]>
Commit | Line | Data |
---|---|---|
2c08214a | 1 | <?php |
09e8bdfd AD |
2 | define('DAEMON_UPDATE_LOGIN_LIMIT', 30); |
3 | define('DAEMON_FEED_LIMIT', 100); | |
0f556e3e | 4 | define('DAEMON_SLEEP_INTERVAL', 60); |
09e8bdfd | 5 | |
79178062 AD |
6 | function update_feedbrowser_cache($link) { |
7 | ||
8 | $result = db_query($link, "SELECT feed_url, site_url, title, COUNT(id) AS subscribers | |
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 | ||
14 | db_query($link, "BEGIN"); | |
15 | ||
16 | db_query($link, "DELETE FROM ttrss_feedbrowser_cache"); | |
17 | ||
18 | $count = 0; | |
19 | ||
20 | while ($line = db_fetch_assoc($result)) { | |
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"]); | |
25 | ||
26 | $tmp_result = db_query($link, "SELECT subscribers FROM | |
27 | ttrss_feedbrowser_cache WHERE feed_url = '$feed_url'"); | |
28 | ||
29 | if (db_num_rows($tmp_result) == 0) { | |
30 | ||
31 | db_query($link, "INSERT INTO ttrss_feedbrowser_cache | |
32 | (feed_url, site_url, title, subscribers) VALUES ('$feed_url', | |
33 | '$site_url', '$title', '$subscribers')"); | |
34 | ||
35 | ++$count; | |
36 | ||
37 | } | |
38 | ||
39 | } | |
40 | ||
41 | db_query($link, "COMMIT"); | |
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 $debug Set to false to disable debug output. Default to true. | |
58 | * @return void | |
59 | */ | |
60 | function update_daemon_common($link, $limit = DAEMON_FEED_LIMIT, $from_http = false, $debug = true) { | |
61 | // Process all other feeds using last_updated and interval parameters | |
62 | ||
61c1812f AD |
63 | define('PREFS_NO_CACHE', true); |
64 | ||
2c08214a | 65 | // Test if the user has loggued in recently. If not, it does not update its feeds. |
09e8bdfd | 66 | if (!SINGLE_USER_MODE && DAEMON_UPDATE_LOGIN_LIMIT > 0) { |
2c08214a AD |
67 | if (DB_TYPE == "pgsql") { |
68 | $login_thresh_qpart = "AND ttrss_users.last_login >= NOW() - INTERVAL '".DAEMON_UPDATE_LOGIN_LIMIT." days'"; | |
69 | } else { | |
70 | $login_thresh_qpart = "AND ttrss_users.last_login >= DATE_SUB(NOW(), INTERVAL ".DAEMON_UPDATE_LOGIN_LIMIT." DAY)"; | |
71 | } | |
72 | } else { | |
73 | $login_thresh_qpart = ""; | |
74 | } | |
75 | ||
76 | // Test if the feed need a update (update interval exceded). | |
77 | if (DB_TYPE == "pgsql") { | |
78 | $update_limit_qpart = "AND (( | |
79 | ttrss_feeds.update_interval = 0 | |
80 | AND ttrss_feeds.last_updated < NOW() - CAST((ttrss_user_prefs.value || ' minutes') AS INTERVAL) | |
81 | ) OR ( | |
82 | ttrss_feeds.update_interval > 0 | |
83 | AND ttrss_feeds.last_updated < NOW() - CAST((ttrss_feeds.update_interval || ' minutes') AS INTERVAL) | |
941e48a4 AD |
84 | ) OR ttrss_feeds.last_updated IS NULL |
85 | OR last_updated = '1970-01-01 00:00:00')"; | |
2c08214a AD |
86 | } else { |
87 | $update_limit_qpart = "AND (( | |
88 | ttrss_feeds.update_interval = 0 | |
89 | AND ttrss_feeds.last_updated < DATE_SUB(NOW(), INTERVAL CONVERT(ttrss_user_prefs.value, SIGNED INTEGER) MINUTE) | |
90 | ) OR ( | |
91 | ttrss_feeds.update_interval > 0 | |
92 | AND ttrss_feeds.last_updated < DATE_SUB(NOW(), INTERVAL ttrss_feeds.update_interval MINUTE) | |
941e48a4 AD |
93 | ) OR ttrss_feeds.last_updated IS NULL |
94 | OR last_updated = '1970-01-01 00:00:00')"; | |
2c08214a AD |
95 | } |
96 | ||
97 | // Test if feed is currently being updated by another process. | |
98 | if (DB_TYPE == "pgsql") { | |
99 | $updstart_thresh_qpart = "AND (ttrss_feeds.last_update_started IS NULL OR ttrss_feeds.last_update_started < NOW() - INTERVAL '5 minutes')"; | |
100 | } else { | |
101 | $updstart_thresh_qpart = "AND (ttrss_feeds.last_update_started IS NULL OR ttrss_feeds.last_update_started < DATE_SUB(NOW(), INTERVAL 5 MINUTE))"; | |
102 | } | |
103 | ||
104 | // Test if there is a limit to number of updated feeds | |
105 | $query_limit = ""; | |
106 | if($limit) $query_limit = sprintf("LIMIT %d", $limit); | |
107 | ||
108 | $random_qpart = sql_random_function(); | |
109 | ||
110 | // We search for feed needing update. | |
111 | $result = db_query($link, "SELECT ttrss_feeds.feed_url,ttrss_feeds.id, ttrss_feeds.owner_uid, | |
112 | ".SUBSTRING_FOR_DATE."(ttrss_feeds.last_updated,1,19) AS last_updated, | |
113 | ttrss_feeds.update_interval | |
114 | FROM | |
115 | ttrss_feeds, ttrss_users, ttrss_user_prefs | |
116 | WHERE | |
117 | ttrss_feeds.owner_uid = ttrss_users.id | |
118 | AND ttrss_users.id = ttrss_user_prefs.owner_uid | |
119 | AND ttrss_user_prefs.pref_name = 'DEFAULT_UPDATE_INTERVAL' | |
120 | $login_thresh_qpart $update_limit_qpart | |
121 | $updstart_thresh_qpart | |
122 | ORDER BY $random_qpart $query_limit"); | |
123 | ||
124 | $user_prefs_cache = array(); | |
125 | ||
126 | if($debug) _debug(sprintf("Scheduled %d feeds to update...\n", db_num_rows($result))); | |
127 | ||
128 | // Here is a little cache magic in order to minimize risk of double feed updates. | |
129 | $feeds_to_update = array(); | |
130 | while ($line = db_fetch_assoc($result)) { | |
131 | $feeds_to_update[$line['id']] = $line; | |
132 | } | |
133 | ||
134 | // We update the feed last update started date before anything else. | |
135 | // There is no lag due to feed contents downloads | |
136 | // It prevent an other process to update the same feed. | |
137 | $feed_ids = array_keys($feeds_to_update); | |
138 | if($feed_ids) { | |
139 | db_query($link, sprintf("UPDATE ttrss_feeds SET last_update_started = NOW() | |
140 | WHERE id IN (%s)", implode(',', $feed_ids))); | |
141 | } | |
142 | ||
3c696512 | 143 | expire_cached_files($debug); |
2a91b6ff | 144 | expire_lock_files($debug); |
3c696512 | 145 | |
2c08214a AD |
146 | // For each feed, we call the feed update function. |
147 | while ($line = array_pop($feeds_to_update)) { | |
148 | ||
149 | if($debug) _debug("Feed: " . $line["feed_url"] . ", " . $line["last_updated"]); | |
