2011-12-13 12:06:25 +00:00
|
|
|
<?php
|
2017-05-05 15:10:07 +00:00
|
|
|
class RSSUtils {
|
|
|
|
static function calculate_article_hash($article, $pluginhost) {
|
2014-08-18 17:55:48 +00:00
|
|
|
$tmp = "";
|
|
|
|
|
2020-05-17 14:42:37 +00:00
|
|
|
$ignored_fields = [ "feed", "guid", "guid_hashed", "owner_uid", "force_catchup" ];
|
|
|
|
|
2014-08-18 17:55:48 +00:00
|
|
|
foreach ($article as $k => $v) {
|
2020-05-17 14:42:37 +00:00
|
|
|
if (in_array($k, $ignored_fields))
|
|
|
|
continue;
|
|
|
|
|
2014-08-18 17:55:48 +00:00
|
|
|
if ($k != "feed" && isset($v)) {
|
2021-02-05 20:41:32 +00:00
|
|
|
$x = strip_tags(
|
|
|
|
is_array($v) ? implode(",", array_keys($v)) : $v);
|
2015-12-05 00:10:57 +00:00
|
|
|
|
|
|
|
$tmp .= sha1("$k:" . sha1($x));
|
2014-08-18 17:55:48 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2014-08-18 19:11:29 +00:00
|
|
|
return sha1(implode(",", $pluginhost->get_plugin_names()) . $tmp);
|
2014-08-18 14:32:16 +00:00
|
|
|
}
|
|
|
|
|
2018-01-29 20:24:11 +00:00
|
|
|
// Strips utf8mb4 characters (i.e. emoji) for mysql
|
2021-03-06 12:19:31 +00:00
|
|
|
static function strip_utf8mb4(string $str) {
|
2018-01-30 07:44:31 +00:00
|
|
|
return preg_replace('/[\x{10000}-\x{10FFFF}]/u', "\xEF\xBF\xBD", $str);
|
|
|
|
}
|
2018-01-29 20:24:11 +00:00
|
|
|
|
2019-03-06 17:12:44 +00:00
|
|
|
static function cleanup_feed_browser() {
|
|
|
|
$pdo = Db::pdo();
|
|
|
|
$pdo->query("DELETE FROM ttrss_feedbrowser_cache");
|
|
|
|
}
|
|
|
|
|
2020-09-29 09:46:55 +00:00
|
|
|
static function cleanup_feed_icons() {
|
|
|
|
$pdo = Db::pdo();
|
|
|
|
$sth = $pdo->prepare("SELECT id FROM ttrss_feeds WHERE id = ?");
|
|
|
|
|
2021-02-22 19:35:27 +00:00
|
|
|
// check icon files once every Config::get(Config::CACHE_MAX_DAYS) days
|
|
|
|
$icon_files = array_filter(glob(Config::get(Config::ICONS_DIR) . "/*.ico"),
|
|
|
|
function($f) { return filemtime($f) < time() - 86400 * Config::get(Config::CACHE_MAX_DAYS); });
|
2020-09-29 09:46:55 +00:00
|
|
|
|
|
|
|
foreach ($icon_files as $icon) {
|
|
|
|
$feed_id = basename($icon, ".ico");
|
|
|
|
|
|
|
|
$sth->execute([$feed_id]);
|
|
|
|
|
|
|
|
if ($sth->fetch()) {
|
|
|
|
@touch($icon);
|
|
|
|
} else {
|
|
|
|
Debug::log("Removing orphaned feed icon: $icon");
|
|
|
|
unlink($icon);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2021-03-06 12:19:31 +00:00
|
|
|
static function update_daemon_common(int $limit = 0, array $options = []) {
|
2021-02-22 19:35:27 +00:00
|
|
|
if (!$limit) $limit = Config::get(Config::DAEMON_FEED_LIMIT);
|
|
|
|
|
2021-03-04 06:22:24 +00:00
|
|
|
if (Config::get_schema_version() != Config::SCHEMA_VERSION) {
|
2013-04-17 08:10:35 +00:00
|
|
|
die("Schema version is wrong, please upgrade the database.\n");
|
|
|
|
}
|
|
|
|
|
2017-12-02 05:38:57 +00:00
|
|
|
$pdo = Db::pdo();
|
|
|
|
|
2021-02-22 19:35:27 +00:00
|
|
|
if (!Config::get(Config::SINGLE_USER_MODE) && Config::get(Config::DAEMON_UPDATE_LOGIN_LIMIT) > 0) {
|
2021-02-25 11:20:54 +00:00
|
|
|
$login_limit = (int) Config::get(Config::DAEMON_UPDATE_LOGIN_LIMIT);
|
|
|
|
|
2021-02-22 18:47:48 +00:00
|
|
|
if (Config::get(Config::DB_TYPE) == "pgsql") {
|
2021-02-25 11:20:54 +00:00
|
|
|
$login_thresh_qpart = "AND last_login >= NOW() - INTERVAL '$login_limit days'";
|
2011-12-13 12:06:25 +00:00
|
|
|
} else {
|
2021-02-25 11:20:54 +00:00
|
|
|
$login_thresh_qpart = "AND last_login >= DATE_SUB(NOW(), INTERVAL $login_limit DAY)";
|
2011-12-13 12:06:25 +00:00
|
|
|
}
|
|
|
|
} else {
|
|
|
|
$login_thresh_qpart = "";
|
|
|
|
}
|
|
|
|
|
2021-02-25 11:20:54 +00:00
|
|
|
$default_interval = (int) Prefs::get_default(Prefs::DEFAULT_UPDATE_INTERVAL);
|
|
|
|
|
2021-02-22 18:47:48 +00:00
|
|
|
if (Config::get(Config::DB_TYPE) == "pgsql") {
|
2011-12-13 12:06:25 +00:00
|
|
|
$update_limit_qpart = "AND ((
|
2021-02-25 11:20:54 +00:00
|
|
|
update_interval = 0
|
2021-02-26 18:48:20 +00:00
|
|
|
AND (p.value IS NULL OR p.value != '-1')
|
2021-02-25 11:20:54 +00:00
|
|
|
AND last_updated < NOW() - CAST((COALESCE(p.value, '$default_interval') || ' minutes') AS INTERVAL)
|
2011-12-13 12:06:25 +00:00
|
|
|
) OR (
|
2021-02-25 11:20:54 +00:00
|
|
|
update_interval > 0
|
|
|
|
AND last_updated < NOW() - CAST((update_interval || ' minutes') AS INTERVAL)
|
|
|
|
) OR (
|
|
|
|
update_interval >= 0
|
2021-02-26 18:48:20 +00:00
|
|
|
AND (p.value IS NULL OR p.value != '-1')
|
2021-02-25 11:20:54 +00:00
|
|
|
AND (last_updated = '1970-01-01 00:00:00' OR last_updated IS NULL)
|
|
|
|
))";
|
2011-12-13 12:06:25 +00:00
|
|
|
} else {
|
|
|
|
$update_limit_qpart = "AND ((
|
2021-02-25 11:20:54 +00:00
|
|
|
update_interval = 0
|
2021-02-26 18:48:20 +00:00
|
|
|
AND (p.value IS NULL OR p.value != '-1')
|
2021-02-25 11:20:54 +00:00
|
|
|
AND last_updated < DATE_SUB(NOW(), INTERVAL CONVERT(COALESCE(p.value, '$default_interval'), SIGNED INTEGER) MINUTE)
|
|
|
|
) OR (
|
|
|
|
update_interval > 0
|
|
|
|
AND last_updated < DATE_SUB(NOW(), INTERVAL update_interval MINUTE)
|
2011-12-13 12:06:25 +00:00
|
|
|
) OR (
|
2021-02-25 11:20:54 +00:00
|
|
|
update_interval >= 0
|
2021-02-26 18:48:20 +00:00
|
|
|
AND (p.value IS NULL OR p.value != '-1')
|
2021-02-25 11:20:54 +00:00
|
|
|
AND (last_updated = '1970-01-01 00:00:00' OR last_updated IS NULL)
|
|
|
|
))";
|
2011-12-13 12:06:25 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
// Test if feed is currently being updated by another process.
|
2021-02-22 18:47:48 +00:00
|
|
|
if (Config::get(Config::DB_TYPE) == "pgsql") {
|
2020-09-22 17:33:51 +00:00
|
|
|
$updstart_thresh_qpart = "AND (last_update_started IS NULL OR last_update_started < NOW() - INTERVAL '10 minutes')";
|
2011-12-13 12:06:25 +00:00
|
|
|
} else {
|
2020-09-22 17:33:51 +00:00
|
|
|
$updstart_thresh_qpart = "AND (last_update_started IS NULL OR last_update_started < DATE_SUB(NOW(), INTERVAL 10 MINUTE))";
|
2011-12-13 12:06:25 +00:00
|
|
|
}
|
|
|
|
|
2017-06-13 17:09:33 +00:00
|
|
|
$query_limit = $limit ? sprintf("LIMIT %d", $limit) : "";
|
2011-12-13 12:06:25 +00:00
|
|
|
|
2015-12-31 11:54:52 +00:00
|
|
|
// Update the least recently updated feeds first
|
|
|
|
$query_order = "ORDER BY last_updated";
|
|
|
|
|
2021-02-25 11:20:54 +00:00
|
|
|
if (Config::get(Config::DB_TYPE) == "pgsql")
|
|
|
|
$query_order .= " NULLS FIRST";
|
|
|
|
|
|
|
|
$query = "SELECT f.feed_url, f.last_updated
|
2011-12-13 12:06:25 +00:00
|
|
|
FROM
|
2021-02-25 11:20:54 +00:00
|
|
|
ttrss_feeds f, ttrss_users u LEFT JOIN ttrss_user_prefs2 p ON
|
|
|
|
(p.owner_uid = u.id AND profile IS NULL AND pref_name = 'DEFAULT_UPDATE_INTERVAL')
|
2013-05-02 06:26:32 +00:00
|
|
|
WHERE
|
2021-11-10 17:44:51 +00:00
|
|
|
f.owner_uid = u.id AND
|
|
|
|
u.access_level NOT IN (".sprintf("%d, %d", UserHelper::ACCESS_LEVEL_DISABLED, UserHelper::ACCESS_LEVEL_READONLY).")
|
2021-02-25 11:20:54 +00:00
|
|
|
$login_thresh_qpart
|
|
|
|
$update_limit_qpart
|
2013-03-29 08:21:40 +00:00
|
|
|
$updstart_thresh_qpart
|
2015-12-31 11:54:52 +00:00
|
|
|
$query_order $query_limit";
|
2013-04-24 09:43:40 +00:00
|
|
|
|
2021-02-25 11:20:54 +00:00
|
|
|
//print "$query\n";
|
|
|
|
|
2017-12-02 05:38:57 +00:00
|
|
|
$res = $pdo->query($query);
|
2011-12-13 12:06:25 +00:00
|
|
|
|
|
|
|
$feeds_to_update = array();
|
2017-12-02 05:38:57 +00:00
|
|
|
while ($line = $res->fetch()) {
|
2017-06-13 17:09:33 +00:00
|
|
|
array_push($feeds_to_update, $line['feed_url']);
|
2011-12-13 12:06:25 +00:00
|
|
|
}
|
|
|
|
|
2018-11-30 05:34:29 +00:00
|
|
|
Debug::log(sprintf("Scheduled %d feeds to update...", count($feeds_to_update)));
|
2017-12-02 05:38:57 +00:00
|
|
|
|
2017-06-13 17:09:33 +00:00
|
|
|
// Update last_update_started before actually starting the batch
|
|
|
|
// in order to minimize collision risk for parallel daemon tasks
|
|
|
|
if (count($feeds_to_update) > 0) {
|
2017-12-02 05:38:57 +00:00
|
|
|
$feeds_qmarks = arr_qmarks($feeds_to_update);
|
2013-03-29 08:21:40 +00:00
|
|
|
|
2017-12-02 05:38:57 +00:00
|
|
|
$tmph = $pdo->prepare("UPDATE ttrss_feeds SET last_update_started = NOW()
|
|
|
|
WHERE feed_url IN ($feeds_qmarks)");
|
|
|
|
$tmph->execute($feeds_to_update);
|
2011-12-13 12:06:25 +00:00
|
|
|
}
|
|
|
|
|
2013-03-30 12:23:34 +00:00
|
|
|
$nf = 0;
|
2014-08-19 06:32:01 +00:00
|
|
|
$bstarted = microtime(true);
|
2013-03-30 12:23:34 +00:00
|
|
|
|
2021-02-25 11:20:54 +00:00
|
|
|
$batch_owners = [];
|
2016-08-11 12:01:01 +00:00
|
|
|
|
2021-02-25 11:20:54 +00:00
|
|
|
$user_query = "SELECT f.id,
|
2020-09-28 05:02:38 +00:00
|
|
|
last_updated,
|
2021-02-25 11:20:54 +00:00
|
|
|
f.owner_uid,
|
2021-02-25 11:23:56 +00:00
|
|
|
u.login AS owner,
|
2021-02-25 11:20:54 +00:00
|
|
|
f.title
|
|
|
|
FROM ttrss_feeds f, ttrss_users u LEFT JOIN ttrss_user_prefs2 p ON
|
|
|
|
(p.owner_uid = u.id AND profile IS NULL AND pref_name = 'DEFAULT_UPDATE_INTERVAL')
|
|
|
|
WHERE
|
2021-11-10 17:44:51 +00:00
|
|
|
f.owner_uid = u.id AND
|
|
|
|
u.access_level NOT IN (".sprintf("%d, %d", UserHelper::ACCESS_LEVEL_DISABLED, UserHelper::ACCESS_LEVEL_READONLY).")
