// since we have the data cached, we can deal with other feeds with the same url
- $tmp_result = db_query("SELECT DISTINCT ttrss_feeds.id,last_updated
+ $tmp_result = db_query("SELECT DISTINCT ttrss_feeds.id,last_updated,ttrss_feeds.owner_uid
FROM ttrss_feeds, ttrss_users, ttrss_user_prefs WHERE
ttrss_user_prefs.owner_uid = ttrss_feeds.owner_uid AND
ttrss_users.id = ttrss_user_prefs.owner_uid AND
if (db_num_rows($tmp_result) > 0) {
while ($tline = db_fetch_assoc($tmp_result)) {
- if($debug) _debug(" => " . $tline["last_updated"] . ", " . $tline["id"]);
+ if($debug) _debug(" => " . $tline["last_updated"] . ", " . $tline["id"] . " " . $tline["owner_uid"]);
update_rss_feed($tline["id"], true);
++$nf;
}
} // function update_daemon_common
// ignore_daemon is not used
- function update_rss_feed($feed, $ignore_daemon = false, $no_cache = false,
- $override_url = false) {
-
- require_once "lib/simplepie/simplepie.inc";
+ function update_rss_feed($feed, $ignore_daemon = false, $no_cache = false) {
$debug_enabled = defined('DAEMON_EXTENDED_DEBUG') || $_REQUEST['xdebug'];
$result = db_query("SELECT id,update_interval,auth_login,
feed_url,auth_pass,cache_images,last_updated,
mark_unread_on_update, owner_uid,
- pubsub_state, auth_pass_encrypted
+ pubsub_state, auth_pass_encrypted,
+ (SELECT max(date_entered) FROM
+ ttrss_entries, ttrss_user_entries where ref_id = id AND feed_id = '$feed') AS last_article_timestamp
FROM ttrss_feeds WHERE id = '$feed'");
if (db_num_rows($result) == 0) {
}
$last_updated = db_fetch_result($result, 0, "last_updated");
+ $last_article_timestamp = @strtotime(db_fetch_result($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"));
$feed = db_escape_string($feed);
- if ($override_url) $fetch_url = $override_url;
-
$date_feed_processed = date('Y-m-d H:i');
- $cache_filename = CACHE_DIR . "/simplepie/" . sha1($fetch_url) . ".feed";
-
- // Ignore cache if new feed or manual update.
- $cache_age = ($no_cache || is_null($last_updated) || strpos($last_updated, '1970-01-01') === 0) ? 30 : get_feed_update_interval($feed) * 60;
+ $cache_filename = CACHE_DIR . "/simplepie/" . sha1($fetch_url) . ".xml";
- _debug("cache filename: $cache_filename exists: " . file_exists($cache_filename), $debug_enabled);
- _debug("cache age: $cache_age; no cache: $no_cache", $debug_enabled);
+ $pluginhost = new PluginHost();
+ $pluginhost->set_debug($debug_enabled);
+ $user_plugins = get_pref("_ENABLED_PLUGINS", $owner_uid);
- $cached_feed_data_hash = false;
+ $pluginhost->load(PLUGINS, PluginHost::KIND_ALL);
+ $pluginhost->load($user_plugins, PluginHost::KIND_USER, $owner_uid);
+ $pluginhost->load_data();
$rss = false;
$rss_hash = false;
- $cache_timestamp = file_exists($cache_filename) ? filemtime($cache_filename) : 0;
- $last_updated_timestamp = strtotime($last_updated);
$force_refetch = isset($_REQUEST["force_refetch"]);
if (file_exists($cache_filename) &&
is_readable($cache_filename) &&
!$auth_login && !$auth_pass &&
- filemtime($cache_filename) > time() - $cache_age) {
+ filemtime($cache_filename) > time() - 30) {
- _debug("using local cache.", $debug_enabled);
+ _debug("using local cache.", $debug_enabled);
- if ($cache_timestamp > $last_updated_timestamp) {
- @$rss_data = file_get_contents($cache_filename);
+ @$feed_data = file_get_contents($cache_filename);
+
+ if ($feed_data) {
+ $rss_hash = sha1($feed_data);
+ }
- if ($rss_data) {
- $rss_hash = sha1($rss_data);
- @$rss = unserialize($rss_data);
- }
- } else if (!$force_refetch) {