150 | ||
151 | update_rss_feed($link, $line["id"], true); | |
152 | ||
153 | sleep(1); // prevent flood (FIXME make this an option?) | |
154 | } | |
155 | ||
156 | // Send feed digests by email if needed. | |
8437c066 | 157 | send_headlines_digests($link, $debug); |
2c08214a AD |
158 | |
159 | } // function update_daemon_common | |
160 | ||
87764a50 | 161 | // ignore_daemon is not used |
0e4a7d7a | 162 | function update_rss_feed($link, $feed, $ignore_daemon = false, $no_cache = false, |
2c08214a AD |
163 | $override_url = false) { |
164 | ||
165 | require_once "lib/simplepie/simplepie.inc"; | |
2c08214a | 166 | |
2c08214a AD |
167 | $debug_enabled = defined('DAEMON_EXTENDED_DEBUG') || $_REQUEST['xdebug']; |
168 | ||
2c08214a AD |
169 | if ($debug_enabled) { |
170 | _debug("update_rss_feed: start"); | |
171 | } | |
172 | ||
87764a50 | 173 | $result = db_query($link, "SELECT id,update_interval,auth_login, |
19b3992b | 174 | feed_url,auth_pass,cache_images,last_updated,cache_content, |
87764a50 AD |
175 | mark_unread_on_update, owner_uid, update_on_checksum_change, |
176 | pubsub_state | |
177 | FROM ttrss_feeds WHERE id = '$feed'"); | |
2c08214a AD |
178 | |
179 | if (db_num_rows($result) == 0) { | |
180 | if ($debug_enabled) { | |
181 | _debug("update_rss_feed: feed $feed NOT FOUND/SKIPPED"); | |
182 | } | |
183 | return false; | |
184 | } | |
185 | ||
2c08214a AD |
186 | $last_updated = db_fetch_result($result, 0, "last_updated"); |
187 | $owner_uid = db_fetch_result($result, 0, "owner_uid"); | |
188 | $mark_unread_on_update = sql_bool_to_bool(db_fetch_result($result, | |
189 | 0, "mark_unread_on_update")); | |
190 | $update_on_checksum_change = sql_bool_to_bool(db_fetch_result($result, | |
191 | 0, "update_on_checksum_change")); | |
192 | $pubsub_state = db_fetch_result($result, 0, "pubsub_state"); | |
193 | ||
194 | db_query($link, "UPDATE ttrss_feeds SET last_update_started = NOW() | |
195 | WHERE id = '$feed'"); | |
196 | ||
197 | $auth_login = db_fetch_result($result, 0, "auth_login"); | |
198 | $auth_pass = db_fetch_result($result, 0, "auth_pass"); | |
199 | ||
2c08214a | 200 | $cache_images = sql_bool_to_bool(db_fetch_result($result, 0, "cache_images")); |
87764a50 | 201 | $cache_content = sql_bool_to_bool(db_fetch_result($result, 0, "cache_content")); |
2c08214a AD |
202 | $fetch_url = db_fetch_result($result, 0, "feed_url"); |
203 | ||
2c08214a AD |
204 | $feed = db_escape_string($feed); |
205 | ||
206 | if ($auth_login && $auth_pass ){ | |
207 | $url_parts = array(); | |
208 | preg_match("/(^[^:]*):\/\/(.*)/", $fetch_url, $url_parts); | |
209 | ||
210 | if ($url_parts[1] && $url_parts[2]) { | |
211 | $fetch_url = $url_parts[1] . "://$auth_login:$auth_pass@" . $url_parts[2]; | |
212 | } | |
213 | ||
214 | } | |
215 | ||
216 | if ($override_url) | |
217 | $fetch_url = $override_url; | |
218 | ||
219 | if ($debug_enabled) { | |
220 | _debug("update_rss_feed: fetching [$fetch_url]..."); | |
221 | } | |
222 | ||
0e4a7d7a AD |
223 | // Ignore cache if new feed or manual update. |
224 | $cache_age = (is_null($last_updated) || $last_updated == '1970-01-01 00:00:00') ? | |
225 | -1 : get_feed_update_interval($link, $feed) * 60; | |
2c08214a | 226 | |
19b3992b | 227 | $simplepie_cache_dir = CACHE_DIR . "/simplepie"; |
2c08214a | 228 | |
19b3992b AD |
229 | if (!is_dir($simplepie_cache_dir)) { |
230 | mkdir($simplepie_cache_dir); | |
231 | } | |
2c08214a | 232 | |
19b3992b AD |
233 | $rss = new SimplePie(); |
234 | $rss->set_useragent(SELF_USER_AGENT); | |
235 | $rss->set_timeout($no_cache ? 15 : 60); | |
236 | $rss->set_feed_url($fetch_url); | |
237 | $rss->set_output_encoding('UTF-8'); | |
238 | //$rss->force_feed(true); | |
2c08214a | 239 | |
19b3992b AD |
240 | if ($debug_enabled) { |
241 | _debug("feed update interval (sec): " . | |
242 | get_feed_update_interval($link, $feed)*60); | |
243 | } | |
2c08214a | 244 | |
19b3992b | 245 | $rss->enable_cache(!$no_cache); |
2c08214a | 246 | |
19b3992b AD |
247 | if (!$no_cache) { |
248 | $rss->set_cache_location($simplepie_cache_dir); | |
249 | $rss->set_cache_duration($cache_age); | |
2c08214a AD |
250 | } |
251 | ||
19b3992b AD |
252 | $rss->init(); |
253 | ||
2c08214a AD |
254 | // print_r($rss); |
255 | ||
256 | if ($debug_enabled) { | |
257 | _debug("update_rss_feed: fetch done, parsing..."); | |
258 | } | |
259 | ||
260 | $feed = db_escape_string($feed); | |
261 | ||
19b3992b | 262 | if (!$rss->error()) { |
2c08214a AD |
263 | |
264 | if ($debug_enabled) { | |
265 | _debug("update_rss_feed: processing feed data..."); | |
266 | } | |
267 | ||
268 | // db_query($link, "BEGIN"); | |
269 | ||
382268c6 AD |
270 | if (DB_TYPE == "pgsql") { |
271 | $favicon_interval_qpart = "favicon_last_checked < NOW() - INTERVAL '12 hour'"; | |
272 | } else { | |
273 | $favicon_interval_qpart = "favicon_last_checked < DATE_SUB(NOW(), INTERVAL 12 HOUR)"; | |
274 | } | |
275 | ||
276 | $result = db_query($link, "SELECT title,icon_url,site_url,owner_uid, | |
277 | (favicon_last_checked IS NULL OR $favicon_interval_qpart) AS | |
278 | favicon_needs_check | |
2c08214a AD |
279 | FROM ttrss_feeds WHERE id = '$feed'"); |
280 | ||
281 | $registered_title = db_fetch_result($result, 0, "title"); | |
282 | $orig_icon_url = db_fetch_result($result, 0, "icon_url"); | |
283 | $orig_site_url = db_fetch_result($result, 0, "site_url"); | |
382268c6 AD |
284 | $favicon_needs_check = sql_bool_to_bool(db_fetch_result($result, 0, |
285 | "favicon_needs_check")); | |
2c08214a AD |
286 | |
287 | $owner_uid = db_fetch_result($result, 0, "owner_uid"); | |
288 | ||
19b3992b | 289 | $site_url = db_escape_string(mb_substr(rewrite_relative_url($fetch_url, $rss->get_link()), 0, 250)); |
2c08214a AD |
290 | |
291 | if ($debug_enabled) { | |
292 | _debug("update_rss_feed: checking favicon..."); | |
293 | } | |
294 | ||
f2798eb6 AD |
