it here since they're loaded from filters) | |
979 | ||
980 | foreach ($article_filters as $f) { | |
6aff7845 | 981 | if ($f["type"] == "tag") { |
2c08214a | 982 | |
6aff7845 | 983 | $manual_tags = trim_array(explode(",", $f["param"])); |
2c08214a AD |
984 | |
985 | foreach ($manual_tags as $tag) { | |
986 | if (tag_is_valid($tag)) { | |
987 | array_push($entry_tags, $tag); | |
988 | } | |
989 | } | |
990 | } | |
991 | } | |
992 | ||
993 | // Skip boring tags | |
994 | ||
6322ac79 | 995 | $boring_tags = trim_array(explode(",", mb_strtolower(get_pref( |
2c08214a AD |
996 | 'BLACKLISTED_TAGS', $owner_uid, ''), 'utf-8'))); |
997 | ||
998 | $filtered_tags = array(); | |
999 | $tags_to_cache = array(); | |
1000 | ||
1001 | if ($entry_tags && is_array($entry_tags)) { | |
1002 | foreach ($entry_tags as $tag) { | |
1003 | if (array_search($tag, $boring_tags) === false) { | |
1004 | array_push($filtered_tags, $tag); | |
1005 | } | |
1006 | } | |
1007 | } | |
1008 | ||
1009 | $filtered_tags = array_unique($filtered_tags); | |
1010 | ||
1011 | if ($debug_enabled) { | |
68cccafc | 1012 | _debug("filtered article tags:", $debug_enabled); |
2c08214a AD |
1013 | print_r($filtered_tags); |
1014 | } | |
1015 | ||
1016 | // Save article tags in the database | |
1017 | ||
1018 | if (count($filtered_tags) > 0) { | |
1019 | ||
a42c55f0 | 1020 | db_query("BEGIN"); |
2c08214a AD |
1021 | |
1022 | foreach ($filtered_tags as $tag) { | |
1023 | ||
1024 | $tag = sanitize_tag($tag); | |
a42c55f0 | 1025 | $tag = db_escape_string($tag); |
2c08214a AD |
1026 | |
1027 | if (!tag_is_valid($tag)) continue; | |
1028 | ||
a42c55f0 | 1029 | $result = db_query("SELECT id FROM ttrss_tags |
2c08214a AD |
1030 | WHERE tag_name = '$tag' AND post_int_id = '$entry_int_id' AND |
1031 | owner_uid = '$owner_uid' LIMIT 1"); | |
1032 | ||
1033 | if ($result && db_num_rows($result) == 0) { | |
1034 | ||
a42c55f0 | 1035 | db_query("INSERT INTO ttrss_tags |
2c08214a AD |
1036 | (owner_uid,tag_name,post_int_id) |
1037 | VALUES ('$owner_uid','$tag', '$entry_int_id')"); | |
1038 | } | |
1039 | ||
1040 | array_push($tags_to_cache, $tag); | |
1041 | } | |
1042 | ||
1043 | /* update the cache */ | |
1044 | ||
1045 | $tags_to_cache = array_unique($tags_to_cache); | |
1046 | ||
a42c55f0 | 1047 | $tags_str = db_escape_string(join(",", $tags_to_cache)); |
2c08214a | 1048 | |
a42c55f0 | 1049 | db_query("UPDATE ttrss_user_entries |
2c08214a AD |
1050 | SET tag_cache = '$tags_str' WHERE ref_id = '$entry_ref_id' |
1051 | AND owner_uid = $owner_uid"); | |
1052 | ||
a42c55f0 | 1053 | db_query("COMMIT"); |
2c08214a AD |
1054 | } |
1055 | ||
a42c55f0 | 1056 | if (get_pref("AUTO_ASSIGN_LABELS", $owner_uid, false)) { |
68cccafc | 1057 | _debug("auto-assigning labels...", $debug_enabled); |
b24504b1 AD |
1058 | |
1059 | foreach ($labels as $label) { | |
227d5e41 | 1060 | $caption = preg_quote($label["caption"]); |
b24504b1 | 1061 | |
9811276d | 1062 | if ($caption && preg_match("/\b$caption\b/i", "$tags_str " . strip_tags($entry_content) . " $entry_title")) { |
b24504b1 | 1063 | if (!labels_contains_caption($article_labels, $caption)) { |
a42c55f0 | 1064 | label_add_article($entry_ref_id, $caption, $owner_uid); |
b24504b1 AD |
1065 | } |
1066 | } | |
1067 | } | |
1068 | } | |
1069 | ||
68cccafc | 1070 | _debug("article processed", $debug_enabled); |
2c08214a AD |
1071 | } |
1072 | ||
1073 | if (!$last_updated) { | |
68cccafc | 1074 | _debug("new feed, catching it up...", $debug_enabled); |