|
2021-02-25 11:20:54 +00:00
|
|
|
AND feed_url = :feed
|
2013-03-29 08:21:40 +00:00
|
|
|
$login_thresh_qpart
|
2021-02-25 11:20:54 +00:00
|
|
|
$update_limit_qpart
|
|
|
|
ORDER BY f.id $query_limit";
|
|
|
|
|
|
|
|
//print "$user_query\n";
|
|
|
|
|
|
|
|
// since we have feed xml cached, we can deal with other feeds with the same url
|
|
|
|
$usth = $pdo->prepare($user_query);
|
2013-03-29 08:21:40 +00:00
|
|
|
|
2017-12-02 05:38:57 +00:00
|
|
|
foreach ($feeds_to_update as $feed) {
|
2018-11-30 05:34:29 +00:00
|
|
|
Debug::log("Base feed: $feed");
|
2017-12-02 05:38:57 +00:00
|
|
|
|
2021-02-25 11:20:54 +00:00
|
|
|
$usth->execute(["feed" => $feed]);
|
2017-12-02 05:38:57 +00:00
|
|
|
|
|
|
|
if ($tline = $usth->fetch()) {
|
2021-02-25 11:23:56 +00:00
|
|
|
Debug::log(sprintf("=> %s (ID: %d, U: %s [%d]), last updated: %s", $tline["title"], $tline["id"],
|
|
|
|
$tline["owner"], $tline["owner_uid"],
|
2020-09-28 05:02:38 +00:00
|
|
|
$tline["last_updated"] ? $tline["last_updated"] : "never"));
|
2013-12-18 09:55:50 +00:00
|
|
|
|
2021-02-25 11:20:54 +00:00
|
|
|
if (!in_array($tline["owner_uid"], $batch_owners))
|
2017-12-02 05:38:57 +00:00
|
|
|
array_push($batch_owners, $tline["owner_uid"]);
|
2016-08-11 12:01:01 +00:00
|
|
|
|
2017-12-02 05:38:57 +00:00
|
|
|
$fstarted = microtime(true);
|
2018-09-03 10:03:35 +00:00
|
|
|
|
2020-09-27 12:58:13 +00:00
|
|
|
$quiet = (isset($options["quiet"])) ? "--quiet" : "";
|
|
|
|
$log = function_exists("flock") && isset($options['log']) ? '--log '.$options['log'] : '';
|
|
|
|
$log_level = isset($options['log-level']) ? '--log-level '.$options['log-level'] : '';
|
|
|
|
|
2020-09-28 18:19:53 +00:00
|
|
|
/* shared hosting may have this disabled and it's not strictly required */
|
|
|
|
if (self::function_enabled('passthru')) {
|
|
|
|
$exit_code = 0;
|
2020-09-27 12:58:13 +00:00
|
|
|
|
2021-02-22 18:47:48 +00:00
|
|
|
passthru(Config::get(Config::PHP_EXECUTABLE) . " update.php --update-feed " . $tline["id"] . " --pidlock feed-" . $tline["id"] . " $quiet $log $log_level", $exit_code);
|
2018-11-03 12:00:58 +00:00
|
|
|
|
2020-09-28 18:19:53 +00:00
|
|
|
Debug::log(sprintf("<= %.4f (sec) exit code: %d", microtime(true) - $fstarted, $exit_code));
|
2014-08-19 06:32:01 +00:00
|
|
|
|
2020-09-28 18:19:53 +00:00
|
|
|
// -1 can be caused by a SIGCHLD handler which daemon master process installs (not every setup, apparently)
|
|
|
|
if ($exit_code != 0 && $exit_code != -1) {
|
2020-10-18 13:12:58 +00:00
|
|
|
$festh = $pdo->prepare("SELECT last_error FROM ttrss_feeds WHERE id = ?");
|
|
|
|
$festh->execute([$tline["id"]]);
|
2020-09-27 13:42:42 +00:00
|
|
|
|
2020-10-18 13:12:58 +00:00
|
|
|
if ($ferow = $festh->fetch()) {
|
|
|
|
$error_message = $ferow["last_error"];
|
2020-09-28 18:19:53 +00:00
|
|
|
} else {
|
|
|
|
$error_message = "N/A";
|
|
|
|
}
|
2020-09-28 05:02:38 +00:00
|
|
|
|
2020-09-28 18:19:53 +00:00
|
|
|
Debug::log("!! Last error: $error_message");
|
|
|
|
|
2021-02-25 12:49:30 +00:00
|
|
|
Logger::log(E_USER_NOTICE,
|
2020-09-28 18:19:53 +00:00
|
|
|
sprintf("Update process for feed %d (%s, owner UID: %d) failed with exit code: %d (%s).",
|
|
|
|
$tline["id"], clean($tline["title"]), $tline["owner_uid"], $exit_code, clean($error_message)));
|
2020-10-18 13:12:58 +00:00
|
|
|
|
|
|
|
$combined_error_message = sprintf("Update process failed with exit code: %d (%s)",
|
|
|
|
$exit_code, clean($error_message));
|
|
|
|
|
|
|
|
# mark failed feed as having an update error (unless it is already marked)
|
|
|
|
$fusth = $pdo->prepare("UPDATE ttrss_feeds SET last_error = ? WHERE id = ? AND last_error = ''");
|
|
|
|
$fusth->execute([$combined_error_message, $tline["id"]]);
|
2020-09-28 05:02:38 +00:00
|
|
|
}
|
|
|
|
|
2020-09-28 18:19:53 +00:00
|
|
|
} else {
|
|
|
|
try {
|
|
|
|
if (!self::update_rss_feed($tline["id"], true)) {
|
2021-02-25 12:49:30 +00:00
|
|
|
Logger::log(E_USER_NOTICE,
|
2020-09-28 18:19:53 +00:00
|
|
|
sprintf("Update request for feed %d (%s, owner UID: %d) failed: %s.",
|
2021-02-28 07:12:57 +00:00
|
|
|
$tline["id"], clean($tline["title"]), $tline["owner_uid"], clean(UrlHelper::$fetch_last_error)));
|
2020-09-28 18:19:53 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
Debug::log(sprintf("<= %.4f (sec) (not using a separate process)", microtime(true) - $fstarted));
|
|
|
|
|
|
|
|
} catch (PDOException $e) {
|
2021-02-25 12:49:30 +00:00
|
|
|
Logger::log_error(E_USER_WARNING, $e->getMessage(), $e->getFile(), $e->getLine(), $e->getTraceAsString());
|
2020-09-28 05:02:38 +00:00
|
|
|
|
2020-09-28 18:19:53 +00:00
|
|
|
try {
|
|
|
|
$pdo->rollback();
|
|
|
|
} catch (PDOException $e) {
|
|
|
|
// it doesn't matter if there wasn't actually anything to rollback, PDO Exception can be
|
|
|
|
// thrown outside of an active transaction during feed update
|
|
|
|
}
|
|
|
|
}
|
2020-09-27 13:42:42 +00:00
|
|
|
}
|
2014-08-19 06:32:01 +00:00
|
|
|
|
2017-12-02 05:38:57 +00:00
|
|
|
++$nf;
|
2013-03-29 08:21:40 +00:00
|
|
|
}
|
2011-12-13 12:06:25 +00:00
|
|
|
}
|
|
|
|
|
2014-08-19 06:32:01 +00:00
|
|
|
if ($nf > 0) {
|
2018-11-30 05:34:29 +00:00
|
|
|
Debug::log(sprintf("Processed %d feeds in %.4f (sec), %.4f (sec/feed avg)", $nf,
|
2014-08-19 06:32:01 +00:00
|
|
|
microtime(true) - $bstarted, (microtime(true) - $bstarted) / $nf));
|
|
|
|
}
|
|
|
|
|
2016-08-11 12:01:01 +00:00
|
|
|
foreach ($batch_owners as $owner_uid) {
|
2018-11-30 05:34:29 +00:00
|
|
|
Debug::log("Running housekeeping tasks for user $owner_uid...");
|
2016-08-11 12:01:01 +00:00
|
|
|
|
2020-09-22 11:54:15 +00:00
|
|
|
self::housekeeping_user($owner_uid);
|
2016-08-11 12:01:01 +00:00
|
|
|
}
|
|
|
|
|
2011-12-13 12:06:25 +00:00
|
|
|
// Send feed digests by email if needed.
|
2018-11-30 05:34:29 +00:00
|
|
|
Digest::send_headlines_digests();
|
2011-12-13 12:06:25 +00:00
|
|
|
|
2013-03-30 12:23:34 +00:00
|
|
|
return $nf;
|
2017-04-26 12:29:22 +00:00
|
|
|
}
|
2011-12-13 12:06:25 +00:00
|
|
|
|
2021-03-17 01:36:58 +00:00
|
|
|
/** this is used when subscribing */
|
|
|
|
static function update_basic_info(int $feed_id) {
|
|
|
|
$feed = ORM::for_table('ttrss_feeds')
|
|
|
|
->select_many('id', 'owner_uid', 'feed_url', 'auth_pass', 'auth_login', 'title', 'site_url')
|
|
|
|
->find_one($feed_id);
|
2015-06-17 20:11:47 +00:00
|
|
|
|
2021-03-17 01:36:58 +00:00
|
|
|
if ($feed) {
|
2017-12-02 06:55:53 +00:00
|
|
|
$pluginhost = new PluginHost();
|
2021-03-17 01:36:58 +00:00
|
|
|
$user_plugins = get_pref(Prefs::_ENABLED_PLUGINS, $feed->owner_uid);
|
2015-06-17 20:11:47 +00:00
|
|
|
|
2021-02-22 18:47:48 +00:00
|
|
|
$pluginhost->load(Config::get(Config::PLUGINS), PluginHost::KIND_ALL);
|
2021-03-17 01:36:58 +00:00
|
|
|
$pluginhost->load((string)$user_plugins, PluginHost::KIND_USER, $feed->owner_uid);
|
2021-01-15 05:32:06 +00:00
|
|
|
//$pluginhost->load_data();
|
2017-12-02 06:55:53 +00:00
|
|
|
|
2021-02-08 13:14:48 +00:00
|
|
|
$basic_info = [];
|
|
|
|
|
|
|
|
$pluginhost->run_hooks_callback(PluginHost::HOOK_FEED_BASIC_INFO, function ($result) use (&$basic_info) {
|
|
|
|
$basic_info = $result;
|
2021-03-17 01:36:58 +00:00
|
|
|
}, $basic_info, $feed->feed_url, $feed->owner_uid, $feed_id, $feed->auth_login, $feed->auth_pass);
|
2015-06-17 20:11:47 +00:00
|
|
|
|
2017-12-02 06:55:53 +00:00
|
|
|
if (!$basic_info) {
|
2021-03-17 01:36:58 +00:00
|
|
|
$feed_data = UrlHelper::fetch([
|
|
|
|
'url' => $feed->feed_url,
|
|
|
|
'login' => $feed->auth_login,
|
|
|
|
'pass' => $feed->auth_pass,
|
|
|
|
'timeout' => Config::get(Config::FEED_FETCH_TIMEOUT),
|
|
|
|
]);
|
2017-09-25 00:37:49 +00:00
|
|
|
|
2017-12-02 06:55:53 +00:00
|
|
|
$feed_data = trim($feed_data);
|
2015-06-17 20:11:47 +00:00
|
|
|
|
2017-12-02 06:55:53 +00:00
|
|
|
$rss = new FeedParser($feed_data);
|
|
|
|
$rss->init();
|
2015-06-17 20:11:47 +00:00
|
|
|
|
2017-12-02 06:55:53 +00:00
|
|
|
if (!$rss->error()) {
|
2021-03-17 01:36:58 +00:00
|
|
|
$basic_info = [
|
2018-12-26 06:58:28 +00:00
|
|
|
'title' => mb_substr(clean($rss->get_title()), 0, 199),
|
2021-03-17 01:36:58 +00:00
|
|
|
'site_url' => mb_substr(UrlHelper::rewrite_relative($feed->feed_url, clean($rss->get_link())), 0, 245),
|
|
|
|
];
|
2017-12-02 06:55:53 +00:00
|
|
|
}
|
2017-09-27 01:37:05 +00:00
|
|
|
}
|
2015-06-17 20:11:47 +00:00
|
|
|
|
2017-12-02 06:55:53 +00:00
|
|
|
if ($basic_info && is_array($basic_info)) {
|
2021-03-17 01:36:58 +00:00
|
|
|
if (!empty($basic_info['title']) && (!$feed->title || $feed->title == '[Unknown]')) {
|
|
|
|
$feed->title = $basic_info['title'];
|
|
|
|
}
|
2017-12-02 06:55:53 +00:00
|
|
|
|
2021-03-17 01:36:58 +00:00
|
|
|
if (!empty($basic_info['site_url']) && $feed->site_url != $basic_info['site_url']) {
|
|
|
|
$feed->site_url = $basic_info['site_url'];
|
2017-12-02 06:55:53 +00:00
|
|
|
}
|
2021-03-17 01:36:58 +00:00
|
|
|
|
|
|
|
$feed->save();
|
2015-06-17 20:11:47 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2021-03-06 12:19:31 +00:00
|
|
|
static function update_rss_feed(int $feed, bool $no_cache = false) : bool {
|
2011-12-13 12:06:25 +00:00
|
|
|
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log("start", Debug::LOG_VERBOSE);
|
2011-12-13 12:06:25 +00:00
|
|
|
|
2017-12-02 06:55:53 +00:00
|
|
|
$pdo = Db::pdo();
|
|
|
|
|
2021-03-06 08:17:15 +00:00
|
|
|
if (Config::get(Config::DB_TYPE) == "pgsql") {
|
|
|
|
$favicon_interval_qpart = "favicon_last_checked < NOW() - INTERVAL '12 hour'";
|
|
|
|
} else {
|
|
|
|
$favicon_interval_qpart = "favicon_last_checked < DATE_SUB(NOW(), INTERVAL 12 HOUR)";
|
2016-11-30 11:03:21 +00:00
|
|
|
}
|
|
|
|
|
2021-03-06 08:17:15 +00:00
|
|
|
$feed_obj = ORM::for_table('ttrss_feeds')
|
|
|
|
->select_expr("ttrss_feeds.*,
|
|
|
|
".SUBSTRING_FOR_DATE."(last_unconditional, 1, 19) AS last_unconditional,
|
|
|
|
(favicon_is_custom IS NOT TRUE AND
|
|
|
|
(favicon_last_checked IS NULL OR $favicon_interval_qpart)) AS favicon_needs_check")
|
|
|
|
->find_one($feed);
|
|
|
|
|
|
|
|
if ($feed_obj) {
|
|
|
|
$feed_obj->last_update_started = Db::NOW();
|
|
|
|
$feed_obj->save();
|
|
|
|
|
|
|
|
$feed_language = mb_strtolower($feed_obj->feed_language);
|
|
|
|
|
|
|
|
if (!$feed_language) $feed_language = mb_strtolower(get_pref(Prefs::DEFAULT_SEARCH_LANGUAGE, $feed_obj->owner_uid));
|
|
|
|
if (!$feed_language) $feed_language = 'simple';
|
|
|
|
|
2021-11-10 17:44:51 +00:00
|
|
|
$user = ORM::for_table('ttrss_users')->find_one($feed_obj->owner_uid);
|
|
|
|
|
|
|
|
if ($user) {
|
|
|
|
if ($user->access_level == UserHelper::ACCESS_LEVEL_READONLY) {
|
|
|
|
Debug::log("error: denied update for $feed: permission denied by owner access level");
|
|
|
|
return false;
|
|
|
|
}
|
|
|
|
} else {
|
|
|
|
// this would indicate database corruption of some kind
|
|
|
|
Debug::log("error: owner not found for feed: $feed");
|
|
|
|
return false;
|
|
|
|
}
|
|
|
|
|
2021-03-06 08:17:15 +00:00
|
|
|
} else {
|
|
|
|
Debug::log("error: feeds table record not found for feed: $feed");
|
2021-03-06 12:19:31 +00:00
|
|
|
return false;
|
2021-03-06 08:17:15 +00:00
|
|
|
}
|
2015-07-04 16:33:46 +00:00
|
|
|
|
|
|
|
// feed was batch-subscribed or something, we need to get basic info
|
|
|
|
// this is not optimal currently as it fetches stuff separately TODO: optimize
|
2021-03-06 08:17:15 +00:00
|
|
|
if ($feed_obj->title == "[Unknown]" || empty($feed_obj->title) || empty($feed_obj->site_url)) {
|
|
|
|
Debug::log("setting basic feed info for $feed...");
|
2021-03-06 12:19:31 +00:00
|
|
|
self::update_basic_info($feed);
|
2015-07-04 16:33:46 +00:00
|
|
|
}
|
|
|
|
|
2013-03-29 07:31:10 +00:00
|
|
|
$date_feed_processed = date('Y-m-d H:i');
|
2011-12-13 12:06:25 +00:00
|
|
|
|
2021-03-06 08:17:15 +00:00
|
|
|
$cache_filename = Config::get(Config::CACHE_DIR) . "/feeds/" . sha1($feed_obj->feed_url) . ".xml";
|
2013-03-29 07:31:10 +00:00
|
|
|
|
2013-05-20 11:26:53 +00:00
|
|
|
$pluginhost = new PluginHost();
|
2021-03-06 08:17:15 +00:00
|
|
|
$user_plugins = get_pref(Prefs::_ENABLED_PLUGINS, $feed_obj->owner_uid);
|
2013-05-20 11:26:53 +00:00
|
|
|
|
2021-02-22 18:47:48 +00:00
|
|
|
$pluginhost->load(Config::get(Config::PLUGINS), PluginHost::KIND_ALL);
|
2021-03-06 08:17:15 +00:00
|
|
|
$pluginhost->load((string)$user_plugins, PluginHost::KIND_USER, $feed_obj->owner_uid);
|
2013-05-20 11:26:53 +00:00
|
|
|
|
2017-04-26 12:29:22 +00:00
|
|
|
$rss_hash = false;
|
2013-03-29 07:50:58 +00:00
|
|
|
|
2017-04-26 12:29:22 +00:00
|
|
|
$force_refetch = isset($_REQUEST["force_refetch"]);
|
|
|
|
$feed_data = "";
|
2013-04-15 11:08:00 +00:00
|
|
|
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log("running HOOK_FETCH_FEED handlers...", Debug::LOG_VERBOSE);
|
2019-11-14 03:39:45 +00:00
|
|
|
|
2021-02-08 13:52:47 +00:00
|
|
|
$start_ts = microtime(true);
|
|
|
|
$last_article_timestamp = 0;
|
2021-03-06 08:17:15 +00:00
|
|
|
|
|
|
|
$hff_owner_uid = $feed_obj->owner_uid;
|
|
|
|
$hff_feed_url = $feed_obj->feed_url;
|
|
|
|
|
2021-02-08 13:52:47 +00:00
|
|
|
$pluginhost->chain_hooks_callback(PluginHost::HOOK_FETCH_FEED,
|
|
|
|
function ($result, $plugin) use (&$feed_data, $start_ts) {
|
|
|
|
$feed_data = $result;
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log(sprintf("=== %.4f (sec) %s", microtime(true) - $start_ts, get_class($plugin)), Debug::LOG_VERBOSE);