295 | if ($favicon_needs_check) { |
296 | check_feed_favicon($site_url, $feed, $link); | |
297 | ||
298 | db_query($link, "UPDATE ttrss_feeds SET favicon_last_checked = NOW() | |
299 | WHERE id = '$feed'"); | |
300 | } | |
2c08214a AD |
301 | |
302 | if (!$registered_title || $registered_title == "[Unknown]") { | |
303 | ||
19b3992b | 304 | $feed_title = db_escape_string($rss->get_title()); |
2c08214a AD |
305 | |
306 | if ($debug_enabled) { | |
307 | _debug("update_rss_feed: registering title: $feed_title"); | |
308 | } | |
309 | ||
310 | db_query($link, "UPDATE ttrss_feeds SET | |
311 | title = '$feed_title' WHERE id = '$feed'"); | |
312 | } | |
313 | ||
0cf81637 | 314 | if ($site_url && $orig_site_url != $site_url) { |
2c08214a AD |
315 | db_query($link, "UPDATE ttrss_feeds SET |
316 | site_url = '$site_url' WHERE id = '$feed'"); | |
317 | } | |
318 | ||
19b3992b AD |
319 | $icon_url = db_escape_string(mb_substr( |
320 | rewrite_relative_url($fetch_url, $rss->get_image_url()), 0, 250)); | |
2c08214a AD |
321 | |
322 | if ($icon_url && $orig_icon_url != $icon_url) { | |
323 | db_query($link, "UPDATE ttrss_feeds SET icon_url = '$icon_url' WHERE id = '$feed'"); | |
324 | } | |
325 | ||
326 | if ($debug_enabled) { | |
b24504b1 | 327 | _debug("update_rss_feed: loading filters & labels..."); |
2c08214a AD |
328 | } |
329 | ||
330 | $filters = load_filters($link, $feed, $owner_uid); | |
b24504b1 | 331 | $labels = get_all_labels($link, $owner_uid); |
2c08214a | 332 | |
67bd0b1f AD |
333 | if ($debug_enabled) { |
334 | //print_r($filters); | |
335 | _debug("update_rss_feed: " . count($filters) . " filters loaded."); | |
336 | } | |
2c08214a | 337 | |
19b3992b | 338 | $items = $rss->get_items(); |
2c08214a | 339 | |
19b3992b | 340 | if (!is_array($items)) { |
2c08214a | 341 | if ($debug_enabled) { |
19b3992b | 342 | _debug("update_rss_feed: no articles found."); |
2c08214a AD |
343 | } |
344 | ||
345 | db_query($link, "UPDATE ttrss_feeds | |
346 | SET last_updated = NOW(), last_error = '' WHERE id = '$feed'"); | |
347 | ||
348 | return; // no articles | |
349 | } | |
350 | ||
351 | if ($pubsub_state != 2 && PUBSUBHUBBUB_ENABLED) { | |
352 | ||
353 | if ($debug_enabled) _debug("update_rss_feed: checking for PUSH hub..."); | |
354 | ||
355 | $feed_hub_url = false; | |
2c08214a | 356 | |
19b3992b | 357 | $links = $rss->get_links('hub'); |
2c08214a | 358 | |
19b3992b AD |
359 | if ($links && is_array($links)) { |
360 | foreach ($links as $l) { | |
361 | $feed_hub_url = $l; | |
362 | break; | |
2c08214a AD |
363 | } |
364 | } | |
365 | ||
366 | if ($debug_enabled) _debug("update_rss_feed: feed hub url: $feed_hub_url"); | |
367 | ||
368 | if ($feed_hub_url && function_exists('curl_init') && | |
369 | !ini_get("open_basedir")) { | |
370 | ||
371 | require_once 'lib/pubsubhubbub/subscriber.php'; | |
372 | ||
373 | $callback_url = get_self_url_prefix() . | |
374 | "/public.php?op=pubsub&id=$feed"; | |
375 | ||
376 | $s = new Subscriber($feed_hub_url, $callback_url); | |
377 | ||
378 | $rc = $s->subscribe($fetch_url); | |
379 | ||
380 | if ($debug_enabled) | |
381 | _debug("update_rss_feed: feed hub url found, subscribe request sent."); | |
382 | ||
383 | db_query($link, "UPDATE ttrss_feeds SET pubsub_state = 1 | |
384 | WHERE id = '$feed'"); | |
385 | } | |
386 | } | |
387 | ||
388 | if ($debug_enabled) { | |
389 | _debug("update_rss_feed: processing articles..."); | |
390 | } | |
391 | ||
19b3992b | 392 | foreach ($items as $item) { |
2c08214a AD |
393 | if ($_REQUEST['xdebug'] == 2) { |
394 | print_r($item); | |
395 | } | |
396 | ||
19b3992b AD |
397 | $entry_guid = $item->get_id(); |
398 | if (!$entry_guid) $entry_guid = $item->get_link(); | |
399 | if (!$entry_guid) $entry_guid = make_guid_from_title($item->get_title()); | |
2c08214a | 400 | |
765509c5 | 401 | if ($cache_content) { |
d5e9cf28 | 402 | $entry_guid = "ccache:$entry_guid"; |
765509c5 AD |
403 | } |
404 | ||
405 | if ($auth_login || $auth_pass) { | |
406 | $entry_guid = "auth,$owner_uid:$entry_guid"; | |
407 | } | |
408 | ||
2c08214a AD |
409 | if ($debug_enabled) { |
410 | _debug("update_rss_feed: guid $entry_guid"); | |
411 | } | |
412 | ||
413 | if (!$entry_guid) continue; | |
414 | ||
415 | $entry_timestamp = ""; | |
416 | ||
19b3992b | 417 | $entry_timestamp = strtotime($item->get_date()); |
2c08214a | 418 | |
19b3992b | 419 | if ($entry_timestamp == -1 || !$entry_timestamp) { |
2c08214a AD |
420 | $entry_timestamp = time(); |
421 | $no_orig_date = 'true'; | |
422 | } else { | |
423 | $no_orig_date = 'false'; | |
424 | } | |
425 | ||
426 | $entry_timestamp_fmt = strftime("%Y/%m/%d %H:%M:%S", $entry_timestamp); | |
427 | ||
428 | if ($debug_enabled) { | |
429 | _debug("update_rss_feed: date $entry_timestamp [$entry_timestamp_fmt]"); | |
430 | } | |
431 | ||
19b3992b | 432 | $entry_title = $item->get_title(); |
1b35d30c AD |
433 | |
434 | $entry_link = rewrite_relative_url($site_url, htmlspecialchars_decode($item->get_link())); | |
2c08214a AD |
435 | |
436 | if ($debug_enabled) { | |
437 | _debug("update_rss_feed: title $entry_title"); | |
438 | _debug("update_rss_feed: link $entry_link"); | |
439 | } | |
440 | ||
441 | if (!$entry_title) $entry_title = date("Y-m-d H:i:s", $entry_timestamp);; | |
442 | ||
19b3992b AD |
443 | $entry_content = $item->get_content(); |
444 | if (!$entry_content) $entry_content = $item->get_description(); | |
2c08214a | 445 | |
c5867798 | 446 | if ($cache_images && is_writable(CACHE_DIR . '/images')) |
3c696512 AD |
447 | $entry_content = cache_images($entry_content, $site_url, $debug_enabled); |
448 | ||
2c08214a AD |
449 | if ($_REQUEST["xdebug"] == 2) { |
450 | print "update_rss_feed: content: "; | |
487f0750 | 451 | print $entry_content; |
3c696512 | 452 | print "\n"; |
2c08214a AD |
453 | } |
454 | ||
87764a50 | 455 | $entry_cached_content = ""; |