- _debug("local cache valid and older than last_updated, nothing to do.", $debug_enabled);
- return;
- }
} else {
_debug("local cache will not be used for this feed", $debug_enabled);
- $cache_timestamp = 0;
}
if (!$rss) {
+ foreach ($pluginhost->get_hooks(PluginHost::HOOK_FETCH_FEED) as $plugin) {
+ $feed_data = $plugin->hook_fetch_feed($feed_data, $fetch_url, $owner_uid);
+ }
+
if (!$feed_data) {
- _debug("fetching [$fetch_url] (ts: $cache_timestamp/$last_updated_timestamp)", $debug_enabled);
+ _debug("fetching [$fetch_url]...", $debug_enabled);
+ _debug("If-Modified-Since: ".gmdate('D, d M Y H:i:s \G\M\T', $last_article_timestamp), $debug_enabled);
$feed_data = fetch_file_contents($fetch_url, false,
$auth_login, $auth_pass, false,
$no_cache ? FEED_FETCH_NO_CACHE_TIMEOUT : FEED_FETCH_TIMEOUT,
- $force_refetch ? 0 : max($last_updated_timestamp, $cache_timestamp));
+ $force_refetch ? 0 : $last_article_timestamp);
global $fetch_curl_used;
_debug("fetch done.", $debug_enabled);
- if ($feed_data) {
+ /* if ($feed_data) {
$error = verify_feed_xml($feed_data);
if ($error) {
if ($error) $feed_data = '';
}
}
- }
+ } */
}
if (!$feed_data) {
}
}
- $pluginhost = new PluginHost();
- $pluginhost->set_debug($debug_enabled, $debug_enabled);
- $user_plugins = get_pref("_ENABLED_PLUGINS", $owner_uid);
-
- $pluginhost->load(PLUGINS, PluginHost::KIND_ALL);
- $pluginhost->load($user_plugins, PluginHost::KIND_USER, $owner_uid);
- $pluginhost->load_data();
-
foreach ($pluginhost->get_hooks(PluginHost::HOOK_FEED_FETCHED) as $plugin) {
- $feed_data = $plugin->hook_feed_fetched($feed_data);
+ $feed_data = $plugin->hook_feed_fetched($feed_data, $fetch_url, $owner_uid);
}
+ // set last update to now so if anything *simplepie* crashes later we won't be
+ // continuously failing on the same feed
+ //db_query("UPDATE ttrss_feeds SET last_updated = NOW() WHERE id = '$feed'");
+
if (!$rss) {
- $rss = new SimplePie();
- $rss->set_sanitize_class("SanitizeDummy");
- // simplepie ignores the above and creates default sanitizer anyway,
- // so let's override it...
- $rss->sanitize = new SanitizeDummy();
- $rss->set_output_encoding('UTF-8');
- $rss->set_raw_data($feed_data);
- $rss->enable_cache(false);
-
- @$rss->init();
+ $rss = new FeedParser($feed_data);
+ $rss->init();
}
// print_r($rss);
// cache data for later
if (!$auth_pass && !$auth_login && is_writable(CACHE_DIR . "/simplepie")) {
- $rss_data = serialize($rss);
$new_rss_hash = sha1($rss_data);
- if ($new_rss_hash != $rss_hash) {
+ if ($new_rss_hash != $rss_hash && count($rss->get_items()) > 0 ) {
_debug("saving $cache_filename", $debug_enabled);
- @file_put_contents($cache_filename, serialize($rss));
+ @file_put_contents($cache_filename, $feed_data);
}
}
$site_url = db_escape_string(mb_substr(rewrite_relative_url($fetch_url, $rss->get_link()), 0, 245));
+ _debug("site_url: $site_url", $debug_enabled);
+ _debug("feed_title: " . $rss->get_title(), $debug_enabled);
+
if ($favicon_needs_check || $force_refetch) {
/* terrible hack: if we crash on floicon shit here, we won't check
_debug("checking favicon...", $debug_enabled);
- check_feed_favicon($site_url, $feed, $link);
+ check_feed_favicon($site_url, $feed);
$favicon_modified_new = @filemtime($favicon_file);
if ($favicon_modified_new > $favicon_modified)
$feed_title = db_escape_string($rss->get_title());
- _debug("registering title: $feed_title", $debug_enabled);
+ if ($feed_title) {
+ _debug("registering title: $feed_title", $debug_enabled);