a42c55f0 | 1075 | catchup_feed($feed, false, $owner_uid); |
2c08214a AD |
1076 | } |
1077 | ||
68cccafc | 1078 | _debug("purging feed...", $debug_enabled); |
2c08214a | 1079 | |
a42c55f0 | 1080 | purge_feed($feed, 0, $debug_enabled); |
2c08214a | 1081 | |
a42c55f0 | 1082 | db_query("UPDATE ttrss_feeds |
f2798eb6 | 1083 | SET last_updated = NOW(), last_error = '' WHERE id = '$feed'"); |
2c08214a | 1084 | |
a42c55f0 | 1085 | // db_query("COMMIT"); |
2c08214a AD |
1086 | |
1087 | } else { | |
1088 | ||
a42c55f0 | 1089 | $error_msg = db_escape_string(mb_substr($rss->error(), 0, 245)); |
2c08214a | 1090 | |
68cccafc | 1091 | _debug("error fetching feed: $error_msg", $debug_enabled); |
2c08214a | 1092 | |
6322ac79 | 1093 | db_query( |
2c08214a AD |
1094 | "UPDATE ttrss_feeds SET last_error = '$error_msg', |
1095 | last_updated = NOW() WHERE id = '$feed'"); | |
1096 | } | |
1097 | ||
19b3992b | 1098 | unset($rss); |
2c08214a | 1099 | |
68cccafc | 1100 | _debug("done", $debug_enabled); |
2c08214a AD |
1101 | } |
1102 | ||
3c696512 AD |
1103 | function cache_images($html, $site_url, $debug) { |
1104 | $cache_dir = CACHE_DIR . "/images"; | |
1105 | ||
1106 | libxml_use_internal_errors(true); | |
1107 | ||
1108 | $charset_hack = '<head> | |
1109 | <meta http-equiv="Content-Type" content="text/html; charset=utf-8"/> | |
1110 | </head>'; | |
1111 | ||
1112 | $doc = new DOMDocument(); | |
1113 | $doc->loadHTML($charset_hack . $html); | |
1114 | $xpath = new DOMXPath($doc); | |
1115 | ||
1116 | $entries = $xpath->query('(//img[@src])'); | |
1117 | ||
1118 | foreach ($entries as $entry) { | |
1119 | if ($entry->hasAttribute('src')) { | |
1120 | $src = rewrite_relative_url($site_url, $entry->getAttribute('src')); | |
1121 | ||
1122 | $local_filename = CACHE_DIR . "/images/" . sha1($src) . ".png"; | |
1123 | ||
1124 | if ($debug) _debug("cache_images: downloading: $src to $local_filename"); | |
1125 | ||
1126 | if (!file_exists($local_filename)) { | |
1127 | $file_content = fetch_file_contents($src); | |
1128 | ||
b8379e69 | 1129 | if ($file_content && strlen($file_content) > 1024) { |
3c696512 AD |
1130 | file_put_contents($local_filename, $file_content); |
1131 | } | |
1132 | } | |
1133 | ||
1134 | if (file_exists($local_filename)) { | |
1135 | $entry->setAttribute('src', SELF_URL_PATH . '/image.php?url=' . | |
487f0750 | 1136 | base64_encode($src)); |
3c696512 AD |
1137 | } |
1138 | } | |
1139 | } | |
1140 | ||
1141 | $node = $doc->getElementsByTagName('body')->item(0); | |
1142 | ||
cc38c8e5 | 1143 | return $doc->saveXML($node); |
3c696512 AD |
1144 | } |
1145 | ||
a42c55f0 | 1146 | function expire_error_log($debug) { |
e2261e17 AD |
1147 | if ($debug) _debug("Removing old error log entries..."); |
1148 | ||
1149 | if (DB_TYPE == "pgsql") { | |
a42c55f0 | 1150 | db_query("DELETE FROM ttrss_error_log |
e2261e17 AD |
1151 | WHERE created_at < NOW() - INTERVAL '7 days'"); |
1152 | } else { | |
a42c55f0 | 1153 | db_query("DELETE FROM ttrss_error_log |
e2261e17 AD |
1154 | WHERE created_at < DATE_SUB(NOW(), INTERVAL 7 DAY)"); |
1155 | } | |
1156 | ||
1157 | } | |
1158 | ||
2a91b6ff | 1159 | function expire_lock_files($debug) { |
65465085 | 1160 | //if ($debug) _debug("Removing old lock files..."); |
2a91b6ff AD |
1161 | |
1162 | $num_deleted = 0; | |
1163 | ||
1164 | if (is_writable(LOCK_DIRECTORY)) { | |
1165 | $files = glob(LOCK_DIRECTORY . "/*.lock"); | |
1166 | ||
1167 | if ($files) { | |
1168 | foreach ($files as $file) { | |