2c08214a | 456 | |
19b3992b | 457 | $entry_comments = $item->data["comments"]; |
2c08214a | 458 | |
19b3992b AD |
459 | if ($item->get_author()) { |
460 | $entry_author_item = $item->get_author(); | |
461 | $entry_author = $entry_author_item->get_name(); | |
462 | if (!$entry_author) $entry_author = $entry_author_item->get_email(); | |
2c08214a | 463 | |
19b3992b | 464 | $entry_author = db_escape_string($entry_author); |
2c08214a AD |
465 | } |
466 | ||
19b3992b | 467 | $entry_guid = db_escape_string(mb_substr($entry_guid, 0, 250)); |
2c08214a AD |
468 | |
469 | $result = db_query($link, "SELECT id FROM ttrss_entries | |
470 | WHERE guid = '$entry_guid'"); | |
471 | ||
19b3992b AD |
472 | $entry_comments = db_escape_string(mb_substr($entry_comments, 0, 250)); |
473 | $entry_author = db_escape_string(mb_substr($entry_author, 0, 250)); | |
2c08214a | 474 | |
19b3992b | 475 | $num_comments = $item->get_item_tags('http://purl.org/rss/1.0/modules/slash/', 'comments'); |
83e6e313 | 476 | |
19b3992b AD |
477 | if (is_array($num_comments) && is_array($num_comments[0])) { |
478 | $num_comments = (int) $num_comments[0]["data"]; | |
2c08214a | 479 | } else { |
19b3992b | 480 | $num_comments = 0; |
2c08214a AD |
481 | } |
482 | ||
2c08214a | 483 | if ($debug_enabled) { |
83e6e313 | 484 | _debug("update_rss_feed: num_comments: $num_comments"); |
2c08214a AD |
485 | _debug("update_rss_feed: looking for tags [1]..."); |
486 | } | |
487 | ||
488 | // parse <category> entries into tags | |
489 | ||
490 | $additional_tags = array(); | |
491 | ||
19b3992b | 492 | $additional_tags_src = $item->get_categories(); |
2c08214a | 493 | |
19b3992b AD |
494 | if (is_array($additional_tags_src)) { |
495 | foreach ($additional_tags_src as $tobj) { | |
496 | array_push($additional_tags, $tobj->get_term()); | |
2c08214a | 497 | } |
19b3992b | 498 | } |
2c08214a | 499 | |
19b3992b AD |
500 | if ($debug_enabled) { |
501 | _debug("update_rss_feed: category tags:"); | |
502 | print_r($additional_tags); | |
2c08214a AD |
503 | } |
504 | ||
505 | if ($debug_enabled) { | |
506 | _debug("update_rss_feed: looking for tags [2]..."); | |
507 | } | |
508 | ||
509 | /* taaaags */ | |
510 | // <a href="..." rel="tag">Xorg</a>, // | |
511 | ||
512 | $entry_tags = null; | |
513 | ||
514 | preg_match_all("/<a.*?rel=['\"]tag['\"].*?\>([^<]+)<\/a>/i", | |
2bbd6994 | 515 | $entry_content, $entry_tags); |
2c08214a AD |
516 | |
517 | $entry_tags = $entry_tags[1]; | |
518 | ||
519 | $entry_tags = array_merge($entry_tags, $additional_tags); | |
520 | $entry_tags = array_unique($entry_tags); | |
521 | ||
522 | for ($i = 0; $i < count($entry_tags); $i++) | |
523 | $entry_tags[$i] = mb_strtolower($entry_tags[$i], 'utf-8'); | |
524 | ||
525 | if ($debug_enabled) { | |
6aff7845 AD |
526 | //_debug("update_rss_feed: unfiltered tags found:"); |
527 | //print_r($entry_tags); | |
2c08214a AD |
528 | } |
529 | ||
2c08214a AD |
530 | if ($debug_enabled) { |
531 | _debug("update_rss_feed: done collecting data [TITLE:$entry_title]"); | |
532 | } | |
533 | ||
cc85704f | 534 | // TODO: less memory-hungry implementation |
19c73507 AD |
535 | global $pluginhost; |
536 | ||
19b3992b AD |
537 | if ($debug_enabled) { |
538 | _debug("update_rss_feed: applying plugin filters.."); | |
539 | } | |
cc85704f | 540 | |
19b3992b AD |
541 | $article = array("owner_uid" => $owner_uid, |
542 | "title" => $entry_title, | |
543 | "content" => $entry_content, | |
544 | "link" => $entry_link, | |
545 | "tags" => $entry_tags, | |
546 | "author" => $entry_author); | |
cc85704f | 547 | |
19b3992b AD |
548 | foreach ($pluginhost->get_hooks($pluginhost::HOOK_ARTICLE_FILTER) as $plugin) { |
549 | $article = $plugin->hook_article_filter($article); | |
cc85704f AD |
550 | } |
551 | ||
19b3992b | 552 | $entry_tags = $article["tags"]; |
1b35d30c AD |
553 | $entry_content = db_escape_string($article["content"], false); |
554 | $entry_title = db_escape_string($article["title"]); | |
555 | $entry_author = db_escape_string($article["author"]); | |
556 | $entry_link = db_escape_string($article["link"]); | |
2bbd6994 | 557 | |
cc85704f AD |
558 | $content_hash = "SHA1:" . sha1(strip_tags($entry_content)); |
559 | ||
2c08214a AD |
560 | db_query($link, "BEGIN"); |
561 | ||
562 | if (db_num_rows($result) == 0) { | |
563 | ||
564 | if ($debug_enabled) { | |
565 | _debug("update_rss_feed: base guid not found"); | |
566 | } | |
567 | ||
87764a50 AD |
568 | if ($cache_content) { |
569 | if ($debug_enabled) { | |
8054439f | 570 | _debug("update_rss_feed: caching content (initial)..."); |
87764a50 AD |
571 | } |
572 | ||
573 | $entry_cached_content = cache_content($link, $entry_link, $auth_login, $auth_pass); | |
574 | ||
575 | if ($cache_images && is_writable(CACHE_DIR . '/images')) | |
576 | $entry_cached_content = cache_images($entry_cached_content, $site_url, $debug_enabled); | |
577 | ||
578 | $entry_cached_content = db_escape_string($entry_cached_content, false); | |
87764a50 AD |
579 | } |
580 | ||
2c08214a AD |
581 | // base post entry does not exist, create it |
582 | ||
583 | $result = db_query($link, | |
584 | "INSERT INTO ttrss_entries | |
585 | (title, | |
586 | guid, | |
587 | link, | |
588 | updated, | |
589 | content, | |
590 | content_hash, | |
87764a50 | 591 | cached_content, |
2c08214a AD |
592 | no_orig_date, |
593 | date_updated, | |
594 | date_entered, | |
595 | comments, | |
596 | num_comments, | |
597 | author) | |
598 | VALUES | |
599 | ('$entry_title', | |
600 | '$entry_guid', | |
601 | '$entry_link', | |
602 | '$entry_timestamp_fmt', | |
603 | '$entry_content', | |
604 | '$content_hash', | |
cb93a5de | 605 | '$entry_cached_content', |
2c08214a AD |
606 | $no_orig_date, |
607 | NOW(), | |
608 | NOW(), | |
609 | '$entry_comments', | |
610 | '$num_comments', | |
611 | '$entry_author')"); | |
