$nf = 0;
$bstarted = microtime(true);
+ $batch_owners = array();
+
// For each feed, we call the feed update function.
foreach ($feeds_to_update as $feed) {
if($debug) _debug("Base feed: $feed");
ttrss_users.id = ttrss_user_prefs.owner_uid AND
ttrss_user_prefs.pref_name = 'DEFAULT_UPDATE_INTERVAL' AND
ttrss_user_prefs.profile IS NULL AND
- feed_url = '".db_escape_string($feed)."' AND
- (ttrss_feeds.update_interval > 0 OR
- ttrss_user_prefs.value != '-1')
+ feed_url = '".db_escape_string($feed)."'
+ $update_limit_qpart
$login_thresh_qpart
ORDER BY ttrss_feeds.id $query_limit");
while ($tline = db_fetch_assoc($tmp_result)) {
if($debug) _debug(" => " . $tline["last_updated"] . ", " . $tline["id"] . " " . $tline["owner_uid"]);
+ if (array_search($tline["owner_uid"], $batch_owners) === FALSE)
+ array_push($batch_owners, $tline["owner_uid"]);
+
$fstarted = microtime(true);
$rss = update_rss_feed($tline["id"], true, false);
_debug_suppress(false);
microtime(true) - $bstarted, (microtime(true) - $bstarted) / $nf));
}
+ foreach ($batch_owners as $owner_uid) {
+ _debug("Running housekeeping tasks for user $owner_uid...");
+
+ housekeeping_user($owner_uid);
+ }
+
require_once "digest.php";
// Send feed digests by email if needed.
feed_url,auth_pass,cache_images,
mark_unread_on_update, owner_uid,
pubsub_state, auth_pass_encrypted,
- feed_language,
- (SELECT max(date_entered) FROM
- ttrss_entries, ttrss_user_entries where ref_id = id AND feed_id = '$feed') AS last_article_timestamp
+ feed_language
FROM ttrss_feeds WHERE id = '$feed'");
if (db_num_rows($result) == 0) {
_debug("feed $feed NOT FOUND/SKIPPED", $debug_enabled);
+ user_error("Attempt to update unknown/invalid feed $feed", E_USER_WARNING);
return false;
}
- $last_article_timestamp = @strtotime(db_fetch_result($result, 0, "last_article_timestamp"));
-
- if (defined('_DISABLE_HTTP_304'))
- $last_article_timestamp = 0;
-
$owner_uid = db_fetch_result($result, 0, "owner_uid");
$mark_unread_on_update = sql_bool_to_bool(db_fetch_result($result,
0, "mark_unread_on_update"));
$rss_hash = false;
$force_refetch = isset($_REQUEST["force_refetch"]);
+ $feed_data = "";
foreach ($pluginhost->get_hooks(PluginHost::HOOK_FETCH_FEED) as $plugin) {
- $feed_data = $plugin->hook_fetch_feed($feed_data, $fetch_url, $owner_uid, $feed, $last_article_timestamp, $auth_login, $auth_pass);
+ $feed_data = $plugin->hook_fetch_feed($feed_data, $fetch_url, $owner_uid, $feed, 0, $auth_login, $auth_pass);
}
// try cache
// fetch feed from source
if (!$feed_data) {
_debug("fetching [$fetch_url]...", $debug_enabled);
- _debug("If-Modified-Since: ".gmdate('D, d M Y H:i:s \G\M\T', $last_article_timestamp), $debug_enabled);
+
+ if (ini_get("open_basedir") && function_exists("curl_init")) {
+ _debug("not using CURL due to open_basedir restrictions");
+ }
$feed_data = fetch_file_contents($fetch_url, false,
$auth_login, $auth_pass, false,
$no_cache ? FEED_FETCH_NO_CACHE_TIMEOUT : FEED_FETCH_TIMEOUT,
- $force_refetch ? 0 : $last_article_timestamp);
+ 0);
global $fetch_curl_used;
if ($_REQUEST["xdebug"] == 2) {
print "content: ";
- print $entry_content;
+ print htmlspecialchars($entry_content);
print "\n";
}
- $entry_comments = $item->get_comments_url();
- $entry_author = $item->get_author();
+ $entry_comments = db_escape_string(mb_substr($item->get_comments_url(), 0, 245));