|
2021-02-08 13:52:47 +00:00
|
|
|
},
|
2021-03-06 08:17:15 +00:00
|
|
|
$feed_data, $hff_feed_url, $hff_owner_uid, $feed, $last_article_timestamp, $auth_login, $auth_pass);
|
2019-11-14 03:39:45 +00:00
|
|
|
|
|
|
|
if ($feed_data) {
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log("feed data has been modified by a plugin.", Debug::LOG_VERBOSE);
|
2019-11-14 03:39:45 +00:00
|
|
|
} else {
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log("feed data has not been modified by a plugin.", Debug::LOG_VERBOSE);
|
2017-04-26 12:29:22 +00:00
|
|
|
}
|
2011-12-13 12:06:25 +00:00
|
|
|
|
2017-04-26 12:29:22 +00:00
|
|
|
// try cache
|
|
|
|
if (!$feed_data &&
|
|
|
|
is_readable($cache_filename) &&
|
2021-03-06 08:17:15 +00:00
|
|
|
!$feed_obj->auth_login && !$feed_obj->auth_pass &&
|
2017-04-26 12:29:22 +00:00
|
|
|
filemtime($cache_filename) > time() - 30) {
|
2013-03-27 09:34:04 +00:00
|
|
|
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log("using local cache: {$cache_filename}.", Debug::LOG_VERBOSE);
|
2013-05-01 21:36:17 +00:00
|
|
|
|
2021-03-06 08:17:15 +00:00
|
|
|
$feed_data = file_get_contents($cache_filename);
|
2013-03-29 07:31:10 +00:00
|
|
|
|
2017-04-26 12:29:22 +00:00
|
|
|
if ($feed_data) {
|
|
|
|
$rss_hash = sha1($feed_data);
|
2013-10-25 10:42:43 +00:00
|
|
|
}
|
2013-05-20 11:26:53 +00:00
|
|
|
|
2017-04-26 12:29:22 +00:00
|
|
|
} else {
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log("local cache will not be used for this feed", Debug::LOG_VERBOSE);
|
2017-04-26 12:29:22 +00:00
|
|
|
}
|
2016-01-13 15:12:31 +00:00
|
|
|
|
2017-04-26 12:29:22 +00:00
|
|
|
// fetch feed from source
|
|
|
|
if (!$feed_data) {
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log("last unconditional update request: {$feed_obj->last_unconditional}", Debug::LOG_VERBOSE);
|
2016-01-13 15:12:31 +00:00
|
|
|
|
2017-04-26 12:29:22 +00:00
|
|
|
if (ini_get("open_basedir") && function_exists("curl_init")) {
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log("not using CURL due to open_basedir restrictions", Debug::LOG_VERBOSE);
|
2017-04-26 12:29:22 +00:00
|
|
|
}
|
2013-04-18 07:35:58 +00:00
|
|
|
|
2021-03-06 08:17:15 +00:00
|
|
|
if (time() - strtotime($feed_obj->last_unconditional) > Config::get(Config::MAX_CONDITIONAL_INTERVAL)) {
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log("maximum allowed interval for conditional requests exceeded, forcing refetch", Debug::LOG_VERBOSE);
|
2017-11-30 10:12:28 +00:00
|
|
|
|
2018-01-30 07:44:31 +00:00
|
|
|
$force_refetch = true;
|
|
|
|
} else {
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log("stored last modified for conditional request: {$feed_obj->last_modified}", Debug::LOG_VERBOSE);
|
2018-01-30 07:44:31 +00:00
|
|
|
}
|
2017-08-17 11:40:21 +00:00
|
|
|
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log("fetching {$feed_obj->feed_url} (force_refetch: $force_refetch)...", Debug::LOG_VERBOSE);
|
2017-08-17 11:40:21 +00:00
|
|
|
|
2020-09-22 06:04:33 +00:00
|
|
|
$feed_data = UrlHelper::fetch([
|
2021-03-06 08:17:15 +00:00
|
|
|
"url" => $feed_obj->feed_url,
|
|
|
|
"login" => $feed_obj->auth_login,
|
|
|
|
"pass" => $feed_obj->auth_pass,
|
2021-02-22 19:35:27 +00:00
|
|
|
"timeout" => $no_cache ? Config::get(Config::FEED_FETCH_NO_CACHE_TIMEOUT) : Config::get(Config::FEED_FETCH_TIMEOUT),
|
2021-03-06 08:17:15 +00:00
|
|
|
"last_modified" => $force_refetch ? "" : $feed_obj->last_modified
|
2017-08-17 11:40:21 +00:00
|
|
|
]);
|
2013-04-18 07:35:58 +00:00
|
|
|
|
2017-04-26 12:29:22 +00:00
|
|
|
$feed_data = trim($feed_data);
|
2013-05-08 15:22:33 +00:00
|
|
|
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log("fetch done.", Debug::LOG_VERBOSE);
|
|
|
|
Debug::log(sprintf("effective URL (after redirects): %s (IP: %s) ", UrlHelper::$fetch_effective_url, UrlHelper::$fetch_effective_ip_addr), Debug::LOG_VERBOSE);
|
|
|
|
Debug::log("server last modified: " . UrlHelper::$fetch_last_modified, Debug::LOG_VERBOSE);
|
2017-08-17 11:40:21 +00:00
|
|
|
|
2021-03-06 08:17:15 +00:00
|
|
|
if ($feed_data && UrlHelper::$fetch_last_modified != $feed_obj->last_modified) {
|
|
|
|
$feed_obj->last_modified = substr(UrlHelper::$fetch_last_modified, 0, 245);
|
|
|
|
$feed_obj->save();
|
2017-08-17 11:40:21 +00:00
|
|
|
}
|
2014-03-14 07:29:58 +00:00
|
|
|
|
2017-04-26 12:29:22 +00:00
|
|
|
// cache vanilla feed data for re-use
|
2021-03-06 08:17:15 +00:00
|
|
|
if ($feed_data && !$feed_obj->auth_pass && !$feed_obj->auth_login && is_writable(Config::get(Config::CACHE_DIR) . "/feeds")) {
|
2017-04-26 12:29:22 +00:00
|
|
|
$new_rss_hash = sha1($feed_data);
|
|
|
|
|
|
|
|
if ($new_rss_hash != $rss_hash) {
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log("saving to local cache: $cache_filename", Debug::LOG_VERBOSE);
|
2021-03-06 08:17:15 +00:00
|
|
|
file_put_contents($cache_filename, $feed_data);
|
2014-03-14 07:29:58 +00:00
|
|
|
}
|
2013-03-29 07:50:58 +00:00
|
|
|
}
|
2017-04-26 12:29:22 +00:00
|
|
|
}
|
2013-02-23 14:38:50 +00:00
|
|
|
|
2017-04-26 12:29:22 +00:00
|
|
|
if (!$feed_data) {
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log(sprintf("unable to fetch: %s [%s]", UrlHelper::$fetch_last_error, UrlHelper::$fetch_last_error_code), Debug::LOG_VERBOSE);
|
2013-03-29 07:31:10 +00:00
|
|
|
|
2017-04-26 12:29:22 +00:00
|
|
|
// If-Modified-Since
|
2021-02-28 07:12:57 +00:00
|
|
|
if (UrlHelper::$fetch_last_error_code == 304) {
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log("source claims data not modified, nothing to do.", Debug::LOG_VERBOSE);
|
2017-12-02 06:55:53 +00:00
|
|
|
$error_message = "";
|
2013-03-29 07:50:58 +00:00
|
|
|
|
2021-03-06 08:17:15 +00:00
|
|
|
$feed_obj->set([
|
|
|
|
'last_error' => '',
|
|
|
|
'last_successful_update' => Db::NOW(),
|
|
|
|
'last_updated' => Db::NOW(),
|
|
|
|
]);
|
|
|
|
|
|
|
|
$feed_obj->save();
|
|
|
|
|
2020-09-30 13:35:50 +00:00
|
|
|
} else {
|
2021-02-28 07:12:57 +00:00
|
|
|
$error_message = UrlHelper::$fetch_last_error;
|
2020-09-30 13:35:50 +00:00
|
|
|
|
2021-03-06 08:17:15 +00:00
|
|
|
$feed_obj->set([
|
|
|
|
'last_error' => $error_message,
|
|
|
|
'last_updated' => Db::NOW(),
|
|
|
|
]);
|
|
|
|
|
|
|
|
$feed_obj->save();
|
2020-09-30 13:35:50 +00:00
|
|
|
}
|
|
|
|
|
2020-09-27 14:13:36 +00:00
|
|
|
return $error_message == "";
|
2013-03-29 07:31:10 +00:00
|
|
|
}
|
|
|
|
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log("running HOOK_FEED_FETCHED handlers...", Debug::LOG_VERBOSE);
|
2019-11-14 03:39:45 +00:00
|
|
|
$feed_data_checksum = md5($feed_data);
|
|
|
|
|
2021-03-06 08:17:15 +00:00
|
|
|
// because chain_hooks_callback() accepts variables by value
|
|
|
|
$pff_owner_uid = $feed_obj->owner_uid;
|
2021-03-06 12:05:45 +00:00
|
|
|
$pff_feed_url = $feed_obj->feed_url;
|
2021-03-06 08:17:15 +00:00
|
|
|
|
2021-02-08 13:52:47 +00:00
|
|
|
$start_ts = microtime(true);
|
|
|
|
$pluginhost->chain_hooks_callback(PluginHost::HOOK_FEED_FETCHED,
|
|
|
|
function ($result, $plugin) use (&$feed_data, $start_ts) {
|
|
|
|
$feed_data = $result;
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log(sprintf("=== %.4f (sec) %s", microtime(true) - $start_ts, get_class($plugin)), Debug::LOG_VERBOSE);
|
2021-02-08 13:52:47 +00:00
|
|
|
},
|
2021-03-06 12:05:45 +00:00
|
|
|
$feed_data, $pff_feed_url, $pff_owner_uid, $feed);
|
2019-11-14 03:39:45 +00:00
|
|
|
|
|
|
|
if (md5($feed_data) != $feed_data_checksum) {
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log("feed data has been modified by a plugin.", Debug::LOG_VERBOSE);
|
2019-11-14 03:39:45 +00:00
|
|
|
} else {
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log("feed data has not been modified by a plugin.", Debug::LOG_VERBOSE);
|
2013-02-23 14:38:50 +00:00
|
|
|
}
|
|
|
|
|
2017-04-27 10:08:43 +00:00
|
|
|
$rss = new FeedParser($feed_data);
|
|
|
|
$rss->init();
|
2011-12-13 12:06:25 +00:00
|
|
|
|
2012-12-24 09:45:34 +00:00
|
|
|
if (!$rss->error()) {
|
2011-12-13 12:06:25 +00:00
|
|
|
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log("running HOOK_FEED_PARSED handlers...", Debug::LOG_VERBOSE);
|
2019-11-14 03:39:45 +00:00
|
|
|
|
2012-12-25 06:02:08 +00:00
|
|
|
// We use local pluginhost here because we need to load different per-user feed plugins
|
2019-11-14 03:39:45 +00:00
|
|
|
|
2021-02-08 13:52:47 +00:00
|
|
|
$start_ts = microtime(true);
|
|
|
|
$pluginhost->chain_hooks_callback(PluginHost::HOOK_FEED_PARSED,
|
|
|
|
function($result, $plugin) use ($start_ts) {
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log(sprintf("=== %.4f (sec) %s", microtime(true) - $start_ts, get_class($plugin)), Debug::LOG_VERBOSE);
|
2021-02-08 13:52:47 +00:00
|
|
|
},
|
|
|
|
$rss, $feed);
|
2012-12-24 10:13:03 +00:00
|
|
|
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log("language: $feed_language", Debug::LOG_VERBOSE);
|
|
|
|
Debug::log("processing feed data...", Debug::LOG_VERBOSE);
|
2011-12-13 12:06:25 +00:00
|
|
|
|
2021-03-06 12:05:45 +00:00
|
|
|
$site_url = mb_substr(rewrite_relative_url($feed_obj->feed_url, clean($rss->get_link())), 0, 245);
|
2011-12-13 12:06:25 +00:00
|
|
|
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log("site_url: $site_url", Debug::LOG_VERBOSE);
|
|
|
|
Debug::log("feed_title: {$rss->get_title()}", Debug::LOG_VERBOSE);
|
2021-03-06 08:17:15 +00:00
|
|
|
|
|
|
|
Debug::log("favicon: needs check: {$feed_obj->favicon_needs_check} is custom: {$feed_obj->favicon_is_custom} avg color: {$feed_obj->favicon_avg_color}",
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::LOG_VERBOSE);
|
2013-05-01 13:04:57 +00:00
|
|
|
|
2021-03-06 08:17:15 +00:00
|
|
|
if ($feed_obj->favicon_needs_check || $force_refetch) {
|
2013-04-23 18:30:06 +00:00
|
|
|
|
|
|
|
/* terrible hack: if we crash on floicon shit here, we won't check
|
2013-04-23 18:33:23 +00:00
|
|
|
* the icon avgcolor again (unless the icon got updated) */
|
2013-04-23 18:30:06 +00:00
|
|
|
|
2021-02-22 19:35:27 +00:00
|
|
|
$favicon_file = Config::get(Config::ICONS_DIR) . "/$feed.ico";
|
2021-02-06 07:23:45 +00:00
|
|
|
$favicon_modified = file_exists($favicon_file) ? filemtime($favicon_file) : -1;
|
2013-04-23 18:33:23 +00:00
|
|
|
|
2021-03-06 08:17:15 +00:00
|
|
|
if (!$feed_obj->favicon_is_custom) {
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log("favicon: trying to update favicon...", Debug::LOG_VERBOSE);
|
|
|
|
self::update_favicon($site_url, $feed);
|
2013-04-15 11:08:00 +00:00
|
|
|
|
2021-03-06 08:17:15 +00:00
|
|
|
if ((file_exists($favicon_file) ? filemtime($favicon_file) : -1) > $favicon_modified)
|
|
|
|
$feed_obj->favicon_avg_color = null;
|
|
|
|
}
|
2013-04-15 11:08:00 +00:00
|
|
|
|
2021-03-06 12:19:31 +00:00
|
|
|
if (is_readable($favicon_file) && function_exists("imagecreatefromstring") && empty($feed_obj->favicon_avg_color)) {
|
2017-05-05 15:10:07 +00:00
|
|
|
require_once "colors.php";
|
2013-04-15 11:08:00 +00:00
|
|
|
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log("favicon: trying to calculate average color...", Debug::LOG_VERBOSE);
|
2013-04-24 15:01:51 +00:00
|
|
|
|
2021-03-06 08:17:15 +00:00
|
|
|
$feed_obj->favicon_avg_color = 'fail';
|
|
|
|
$feed_obj->save();
|
2017-12-02 06:55:53 +00:00
|
|
|
|
2021-03-06 08:17:15 +00:00
|
|
|
$feed_obj->favicon_avg_color = \Colors\calculate_avg_color($favicon_file);
|
|
|
|
$feed_obj->save();
|
2013-04-15 11:17:12 +00:00
|
|
|
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log("favicon: avg color: {$feed_obj->favicon_avg_color}", Debug::LOG_VERBOSE);
|
2013-04-15 11:08:00 +00:00
|
|
|
|
2021-03-06 08:17:15 +00:00
|
|
|
} else if ($feed_obj->favicon_avg_color == 'fail') {
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log("floicon failed $favicon_file, not trying to recalculate avg color", Debug::LOG_VERBOSE);
|
2021-03-06 08:17:15 +00:00
|
|
|
}
|
2012-05-24 12:45:54 +00:00
|
|
|
}
|
2011-12-13 12:06:25 +00:00
|
|
|
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log("loading filters & labels...", Debug::LOG_VERBOSE);
|
2011-12-13 12:06:25 +00:00
|
|
|
|
2021-03-06 08:17:15 +00:00
|
|
|
$filters = self::load_filters($feed, $feed_obj->owner_uid);
|
2011-12-13 12:06:25 +00:00
|
|
|
|
2021-03-06 12:19:31 +00:00
|
|
|
if (Debug::get_loglevel() >= Debug::LOG_EXTENDED) {
|
2018-01-30 07:44:31 +00:00
|
|
|
print_r($filters);
|
|
|
|
}
|
2017-07-02 17:37:52 +00:00
|
|
|
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log("" . count($filters) . " filters loaded.", Debug::LOG_VERBOSE);
|
2011-12-13 12:06:25 +00:00
|
|
|
|
2012-12-24 09:45:34 +00:00
|
|
|
$items = $rss->get_items();
|
2011-12-13 12:06:25 +00:00
|
|
|
|
2012-12-24 09:45:34 +00:00
|
|
|
if (!is_array($items)) {
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log("no articles found.", Debug::LOG_VERBOSE);
|
2021-03-06 08:17:15 +00:00
|
|
|
|
|
|
|
$feed_obj->set([
|
|
|
|
'last_updated' => Db::NOW(),
|
|
|
|
'last_unconditional' => Db::NOW(),
|
|
|
|
'last_error' => '',
|
|
|
|
]);
|
|
|
|
|
|
|
|
$feed_obj->save();
|
2011-12-13 12:06:25 +00:00
|
|
|
|
2017-12-02 06:55:53 +00:00
|
|
|
return true; // no articles
|
2011-12-13 12:06:25 +00:00
|
|
|
}
|
|
|
|