e8291805 AD |
612 | |
613 | $article_labels = array(); | |
614 | ||
2c08214a AD |
615 | } else { |
616 | // we keep encountering the entry in feeds, so we need to | |
617 | // update date_updated column so that we don't get horrible | |
618 | // dupes when the entry gets purged and reinserted again e.g. | |
619 | // in the case of SLOW SLOW OMG SLOW updating feeds | |
620 | ||
621 | $base_entry_id = db_fetch_result($result, 0, "id"); | |
622 | ||
623 | db_query($link, "UPDATE ttrss_entries SET date_updated = NOW() | |
624 | WHERE id = '$base_entry_id'"); | |
e8291805 AD |
625 | |
626 | $article_labels = get_article_labels($link, $base_entry_id, $owner_uid); | |
2c08214a AD |
627 | } |
628 | ||
629 | // now it should exist, if not - bad luck then | |
630 | ||
631 | $result = db_query($link, "SELECT | |
632 | id,content_hash,no_orig_date,title, | |
633 | ".SUBSTRING_FOR_DATE."(date_updated,1,19) as date_updated, | |
634 | ".SUBSTRING_FOR_DATE."(updated,1,19) as updated, | |
130b0781 | 635 | num_comments, cached_content |
2c08214a AD |
636 | FROM |
637 | ttrss_entries | |
638 | WHERE guid = '$entry_guid'"); | |
639 | ||
640 | $entry_ref_id = 0; | |
641 | $entry_int_id = 0; | |
642 | ||
643 | if (db_num_rows($result) == 1) { | |
644 | ||
645 | if ($debug_enabled) { | |
646 | _debug("update_rss_feed: base guid found, checking for user record"); | |
647 | } | |
648 | ||
649 | // this will be used below in update handler | |
650 | $orig_content_hash = db_fetch_result($result, 0, "content_hash"); | |
651 | $orig_title = db_fetch_result($result, 0, "title"); | |
652 | $orig_num_comments = db_fetch_result($result, 0, "num_comments"); | |
130b0781 | 653 | $orig_cached_content = trim(db_fetch_result($result, 0, "cached_content")); |
2c08214a AD |
654 | $orig_date_updated = strtotime(db_fetch_result($result, |
655 | 0, "date_updated")); | |
656 | ||
657 | $ref_id = db_fetch_result($result, 0, "id"); | |
658 | $entry_ref_id = $ref_id; | |
659 | ||
660 | // check for user post link to main table | |
661 | ||
662 | // do we allow duplicate posts with same GUID in different feeds? | |
663 | if (get_pref($link, "ALLOW_DUPLICATE_POSTS", $owner_uid, false)) { | |
664 | $dupcheck_qpart = "AND (feed_id = '$feed' OR feed_id IS NULL)"; | |
665 | } else { | |
666 | $dupcheck_qpart = ""; | |
667 | } | |
668 | ||
669 | /* Collect article tags here so we could filter by them: */ | |
670 | ||
671 | $article_filters = get_article_filters($filters, $entry_title, | |
4021d61a | 672 | $entry_content, $entry_link, $entry_timestamp, $entry_author, |
2c08214a AD |
673 | $entry_tags); |
674 | ||
675 | if ($debug_enabled) { | |
676 | _debug("update_rss_feed: article filters: "); | |
677 | if (count($article_filters) != 0) { | |
678 | print_r($article_filters); | |
679 | } | |
680 | } | |
681 | ||
682 | if (find_article_filter($article_filters, "filter")) { | |
683 | db_query($link, "COMMIT"); // close transaction in progress | |
684 | continue; | |
685 | } | |
686 | ||
687 | $score = calculate_article_score($article_filters); | |
688 | ||
689 | if ($debug_enabled) { | |
690 | _debug("update_rss_feed: initial score: $score"); | |
691 | } | |
692 | ||
693 | $query = "SELECT ref_id, int_id FROM ttrss_user_entries WHERE | |
694 | ref_id = '$ref_id' AND owner_uid = '$owner_uid' | |
695 | $dupcheck_qpart"; | |
696 | ||
697 | // if ($_REQUEST["xdebug"]) print "$query\n"; | |
698 | ||
699 | $result = db_query($link, $query); | |
700 | ||
701 | // okay it doesn't exist - create user entry | |
702 | if (db_num_rows($result) == 0) { | |
703 | ||
704 | if ($debug_enabled) { | |
705 | _debug("update_rss_feed: user record not found, creating..."); | |
706 | } | |
707 | ||
708 | if ($score >= -500 && !find_article_filter($article_filters, 'catchup')) { | |
709 | $unread = 'true'; | |
710 | $last_read_qpart = 'NULL'; | |
711 | } else { | |
712 | $unread = 'false'; | |
713 | $last_read_qpart = 'NOW()'; | |
714 | } | |
715 | ||
716 | if (find_article_filter($article_filters, 'mark') || $score > 1000) { | |
717 | $marked = 'true'; | |
718 | } else { | |
719 | $marked = 'false'; | |
720 | } | |
721 | ||
722 | if (find_article_filter($article_filters, 'publish')) { | |
723 | $published = 'true'; | |
724 | } else { | |
725 | $published = 'false'; | |
726 | } | |
727 | ||
2ea9bbfd AD |
728 | // N-grams |
729 | ||
730 | if (DB_TYPE == "pgsql" and defined('_NGRAM_TITLE_DUPLICATE_THRESHOLD')) { | |
731 | ||
732 | $result = db_query($link, "SELECT COUNT(*) AS similar FROM | |
733 | ttrss_entries,ttrss_user_entries | |
734 | WHERE ref_id = id AND updated >= NOW() - INTERVAL '7 day' | |
735 | AND similarity(title, '$entry_title') >= "._NGRAM_TITLE_DUPLICATE_THRESHOLD." | |
736 | AND owner_uid = $owner_uid"); | |
737 | ||
738 | $ngram_similar = db_fetch_result($result, 0, "similar"); | |
739 | ||
740 | if ($debug_enabled) { | |
741 | _debug("update_rss_feed: N-gram similar results: $ngram_similar"); | |
742 | } | |
743 | ||
744 | if ($ngram_similar > 0) { | |
745 | $unread = 'false'; | |
746 | } | |
747 | } | |
748 | ||
2c08214a AD |
749 | $result = db_query($link, |
750 | "INSERT INTO ttrss_user_entries | |
751 | (ref_id, owner_uid, feed_id, unread, last_read, marked, | |
752 | published, score, tag_cache, label_cache, uuid) | |
753 | VALUES ('$ref_id', '$owner_uid', '$feed', $unread, | |
754 | $last_read_qpart, $marked, $published, '$score', '', '', '')"); | |
755 | ||
756 | if (PUBSUBHUBBUB_HUB && $published == 'true') { | |
757 | $rss_link = get_self_url_prefix() . | |
758 | "/public.php?op=rss&id=-2&key=" . | |
759 | get_feed_access_key($link, -2, false, $owner_uid); | |
760 | ||
761 | $p = new Publisher(PUBSUBHUBBUB_HUB); | |
762 | ||
763 | $pubsub_result = $p->publish_update($rss_link); | |
764 | } | |
765 | ||
766 | $result = db_query($link, | |
767 | "SELECT int_id FROM ttrss_user_entries WHERE | |