+ $num_comments = (int) $item->get_comments_count();
+ $entry_author = $item->get_author(); // escaped later
$entry_guid = db_escape_string(mb_substr($entry_guid, 0, 245));
- $entry_comments = db_escape_string(mb_substr(trim($entry_comments), 0, 245));
- $entry_author = db_escape_string(mb_substr(trim($entry_author), 0, 245));
-
- $num_comments = (int) $item->get_comments_count();
-
_debug("author $entry_author", $debug_enabled);
_debug("num_comments: $num_comments", $debug_enabled);
_debug("looking for tags...", $debug_enabled);
"language" => $entry_language,
"feed" => array("id" => $feed,
"fetch_url" => $fetch_url,
- "site_url" => $site_url)
+ "site_url" => $site_url,
+ "cache_images" => $cache_images)
);
$entry_plugin_data = "";
db_query("UPDATE ttrss_entries SET date_updated = NOW()
WHERE id = '$base_entry_id'");
- // if we allow duplicate posts, we have to continue to
- // create the user entries for this feed (if needed)
- if (get_pref("ALLOW_DUPLICATE_POSTS", $owner_uid, false)) {
-
- $query = "SELECT int_id FROM ttrss_user_entries WHERE
- ref_id = '$base_entry_id' AND owner_uid = '$owner_uid'
- AND (feed_id = '$feed' OR feed_id IS NULL) LIMIT 1";
-
- $result = db_query($query);
-
- if (db_num_rows($result) == 0) {
- _debug("allow duplicate posts is enabled and user record is not found, continuing.");
- } else {
- continue;
- }
-
- } else {
- continue;
- }
+ continue;
}
_debug("hash differs, applying plugin filters:", $debug_enabled);
$entry_plugin_data .= mb_strtolower(get_class($plugin)) . ",";
}
+ if ($_REQUEST["xdebug"] == 2) {
+ print "processed content: ";
+ print htmlspecialchars($article["content"]);
+ print "\n";
+ }
+
$entry_plugin_data = db_escape_string($entry_plugin_data);
_debug("plugin data: $entry_plugin_data", $debug_enabled);
foreach ($article as $k => $v) {
// i guess we'll have to take the risk of 4byte unicode labels & tags here
- if (!is_array($article[$k])) {
+ if (is_string($article[$k])) {
$article[$k] = preg_replace('/[\x{10000}-\x{10FFFF}]/u', "\xEF\xBF\xBD", $v);
}
}
/* Collect article tags here so we could filter by them: */
+ $matched_rules = array();
+
$article_filters = get_article_filters($filters, $article["title"],
$article["content"], $article["link"], 0, $article["author"],
- $article["tags"]);
+ $article["tags"], $matched_rules);
if ($debug_enabled) {
- _debug("article filters: ", $debug_enabled);
+ _debug("matched filter rules: ", $debug_enabled);
+
+ if (count($matched_rules) != 0) {
+ print_r($matched_rules);
+ }
+
+ _debug("filter actions: ", $debug_enabled);
+
if (count($article_filters) != 0) {
print_r($article_filters);
}
$entry_tags = $article["tags"];
$entry_guid = db_escape_string($entry_guid);
$entry_title = db_escape_string($article["title"]);
- $entry_author = db_escape_string($article["author"]);
+ $entry_author = db_escape_string(mb_substr($article["author"], 0, 245));
$entry_link = db_escape_string($article["link"]);
$entry_content = $article["content"]; // escaped below
$entry_force_catchup = $article["force_catchup"];
if ($debug_enabled) {
_debug("article labels:", $debug_enabled);
- print_r($article_labels);
+
+ if (count($article_labels) != 0) {
+ print_r($article_labels);
+ }
}
_debug("force catchup: $entry_force_catchup");
// check for user post link to main table
- // do we allow duplicate posts with same GUID in different feeds?