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log("processing articles...", Debug::LOG_VERBOSE);
|
2011-12-13 12:06:25 +00:00
|
|
|
|
2015-06-17 20:25:21 +00:00
|
|
|
$tstart = time();
|
|
|
|
|
2012-12-24 09:45:34 +00:00
|
|
|
foreach ($items as $item) {
|
2017-12-02 08:16:33 +00:00
|
|
|
$pdo->beginTransaction();
|
|
|
|
|
2021-03-06 08:17:15 +00:00
|
|
|
Debug::log("=================================================================================================================================",
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::LOG_VERBOSE);
|
2021-03-06 08:17:15 +00:00
|
|
|
|
2018-11-30 05:34:29 +00:00
|
|
|
if (Debug::get_loglevel() >= 3) {
|
2011-12-13 12:06:25 +00:00
|
|
|
print_r($item);
|
|
|
|
}
|
|
|
|
|
2015-06-17 20:25:21 +00:00
|
|
|
if (ini_get("max_execution_time") > 0 && time() - $tstart >= ini_get("max_execution_time") * 0.7) {
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log("looks like there's too many articles to process at once, breaking out.", Debug::LOG_VERBOSE);
|
2017-12-02 08:16:33 +00:00
|
|
|
$pdo->commit();
|
2015-06-17 20:25:21 +00:00
|
|
|
break;
|
|
|
|
}
|
|
|
|
|
2017-12-02 06:55:53 +00:00
|
|
|
$entry_guid = strip_tags($item->get_id());
|
|
|
|
if (!$entry_guid) $entry_guid = strip_tags($item->get_link());
|
2020-09-22 11:54:15 +00:00
|
|
|
if (!$entry_guid) $entry_guid = self::make_guid_from_title($item->get_title());
|
2017-12-02 08:16:33 +00:00
|
|
|
|
|
|
|
if (!$entry_guid) {
|
|
|
|
$pdo->commit();
|
|
|
|
continue;
|
|
|
|
}
|
2011-12-13 12:06:25 +00:00
|
|
|
|
2021-03-06 08:17:15 +00:00
|
|
|
$entry_guid_hashed_compat = 'SHA1:' . sha1("{$feed_obj->owner_uid},$entry_guid");
|
|
|
|
$entry_guid_hashed = json_encode(["ver" => 2, "uid" => $feed_obj->owner_uid, "hash" => 'SHA1:' . sha1($entry_guid)]);
|
|
|
|
$entry_guid = "$feed_obj->owner_uid,$entry_guid";
|
2013-01-18 13:42:26 +00:00
|
|
|
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log("guid $entry_guid (hash: $entry_guid_hashed compat: $entry_guid_hashed_compat)", Debug::LOG_VERBOSE);
|
2013-04-15 15:28:11 +00:00
|
|
|
|
2019-09-30 08:41:07 +00:00
|
|
|
$entry_timestamp = (int)$item->get_date();
|
2013-05-01 13:38:16 +00:00
|
|
|
|
2021-03-06 08:17:15 +00:00
|
|
|
Debug::log(sprintf("orig date: %s (%s)", $item->get_date(), date("Y-m-d H:i:s", $item->get_date())),
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::LOG_VERBOSE);
|
2011-12-13 12:06:25 +00:00
|
|
|
|
2017-12-02 06:55:53 +00:00
|
|
|
$entry_title = strip_tags($item->get_title());
|
2012-12-24 09:53:38 +00:00
|
|
|
|
2021-06-19 12:36:04 +00:00
|
|
|
$entry_link = UrlHelper::rewrite_relative($site_url, clean($item->get_link()), "a", "href");
|
2011-12-13 12:06:25 +00:00
|
|
|
|
2018-08-12 15:12:34 +00:00
|
|
|
$entry_language = mb_substr(trim($item->get_language()), 0, 2);
|
2018-08-12 13:54:13 +00:00
|
|
|
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log("title $entry_title", Debug::LOG_VERBOSE);
|
|
|
|
Debug::log("link $entry_link", Debug::LOG_VERBOSE);
|
|
|
|
Debug::log("language $entry_language", Debug::LOG_VERBOSE);
|
2011-12-13 12:06:25 +00:00
|
|
|
|
|
|
|
if (!$entry_title) $entry_title = date("Y-m-d H:i:s", $entry_timestamp);;
|
|
|
|
|
2012-12-24 09:45:34 +00:00
|
|
|
$entry_content = $item->get_content();
|
|
|
|
if (!$entry_content) $entry_content = $item->get_description();
|
2011-12-13 12:06:25 +00:00
|
|
|
|
2018-11-30 05:34:29 +00:00
|
|
|
if (Debug::get_loglevel() >= 3) {
|
2013-04-19 09:24:51 +00:00
|
|
|
print "content: ";
|
2016-01-22 23:17:24 +00:00
|
|
|
print htmlspecialchars($entry_content);
|
2011-12-26 09:11:08 +00:00
|
|
|
print "\n";
|
2011-12-13 12:06:25 +00:00
|
|
|
}
|
|
|
|
|
2017-12-02 06:55:53 +00:00
|
|
|
$entry_comments = mb_substr(strip_tags($item->get_comments_url()), 0, 245);
|
2016-11-25 11:04:41 +00:00
|
|
|
$num_comments = (int) $item->get_comments_count();
|
2011-12-13 12:06:25 +00:00
|
|
|
|
2017-12-02 06:55:53 +00:00
|
|
|
$entry_author = strip_tags($item->get_author());
|
|
|
|
$entry_guid = mb_substr($entry_guid, 0, 245);
|
2011-12-13 12:06:25 +00:00
|
|
|
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log("author $entry_author", Debug::LOG_VERBOSE);
|
|
|
|
Debug::log("looking for tags...", Debug::LOG_VERBOSE);
|
2011-12-13 12:06:25 +00:00
|
|
|
|
2019-11-20 15:56:34 +00:00
|
|
|
$entry_tags = $item->get_categories();
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log("tags found: " . join(", ", $entry_tags), Debug::LOG_VERBOSE);
|
2013-05-01 16:33:59 +00:00
|
|
|
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log("done collecting data.", Debug::LOG_VERBOSE);
|
2011-12-13 12:06:25 +00:00
|
|
|
|
2017-12-02 06:55:53 +00:00
|
|
|
$sth = $pdo->prepare("SELECT id, content_hash, lang FROM ttrss_entries
|
2020-05-17 11:01:16 +00:00
|
|
|
WHERE guid IN (?, ?, ?)");
|
|
|
|
$sth->execute([$entry_guid, $entry_guid_hashed, $entry_guid_hashed_compat]);
|
2013-02-21 18:24:06 +00:00
|
|
|
|
2017-12-02 06:55:53 +00:00
|
|
|
if ($row = $sth->fetch()) {
|
|
|
|
$base_entry_id = $row["id"];
|
|
|
|
$entry_stored_hash = $row["content_hash"];
|
2021-03-06 08:17:15 +00:00
|
|
|
$article_labels = Article::_get_labels($base_entry_id, $feed_obj->owner_uid);
|
2015-12-07 08:54:24 +00:00
|
|
|
|
2021-03-06 08:17:15 +00:00
|
|
|
$existing_tags = Article::_get_tags($base_entry_id, $feed_obj->owner_uid);
|
2015-12-07 08:54:24 +00:00
|
|
|
$entry_tags = array_unique(array_merge($entry_tags, $existing_tags));
|
2013-02-21 18:24:06 +00:00
|
|
|
} else {
|
2014-08-18 14:32:16 +00:00
|
|
|
$base_entry_id = false;
|
|
|
|
$entry_stored_hash = "";
|
2015-03-03 21:26:52 +00:00
|
|
|
$article_labels = array();
|
2013-02-21 18:24:06 +00:00
|
|
|
}
|
|
|
|
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log("looking for enclosures...", Debug::LOG_VERBOSE);
|
2020-06-24 19:54:14 +00:00
|
|
|
|
|
|
|
// enclosures
|
|
|
|
|
|
|
|
$enclosures = array();
|
|
|
|
|
2021-03-05 06:35:17 +00:00
|
|
|
$encs = $item->get_enclosures();
|
2020-06-24 19:54:14 +00:00
|
|
|
|
|
|
|
if (is_array($encs)) {
|
|
|
|
foreach ($encs as $e) {
|
|
|
|
|
2021-02-08 13:52:47 +00:00
|
|
|
$pluginhost->chain_hooks_callback(PluginHost::HOOK_ENCLOSURE_IMPORTED,
|
|
|
|
function ($result) use (&$e) {
|
|
|
|
$e = $result;
|
|
|
|
},
|
|
|
|
$e, $feed);
|
2020-06-24 19:54:14 +00:00
|
|
|
|
|
|
|
$e_item = array(
|
|
|
|
rewrite_relative_url($site_url, $e->link),
|
|
|
|
$e->type, $e->length, $e->title, $e->width, $e->height);
|
|
|
|
|
|
|
|
// Yet another episode of "mysql utf8_general_ci is gimped"
|
2021-02-22 19:35:27 +00:00
|
|
|
if (Config::get(Config::DB_TYPE) == "mysql" && Config::get(Config::MYSQL_CHARSET) != "UTF8MB4") {
|
2020-06-24 19:54:14 +00:00
|
|
|
for ($i = 0; $i < count($e_item); $i++) {
|
|
|
|
if (is_string($e_item[$i])) {
|
2020-09-22 11:54:15 +00:00
|
|
|
$e_item[$i] = self::strip_utf8mb4($e_item[$i]);
|
2020-06-24 19:54:14 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
array_push($enclosures, $e_item);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2021-03-06 08:17:15 +00:00
|
|
|
$article = array("owner_uid" => $feed_obj->owner_uid, // read only
|
2013-02-21 18:24:06 +00:00
|
|
|
"guid" => $entry_guid, // read only
|
2015-06-17 12:15:04 +00:00
|
|
|
"guid_hashed" => $entry_guid_hashed, // read only
|
2012-12-24 09:45:34 +00:00
|
|
|
"title" => $entry_title,
|
|
|
|
"content" => $entry_content,
|
|
|
|
"link" => $entry_link,
|
2015-03-03 21:26:52 +00:00
|
|
|
"labels" => $article_labels, // current limitation: can add labels to article, can't remove them
|
2012-12-24 09:45:34 +00:00
|
|
|
"tags" => $entry_tags,
|
2013-02-23 08:07:46 +00:00
|
|
|
"author" => $entry_author,
|
2015-01-19 09:51:59 +00:00
|
|
|
"force_catchup" => false, // ugly hack for the time being
|
2015-06-15 17:01:07 +00:00
|
|
|
"score_modifier" => 0, // no previous value, plugin should recalculate score modifier based on content if needed
|
2015-06-19 07:12:47 +00:00
|
|
|
"language" => $entry_language,
|
2019-09-11 11:04:59 +00:00
|
|
|
"timestamp" => $entry_timestamp,
|
2019-10-25 11:37:00 +00:00
|
|
|
"num_comments" => $num_comments,
|
2020-06-24 19:54:14 +00:00
|
|
|
"enclosures" => $enclosures,
|
2013-09-02 06:03:04 +00:00
|
|
|
"feed" => array("id" => $feed,
|
2021-03-06 12:05:45 +00:00
|
|
|
"fetch_url" => $feed_obj->feed_url,
|
2016-08-06 18:16:39 +00:00
|
|
|
"site_url" => $site_url,
|
2021-03-06 08:17:15 +00:00
|
|
|
"cache_images" => $feed_obj->cache_images)
|
2017-05-05 15:10:07 +00:00
|
|
|
);
|
2012-11-26 10:33:18 +00:00
|
|
|
|
2014-08-18 14:32:16 +00:00
|
|
|
$entry_plugin_data = "";
|
2020-09-22 11:54:15 +00:00
|
|
|
$entry_current_hash = self::calculate_article_hash($article, $pluginhost);
|
2014-08-18 14:32:16 +00:00
|
|
|
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log("article hash: $entry_current_hash [stored=$entry_stored_hash]", Debug::LOG_VERBOSE);
|
2014-08-18 14:32:16 +00:00
|
|
|
|
2014-08-21 09:20:41 +00:00
|
|
|
if ($entry_current_hash == $entry_stored_hash && !isset($_REQUEST["force_rehash"])) {
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log("stored article seems up to date [IID: $base_entry_id], updating timestamp only.", Debug::LOG_VERBOSE);
|
2014-08-18 14:32:16 +00:00
|
|
|
|
|
|
|
// we keep encountering the entry in feeds, so we need to
|
|
|
|
// update date_updated column so that we don't get horrible
|
|
|
|
// dupes when the entry gets purged and reinserted again e.g.
|
|
|
|
// in the case of SLOW SLOW OMG SLOW updating feeds
|
|
|
|
|
2021-03-06 08:17:15 +00:00
|
|
|
$entry_obj = ORM::for_table('ttrss_entries')
|
|
|
|
->find_one($base_entry_id)
|
|
|
|
->set('date_updated', Db::NOW())
|
|
|
|
->save();
|
|
|
|
|
2017-12-02 08:16:33 +00:00
|
|
|
$pdo->commit();
|
2021-03-06 12:19:31 +00:00
|
|
|
|
2016-01-07 19:01:35 +00:00
|
|
|
continue;
|
2014-08-18 14:32:16 +00:00
|
|
|
}
|
|
|
|
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log("hash differs, running HOOK_ARTICLE_FILTER handlers...", Debug::LOG_VERBOSE);
|
2014-08-18 14:32:16 +00:00
|
|
|
|
2021-02-08 11:24:45 +00:00
|
|
|
$start_ts = microtime(true);
|
|
|
|
|
2021-02-08 13:14:48 +00:00
|
|
|
$pluginhost->chain_hooks_callback(PluginHost::HOOK_ARTICLE_FILTER,
|
|
|
|
function ($result, $plugin) use (&$article, &$entry_plugin_data, $start_ts) {
|
|
|
|
$article = $result;
|
2014-08-12 15:36:29 +00:00
|
|
|
|
2021-02-08 13:14:48 +00:00
|
|
|
$entry_plugin_data .= mb_strtolower(get_class($plugin)) . ",";
|
2014-08-18 14:32:16 +00:00
|
|
|
|
2021-02-08 13:14:48 +00:00
|
|
|
Debug::log(sprintf("=== %.4f (sec) %s", microtime(true) - $start_ts, get_class($plugin)),
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::LOG_VERBOSE);
|
2021-02-08 13:14:48 +00:00
|
|
|
},
|
|
|
|
$article);
|
2012-11-26 10:33:18 +00:00
|
|
|
|
2021-02-08 11:24:45 +00:00
|
|
|
if (Debug::get_loglevel() >= 3) {
|
2016-01-22 23:17:24 +00:00
|
|
|
print "processed content: ";
|
|
|
|
print htmlspecialchars($article["content"]);
|
|
|
|
print "\n";
|
|
|
|
}
|
|
|
|
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log("plugin data: {$entry_plugin_data}", Debug::LOG_VERBOSE);
|
2014-08-18 14:32:16 +00:00
|
|
|
|
2015-06-16 05:17:49 +00:00
|
|
|
// Workaround: 4-byte unicode requires utf8mb4 in MySQL. See https://tt-rss.org/forum/viewtopic.php?f=1&t=3377&p=20077#p20077
|
2021-02-22 19:35:27 +00:00
|
|
|
if (Config::get(Config::DB_TYPE) == "mysql" && Config::get(Config::MYSQL_CHARSET) != "UTF8MB4") {
|
2015-06-16 05:17:49 +00:00
|
|
|
foreach ($article as $k => $v) {
|
2015-07-06 12:32:26 +00:00
|
|
|
// i guess we'll have to take the risk of 4byte unicode labels & tags here
|
2016-07-26 07:51:22 +00:00
|
|
|
if (is_string($article[$k])) {
|
2020-09-22 11:54:15 +00:00
|
|
|
$article[$k] = self::strip_utf8mb4($v);
|
2015-07-06 12:32:26 +00:00
|
|
|
}
|
2015-06-16 05:17:49 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2015-08-11 20:28:41 +00:00
|
|
|
/* Collect article tags here so we could filter by them: */
|
|
|
|
|
2018-12-06 16:15:00 +00:00
|
|
|
$matched_rules = [];
|
|
|
|
$matched_filters = [];
|
2016-08-26 11:45:34 +00:00
|
|
|
|
2020-09-22 11:54:15 +00:00
|
|
|
$article_filters = self::get_article_filters($filters, $article["title"],
|
2017-04-26 12:29:22 +00:00
|
|
|
$article["content"], $article["link"], $article["author"],
|
2018-12-06 16:15:00 +00:00
|
|
|
$article["tags"], $matched_rules, $matched_filters);
|
|
|
|
|
|
|
|
// $article_filters should be renamed to something like $filter_actions; actual filter objects are in $matched_filters
|
2021-02-08 13:52:47 +00:00
|
|
|
$pluginhost->run_hooks(PluginHost::HOOK_FILTER_TRIGGERED,
|
2021-03-06 08:17:15 +00:00
|
|
|
$feed, $feed_obj->owner_uid, $article, $matched_filters, $matched_rules, $article_filters);
|
2015-08-11 20:28:41 +00:00
|
|
|
|
2018-12-06 20:12:20 +00:00
|
|
|
$matched_filter_ids = array_map(function($f) { return $f['id']; }, $matched_filters);
|
2018-12-06 16:37:20 +00:00
|
|
|
|
2018-12-06 20:12:20 +00:00
|
|
|
if (count($matched_filter_ids) > 0) {
|
2021-03-06 08:17:15 +00:00
|
|
|
$filter_objs = ORM::for_table('ttrss_filters2')
|
|
|
|
->where('owner_uid', $feed_obj->owner_uid)
|
|
|
|
->where_in('id', $matched_filter_ids);
|
|
|
|
|
|
|
|