768 | ref_id = '$ref_id' AND owner_uid = '$owner_uid' AND | |
769 | feed_id = '$feed' LIMIT 1"); | |
770 | ||
771 | if (db_num_rows($result) == 1) { | |
772 | $entry_int_id = db_fetch_result($result, 0, "int_id"); | |
773 | } | |
774 | } else { | |
775 | if ($debug_enabled) { | |
776 | _debug("update_rss_feed: user record FOUND"); | |
777 | } | |
778 | ||
779 | $entry_ref_id = db_fetch_result($result, 0, "ref_id"); | |
780 | $entry_int_id = db_fetch_result($result, 0, "int_id"); | |
781 | } | |
782 | ||
783 | if ($debug_enabled) { | |
784 | _debug("update_rss_feed: RID: $entry_ref_id, IID: $entry_int_id"); | |
785 | } | |
786 | ||
787 | $post_needs_update = false; | |
788 | $update_insignificant = false; | |
130b0781 | 789 | $cached_content_needs_update = false; |
2c08214a AD |
790 | |
791 | if ($orig_num_comments != $num_comments) { | |
792 | $post_needs_update = true; | |
793 | $update_insignificant = true; | |
794 | } | |
795 | ||
796 | if ($content_hash != $orig_content_hash) { | |
797 | $post_needs_update = true; | |
798 | $update_insignificant = false; | |
130b0781 AD |
799 | $cached_content_needs_update = true; |
800 | } | |
87764a50 | 801 | |
130b0781 AD |
802 | if ($cache_content) { |
803 | if ($debug_enabled) { | |
804 | _debug("update_rss_feed: caching content because original checksum changed..."); | |
805 | } | |
87764a50 | 806 | |
130b0781 | 807 | $entry_cached_content = cache_content($link, $entry_link, $auth_login, $auth_pass); |
87764a50 | 808 | |
130b0781 | 809 | if ($entry_cached_content) { |
87764a50 AD |
810 | if ($cache_images && is_writable(CACHE_DIR . '/images')) |
811 | $entry_cached_content = cache_images($entry_cached_content, $site_url, $debug_enabled); | |
812 | ||
813 | $entry_cached_content = db_escape_string($entry_cached_content, false); | |
130b0781 AD |
814 | $post_needs_update = true; |
815 | } else { | |
816 | $entry_cached_content = db_escape_string($orig_cached_content); | |
87764a50 | 817 | } |
130b0781 AD |
818 | } else { |
819 | $entry_cached_content = db_escape_string($orig_cached_content); | |
2c08214a AD |
820 | } |
821 | ||
822 | if (db_escape_string($orig_title) != $entry_title) { | |
823 | $post_needs_update = true; | |
824 | $update_insignificant = false; | |
825 | } | |
826 | ||
827 | // if post needs update, update it and mark all user entries | |
828 | // linking to this post as updated | |
829 | if ($post_needs_update) { | |
830 | ||
831 | if (defined('DAEMON_EXTENDED_DEBUG')) { | |
832 | _debug("update_rss_feed: post $entry_guid needs update..."); | |
833 | } | |
834 | ||
835 | // print "<!-- post $orig_title needs update : $post_needs_update -->"; | |
836 | ||
837 | db_query($link, "UPDATE ttrss_entries | |
838 | SET title = '$entry_title', content = '$entry_content', | |
839 | content_hash = '$content_hash', | |
87764a50 | 840 | cached_content = '$entry_cached_content', |
2c08214a AD |
841 | updated = '$entry_timestamp_fmt', |
842 | num_comments = '$num_comments' | |
843 | WHERE id = '$ref_id'"); | |
844 | ||
845 | if (!$update_insignificant) { | |
846 | if ($mark_unread_on_update) { | |
847 | db_query($link, "UPDATE ttrss_user_entries | |
848 | SET last_read = null, unread = true WHERE ref_id = '$ref_id'"); | |
849 | } else if ($update_on_checksum_change) { | |
850 | db_query($link, "UPDATE ttrss_user_entries | |
851 | SET last_read = null WHERE ref_id = '$ref_id' | |
852 | AND unread = false"); | |
853 | } | |
854 | } | |
855 | } | |
856 | } | |
857 | ||
858 | db_query($link, "COMMIT"); | |
859 | ||
860 | if ($debug_enabled) { | |
861 | _debug("update_rss_feed: assigning labels..."); | |
862 | } | |
863 | ||
92c14e9d | 864 | assign_article_to_label_filters($link, $entry_ref_id, $article_filters, |
b24504b1 | 865 | $owner_uid, $article_labels); |
2c08214a AD |
866 | |
867 | if ($debug_enabled) { | |
868 | _debug("update_rss_feed: looking for enclosures..."); | |
869 | } | |
870 | ||
871 | // enclosures | |
872 | ||
873 | $enclosures = array(); | |
874 | ||
19b3992b | 875 | $encs = $item->get_enclosures(); |
2c08214a | 876 | |
19b3992b AD |
877 | if (is_array($encs)) { |
878 | foreach ($encs as $e) { | |
879 | $e_item = array( | |
880 | $e->link, $e->type, $e->length); | |
2c08214a | 881 | array_push($enclosures, $e_item); |
2c08214a AD |
882 | } |
883 | } | |
884 | ||
2c08214a AD |
885 | if ($debug_enabled) { |
886 | _debug("update_rss_feed: article enclosures:"); | |
887 | print_r($enclosures); | |
888 | } | |
889 | ||
890 | db_query($link, "BEGIN"); | |
891 | ||
892 | foreach ($enclosures as $enc) { | |
893 | $enc_url = db_escape_string($enc[0]); | |
894 | $enc_type = db_escape_string($enc[1]); | |
895 | $enc_dur = db_escape_string($enc[2]); | |
896 | ||
897 | $result = db_query($link, "SELECT id FROM ttrss_enclosures | |
898 | WHERE content_url = '$enc_url' AND post_id = '$entry_ref_id'"); | |
899 | ||
900 | if (db_num_rows($result) == 0) { | |
901 | db_query($link, "INSERT INTO ttrss_enclosures | |
902 | (content_url, content_type, title, duration, post_id) VALUES | |
903 | ('$enc_url', '$enc_type', '', '$enc_dur', '$entry_ref_id')"); | |
904 | } | |
905 | } | |
906 | ||
907 | db_query($link, "COMMIT"); | |
908 | ||
909 | // check for manual tags (we have to do it here since they're loaded from filters) | |
910 | ||
911 | foreach ($article_filters as $f) { | |
6aff7845 | 912 | if ($f["type"] == "tag") { |
2c08214a | 913 | |
6aff7845 | 914 | $manual_tags = trim_array(explode(",", $f["param"])); |
2c08214a AD |
915 | |
916 | foreach ($manual_tags as $tag) { | |
917 | if (tag_is_valid($tag)) { | |
918 | array_push($entry_tags, $tag); | |
919 | } | |
920 | } | |
921 | } | |
922 | } | |
923 | ||
924 | // Skip boring tags | |
925 | ||
926 | $boring_tags = trim_array(explode(",", mb_strtolower(get_pref($link, | |
927 | 'BLACKLISTED_TAGS', $owner_uid, ''), 'utf-8'))); | |