11344971 | 1169 | if (!file_is_locked(basename($file)) && time() - filemtime($file) > 86400*2) { |
2a91b6ff AD |
1170 | unlink($file); |
1171 | ++$num_deleted; | |
1172 | } | |
1173 | } | |
1174 | } | |
1175 | } | |
1176 | ||
65465085 | 1177 | if ($debug) _debug("Removed $num_deleted old lock files."); |
2a91b6ff AD |
1178 | } |
1179 | ||
3c696512 | 1180 | function expire_cached_files($debug) { |
3306daec | 1181 | foreach (array("simplepie", "images", "export", "upload") as $dir) { |
3c696512 | 1182 | $cache_dir = CACHE_DIR . "/$dir"; |
2c08214a | 1183 | |
65465085 | 1184 | // if ($debug) _debug("Expiring $cache_dir"); |
2c08214a | 1185 | |
3c696512 AD |
1186 | $num_deleted = 0; |
1187 | ||
1188 | if (is_writable($cache_dir)) { | |
1189 | $files = glob("$cache_dir/*"); | |
1190 | ||
2a91b6ff | 1191 | if ($files) { |
2ab20c31 AD |
1192 | foreach ($files as $file) { |
1193 | if (time() - filemtime($file) > 86400*7) { | |
1194 | unlink($file); | |
3c696512 | 1195 | |
2ab20c31 AD |
1196 | ++$num_deleted; |
1197 | } | |
3c696512 AD |
1198 | } |
1199 | } | |
2a91b6ff | 1200 | } |
3c696512 | 1201 | |
65465085 | 1202 | if ($debug) _debug("$cache_dir: removed $num_deleted files."); |
3c696512 AD |
1203 | } |
1204 | } | |
2c08214a | 1205 | |
a3e0bdcf AD |
1206 | /** |
1207 | * Source: http://www.php.net/manual/en/function.parse-url.php#104527 | |
1208 | * Returns the url query as associative array | |
1209 | * | |
1210 | * @param string query | |
1211 | * @return array params | |
1212 | */ | |
1213 | function convertUrlQuery($query) { | |
1214 | $queryParts = explode('&', $query); | |
1215 | ||
1216 | $params = array(); | |
1217 | ||
1218 | foreach ($queryParts as $param) { | |
1219 | $item = explode('=', $param); | |
1220 | $params[$item[0]] = $item[1]; | |
1221 | } | |
1222 | ||
1223 | return $params; | |
1224 | } | |
92c14e9d | 1225 | |
48cb2536 | 1226 | function get_article_filters($filters, $title, $content, $link, $timestamp, $author, $tags) { |
92c14e9d AD |
1227 | $matches = array(); |
1228 | ||
1229 | foreach ($filters as $filter) { | |
1230 | $match_any_rule = $filter["match_any_rule"]; | |
a3a896a1 | 1231 | $inverse = $filter["inverse"]; |
92c14e9d AD |
1232 | $filter_match = false; |
1233 | ||
1234 | foreach ($filter["rules"] as $rule) { | |
1235 | $match = false; | |
ffa1bd7b | 1236 | $reg_exp = str_replace('/', '\/', $rule["reg_exp"]); |
a3a896a1 | 1237 | $rule_inverse = $rule["inverse"]; |
92c14e9d AD |
1238 | |
1239 | if (!$reg_exp) | |
1240 | continue; | |
1241 | ||
1242 | switch ($rule["type"]) { | |
1243 | case "title": | |
1244 | $match = @preg_match("/$reg_exp/i", $title); | |
1245 | break; | |
1246 | case "content": | |
d03ae73e AD |
1247 | // we don't need to deal with multiline regexps |
1248 | $content = preg_replace("/[\r\n\t]/", "", $content); | |
1249 | ||
92c14e9d AD |
1250 | $match = @preg_match("/$reg_exp/i", $content); |
1251 | break; | |
1252 | case "both": | |
d03ae73e AD |
1253 | // we don't need to deal with multiline regexps |
1254 | $content = preg_replace("/[\r\n\t]/", "", $content); | |
1255 | ||
72d1d067 | 1256 | $match = (@preg_match("/$reg_exp/i", $title) || @preg_match("/$reg_exp/i", $content)); |
92c14e9d AD |
1257 | break; |
1258 | case "link": | |
1259 | $match = @preg_match("/$reg_exp/i", $link); | |
1260 | break; | |
1261 | case "author": | |
1262 | $match = @preg_match("/$reg_exp/i", $author); | |
1263 | break; | |
1264 | case "tag": | |
7b80b5e1 DK |
1265 | foreach ($tags as $tag) { |
1266 | if (@preg_match("/$reg_exp/i", $tag)) { | |