foreach ($filter_objs as $filter_obj) {
|
|
|
|
$filter_obj->set('last_triggered', Db::NOW());
|
|
|
|
$filter_obj->save();
|
|
|
|
}
|
2018-12-06 20:05:02 +00:00
|
|
|
}
|
2018-12-06 16:37:20 +00:00
|
|
|
|
2021-03-06 12:19:31 +00:00
|
|
|
if (Debug::get_loglevel() >= Debug::LOG_EXTENDED) {
|
|
|
|
Debug::log("matched filters: ", Debug::LOG_VERBOSE);
|
2018-12-06 16:15:00 +00:00
|
|
|
|
2020-10-29 04:52:07 +00:00
|
|
|
if (count($matched_filters) != 0) {
|
2018-12-06 16:15:00 +00:00
|
|
|
print_r($matched_filters);
|
|
|
|
}
|
|
|
|
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log("matched filter rules: ", Debug::LOG_VERBOSE);
|
2016-08-26 11:45:34 +00:00
|
|
|
|
|
|
|
if (count($matched_rules) != 0) {
|
|
|
|
print_r($matched_rules);
|
|
|
|
}
|
|
|
|
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log("filter actions: ", Debug::LOG_VERBOSE);
|
2016-08-26 11:45:34 +00:00
|
|
|
|
2015-08-11 20:28:41 +00:00
|
|
|
if (count($article_filters) != 0) {
|
|
|
|
print_r($article_filters);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2020-09-22 11:54:15 +00:00
|
|
|
$plugin_filter_names = self::find_article_filters($article_filters, "plugin");
|
2015-08-11 20:28:41 +00:00
|
|
|
$plugin_filter_actions = $pluginhost->get_filter_actions();
|
|
|
|
|
|
|
|
if (count($plugin_filter_names) > 0) {
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log("applying plugin filter actions...", Debug::LOG_VERBOSE);
|
2015-08-11 20:28:41 +00:00
|
|
|
|
|
|
|
foreach ($plugin_filter_names as $pfn) {
|
|
|
|
list($pfclass,$pfaction) = explode(":", $pfn["param"]);
|
|
|
|
|
|
|
|
if (isset($plugin_filter_actions[$pfclass])) {
|
|
|
|
$plugin = $pluginhost->get_plugin($pfclass);
|
|
|
|
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log("... $pfclass: $pfaction", Debug::LOG_VERBOSE);
|
2015-08-11 20:28:41 +00:00
|
|
|
|
|
|
|
if ($plugin) {
|
|
|
|
$start = microtime(true);
|
|
|
|
$article = $plugin->hook_article_filter_action($article, $pfaction);
|
|
|
|
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log(sprintf("=== %.4f (sec)", microtime(true) - $start), Debug::LOG_VERBOSE);
|
2015-08-11 20:28:41 +00:00
|
|
|
} else {
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log("??? $pfclass: plugin object not found.", Debug::LOG_VERBOSE);
|
2015-08-11 20:28:41 +00:00
|
|
|
}
|
|
|
|
} else {
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log("??? $pfclass: filter plugin not registered.", Debug::LOG_VERBOSE);
|
2015-08-11 20:28:41 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2012-12-24 09:45:34 +00:00
|
|
|
$entry_tags = $article["tags"];
|
2017-12-02 06:55:53 +00:00
|
|
|
$entry_title = strip_tags($article["title"]);
|
|
|
|
$entry_author = mb_substr(strip_tags($article["author"]), 0, 245);
|
|
|
|
$entry_link = strip_tags($article["link"]);
|
2013-03-18 20:42:59 +00:00
|
|
|
$entry_content = $article["content"]; // escaped below
|
2015-01-19 09:51:59 +00:00
|
|
|
$entry_force_catchup = $article["force_catchup"];
|
2015-03-03 21:26:52 +00:00
|
|
|
$article_labels = $article["labels"];
|
2015-06-15 17:01:07 +00:00
|
|
|
$entry_score_modifier = (int) $article["score_modifier"];
|
2017-12-02 06:55:53 +00:00
|
|
|
$entry_language = $article["language"];
|
2019-09-11 11:04:59 +00:00
|
|
|
$entry_timestamp = $article["timestamp"];
|
2019-10-25 11:37:00 +00:00
|
|
|
$num_comments = $article["num_comments"];
|
2020-06-24 19:54:14 +00:00
|
|
|
$enclosures = $article["enclosures"];
|
2019-09-11 11:04:59 +00:00
|
|
|
|
|
|
|
if ($entry_timestamp == -1 || !$entry_timestamp || $entry_timestamp > time()) {
|
|
|
|
$entry_timestamp = time();
|
|
|
|
}
|
|
|
|
|
|
|
|
$entry_timestamp_fmt = strftime("%Y/%m/%d %H:%M:%S", $entry_timestamp);
|
|
|
|
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log("date: $entry_timestamp ($entry_timestamp_fmt)", Debug::LOG_VERBOSE);
|
|
|
|
Debug::log("num_comments: $num_comments", Debug::LOG_VERBOSE);
|
2015-03-03 21:26:52 +00:00
|
|
|
|
2021-03-06 12:19:31 +00:00
|
|
|
if (Debug::get_loglevel() >= Debug::LOG_EXTENDED) {
|
|
|
|
Debug::log("article labels:", Debug::LOG_VERBOSE);
|
2016-08-26 11:45:34 +00:00
|
|
|
|
|
|
|
if (count($article_labels) != 0) {
|
|
|
|
print_r($article_labels);
|
|
|
|
}
|
2015-03-03 21:26:52 +00:00
|
|
|
}
|
2015-01-19 09:51:59 +00:00
|
|
|
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log("force catchup: $entry_force_catchup", Debug::LOG_VERBOSE);
|
2013-03-18 20:42:59 +00:00
|
|
|
|
2021-03-06 08:17:15 +00:00
|
|
|
if ($feed_obj->cache_images)
|
2020-09-22 11:54:15 +00:00
|
|
|
self::cache_media($entry_content, $site_url);
|
2013-02-04 09:00:50 +00:00
|
|
|
|
2017-12-02 06:55:53 +00:00
|
|
|
$csth = $pdo->prepare("SELECT id FROM ttrss_entries
|
2020-05-17 11:01:16 +00:00
|
|
|
WHERE guid IN (?, ?, ?)");
|
|
|
|
$csth->execute([$entry_guid, $entry_guid_hashed, $entry_guid_hashed_compat]);
|
2012-12-30 08:22:29 +00:00
|
|
|
|
2017-12-02 06:55:53 +00:00
|
|
|
if (!$row = $csth->fetch()) {
|
2011-12-13 12:06:25 +00:00
|
|
|
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log("base guid [$entry_guid or $entry_guid_hashed] not found, creating...", Debug::LOG_VERBOSE);
|
2011-12-13 12:06:25 +00:00
|
|
|
|
|
|
|
// base post entry does not exist, create it
|
|
|
|
|
2017-12-02 06:55:53 +00:00
|
|
|
$usth = $pdo->prepare(
|
2011-12-13 12:06:25 +00:00
|
|
|
"INSERT INTO ttrss_entries
|
2018-11-03 12:00:58 +00:00
|
|
|
(title,
|
2011-12-13 12:06:25 +00:00
|
|
|
guid,
|
|
|
|
link,
|
|
|
|
updated,
|
|
|
|
content,
|
|
|
|
content_hash,
|
|
|
|
no_orig_date,
|
|
|
|
date_updated,
|
|
|
|
date_entered,
|
|
|
|
comments,
|
|
|
|
num_comments,
|
2013-02-21 18:24:06 +00:00
|
|
|
plugin_data,
|
2013-07-31 06:30:14 +00:00
|
|
|
lang,
|
2011-12-13 12:06:25 +00:00
|
|
|
author)
|
|
|
|
VALUES
|
2017-12-02 06:55:53 +00:00
|
|
|
(?, ?, ?, ?, ?, ?,
|
2014-02-19 11:17:29 +00:00
|
|
|
false,
|
2011-12-13 12:06:25 +00:00
|
|
|
NOW(),
|
2017-12-02 06:55:53 +00:00
|
|
|
?, ?, ?, ?, ?, ?)");
|
|
|
|
|
|
|
|
$usth->execute([$entry_title,
|
|
|
|
$entry_guid_hashed,
|
|
|
|
$entry_link,
|
|
|
|
$entry_timestamp_fmt,
|
2017-12-03 10:32:24 +00:00
|
|
|
"$entry_content",
|
2017-12-02 06:55:53 +00:00
|
|
|
$entry_current_hash,
|
|
|
|
$date_feed_processed,
|
|
|
|
$entry_comments,
|
2017-12-03 06:35:59 +00:00
|
|
|
(int)$num_comments,
|
2017-12-02 06:55:53 +00:00
|
|
|
$entry_plugin_data,
|
2017-12-03 10:32:24 +00:00
|
|
|
"$entry_language",
|
|
|
|
"$entry_author"]);
|
2012-10-30 11:10:52 +00:00
|
|
|
|
2011-12-13 12:06:25 +00:00
|
|
|
}
|
|
|
|
|
2020-05-17 11:01:16 +00:00
|
|
|
$csth->execute([$entry_guid, $entry_guid_hashed, $entry_guid_hashed_compat]);
|
2011-12-13 12:06:25 +00:00
|
|
|
|
|
|
|
$entry_ref_id = 0;
|
|
|
|
$entry_int_id = 0;
|
|
|
|
|
2017-12-02 06:55:53 +00:00
|
|
|
if ($row = $csth->fetch()) {
|
2011-12-13 12:06:25 +00:00
|
|
|
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log("base guid found, checking for user record", Debug::LOG_VERBOSE);
|
2011-12-13 12:06:25 +00:00
|
|
|
|
2017-12-02 06:55:53 +00:00
|
|
|
$ref_id = $row['id'];
|
2011-12-13 12:06:25 +00:00
|
|
|
$entry_ref_id = $ref_id;
|
|
|
|
|
2020-09-22 11:54:15 +00:00
|
|
|
if (self::find_article_filter($article_filters, "filter")) {
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log("article is filtered out, nothing to do.", Debug::LOG_VERBOSE);
|
2017-12-02 08:16:33 +00:00
|
|
|
$pdo->commit();
|
2011-12-13 12:06:25 +00:00
|
|
|
continue;
|
|
|
|
}
|
|
|
|
|
2020-09-22 11:54:15 +00:00
|
|
|
$score = self::calculate_article_score($article_filters) + $entry_score_modifier;
|
2011-12-13 12:06:25 +00:00
|
|
|
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log("initial score: $score [including plugin modifier: $entry_score_modifier]", Debug::LOG_VERBOSE);
|
2011-12-13 12:06:25 +00:00
|
|
|
|
2016-01-07 18:49:31 +00:00
|
|
|
// check for user post link to main table
|
|
|
|
|
2017-12-02 06:55:53 +00:00
|
|
|
$sth = $pdo->prepare("SELECT ref_id, int_id FROM ttrss_user_entries WHERE
|
|
|
|
ref_id = ? AND owner_uid = ?");
|
2021-03-06 08:17:15 +00:00
|
|
|
$sth->execute([$ref_id, $feed_obj->owner_uid]);
|
2011-12-13 12:06:25 +00:00
|
|
|
|
|
|
|
// okay it doesn't exist - create user entry
|
2017-12-02 06:55:53 +00:00
|
|
|
if ($row = $sth->fetch()) {
|
|
|
|
$entry_ref_id = $row["ref_id"];
|
|
|
|
$entry_int_id = $row["int_id"];
|
2011-12-13 12:06:25 +00:00
|
|
|
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log("user record FOUND: RID: $entry_ref_id, IID: $entry_int_id", Debug::LOG_VERBOSE);
|
2017-12-02 06:55:53 +00:00
|
|
|
} else {
|
2018-11-03 12:00:58 +00:00
|
|
|
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log("user record not found, creating...", Debug::LOG_VERBOSE);
|
2011-12-13 12:06:25 +00:00
|
|
|
|
2020-09-22 11:54:15 +00:00
|
|
|
if ($score >= -500 && !self::find_article_filter($article_filters, 'catchup') && !$entry_force_catchup) {
|
2017-12-02 06:55:53 +00:00
|
|
|
$unread = 1;
|
|
|
|
$last_read_qpart = null;
|
2011-12-13 12:06:25 +00:00
|
|
|
} else {
|
2017-12-02 06:55:53 +00:00
|
|
|
$unread = 0;
|
2017-12-18 15:27:23 +00:00
|
|
|
$last_read_qpart = date("Y-m-d H:i"); // we can't use NOW() here because it gets quoted
|
2011-12-13 12:06:25 +00:00
|
|
|
}
|
|
|
|
|
2020-09-22 11:54:15 +00:00
|
|
|
if (self::find_article_filter($article_filters, 'mark') || $score > 1000) {
|
2017-12-02 06:55:53 +00:00
|
|
|
$marked = 1;
|
2011-12-13 12:06:25 +00:00
|
|
|
} else {
|
2017-12-02 06:55:53 +00:00
|
|
|
$marked = 0;
|
2011-12-13 12:06:25 +00:00
|
|
|
}
|
|
|
|
|
2020-09-22 11:54:15 +00:00
|
|
|
if (self::find_article_filter($article_filters, 'publish')) {
|
2017-12-02 06:55:53 +00:00
|
|
|
$published = 1;
|
2011-12-13 12:06:25 +00:00
|
|
|
} else {
|
2017-12-02 06:55:53 +00:00
|
|
|
$published = 0;
|
2011-12-13 12:06:25 +00:00
|
|
|
}
|
|
|
|
|
2018-01-22 22:41:24 +00:00
|
|
|
$last_marked = ($marked == 1) ? 'NOW()' : 'NULL';
|
|
|
|
$last_published = ($published == 1) ? 'NOW()' : 'NULL';
|
2013-03-17 11:32:44 +00:00
|
|
|
|
2017-12-02 06:55:53 +00:00
|
|
|
$sth = $pdo->prepare(
|
2011-12-13 12:06:25 +00:00
|
|
|
"INSERT INTO ttrss_user_entries
|
|
|
|
(ref_id, owner_uid, feed_id, unread, last_read, marked,
|
2013-03-17 11:32:44 +00:00
|
|
|
published, score, tag_cache, label_cache, uuid,
|
|
|
|
last_marked, last_published)
|
2017-12-10 23:05:03 +00:00
|
|
|
VALUES (?, ?, ?, ?, ?, ?, ?, ?, '', '', '', ".$last_marked.", ".$last_published.")");
|
2011-12-13 12:06:25 +00:00
|
|
|
|
2021-03-06 08:17:15 +00:00
|
|
|
$sth->execute([$ref_id, $feed_obj->owner_uid, $feed, $unread, $last_read_qpart, $marked,
|
2017-12-10 23:05:03 +00:00
|
|
|
$published, $score]);
|
2011-12-13 12:06:25 +00:00
|
|
|
|
2017-12-02 06:55:53 +00:00
|
|
|
$sth = $pdo->prepare("SELECT int_id FROM ttrss_user_entries WHERE
|
|
|
|
ref_id = ? AND owner_uid = ? AND
|
|
|
|
feed_id = ? LIMIT 1");
|
2011-12-13 12:06:25 +00:00
|
|
|
|
2021-03-06 08:17:15 +00:00
|
|
|
$sth->execute([$ref_id, $feed_obj->owner_uid, $feed]);
|
2017-12-02 06:55:53 +00:00
|
|
|
|
|
|
|
if ($row = $sth->fetch())
|
|
|
|
$entry_int_id = $row['int_id'];
|
2011-12-13 12:06:25 +00:00
|
|
|
}
|
|
|
|
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log("resulting RID: $entry_ref_id, IID: $entry_int_id", Debug::LOG_VERBOSE);
|
2011-12-13 12:06:25 +00:00
|
|
|
|
2021-02-22 18:47:48 +00:00
|
|
|
if (Config::get(Config::DB_TYPE) == "pgsql")
|
2018-04-03 10:57:27 +00:00
|
|
|
$tsvector_qpart = "tsvector_combined = to_tsvector(:ts_lang, :ts_content),";
|
|
|
|
else
|
2015-08-04 09:52:49 +00:00
|
|
|
$tsvector_qpart = "";
|
2017-12-03 07:10:01 +00:00
|
|
|
|
2017-12-02 06:55:53 +00:00
|
|
|
$sth = $pdo->prepare("UPDATE ttrss_entries
|
2017-12-03 07:10:01 +00:00
|
|
|
SET title = :title,
|
2015-08-04 09:52:49 +00:00
|
|
|
$tsvector_qpart
|
2017-12-03 07:10:01 +00:00
|
|
|
content = :content,
|
|
|
|
content_hash = :content_hash,
|
|
|
|
updated = :updated,
|
2018-05-20 07:41:08 +00:00
|
|
|
date_updated = NOW(),
|
2017-12-03 07:10:01 +00:00
|
|
|
num_comments = :num_comments,
|
|
|
|
plugin_data = :plugin_data,
|
|
|
|
author = :author,
|
2018-11-03 12:00:58 +00:00
|
|
|
lang = :lang
|
2017-12-03 07:10:01 +00:00
|
|
|
WHERE id = :id");
|
|
|
|
|
2018-04-03 10:57:27 +00:00
|
|
|
$params = [":title" => $entry_title,
|
2017-12-03 10:32:24 +00:00
|
|
|
":content" => "$entry_content",
|
2017-12-03 07:10:01 +00:00
|
|
|
":content_hash" => $entry_current_hash,
|
|
|
|
":updated" => $entry_timestamp_fmt,
|
|
|
|
":num_comments" => (int)$num_comments,
|
|
|
|
":plugin_data" => $entry_plugin_data,
|
2017-12-03 10:32:24 +00:00
|
|
|
":author" => "$entry_author",
|
2017-12-03 07:10:01 +00:00
|
|
|
":lang" => $entry_language,
|
2018-04-03 10:57:27 +00:00
|
|
|
":id" => $ref_id];
|
|
|
|
|
2021-02-22 18:47:48 +00:00
|
|
|
if (Config::get(Config::DB_TYPE) == "pgsql") {
|
2018-04-03 10:57:27 +00:00
|
|
|
$params[":ts_lang"] = $feed_language;
|
2018-07-10 11:08:55 +00:00
|
|
|
$params[":ts_content"] = mb_substr(strip_tags($entry_title . " " . $entry_content), 0, 900000);
|
2018-04-03 10:57:27 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
$sth->execute($params);
|
2014-08-18 14:32:16 +00:00
|
|
|
|
2015-06-17 12:15:04 +00:00
|