- db_query("UPDATE ttrss_feeds SET
- title = '$feed_title' WHERE id = '$feed'");
+ db_query("UPDATE ttrss_feeds SET
+ title = '$feed_title' WHERE id = '$feed'");
+ }
}
if ($site_url && $orig_site_url != $site_url) {
if (!$entry_guid) $entry_guid = $item->get_link();
if (!$entry_guid) $entry_guid = make_guid_from_title($item->get_title());
+ _debug("f_guid $entry_guid", $debug_enabled);
+
if (!$entry_guid) continue;
$entry_guid = "$owner_uid,$entry_guid";
$entry_timestamp = "";
- $entry_timestamp = strtotime($item->get_date());
+ $entry_timestamp = $item->get_date();
+
+ _debug("orig date: " . $item->get_date(), $debug_enabled);
if ($entry_timestamp == -1 || !$entry_timestamp || $entry_timestamp > time()) {
$entry_timestamp = time();
_debug("date $entry_timestamp [$entry_timestamp_fmt]", $debug_enabled);
- $entry_title = html_entity_decode($item->get_title(), ENT_COMPAT, 'UTF-8');
+// $entry_title = html_entity_decode($item->get_title(), ENT_COMPAT, 'UTF-8');
+// $entry_title = decode_numeric_entities($entry_title);
+ $entry_title = $item->get_title();
$entry_link = rewrite_relative_url($site_url, $item->get_link());
print "\n";
}
- $entry_comments = $item->data["comments"];
-
- if ($item->get_author()) {
- $entry_author_item = $item->get_author();
- $entry_author = $entry_author_item->get_name();
- if (!$entry_author) $entry_author = $entry_author_item->get_email();
- }
+ $entry_comments = $item->get_comments_url();
+ $entry_author = $item->get_author();
$entry_guid = db_escape_string(mb_substr($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 = $item->get_item_tags('http://purl.org/rss/1.0/modules/slash/', 'comments');
-
- if (is_array($num_comments) && is_array($num_comments[0])) {
- $num_comments = (int) $num_comments[0]["data"];
- } else {
- $num_comments = 0;
- }
+ $num_comments = (int) $item->get_comments_count();
_debug("author $entry_author", $debug_enabled);
_debug("num_comments: $num_comments", $debug_enabled);
- _debug("looking for tags [1]...", $debug_enabled);
+ _debug("looking for tags...", $debug_enabled);
// parse <category> entries into tags
if (is_array($additional_tags_src)) {
foreach ($additional_tags_src as $tobj) {
- array_push($additional_tags, $tobj->get_term());
+ array_push($additional_tags, $tobj);
}
}
- _debug("category tags:", $debug_enabled);
- _debug("looking for tags [2]...", $debug_enabled);
-
$entry_tags = array_unique($additional_tags);
for ($i = 0; $i < count($entry_tags); $i++)
$entry_tags[$i] = mb_strtolower($entry_tags[$i], 'utf-8');
+ _debug("tags found: " . join(",", $entry_tags), $debug_enabled);
+
_debug("done collecting data.", $debug_enabled);
// TODO: less memory-hungry implementation
foreach ($filter["rules"] as $rule) {
$match = false;
- $reg_exp = $rule["reg_exp"];
+ $reg_exp = str_replace('/', '\/', $rule["reg_exp"]);
$rule_inverse = $rule["inverse"];
if (!$reg_exp)
$match = @preg_match("/$reg_exp/i", $author);
break;
case "tag":
- $tag_string = join(",", $tags);
- $match = @preg_match("/$reg_exp/i", $tag_string);
+ foreach ($tags as $tag) {
+ if (@preg_match("/$reg_exp/i", $tag)) {
+ $match = true;
+ break;
+ }
+ }
break;
}
mb_strtolower(strip_tags($title), 'utf-8'));
}
- function verify_feed_xml($feed_data) {
+ /* function verify_feed_xml($feed_data) {
libxml_use_internal_errors(true);
$doc = new DOMDocument();
$doc->loadXML($feed_data);
$error = libxml_get_last_error();
libxml_clear_errors();
return $error;
- }
+ } */
function housekeeping_common($debug) {
expire_cached_files($debug);