928 | ||
929 | $filtered_tags = array(); | |
930 | $tags_to_cache = array(); | |
931 | ||
932 | if ($entry_tags && is_array($entry_tags)) { | |
933 | foreach ($entry_tags as $tag) { | |
934 | if (array_search($tag, $boring_tags) === false) { | |
935 | array_push($filtered_tags, $tag); | |
936 | } | |
937 | } | |
938 | } | |
939 | ||
940 | $filtered_tags = array_unique($filtered_tags); | |
941 | ||
942 | if ($debug_enabled) { | |
943 | _debug("update_rss_feed: filtered article tags:"); | |
944 | print_r($filtered_tags); | |
945 | } | |
946 | ||
947 | // Save article tags in the database | |
948 | ||
949 | if (count($filtered_tags) > 0) { | |
950 | ||
951 | db_query($link, "BEGIN"); | |
952 | ||
953 | foreach ($filtered_tags as $tag) { | |
954 | ||
955 | $tag = sanitize_tag($tag); | |
956 | $tag = db_escape_string($tag); | |
957 | ||
958 | if (!tag_is_valid($tag)) continue; | |
959 | ||
960 | $result = db_query($link, "SELECT id FROM ttrss_tags | |
961 | WHERE tag_name = '$tag' AND post_int_id = '$entry_int_id' AND | |
962 | owner_uid = '$owner_uid' LIMIT 1"); | |
963 | ||
964 | if ($result && db_num_rows($result) == 0) { | |
965 | ||
966 | db_query($link, "INSERT INTO ttrss_tags | |
967 | (owner_uid,tag_name,post_int_id) | |
968 | VALUES ('$owner_uid','$tag', '$entry_int_id')"); | |
969 | } | |
970 | ||
971 | array_push($tags_to_cache, $tag); | |
972 | } | |
973 | ||
974 | /* update the cache */ | |
975 | ||
976 | $tags_to_cache = array_unique($tags_to_cache); | |
977 | ||
978 | $tags_str = db_escape_string(join(",", $tags_to_cache)); | |
979 | ||
980 | db_query($link, "UPDATE ttrss_user_entries | |
981 | SET tag_cache = '$tags_str' WHERE ref_id = '$entry_ref_id' | |
982 | AND owner_uid = $owner_uid"); | |
983 | ||
984 | db_query($link, "COMMIT"); | |
985 | } | |
986 | ||
b24504b1 AD |
987 | if (get_pref($link, "AUTO_ASSIGN_LABELS", $owner_uid, false)) { |
988 | if ($debug_enabled) { | |
989 | _debug("update_rss_feed: auto-assigning labels..."); | |
990 | } | |
991 | ||
992 | foreach ($labels as $label) { | |
993 | $caption = $label["caption"]; | |
994 | ||
3fc6e71a | 995 | if (preg_match("/\b$caption\b/i", "$tags_str " . strip_tags($entry_content) . " $entry_title")) { |
b24504b1 AD |
996 | if (!labels_contains_caption($article_labels, $caption)) { |
997 | label_add_article($link, $entry_ref_id, $caption, $owner_uid); | |
998 | } | |
999 | } | |
1000 | } | |
1001 | } | |
1002 | ||
2c08214a AD |
1003 | if ($debug_enabled) { |
1004 | _debug("update_rss_feed: article processed"); | |
1005 | } | |
1006 | } | |
1007 | ||
1008 | if (!$last_updated) { | |
1009 | if ($debug_enabled) { | |
1010 | _debug("update_rss_feed: new feed, catching it up..."); | |
1011 | } | |
1012 | catchup_feed($link, $feed, false, $owner_uid); | |
1013 | } | |
1014 | ||
1015 | if ($debug_enabled) { | |
1016 | _debug("purging feed..."); | |
1017 | } | |
1018 | ||
1019 | purge_feed($link, $feed, 0, $debug_enabled); | |
1020 | ||
1021 | db_query($link, "UPDATE ttrss_feeds | |
f2798eb6 | 1022 | SET last_updated = NOW(), last_error = '' WHERE id = '$feed'"); |
2c08214a AD |
1023 | |
1024 | // db_query($link, "COMMIT"); | |
1025 | ||
1026 | } else { | |
1027 | ||
19b3992b | 1028 | $error_msg = mb_substr($rss->error(), 0, 250); |
2c08214a AD |
1029 | |
1030 | if ($debug_enabled) { | |
1031 | _debug("update_rss_feed: error fetching feed: $error_msg"); | |
1032 | } | |
1033 | ||
1034 | $error_msg = db_escape_string($error_msg); | |
1035 | ||
1036 | db_query($link, | |
1037 | "UPDATE ttrss_feeds SET last_error = '$error_msg', | |
1038 | last_updated = NOW() WHERE id = '$feed'"); | |
1039 | } | |
1040 | ||
19b3992b | 1041 | unset($rss); |
2c08214a AD |
1042 | |
1043 | if ($debug_enabled) { | |
1044 | _debug("update_rss_feed: done"); | |
1045 | } | |
1046 | ||
1047 | } | |
1048 | ||
3c696512 AD |
1049 | function cache_images($html, $site_url, $debug) { |
1050 | $cache_dir = CACHE_DIR . "/images"; | |
1051 | ||
1052 | libxml_use_internal_errors(true); | |
1053 | ||
1054 | $charset_hack = '<head> | |
1055 | <meta http-equiv="Content-Type" content="text/html; charset=utf-8"/> | |
1056 | </head>'; | |
1057 | ||
1058 | $doc = new DOMDocument(); | |
1059 | $doc->loadHTML($charset_hack . $html); | |
1060 | $xpath = new DOMXPath($doc); | |
1061 | ||
1062 | $entries = $xpath->query('(//img[@src])'); | |
1063 | ||
1064 | foreach ($entries as $entry) { | |
1065 | if ($entry->hasAttribute('src')) { | |
1066 | $src = rewrite_relative_url($site_url, $entry->getAttribute('src')); | |
1067 | ||
1068 | $local_filename = CACHE_DIR . "/images/" . sha1($src) . ".png"; | |
1069 | ||
1070 | if ($debug) _debug("cache_images: downloading: $src to $local_filename"); | |
1071 | ||
1072 | if (!file_exists($local_filename)) { | |
1073 | $file_content = fetch_file_contents($src); | |
1074 | ||
b8379e69 | 1075 | if ($file_content && strlen($file_content) > 1024) { |
3c696512 AD |
1076 | file_put_contents($local_filename, $file_content); |
1077 | } | |
1078 | } | |
1079 | ||
1080 | if (file_exists($local_filename)) { | |
1081 | $entry->setAttribute('src', SELF_URL_PATH . '/image.php?url=' . | |
487f0750 | 1082 | base64_encode($src)); |
3c696512 AD |
1083 | } |
1084 | } | |
1085 | } | |
1086 | ||
1087 | $node = $doc->getElementsByTagName('body')->item(0); | |
1088 | ||
7b8ff151 | 1089 | return $doc->saveXML($node, LIBXML_NOEMPTYTAG); |
3c696512 AD |
1090 | } |
1091 | ||
2a91b6ff AD |
1092 | function expire_lock_files($debug) { |
1093 | if ($debug) _debug("Removing old lock files..."); | |
1094 | ||
1095 | $num_deleted = 0; | |
1096 | ||
1097 | if (is_writable(LOCK_DIRECTORY)) { | |
1098 | $files = glob(LOCK_DIRECTORY . "/*.lock"); | |
1099 | ||
1100 | if ($files) { | |
1101 | foreach ($files as $file) { | |
1102 | if (!file_is_locked($file) && time() - filemtime($file) > 86400*2) { | |