|
|
// update aux data
|
2017-12-02 06:55:53 +00:00
|
|
|
$sth = $pdo->prepare("UPDATE ttrss_user_entries
|
|
|
|
SET score = ? WHERE ref_id = ?");
|
|
|
|
$sth->execute([$score, $ref_id]);
|
2015-06-17 12:15:04 +00:00
|
|
|
|
2021-03-06 08:17:15 +00:00
|
|
|
if ($feed_obj->mark_unread_on_update &&
|
2019-02-06 19:56:14 +00:00
|
|
|
!$entry_force_catchup &&
|
2020-09-22 11:54:15 +00:00
|
|
|
!self::find_article_filter($article_filters, 'catchup')) {
|
2019-02-06 19:56:14 +00:00
|
|
|
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log("article updated, marking unread as requested.", Debug::LOG_VERBOSE);
|
2015-12-05 00:10:57 +00:00
|
|
|
|
2017-12-02 06:55:53 +00:00
|
|
|
$sth = $pdo->prepare("UPDATE ttrss_user_entries
|
|
|
|
SET last_read = null, unread = true WHERE ref_id = ?");
|
|
|
|
$sth->execute([$ref_id]);
|
2019-02-06 19:56:14 +00:00
|
|
|
} else {
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log("article updated, but we're forbidden to mark it unread.", Debug::LOG_VERBOSE);
|
2011-12-13 12:06:25 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log("assigning labels [other]...", Debug::LOG_VERBOSE);
|
2015-03-03 21:26:52 +00:00
|
|
|
|
|
|
|
foreach ($article_labels as $label) {
|
2021-03-06 08:17:15 +00:00
|
|
|
Labels::add_article($entry_ref_id, $label[1], $feed_obj->owner_uid);
|
2015-03-03 21:26:52 +00:00
|
|
|
}
|
|
|
|
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log("assigning labels [filters]...", Debug::LOG_VERBOSE);
|
2011-12-13 12:06:25 +00:00
|
|
|
|
2020-09-22 11:54:15 +00:00
|
|
|
self::assign_article_to_label_filters($entry_ref_id, $article_filters,
|
2021-03-06 08:17:15 +00:00
|
|
|
$feed_obj->owner_uid, $article_labels);
|
2011-12-13 12:06:25 +00:00
|
|
|
|
2021-03-06 08:17:15 +00:00
|
|
|
if ($feed_obj->cache_images)
|
2020-09-22 11:54:15 +00:00
|
|
|
self::cache_enclosures($enclosures, $site_url);
|
2017-03-23 12:19:25 +00:00
|
|
|
|
2021-03-06 12:19:31 +00:00
|
|
|
if (Debug::get_loglevel() >= Debug::LOG_EXTENDED) {
|
|
|
|
Debug::log("article enclosures:", Debug::LOG_VERBOSE);
|
2011-12-13 12:06:25 +00:00
|
|
|
print_r($enclosures);
|
|
|
|
}
|
|
|
|
|
2017-12-02 06:55:53 +00:00
|
|
|
$esth = $pdo->prepare("SELECT id FROM ttrss_enclosures
|
2018-08-18 16:17:30 +00:00
|
|
|
WHERE content_url = ? AND content_type = ? AND post_id = ?");
|
2011-12-13 12:06:25 +00:00
|
|
|
|
2017-12-02 06:55:53 +00:00
|
|
|
$usth = $pdo->prepare("INSERT INTO ttrss_enclosures
|
|
|
|
(content_url, content_type, title, duration, post_id, width, height) VALUES
|
|
|
|
(?, ?, ?, ?, ?, ?, ?)");
|
2013-08-05 08:26:09 +00:00
|
|
|
|
2011-12-13 12:06:25 +00:00
|
|
|
foreach ($enclosures as $enc) {
|
2017-12-02 06:55:53 +00:00
|
|
|
$enc_url = $enc[0];
|
|
|
|
$enc_type = $enc[1];
|
2017-12-02 08:16:33 +00:00
|
|
|
$enc_dur = (int)$enc[2];
|
2017-12-02 06:55:53 +00:00
|
|
|
$enc_title = $enc[3];
|
2014-07-15 13:32:28 +00:00
|
|
|
$enc_width = intval($enc[4]);
|
|
|
|
$enc_height = intval($enc[5]);
|
2011-12-13 12:06:25 +00:00
|
|
|
|
2018-08-18 16:17:30 +00:00
|
|
|
$esth->execute([$enc_url, $enc_type, $entry_ref_id]);
|
2011-12-13 12:06:25 +00:00
|
|
|
|
2017-12-02 06:55:53 +00:00
|
|
|
if (!$esth->fetch()) {
|
|
|
|
$usth->execute([$enc_url, $enc_type, (string)$enc_title, $enc_dur, $entry_ref_id, $enc_width, $enc_height]);
|
2011-12-13 12:06:25 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
// check for manual tags (we have to do it here since they're loaded from filters)
|
|
|
|
|
|
|
|
foreach ($article_filters as $f) {
|
2012-08-30 14:50:56 +00:00
|
|
|
if ($f["type"] == "tag") {
|
2011-12-13 12:06:25 +00:00
|
|
|
|
2020-12-20 20:12:45 +00:00
|
|
|
$manual_tags = array_map('trim', explode(",", mb_strtolower($f["param"])));
|
2011-12-13 12:06:25 +00:00
|
|
|
|
|
|
|
foreach ($manual_tags as $tag) {
|
2019-11-20 15:56:34 +00:00
|
|
|
array_push($entry_tags, $tag);
|
2011-12-13 12:06:25 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
// Skip boring tags
|
|
|
|
|
2020-12-20 20:12:45 +00:00
|
|
|
$boring_tags = array_map('trim',
|
|
|
|
explode(",", mb_strtolower(
|
2021-03-06 08:17:15 +00:00
|
|
|
get_pref(Prefs::BLACKLISTED_TAGS, $feed_obj->owner_uid))));
|
2011-12-13 12:06:25 +00:00
|
|
|
|
2020-12-20 20:12:45 +00:00
|
|
|
$entry_tags = FeedItem_Common::normalize_categories(
|
|
|
|
array_unique(
|
|
|
|
array_diff($entry_tags, $boring_tags)));
|
2011-12-13 12:06:25 +00:00
|
|
|
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log("filtered tags: " . implode(", ", $entry_tags), Debug::LOG_VERBOSE);
|
2011-12-13 12:06:25 +00:00
|
|
|
|
|
|
|
// Save article tags in the database
|
|
|
|
|
2020-12-20 20:12:45 +00:00
|
|
|
if (count($entry_tags) > 0) {
|
2011-12-13 12:06:25 +00:00
|
|
|
|
2017-12-02 06:55:53 +00:00
|
|
|
$tsth = $pdo->prepare("SELECT id FROM ttrss_tags
|
|
|
|
WHERE tag_name = ? AND post_int_id = ? AND
|
|
|
|
owner_uid = ? LIMIT 1");
|
|
|
|
|
|
|
|
$usth = $pdo->prepare("INSERT INTO ttrss_tags
|
|
|
|
(owner_uid,tag_name,post_int_id)
|
|
|
|
VALUES (?, ?, ?)");
|
2011-12-13 12:06:25 +00:00
|
|
|
|
2020-12-20 20:12:45 +00:00
|
|
|
foreach ($entry_tags as $tag) {
|
2021-03-06 08:17:15 +00:00
|
|
|
$tsth->execute([$tag, $entry_int_id, $feed_obj->owner_uid]);
|
2011-12-13 12:06:25 +00:00
|
|
|
|
2017-12-02 06:55:53 +00:00
|
|
|
if (!$tsth->fetch()) {
|
2021-03-06 08:17:15 +00:00
|
|
|
$usth->execute([$feed_obj->owner_uid, $tag, $entry_int_id]);
|
2017-05-05 15:10:07 +00:00
|
|
|
}
|
2011-12-13 12:06:25 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
/* update the cache */
|
|
|
|
|
2017-12-02 06:55:53 +00:00
|
|
|
$tsth = $pdo->prepare("UPDATE ttrss_user_entries
|
|
|
|
SET tag_cache = ? WHERE ref_id = ?
|
|
|
|
AND owner_uid = ?");
|
2020-12-20 20:12:45 +00:00
|
|
|
|
|
|
|
$tsth->execute([
|
|
|
|
join(",", $entry_tags),
|
|
|
|
$entry_ref_id,
|
2021-03-06 08:17:15 +00:00
|
|
|
$feed_obj->owner_uid
|
2020-12-20 20:12:45 +00:00
|
|
|
]);
|
2011-12-13 12:06:25 +00:00
|
|
|
}
|
|
|
|
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log("article processed.", Debug::LOG_VERBOSE);
|
2017-12-02 08:16:33 +00:00
|
|
|
|
|
|
|
$pdo->commit();
|
2011-12-13 12:06:25 +00:00
|
|
|
}
|
|
|
|
|
2021-03-06 08:17:15 +00:00
|
|
|
Debug::log("=================================================================================================================================",
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::LOG_VERBOSE);
|
2021-03-06 08:17:15 +00:00
|
|
|
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log("purging feed...", Debug::LOG_VERBOSE);
|
2011-12-13 12:06:25 +00:00
|
|
|
|
2021-02-15 12:43:07 +00:00
|
|
|
Feeds::_purge($feed, 0);
|
2011-12-13 12:06:25 +00:00
|
|
|
|
2021-03-06 08:17:15 +00:00
|
|
|
$feed_obj->set([
|
|
|
|
'last_updated' => Db::NOW(),
|
|
|
|
'last_unconditional' => Db::NOW(),
|
|
|
|
'last_successful_update' => Db::NOW(),
|
|
|
|
'last_error' => '',
|
|
|
|
]);
|
|
|
|
|
|
|
|
$feed_obj->save();
|
|
|
|
|
2011-12-13 12:06:25 +00:00
|
|
|
} else {
|
|
|
|
|
2017-12-02 06:55:53 +00:00
|
|
|
$error_msg = mb_substr($rss->error(), 0, 245);
|
2011-12-13 12:06:25 +00:00
|
|
|
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log("fetch error: $error_msg", Debug::LOG_VERBOSE);
|
2013-10-29 08:15:26 +00:00
|
|
|
|
|
|
|
if (count($rss->errors()) > 1) {
|
|
|
|
foreach ($rss->errors() as $error) {
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log("+ $error", Debug::LOG_VERBOSE);
|
2013-10-29 08:15:26 +00:00
|
|
|
}
|
|
|
|
}
|
2011-12-13 12:06:25 +00:00
|
|
|
|
2021-03-06 08:17:15 +00:00
|
|
|
$feed_obj->set([
|
|
|
|
'last_updated' => Db::NOW(),
|
|
|
|
'last_unconditional' => Db::NOW(),
|
|
|
|
'last_error' => $error_msg,
|
|
|
|
]);
|
|
|
|
|
|
|
|
$feed_obj->save();
|
2011-12-13 12:06:25 +00:00
|
|
|
|
2013-10-25 10:42:43 +00:00
|
|
|
unset($rss);
|
2019-05-12 07:13:22 +00:00
|
|
|
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log("update failed.", Debug::LOG_VERBOSE);
|
2017-12-02 06:55:53 +00:00
|
|
|
return false;
|
2013-10-25 10:42:43 +00:00
|
|
|
}
|
2011-12-13 12:06:25 +00:00
|
|
|
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log("update done.", Debug::LOG_VERBOSE);
|
2017-04-26 12:29:22 +00:00
|
|
|
return true;
|
2011-12-13 12:06:25 +00:00
|
|
|
}
|
|
|
|
|
2020-06-15 08:58:59 +00:00
|
|
|
/* TODO: move to DiskCache? */
|
2018-11-30 05:34:29 +00:00
|
|
|
static function cache_enclosures($enclosures, $site_url) {
|
2019-08-14 09:15:56 +00:00
|
|
|
$cache = new DiskCache("images");
|
2017-03-23 12:19:25 +00:00
|
|
|
|
2021-02-15 13:11:30 +00:00
|
|
|
if ($cache->is_writable()) {
|
2019-08-14 09:15:56 +00:00
|
|
|
foreach ($enclosures as $enc) {
|
2017-03-23 12:19:25 +00:00
|
|
|
|
2019-08-14 09:15:56 +00:00
|
|
|
if (preg_match("/(image|audio|video)/", $enc[1])) {
|
|
|
|
$src = rewrite_relative_url($site_url, $enc[0]);
|
2017-03-23 12:19:25 +00:00
|
|
|
|
2019-08-14 09:15:56 +00:00
|
|
|
$local_filename = sha1($src);
|
2017-03-23 12:19:25 +00:00
|
|
|
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log("cache_enclosures: downloading: $src to $local_filename", Debug::LOG_VERBOSE);
|
2017-03-23 12:19:25 +00:00
|
|
|
|
2019-08-14 09:15:56 +00:00
|
|
|
if (!$cache->exists($local_filename)) {
|
2020-09-22 06:04:33 +00:00
|
|
|
$file_content = UrlHelper::fetch(array("url" => $src,
|
2019-11-25 06:48:24 +00:00
|
|
|
"http_referrer" => $src,
|
2021-02-22 19:35:27 +00:00
|
|
|
"max_size" => Config::get(Config::MAX_CACHE_FILE_SIZE)));
|
2017-03-23 12:19:25 +00:00
|
|
|
|
2019-08-14 09:44:50 +00:00
|
|
|
if ($file_content) {
|
2019-08-14 09:15:56 +00:00
|
|
|
$cache->put($local_filename, $file_content);
|
2019-11-25 06:48:24 +00:00
|
|
|
} else {
|
2021-02-28 07:12:57 +00:00
|
|
|
Debug::log("cache_enclosures: failed with ".UrlHelper::$fetch_last_error_code.": ".UrlHelper::$fetch_last_error);
|
2019-08-14 09:15:56 +00:00
|
|
|
}
|
|
|
|
} else if (is_writable($local_filename)) {
|
|
|
|
$cache->touch($local_filename);
|
2017-03-23 12:19:25 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2020-06-15 08:58:59 +00:00
|
|
|
/* TODO: move to DiskCache? */
|
2020-04-29 16:29:36 +00:00
|
|
|
static function cache_media_url($cache, $url, $site_url) {
|
|
|
|
$url = rewrite_relative_url($site_url, $url);
|
|
|
|
$local_filename = sha1($url);
|
|
|
|
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log("cache_media: checking $url", Debug::LOG_VERBOSE);
|
2020-04-29 16:29:36 +00:00
|
|
|
|
|
|
|
if (!$cache->exists($local_filename)) {
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log("cache_media: downloading: $url to $local_filename", Debug::LOG_VERBOSE);
|
2020-04-29 16:29:36 +00:00
|
|
|
|
2020-09-22 06:04:33 +00:00
|
|
|
$file_content = UrlHelper::fetch(array("url" => $url,
|
2020-04-29 16:29:36 +00:00
|
|
|
"http_referrer" => $url,
|
2021-02-22 19:35:27 +00:00
|
|
|
"max_size" => Config::get(Config::MAX_CACHE_FILE_SIZE)));
|
2020-04-29 16:29:36 +00:00
|
|
|
|
|
|
|
if ($file_content) {
|
|
|
|
$cache->put($local_filename, $file_content);
|
|
|
|
} else {
|
2021-02-28 07:12:57 +00:00
|
|
|
Debug::log("cache_media: failed with ".UrlHelper::$fetch_last_error_code.": ".UrlHelper::$fetch_last_error);
|
2020-04-29 16:29:36 +00:00
|
|
|
}
|
2021-02-15 13:11:30 +00:00
|
|
|
} else if ($cache->is_writable($local_filename)) {
|
2020-04-29 16:29:36 +00:00
|
|
|
$cache->touch($local_filename);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2020-06-15 08:58:59 +00:00
|
|
|
/* TODO: move to DiskCache? */
|
2018-11-30 05:34:29 +00:00
|
|
|
static function cache_media($html, $site_url) {
|
2019-08-14 09:15:56 +00:00
|
|
|
$cache = new DiskCache("images");
|
2011-12-26 09:11:08 +00:00
|
|
|
|
2021-02-15 13:11:30 +00:00
|
|
|
if ($html && $cache->is_writable()) {
|
2019-08-14 09:15:56 +00:00
|
|
|
$doc = new DOMDocument();
|
2020-10-01 10:20:07 +00:00
|
|
|
if (@$doc->loadHTML($html)) {
|
2019-08-14 09:15:56 +00:00
|
|
|
$xpath = new DOMXPath($doc);
|
2011-12-26 09:11:08 +00:00
|
|
|
|
2020-04-29 16:29:36 +00:00
|
|
|
$entries = $xpath->query('(//img[@src]|//source[@src|@srcset]|//video[@poster|@src])');
|
2011-12-26 09:11:08 +00:00
|
|
|
|
2019-08-14 09:15:56 +00:00
|
|
|
foreach ($entries as $entry) {
|
2020-03-12 10:08:33 +00:00
|
|
|
foreach (array('src', 'poster') as $attr) {
|
|
|
|
if ($entry->hasAttribute($attr) && strpos($entry->getAttribute($attr), "data:") !== 0) {
|
2020-09-22 11:54:15 +00:00
|
|
|
self::cache_media_url($cache, $entry->getAttribute($attr), $site_url);
|
2020-04-29 16:29:36 +00:00
|
|
|
}
|
|
|
|
}
|
2018-08-16 08:56:51 +00:00
|
|
|
|
2020-04-29 16:29:36 +00:00
|
|
|
if ($entry->hasAttribute("srcset")) {
|
2020-09-22 11:54:15 +00:00
|
|
|
$matches = self::decode_srcset($entry->getAttribute('srcset'));
|
2011-12-26 09:11:08 +00:00
|
|
|
|
2020-06-15 08:58:59 +00:00
|
|
|
for ($i = 0; $i < count($matches); $i++) {
|
2020-09-22 11:54:15 +00:00
|
|
|
self::cache_media_url($cache, $matches[$i]["url"], $site_url);
|
2019-06-20 05:40:02 +00:00
|
|
|
}
|
2011-12-26 09:11:08 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2018-11-30 05:34:29 +00:00
|
|
|
static function expire_error_log() {
|
|
|
|
Debug::log("Removing old error log entries...");
|
2013-04-16 16:34:20 +00:00
|
|
|
|