1267 | $match = true; | |
1268 | break; | |
1269 | } | |
1270 | } | |
92c14e9d AD |
1271 | break; |
1272 | } | |
1273 | ||
a3a896a1 AD |
1274 | if ($rule_inverse) $match = !$match; |
1275 | ||
92c14e9d AD |
1276 | if ($match_any_rule) { |
1277 | if ($match) { | |
1278 | $filter_match = true; | |
1279 | break; | |
1280 | } | |
1281 | } else { | |
1282 | $filter_match = $match; | |
1283 | if (!$match) { | |
1284 | break; | |
1285 | } | |
1286 | } | |
1287 | } | |
1288 | ||
a3a896a1 AD |
1289 | if ($inverse) $filter_match = !$filter_match; |
1290 | ||
92c14e9d AD |
1291 | if ($filter_match) { |
1292 | foreach ($filter["actions"] AS $action) { | |
1293 | array_push($matches, $action); | |
5e736e45 AD |
1294 | |
1295 | // if Stop action encountered, perform no further processing | |
1296 | if ($action["type"] == "stop") return $matches; | |
92c14e9d AD |
1297 | } |
1298 | } | |
1299 | } | |
1300 | ||
1301 | return $matches; | |
1302 | } | |
1303 | ||
1304 | function find_article_filter($filters, $filter_name) { | |
1305 | foreach ($filters as $f) { | |
1306 | if ($f["type"] == $filter_name) { | |
1307 | return $f; | |
1308 | }; | |
1309 | } | |
1310 | return false; | |
1311 | } | |
1312 | ||
1313 | function find_article_filters($filters, $filter_name) { | |
1314 | $results = array(); | |
1315 | ||
1316 | foreach ($filters as $f) { | |
1317 | if ($f["type"] == $filter_name) { | |
1318 | array_push($results, $f); | |
1319 | }; | |
1320 | } | |
1321 | return $results; | |
1322 | } | |
1323 | ||
1324 | function calculate_article_score($filters) { | |
1325 | $score = 0; | |
1326 | ||
1327 | foreach ($filters as $f) { | |
1328 | if ($f["type"] == "score") { | |
1329 | $score += $f["param"]; | |
1330 | }; | |
1331 | } | |
1332 | return $score; | |
1333 | } | |
1334 | ||
b24504b1 AD |
1335 | function labels_contains_caption($labels, $caption) { |
1336 | foreach ($labels as $label) { | |
1337 | if ($label[1] == $caption) { | |
1338 | return true; | |
1339 | } | |
1340 | } | |
1341 | ||
1342 | return false; | |
1343 | } | |
1344 | ||
a42c55f0 | 1345 | function assign_article_to_label_filters($id, $filters, $owner_uid, $article_labels) { |
92c14e9d AD |
1346 | foreach ($filters as $f) { |
1347 | if ($f["type"] == "label") { | |
b24504b1 | 1348 | if (!labels_contains_caption($article_labels, $f["param"])) { |
a42c55f0 | 1349 | label_add_article($id, $f["param"], $owner_uid); |
b24504b1 AD |
1350 | } |
1351 | } | |
92c14e9d AD |
1352 | } |
1353 | } | |
87764a50 | 1354 | |
87d7e850 AD |
1355 | function make_guid_from_title($title) { |
1356 | return preg_replace("/[ \"\',.:;]/", "-", | |
1357 | mb_strtolower(strip_tags($title), 'utf-8')); | |
1358 | } | |
1359 | ||
ebec81a6 AD |
1360 | function verify_feed_xml($feed_data) { |
1361 | libxml_use_internal_errors(true); | |
1362 | $doc = new DOMDocument(); | |
1363 | $doc->loadXML($feed_data); | |
1364 | $error = libxml_get_last_error(); | |
1365 | libxml_clear_errors(); | |
1366 | return $error; | |
1367 | } | |
87d7e850 | 1368 | |
e2cf81e2 AD |
1369 | function housekeeping_common($debug) { |
1370 | expire_cached_files($debug); | |
1371 | expire_lock_files($debug); | |
1372 | expire_error_log($debug); | |
1373 | ||
1374 | $count = update_feedbrowser_cache(); | |
1375 | _debug("Feedbrowser updated, $count feeds processed."); | |
1376 | ||
1377 | purge_orphans( true); | |
1378 | $rc = cleanup_tags( 14, 50000); | |
1379 | ||
1380 | _debug("Cleaned $rc cached tags."); | |
1381 | } | |
2c08214a | 1382 | ?> |