- if (get_pref("ALLOW_DUPLICATE_POSTS", $owner_uid, false)) {
- $dupcheck_qpart = "AND (feed_id = '$feed' OR feed_id IS NULL)";
- } else {
- $dupcheck_qpart = "";
- }
-
$query = "SELECT ref_id, int_id FROM ttrss_user_entries WHERE
- ref_id = '$ref_id' AND owner_uid = '$owner_uid'
- $dupcheck_qpart";
+ ref_id = '$ref_id' AND owner_uid = '$owner_uid'";
// if ($_REQUEST["xdebug"]) print "$query\n";
$published = 'false';
}
- // N-grams
-
- /* if (DB_TYPE == "pgsql" and defined('_NGRAM_TITLE_DUPLICATE_THRESHOLD')) {
-
- $result = db_query("SELECT COUNT(*) AS similar FROM
- ttrss_entries,ttrss_user_entries
- WHERE ref_id = id AND updated >= NOW() - INTERVAL '7 day'
- AND similarity(title, '$entry_title') >= "._NGRAM_TITLE_DUPLICATE_THRESHOLD."
- AND owner_uid = $owner_uid");
-
- $ngram_similar = db_fetch_result($result, 0, "similar");
-
- _debug("N-gram similar results: $ngram_similar", $debug_enabled);
-
- if ($ngram_similar > 0) {
- $unread = 'false';
- }
- } */
-
$last_marked = ($marked == 'true') ? 'NOW()' : 'NULL';
$last_published = ($published == 'true') ? 'NOW()' : 'NULL';
return $params;
}
- function get_article_filters($filters, $title, $content, $link, $timestamp, $author, $tags) {
+ function get_article_filters($filters, $title, $content, $link, $timestamp, $author, $tags, &$matched_rules = false) {
$matches = array();
foreach ($filters as $filter) {
switch ($rule["type"]) {
case "title":
- $match = @preg_match("/$reg_exp/i", $title);
+ $match = @preg_match("/$reg_exp/iu", $title);
break;
case "content":
// we don't need to deal with multiline regexps
$content = preg_replace("/[\r\n\t]/", "", $content);
- $match = @preg_match("/$reg_exp/i", $content);
+ $match = @preg_match("/$reg_exp/iu", $content);
break;
case "both":
// we don't need to deal with multiline regexps
$content = preg_replace("/[\r\n\t]/", "", $content);
- $match = (@preg_match("/$reg_exp/i", $title) || @preg_match("/$reg_exp/i", $content));
+ $match = (@preg_match("/$reg_exp/iu", $title) || @preg_match("/$reg_exp/iu", $content));
break;
case "link":
- $match = @preg_match("/$reg_exp/i", $link);
+ $match = @preg_match("/$reg_exp/iu", $link);
break;
case "author":
- $match = @preg_match("/$reg_exp/i", $author);
+ $match = @preg_match("/$reg_exp/iu", $author);
break;
case "tag":
foreach ($tags as $tag) {
- if (@preg_match("/$reg_exp/i", $tag)) {
+ if (@preg_match("/$reg_exp/iu", $tag)) {
$match = true;
break;
}
if ($inverse) $filter_match = !$filter_match;
if ($filter_match) {
+ if (is_array($matched_rules)) array_push($matched_rules, $rule);
+
foreach ($filter["actions"] AS $action) {
array_push($matches, $action);
_debug("Removed $frows (feeds) $crows (cats) orphaned counter cache entries.");
}
+ function housekeeping_user($owner_uid) {
+ $tmph = new PluginHost();
+
+ load_user_plugins($owner_uid, $tmph);
+
+ $tmph->run_hooks(PluginHost::HOOK_HOUSE_KEEPING, "hook_house_keeping", "");
+ }
+
function housekeeping_common($debug) {
expire_cached_files($debug);
expire_lock_files($debug);
//_debug("Cleaned $rc cached tags.");
PluginHost::getInstance()->run_hooks(PluginHost::HOOK_HOUSE_KEEPING, "hook_house_keeping", "");
-
}
?>
git.wh0rd.org / tt-rss.git / blobdiff