2017-12-02 06:55:53 +00:00
|
|
|
$pdo = Db::pdo();
|
|
|
|
|
2021-02-22 18:47:48 +00:00
|
|
|
if (Config::get(Config::DB_TYPE) == "pgsql") {
|
2017-12-02 06:55:53 +00:00
|
|
|
$pdo->query("DELETE FROM ttrss_error_log
|
2013-04-16 16:34:20 +00:00
|
|
|
WHERE created_at < NOW() - INTERVAL '7 days'");
|
|
|
|
} else {
|
2017-12-02 06:55:53 +00:00
|
|
|
$pdo->query("DELETE FROM ttrss_error_log
|
2013-04-16 16:34:20 +00:00
|
|
|
WHERE created_at < DATE_SUB(NOW(), INTERVAL 7 DAY)");
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2021-01-17 11:55:11 +00:00
|
|
|
// deprecated; table not used
|
2019-03-06 16:06:05 +00:00
|
|
|
static function expire_feed_archive() {
|
|
|
|
$pdo = Db::pdo();
|
|
|
|
|
2021-01-17 11:55:11 +00:00
|
|
|
$pdo->query("DELETE FROM ttrss_archived_feeds");
|
2019-03-06 16:06:05 +00:00
|
|
|
}
|
|
|
|
|
2018-11-30 05:34:29 +00:00
|
|
|
static function expire_lock_files() {
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log("Removing old lock files...", Debug::LOG_VERBOSE);
|
2012-12-12 05:46:02 +00:00
|
|
|
|
|
|
|
$num_deleted = 0;
|
|
|
|
|
2021-02-22 18:47:48 +00:00
|
|
|
if (is_writable(Config::get(Config::LOCK_DIRECTORY))) {
|
|
|
|
$files = glob(Config::get(Config::LOCK_DIRECTORY) . "/*.lock");
|
2012-12-12 05:46:02 +00:00
|
|
|
|
|
|
|
if ($files) {
|
|
|
|
foreach ($files as $file) {
|
2013-04-19 19:25:26 +00:00
|
|
|
if (!file_is_locked(basename($file)) && time() - filemtime($file) > 86400*2) {
|
2012-12-12 05:46:02 +00:00
|
|
|
unlink($file);
|
|
|
|
++$num_deleted;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2018-11-30 05:34:29 +00:00
|
|
|
Debug::log("Removed $num_deleted old lock files.");
|
2012-12-12 05:46:02 +00:00
|
|
|
}
|
|
|
|
|
2012-01-10 08:49:10 +00:00
|
|
|
/**
|
2017-05-05 15:10:07 +00:00
|
|
|
* Source: http://www.php.net/manual/en/function.parse-url.php#104527
|
|
|
|
* Returns the url query as associative array
|
|
|
|
*
|
|
|
|
* @param string query
|
|
|
|
* @return array params
|
|
|
|
*/
|
2021-02-06 14:19:07 +00:00
|
|
|
/* static function convertUrlQuery($query) {
|
2012-01-10 08:49:10 +00:00
|
|
|
$queryParts = explode('&', $query);
|
|
|
|
|
|
|
|
$params = array();
|
|
|
|
|
|
|
|
foreach ($queryParts as $param) {
|
|
|
|
$item = explode('=', $param);
|
|
|
|
$params[$item[0]] = $item[1];
|
|
|
|
}
|
|
|
|
|
|
|
|
return $params;
|
2021-02-06 14:19:07 +00:00
|
|
|
} */
|
2012-10-30 08:11:46 +00:00
|
|
|
|
2018-12-06 16:15:00 +00:00
|
|
|
static function get_article_filters($filters, $title, $content, $link, $author, $tags, &$matched_rules = false, &$matched_filters = false) {
|
2012-10-30 08:11:46 +00:00
|
|
|
$matches = array();
|
|
|
|
|
|
|
|
foreach ($filters as $filter) {
|
2021-03-09 06:31:52 +00:00
|
|
|
$match_any_rule = $filter["match_any_rule"] ?? false;
|
|
|
|
$inverse = $filter["inverse"] ?? false;
|
2012-10-30 08:11:46 +00:00
|
|
|
$filter_match = false;
|
2021-02-06 14:19:07 +00:00
|
|
|
$last_processed_rule = false;
|
2012-10-30 08:11:46 +00:00
|
|
|
|
|
|
|
foreach ($filter["rules"] as $rule) {
|
|
|
|
$match = false;
|
2021-02-06 14:19:07 +00:00
|
|
|
$reg_exp = str_replace('/', '\/', (string)$rule["reg_exp"]);
|
2020-02-28 10:53:45 +00:00
|
|
|
$reg_exp = str_replace("\n", "", $reg_exp); // reg_exp may be formatted with CRs now because of textarea, we need to strip those
|
2021-03-09 10:42:28 +00:00
|
|
|
$rule_inverse = $rule["inverse"] ?? false;
|
2021-02-06 14:19:07 +00:00
|
|
|
$last_processed_rule = $rule;
|
2012-10-30 08:11:46 +00:00
|
|
|
|
2021-02-06 14:19:07 +00:00
|
|
|
if (empty($reg_exp))
|
2012-10-30 08:11:46 +00:00
|
|
|
continue;
|
|
|
|
|
|
|
|
switch ($rule["type"]) {
|
2017-05-05 15:10:07 +00:00
|
|
|
case "title":
|
|
|
|
$match = @preg_match("/$reg_exp/iu", $title);
|
|
|
|
break;
|
|
|
|
case "content":
|
|
|
|
// we don't need to deal with multiline regexps
|
2021-02-06 14:19:07 +00:00
|
|
|
$content = (string)preg_replace("/[\r\n\t]/", "", $content);
|
2012-11-20 17:27:53 +00:00
|
|
|
|
2017-05-05 15:10:07 +00:00
|
|
|
$match = @preg_match("/$reg_exp/iu", $content);
|
|
|
|
break;
|
|
|
|
case "both":
|
|
|
|
// we don't need to deal with multiline regexps
|
2021-02-06 14:19:07 +00:00
|
|
|
$content = (string)preg_replace("/[\r\n\t]/", "", $content);
|
2012-11-20 17:27:53 +00:00
|
|
|
|
2017-05-05 15:10:07 +00:00
|
|
|
$match = (@preg_match("/$reg_exp/iu", $title) || @preg_match("/$reg_exp/iu", $content));
|
|
|
|
break;
|
|
|
|
case "link":
|
|
|
|
$match = @preg_match("/$reg_exp/iu", $link);
|
|
|
|
break;
|
|
|
|
case "author":
|
|
|
|
$match = @preg_match("/$reg_exp/iu", $author);
|
|
|
|
break;
|
|
|
|
case "tag":
|
|
|
|
foreach ($tags as $tag) {
|
|
|
|
if (@preg_match("/$reg_exp/iu", $tag)) {
|
|
|
|
$match = true;
|
|
|
|
break;
|
|
|
|
}
|
2013-04-26 11:54:04 +00:00
|
|
|
}
|
2017-05-05 15:10:07 +00:00
|
|
|
break;
|
2012-10-30 08:11:46 +00:00
|
|
|
}
|
|
|
|
|
2013-03-25 15:46:43 +00:00
|
|
|
if ($rule_inverse) $match = !$match;
|
|
|
|
|
2012-10-30 08:11:46 +00:00
|
|
|
if ($match_any_rule) {
|
|
|
|
if ($match) {
|
|
|
|
$filter_match = true;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
} else {
|
|
|
|
$filter_match = $match;
|
|
|
|
if (!$match) {
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2013-03-25 15:46:43 +00:00
|
|
|
if ($inverse) $filter_match = !$filter_match;
|
|
|
|
|
2012-10-30 08:11:46 +00:00
|
|
|
if ($filter_match) {
|
2021-02-06 14:19:07 +00:00
|
|
|
if (is_array($matched_rules)) array_push($matched_rules, $last_processed_rule);
|
2018-12-06 16:15:00 +00:00
|
|
|
if (is_array($matched_filters)) array_push($matched_filters, $filter);
|
2016-08-26 11:45:34 +00:00
|
|
|
|
2012-10-30 08:11:46 +00:00
|
|
|
foreach ($filter["actions"] AS $action) {
|
|
|
|
array_push($matches, $action);
|
2013-03-31 15:28:36 +00:00
|
|
|
|
|
|
|
// if Stop action encountered, perform no further processing
|
2015-12-03 12:33:47 +00:00
|
|
|
if (isset($action["type"]) && $action["type"] == "stop") return $matches;
|
2012-10-30 08:11:46 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
return $matches;
|
|
|
|
}
|
|
|
|
|
2017-05-05 15:10:07 +00:00
|
|
|
static function find_article_filter($filters, $filter_name) {
|
2012-10-30 08:11:46 +00:00
|
|
|
foreach ($filters as $f) {
|
|
|
|
if ($f["type"] == $filter_name) {
|
|
|
|
return $f;
|
|
|
|
};
|
|
|
|
}
|
|
|
|
return false;
|
|
|
|
}
|
|
|
|
|
2017-05-05 15:10:07 +00:00
|
|
|
static function find_article_filters($filters, $filter_name) {
|
2012-10-30 08:11:46 +00:00
|
|
|
$results = array();
|
|
|
|
|
|
|
|
foreach ($filters as $f) {
|
|
|
|
if ($f["type"] == $filter_name) {
|
|
|
|
array_push($results, $f);
|
|
|
|
};
|
|
|
|
}
|
|
|
|
return $results;
|
|
|
|
}
|
|
|
|
|
2017-05-05 15:10:07 +00:00
|
|
|
static function calculate_article_score($filters) {
|
2012-10-30 08:11:46 +00:00
|
|
|
$score = 0;
|
|
|
|
|
|
|
|
foreach ($filters as $f) {
|
|
|
|
if ($f["type"] == "score") {
|
|
|
|
$score += $f["param"];
|
|
|
|
};
|
|
|
|
}
|
|
|
|
return $score;
|
|
|
|
}
|
|
|
|
|
2017-05-05 15:10:07 +00:00
|
|
|
static function labels_contains_caption($labels, $caption) {
|
2012-10-30 08:58:35 +00:00
|
|
|
foreach ($labels as $label) {
|
|
|
|
if ($label[1] == $caption) {
|
|
|
|
return true;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
return false;
|
|
|
|
}
|
|
|
|
|
2017-05-05 15:10:07 +00:00
|
|
|
static function assign_article_to_label_filters($id, $filters, $owner_uid, $article_labels) {
|
2012-10-30 08:11:46 +00:00
|
|
|
foreach ($filters as $f) {
|
|
|
|
if ($f["type"] == "label") {
|
2020-09-22 11:54:15 +00:00
|
|
|
if (!self::labels_contains_caption($article_labels, $f["param"])) {
|
2017-05-04 12:57:40 +00:00
|
|
|
Labels::add_article($id, $f["param"], $owner_uid);
|
2012-10-30 08:58:35 +00:00
|
|
|
}
|
|
|
|
}
|
2012-10-30 08:11:46 +00:00
|
|
|
}
|
|
|
|
}
|
2012-11-23 09:22:34 +00:00
|
|
|
|
2017-05-05 15:10:07 +00:00
|
|
|
static function make_guid_from_title($title) {
|
2013-01-22 18:32:17 +00:00
|
|
|
return preg_replace("/[ \"\',.:;]/", "-",
|
|
|
|
mb_strtolower(strip_tags($title), 'utf-8'));
|
|
|
|
}
|
|
|
|
|
2020-01-24 11:25:31 +00:00
|
|
|
/* counter cache is no longer used, if called truncate leftover data */
|
2018-11-30 05:34:29 +00:00
|
|
|
static function cleanup_counters_cache() {
|
2017-12-02 06:55:53 +00:00
|
|
|
$pdo = Db::pdo();
|
|
|
|
|
2020-01-24 11:25:31 +00:00
|
|
|
$pdo->query("DELETE FROM ttrss_counters_cache");
|
|
|
|
$pdo->query("DELETE FROM ttrss_cat_counters_cache");
|
2015-03-06 11:14:51 +00:00
|
|
|
}
|
|
|
|
|
2020-09-30 14:03:16 +00:00
|
|
|
static function disable_failed_feeds() {
|
2021-02-23 06:01:27 +00:00
|
|
|
if (Config::get(Config::DAEMON_UNSUCCESSFUL_DAYS_LIMIT) > 0) {
|
2020-09-30 14:03:16 +00:00
|
|
|
|
|
|
|
$pdo = Db::pdo();
|
|
|
|
|
|
|
|
$pdo->beginTransaction();
|
|
|
|
|
2021-02-23 06:01:27 +00:00
|
|
|
$days = Config::get(Config::DAEMON_UNSUCCESSFUL_DAYS_LIMIT);
|
2020-09-30 14:03:16 +00:00
|
|
|
|
2021-02-22 18:47:48 +00:00
|
|
|
if (Config::get(Config::DB_TYPE) == "pgsql") {
|
2020-10-01 12:59:40 +00:00
|
|
|
$interval_query = "last_successful_update < NOW() - INTERVAL '$days days' AND last_updated > NOW() - INTERVAL '1 days'";
|
2021-02-22 18:47:48 +00:00
|
|
|
} else /* if (Config::get(Config::DB_TYPE) == "mysql") */ {
|
2020-10-01 12:59:40 +00:00
|
|
|
$interval_query = "last_successful_update < DATE_SUB(NOW(), INTERVAL $days DAY) AND last_updated > DATE_SUB(NOW(), INTERVAL 1 DAY)";
|
2020-09-30 14:03:16 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
$sth = $pdo->prepare("SELECT id, title, owner_uid
|
|
|
|
FROM ttrss_feeds
|
|
|
|
WHERE update_interval != -1 AND last_successful_update IS NOT NULL AND $interval_query");
|
|
|
|
|
|
|
|
$sth->execute();
|
|
|
|
|
|
|
|
while ($row = $sth->fetch()) {
|
2021-02-25 12:49:30 +00:00
|
|
|
Logger::log(E_USER_NOTICE,
|
2020-09-30 14:03:16 +00:00
|
|
|
sprintf("Auto disabling feed %d (%s, UID: %d) because it failed to update for %d days.",
|
2021-02-23 06:01:27 +00:00
|
|
|
$row["id"], clean($row["title"]), $row["owner_uid"], Config::get(Config::DAEMON_UNSUCCESSFUL_DAYS_LIMIT)));
|
2020-09-30 14:03:16 +00:00
|
|
|
|
2020-10-01 12:59:40 +00:00
|
|
|
Debug::log(sprintf("Auto-disabling feed %d (%s) (failed to update for %d days).", $row["id"],
|
2021-02-23 06:01:27 +00:00
|
|
|
clean($row["title"]), Config::get(Config::DAEMON_UNSUCCESSFUL_DAYS_LIMIT)));
|
2020-09-30 14:03:16 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
$sth = $pdo->prepare("UPDATE ttrss_feeds SET update_interval = -1 WHERE
|
|
|
|
update_interval != -1 AND last_successful_update IS NOT NULL AND $interval_query");
|
|
|
|
$sth->execute();
|
|
|
|
|
|
|
|
$pdo->commit();
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2017-05-05 15:10:07 +00:00
|
|
|
static function housekeeping_user($owner_uid) {
|
2016-08-11 12:01:01 +00:00
|
|
|
$tmph = new PluginHost();
|
|
|
|
|
2020-09-22 06:04:33 +00:00
|
|
|
UserHelper::load_user_plugins($owner_uid, $tmph);
|
2016-08-11 12:01:01 +00:00
|
|
|
|
2021-02-08 11:24:45 +00:00
|
|
|
$tmph->run_hooks(PluginHost::HOOK_HOUSE_KEEPING);
|
2016-08-11 12:01:01 +00:00
|
|
|
}
|
|
|
|
|
2018-11-30 05:34:29 +00:00
|
|
|
static function housekeeping_common() {
|
2019-08-13 11:13:42 +00:00
|
|
|
DiskCache::expire();
|
|
|
|
|
2020-09-22 11:54:15 +00:00
|
|
|
self::expire_lock_files();
|
|
|
|
self::expire_error_log();
|
|
|
|
self::expire_feed_archive();
|
|
|
|
self::cleanup_feed_browser();
|
2020-09-29 09:46:55 +00:00
|
|
|
self::cleanup_feed_icons();
|
2020-09-30 14:03:16 +00:00
|
|
|
self::disable_failed_feeds();
|
2013-04-24 10:54:59 +00:00
|
|
|
|
2021-02-15 12:52:28 +00:00
|
|
|
Article::_purge_orphans();
|
2020-09-22 11:54:15 +00:00
|
|
|
self::cleanup_counters_cache();
|
2013-08-02 10:06:18 +00:00
|
|
|
|
2021-02-08 11:24:45 +00:00
|
|
|
PluginHost::getInstance()->run_hooks(PluginHost::HOOK_HOUSE_KEEPING);
|
2013-04-24 10:54:59 +00:00
|
|
|
}
|
2017-04-26 17:24:18 +00:00
|
|
|
|
2021-03-06 12:19:31 +00:00
|
|
|
static function update_favicon(string $site_url, int $feed) {
|
2021-02-22 19:35:27 +00:00
|
|
|
$icon_file = Config::get(Config::ICONS_DIR) . "/$feed.ico";
|
2017-05-04 12:00:21 +00:00
|
|
|
|
2021-02-21 18:40:43 +00:00
|
|
|
$favicon_url = self::get_favicon_url($site_url);
|
|
|
|
if (!$favicon_url) {
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log("favicon: couldn't find favicon URL in $site_url", Debug::LOG_VERBOSE);
|
2021-02-21 18:40:43 +00:00
|
|
|
return false;
|
|
|
|
}
|
2017-05-04 12:00:21 +00:00
|
|
|
|
2021-02-21 18:40:43 +00:00
|
|
|
// Limiting to "image" type misses those served with text/plain
|
2021-02-21 23:30:31 +00:00
|
|
|
$contents = UrlHelper::fetch([
|
|
|
|
'url' => $favicon_url,
|
2021-02-22 19:35:27 +00:00
|
|
|
'max_size' => Config::get(Config::MAX_FAVICON_FILE_SIZE),
|
2021-02-21 23:30:31 +00:00
|
|
|
//'type' => 'image',
|
|
|
|
]);
|
2021-02-21 18:40:43 +00:00
|
|
|
if (!$contents) {
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log("favicon: fetching $favicon_url failed", Debug::LOG_VERBOSE);
|
2021-02-21 18:40:43 +00:00
|
|
|
return false;
|
|
|
|
}
|
2017-05-04 12:00:21 +00:00
|
|
|
|
2021-02-21 18:40:43 +00:00
|
|
|
// Crude image type matching.