1103 | unlink($file); | |
1104 | ++$num_deleted; | |
1105 | } | |
1106 | } | |
1107 | } | |
1108 | } | |
1109 | ||
1110 | if ($debug) _debug("Removed $num_deleted files."); | |
1111 | } | |
1112 | ||
3c696512 | 1113 | function expire_cached_files($debug) { |
19b3992b | 1114 | foreach (array("simplepie", "images", "export") as $dir) { |
3c696512 | 1115 | $cache_dir = CACHE_DIR . "/$dir"; |
2c08214a | 1116 | |
3c696512 | 1117 | if ($debug) _debug("Expiring $cache_dir"); |
2c08214a | 1118 | |
3c696512 AD |
1119 | $num_deleted = 0; |
1120 | ||
1121 | if (is_writable($cache_dir)) { | |
1122 | $files = glob("$cache_dir/*"); | |
1123 | ||
2a91b6ff | 1124 | if ($files) { |
2ab20c31 AD |
1125 | foreach ($files as $file) { |
1126 | if (time() - filemtime($file) > 86400*7) { | |
1127 | unlink($file); | |
3c696512 | 1128 | |
2ab20c31 AD |
1129 | ++$num_deleted; |
1130 | } | |
3c696512 AD |
1131 | } |
1132 | } | |
2a91b6ff | 1133 | } |
3c696512 AD |
1134 | |
1135 | if ($debug) _debug("Removed $num_deleted files."); | |
1136 | } | |
1137 | } | |
2c08214a | 1138 | |
a3e0bdcf AD |
1139 | /** |
1140 | * Source: http://www.php.net/manual/en/function.parse-url.php#104527 | |
1141 | * Returns the url query as associative array | |
1142 | * | |
1143 | * @param string query | |
1144 | * @return array params | |
1145 | */ | |
1146 | function convertUrlQuery($query) { | |
1147 | $queryParts = explode('&', $query); | |
1148 | ||
1149 | $params = array(); | |
1150 | ||
1151 | foreach ($queryParts as $param) { | |
1152 | $item = explode('=', $param); | |
1153 | $params[$item[0]] = $item[1]; | |
1154 | } | |
1155 | ||
1156 | return $params; | |
1157 | } | |
92c14e9d AD |
1158 | |
1159 | function get_article_filters($filters, $title, $content, $link, $timestamp, $author, $tags) { | |
1160 | $matches = array(); | |
1161 | ||
1162 | foreach ($filters as $filter) { | |
1163 | $match_any_rule = $filter["match_any_rule"]; | |
1164 | $filter_match = false; | |
1165 | ||
1166 | foreach ($filter["rules"] as $rule) { | |
1167 | $match = false; | |
1168 | $reg_exp = $rule["reg_exp"]; | |
1169 | ||
1170 | if (!$reg_exp) | |
1171 | continue; | |
1172 | ||
1173 | switch ($rule["type"]) { | |
1174 | case "title": | |
1175 | $match = @preg_match("/$reg_exp/i", $title); | |
1176 | break; | |
1177 | case "content": | |
d03ae73e AD |
1178 | // we don't need to deal with multiline regexps |
1179 | $content = preg_replace("/[\r\n\t]/", "", $content); | |
1180 | ||
92c14e9d AD |
1181 | $match = @preg_match("/$reg_exp/i", $content); |
1182 | break; | |
1183 | case "both": | |
d03ae73e AD |
1184 | // we don't need to deal with multiline regexps |
1185 | $content = preg_replace("/[\r\n\t]/", "", $content); | |
1186 | ||
72d1d067 | 1187 | $match = (@preg_match("/$reg_exp/i", $title) || @preg_match("/$reg_exp/i", $content)); |
92c14e9d AD |
1188 | break; |
1189 | case "link": | |
1190 | $match = @preg_match("/$reg_exp/i", $link); | |
1191 | break; | |
1192 | case "author": | |
1193 | $match = @preg_match("/$reg_exp/i", $author); | |
1194 | break; | |
1195 | case "tag": | |
1196 | $tag_string = join(",", $tags); | |
1197 | $match = @preg_match("/$reg_exp/i", $tag_string); | |
1198 | break; | |
1199 | } | |
1200 | ||
1201 | if ($match_any_rule) { | |
1202 | if ($match) { | |
1203 | $filter_match = true; | |
1204 | break; | |
1205 | } | |
1206 | } else { | |
1207 | $filter_match = $match; | |
1208 | if (!$match) { | |
1209 | break; | |
1210 | } | |
1211 | } | |
1212 | } | |
1213 | ||
1214 | if ($filter_match) { | |
1215 | foreach ($filter["actions"] AS $action) { | |
1216 | array_push($matches, $action); | |
1217 | } | |
1218 | } | |
1219 | } | |
1220 | ||
1221 | return $matches; | |
1222 | } | |
1223 | ||
1224 | function find_article_filter($filters, $filter_name) { | |
1225 | foreach ($filters as $f) { | |
1226 | if ($f["type"] == $filter_name) { | |
1227 | return $f; | |
1228 | }; | |
1229 | } | |
1230 | return false; | |
1231 | } | |
1232 | ||
1233 | function find_article_filters($filters, $filter_name) { | |
1234 | $results = array(); | |
1235 | ||
1236 | foreach ($filters as $f) { | |
1237 | if ($f["type"] == $filter_name) { | |
1238 | array_push($results, $f); | |
1239 | }; | |
1240 | } | |
1241 | return $results; | |
1242 | } | |
1243 | ||
1244 | function calculate_article_score($filters) { | |
1245 | $score = 0; | |
1246 | ||
1247 | foreach ($filters as $f) { | |
1248 | if ($f["type"] == "score") { | |
1249 | $score += $f["param"]; | |
1250 | }; | |
1251 | } | |
1252 | return $score; | |
1253 | } | |
1254 | ||
b24504b1 AD |
1255 | function labels_contains_caption($labels, $caption) { |
1256 | foreach ($labels as $label) { | |
1257 | if ($label[1] == $caption) { | |
1258 | return true; | |
1259 | } | |
1260 | } | |
1261 | ||
1262 | return false; | |
1263 | } | |
1264 | ||
1265 | function assign_article_to_label_filters($link, $id, $filters, $owner_uid, $article_labels) { | |
92c14e9d AD |
1266 | foreach ($filters as $f) { |
1267 | if ($f["type"] == "label") { | |
b24504b1 AD |
1268 | if (!labels_contains_caption($article_labels, $f["param"])) { |
1269 | label_add_article($link, $id, $f["param"], $owner_uid); | |
1270 | } | |
1271 | } | |
92c14e9d AD |
1272 | } |
1273 | } | |
87764a50 AD |
1274 | |
1275 | function cache_content($link, $url, $login, $pass) { | |
1276 | ||
1277 | $content = fetch_file_contents($url, $login, $pass); | |
1278 | ||
1279 | if ($content) { | |
1280 | $doc = new DOMDocument(); | |
1281 | @$doc->loadHTML($content); | |
1282 | $xpath = new DOMXPath($doc); | |
1283 | ||
1284 | $node = $doc->getElementsByTagName('body')->item(0); | |
1285 | ||
1286 | if ($node) { | |
1287 | $content = $doc->saveXML($node, LIBXML_NOEMPTYTAG); | |
1288 | ||
1289 | return $content; | |
1290 | } | |
1291 | } | |
1292 | ||
1293 | return ""; | |
1294 | } | |
2c08214a | 1295 | ?> |