|
|
|
|
// Patterns gleaned from the file(1) source code.
|
|
|
|
if (preg_match('/^\x00\x00\x01\x00/', $contents)) {
|
|
|
|
// 0 string \000\000\001\000 MS Windows icon resource
|
2021-03-06 12:19:31 +00:00
|
|
|
//error_log("update_favicon: favicon_url=$favicon_url isa MS Windows icon resource");
|
2021-02-21 18:40:43 +00:00
|
|
|
}
|
|
|
|
elseif (preg_match('/^GIF8/', $contents)) {
|
|
|
|
// 0 string GIF8 GIF image data
|
2021-03-06 12:19:31 +00:00
|
|
|
//error_log("update_favicon: favicon_url=$favicon_url isa GIF image");
|
2021-02-21 18:40:43 +00:00
|
|
|
}
|
|
|
|
elseif (preg_match('/^\x89PNG\x0d\x0a\x1a\x0a/', $contents)) {
|
|
|
|
// 0 string \x89PNG\x0d\x0a\x1a\x0a PNG image data
|
2021-03-06 12:19:31 +00:00
|
|
|
//error_log("update_favicon: favicon_url=$favicon_url isa PNG image");
|
2021-02-21 18:40:43 +00:00
|
|
|
}
|
|
|
|
elseif (preg_match('/^\xff\xd8/', $contents)) {
|
|
|
|
// 0 beshort 0xffd8 JPEG image data
|
2021-03-06 12:19:31 +00:00
|
|
|
//error_log("update_favicon: favicon_url=$favicon_url isa JPG image");
|
2021-02-21 18:40:43 +00:00
|
|
|
}
|
|
|
|
elseif (preg_match('/^BM/', $contents)) {
|
|
|
|
// 0 string BM PC bitmap (OS2, Windows BMP files)
|
2021-03-06 12:19:31 +00:00
|
|
|
//error_log("update_favicon, favicon_url=$favicon_url isa BMP image");
|
2021-02-21 18:40:43 +00:00
|
|
|
}
|
|
|
|
else {
|
2021-03-06 12:19:31 +00:00
|
|
|
//error_log("update_favicon: favicon_url=$favicon_url isa UNKNOWN type");
|
|
|
|
Debug::log("favicon $favicon_url type is unknown (not updating)", Debug::LOG_VERBOSE);
|
2021-02-21 18:40:43 +00:00
|
|
|
return false;
|
|
|
|
}
|
2017-05-04 12:00:21 +00:00
|
|
|
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log("favicon: saving to $icon_file", Debug::LOG_VERBOSE);
|
2021-02-21 18:40:43 +00:00
|
|
|
|
|
|
|
$fp = @fopen($icon_file, "w");
|
|
|
|
if (!$fp) {
|
2021-03-06 12:19:31 +00:00
|
|
|
Debug::log("favicon: failed to open $icon_file for writing", Debug::LOG_VERBOSE);
|
2021-02-21 18:40:43 +00:00
|
|
|
return false;
|
2017-05-04 12:00:21 +00:00
|
|
|
}
|
2021-02-21 18:40:43 +00:00
|
|
|
|
|
|
|
fwrite($fp, $contents);
|
|
|
|
fclose($fp);
|
|
|
|
chmod($icon_file, 0644);
|
|
|
|
clearstatcache();
|
|
|
|
|
|
|
|
return $icon_file;
|
2017-05-04 12:00:21 +00:00
|
|
|
}
|
2017-05-05 15:10:07 +00:00
|
|
|
|
2018-12-21 14:50:16 +00:00
|
|
|
static function is_gzipped($feed_data) {
|
2019-11-12 04:11:10 +00:00
|
|
|
return strpos(substr($feed_data, 0, 3),
|
|
|
|
"\x1f" . "\x8b" . "\x08", 0) === 0;
|
2018-12-14 11:55:36 +00:00
|
|
|
}
|
2017-05-05 15:10:07 +00:00
|
|
|
|
2021-03-06 12:19:31 +00:00
|
|
|
static function load_filters(int $feed_id, int $owner_uid) {
|
2019-06-20 05:14:06 +00:00
|
|
|
$filters = array();
|
|
|
|
|
|
|
|
$feed_id = (int) $feed_id;
|
2021-03-02 06:36:44 +00:00
|
|
|
$cat_id = Feeds::_cat_of($feed_id);
|
2019-06-20 05:14:06 +00:00
|
|
|
|
2021-03-01 19:25:41 +00:00
|
|
|
if (!$cat_id)
|
2019-06-20 05:14:06 +00:00
|
|
|
$null_cat_qpart = "cat_id IS NULL OR";
|
|
|
|
else
|
|
|
|
$null_cat_qpart = "";
|
|
|
|
|
|
|
|
$pdo = Db::pdo();
|
|
|
|
|
|
|
|
$sth = $pdo->prepare("SELECT * FROM ttrss_filters2 WHERE
|
|
|
|
owner_uid = ? AND enabled = true ORDER BY order_id, title");
|
|
|
|
$sth->execute([$owner_uid]);
|
|
|
|
|
|
|
|
$check_cats = array_merge(
|
2021-02-15 12:43:07 +00:00
|
|
|
Feeds::_get_parent_cats($cat_id, $owner_uid),
|
2019-06-20 05:14:06 +00:00
|
|
|
[$cat_id]);
|
|
|
|
|
|
|
|
$check_cats_str = join(",", $check_cats);
|
|
|
|
$check_cats_fullids = array_map(function($a) { return "CAT:$a"; }, $check_cats);
|
|
|
|
|
|
|
|
while ($line = $sth->fetch()) {
|
|
|
|
$filter_id = $line["id"];
|
|
|
|
|
|
|
|
$match_any_rule = sql_bool_to_bool($line["match_any_rule"]);
|
|
|
|
|
|
|
|
$sth2 = $pdo->prepare("SELECT
|
|
|
|
r.reg_exp, r.inverse, r.feed_id, r.cat_id, r.cat_filter, r.match_on, t.name AS type_name
|
|
|
|
FROM ttrss_filters2_rules AS r,
|
|
|
|
ttrss_filter_types AS t
|
|
|
|
WHERE
|
|
|
|
(match_on IS NOT NULL OR
|
|
|
|
(($null_cat_qpart (cat_id IS NULL AND cat_filter = false) OR cat_id IN ($check_cats_str)) AND
|
|
|
|
(feed_id IS NULL OR feed_id = ?))) AND
|
|
|
|
filter_type = t.id AND filter_id = ?");
|
|
|
|
$sth2->execute([$feed_id, $filter_id]);
|
|
|
|
|
|
|
|
$rules = array();
|
|
|
|
$actions = array();
|
|
|
|
|
|
|
|
while ($rule_line = $sth2->fetch()) {
|
|
|
|
# print_r($rule_line);
|
|
|
|
|
|
|
|
if ($rule_line["match_on"]) {
|
|
|
|
$match_on = json_decode($rule_line["match_on"], true);
|
|
|
|
|
|
|
|
if (in_array("0", $match_on) || in_array($feed_id, $match_on) || count(array_intersect($check_cats_fullids, $match_on)) > 0) {
|
|
|
|
|
|
|
|
$rule = array();
|
|
|
|
$rule["reg_exp"] = $rule_line["reg_exp"];
|
|
|
|
$rule["type"] = $rule_line["type_name"];
|
|
|
|
$rule["inverse"] = sql_bool_to_bool($rule_line["inverse"]);
|
|
|
|
|
|
|
|
array_push($rules, $rule);
|
|
|
|
} else if (!$match_any_rule) {
|
|
|
|
// this filter contains a rule that doesn't match to this feed/category combination
|
|
|
|
// thus filter has to be rejected
|
|
|
|
|
|
|
|
$rules = [];
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
|
|
|
|
} else {
|
|
|
|
|
|
|
|
$rule = array();
|
|
|
|
$rule["reg_exp"] = $rule_line["reg_exp"];
|
|
|
|
$rule["type"] = $rule_line["type_name"];
|
|
|
|
$rule["inverse"] = sql_bool_to_bool($rule_line["inverse"]);
|
|
|
|
|
|
|
|
array_push($rules, $rule);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
if (count($rules) > 0) {
|
|
|
|
$sth2 = $pdo->prepare("SELECT a.action_param,t.name AS type_name
|
|
|
|
FROM ttrss_filters2_actions AS a,
|
|
|
|
ttrss_filter_actions AS t
|
|
|
|
WHERE
|
|
|
|
action_id = t.id AND filter_id = ?");
|
|
|
|
$sth2->execute([$filter_id]);
|
|
|
|
|
|
|
|
while ($action_line = $sth2->fetch()) {
|
|
|
|
# print_r($action_line);
|
|
|
|
|
|
|
|
$action = array();
|
|
|
|
$action["type"] = $action_line["type_name"];
|
|
|
|
$action["param"] = $action_line["action_param"];
|
|
|
|
|
|
|
|
array_push($actions, $action);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
$filter = [];
|
|
|
|
$filter["id"] = $filter_id;
|
|
|
|
$filter["match_any_rule"] = sql_bool_to_bool($line["match_any_rule"]);
|
|
|
|
$filter["inverse"] = sql_bool_to_bool($line["inverse"]);
|
|
|
|
$filter["rules"] = $rules;
|
|
|
|
$filter["actions"] = $actions;
|
|
|
|
|
|
|
|
if (count($rules) > 0 && count($actions) > 0) {
|
|
|
|
array_push($filters, $filter);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
return $filters;
|
|
|
|
}
|
2019-06-20 05:40:02 +00:00
|
|
|
|
|
|
|
/**
|
|
|
|
* Try to determine the favicon URL for a feed.
|
|
|
|
* adapted from wordpress favicon plugin by Jeff Minard (http://thecodepro.com/)
|
|
|
|
* http://dev.wp-plugins.org/file/favatars/trunk/favatars.php
|
|
|
|
*
|
|
|
|
* @param string $url A feed or page URL
|
|
|
|
* @access public
|
|
|
|
* @return mixed The favicon URL, or false if none was found.
|
|
|
|
*/
|
2021-03-06 12:19:31 +00:00
|
|
|
static function get_favicon_url(string $url) {
|
2019-06-20 05:40:02 +00:00
|
|
|
|
|
|
|
$favicon_url = false;
|
|
|
|
|
2020-09-22 06:04:33 +00:00
|
|
|
if ($html = @UrlHelper::fetch($url)) {
|
2019-06-20 05:40:02 +00:00
|
|
|
|
|
|
|
$doc = new DOMDocument();
|
2020-10-01 10:20:07 +00:00
|
|
|
if (@$doc->loadHTML($html)) {
|
2019-06-20 05:40:02 +00:00
|
|
|
$xpath = new DOMXPath($doc);
|
|
|
|
|
|
|
|
$base = $xpath->query('/html/head/base[@href]');
|
|
|
|
foreach ($base as $b) {
|
|
|
|
$url = rewrite_relative_url($url, $b->getAttribute("href"));
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
|
|
|
|
$entries = $xpath->query('/html/head/link[@rel="shortcut icon" or @rel="icon"]');
|
|
|
|
if (count($entries) > 0) {
|
|
|
|
foreach ($entries as $entry) {
|
|
|
|
$favicon_url = rewrite_relative_url($url, $entry->getAttribute("href"));
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
if (!$favicon_url)
|
|
|
|
$favicon_url = rewrite_relative_url($url, "/favicon.ico");
|
|
|
|
|
|
|
|
return $favicon_url;
|
|
|
|
}
|
|
|
|
|
2020-06-15 08:58:59 +00:00
|
|
|
// https://community.tt-rss.org/t/problem-with-img-srcset/3519
|
|
|
|
static function decode_srcset($srcset) {
|
|
|
|
$matches = [];
|
|
|
|
|
|
|
|
preg_match_all(
|
|
|
|
'/(?:\A|,)\s*(?P<url>(?!,)\S+(?<!,))\s*(?P<size>\s\d+w|\s\d+(?:\.\d+)?(?:[eE][+-]?\d+)?x|)\s*(?=,|\Z)/',
|
|
|
|
$srcset, $matches, PREG_SET_ORDER
|
|
|
|
);
|
|
|
|
|
|
|
|
foreach ($matches as $m) {
|
|
|
|
array_push($matches, [
|
|
|
|
"url" => trim($m["url"]),
|
|
|
|
"size" => trim($m["size"])
|
|
|
|
]);
|
|
|
|
}
|
|
|
|
|
|
|
|
return $matches;
|
|
|
|
}
|
|
|
|
|
|
|
|
static function encode_srcset($matches) {
|
|
|
|
$tokens = [];
|
|
|
|
|
|
|
|
foreach ($matches as $m) {
|
|
|
|
array_push($tokens, trim($m["url"]) . " " . trim($m["size"]));
|
|
|
|
}
|
|
|
|
|
|
|
|
return implode(",", $tokens);
|
|
|
|
}
|
2020-09-28 18:19:53 +00:00
|
|
|
|
|
|
|
static function function_enabled($func) {
|
|
|
|
return !in_array($func,
|
2021-04-12 08:55:19 +00:00
|
|
|
explode(',', str_replace(" ", "", ini_get('disable_functions'))));
|
2020-09-28 18:19:53 +00:00
|
|
|
}
|
2017-09-25 00:37:49 +00:00
|
|
|
}
|