X-Git-Url: https://git.wh0rd.org/?a=blobdiff_plain;ds=sidebyside;f=include%2Frssfuncs.php;h=4c3e861237349a105af107e21225c25f868a9a06;hb=6791af0cfdf481e90368b92c358559dcc6944272;hp=283e485324f3bb1b90d4f831a172a0cfaababb50;hpb=bc7a144dd589302025c3b024d6da71f2e53d6496;p=tt-rss.git diff --git a/include/rssfuncs.php b/include/rssfuncs.php index 283e4853..4c3e8612 100644 --- a/include/rssfuncs.php +++ b/include/rssfuncs.php @@ -1,34 +1,34 @@ 0 OR ttrss_user_prefs.value != '-1') $login_thresh_qpart @@ -172,8 +173,8 @@ if (db_num_rows($tmp_result) > 0) { while ($tline = db_fetch_assoc($tmp_result)) { - if($debug) _debug(" => " . $tline["last_updated"] . ", " . $tline["id"]); - update_rss_feed($link, $tline["id"], true); + if($debug) _debug(" => " . $tline["last_updated"] . ", " . $tline["id"] . " " . $tline["owner_uid"]); + update_rss_feed($tline["id"], true); ++$nf; } } @@ -182,38 +183,38 @@ require_once "digest.php"; // Send feed digests by email if needed. - send_headlines_digests($link, $debug); + send_headlines_digests($debug); return $nf; } // function update_daemon_common // ignore_daemon is not used - function update_rss_feed($link, $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') || $_REQUEST['xdebug']; - if ($debug_enabled) { - _debug("update_rss_feed: start"); - } + _debug("start", $debug_enabled); - $result = db_query($link, "SELECT id,update_interval,auth_login, + $result = db_query("SELECT id,update_interval,auth_login, feed_url,auth_pass,cache_images,last_updated, mark_unread_on_update, owner_uid, - pubsub_state, auth_pass_encrypted + pubsub_state, auth_pass_encrypted, + (SELECT max(date_entered) FROM + ttrss_entries, ttrss_user_entries where ref_id = id AND feed_id = '$feed') AS last_article_timestamp FROM ttrss_feeds WHERE id = '$feed'"); if (db_num_rows($result) == 0) { - if ($debug_enabled) { - _debug("update_rss_feed: feed $feed NOT FOUND/SKIPPED"); - } + _debug("feed $feed NOT FOUND/SKIPPED", $debug_enabled); return false; } $last_updated = db_fetch_result($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")); @@ -221,7 +222,7 @@ $auth_pass_encrypted = sql_bool_to_bool(db_fetch_result($result, 0, "auth_pass_encrypted")); - db_query($link, "UPDATE ttrss_feeds SET last_update_started = NOW() + db_query("UPDATE ttrss_feeds SET last_update_started = NOW() WHERE id = '$feed'"); $auth_login = db_fetch_result($result, 0, "auth_login"); @@ -235,94 +236,104 @@ $cache_images = sql_bool_to_bool(db_fetch_result($result, 0, "cache_images")); $fetch_url = db_fetch_result($result, 0, "feed_url"); - $feed = db_escape_string($link, $feed); - - if ($override_url) $fetch_url = $override_url; + $feed = db_escape_string($feed); $date_feed_processed = date('Y-m-d H:i'); - $cache_filename = CACHE_DIR . "/simplepie/" . sha1($fetch_url) . ".feed"; + $cache_filename = CACHE_DIR . "/simplepie/" . sha1($fetch_url) . ".xml"; - // Ignore cache if new feed or manual update. - $cache_age = ($no_cache || is_null($last_updated) || $last_updated == '1970-01-01 00:00:00') ? - 30 : get_feed_update_interval($link, $feed) * 60; - - if ($debug_enabled) { - _debug("update_rss_feed: cache filename: $cache_filename exists: " . file_exists($cache_filename)); - _debug("update_rss_feed: cache age: $cache_age; no cache: $no_cache"); - } + $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) { - if ($debug_enabled) { - _debug("update_rss_feed: using local cache."); - } + _debug("using local cache.", $debug_enabled); - if ($cache_timestamp > $last_updated_timestamp) { - @$rss_data = file_get_contents($cache_filename); + @$feed_data = file_get_contents($cache_filename); - if ($rss_data) { - $rss_hash = sha1($rss_data); - @$rss = unserialize($rss_data); - } - } else if (!$force_refetch) { - if ($debug_enabled) { - _debug("update_rss_feed: local cache valid and older than last_updated, nothing to do."); - } - return; - } + if ($feed_data) { + $rss_hash = sha1($feed_data); + } + + } else { + _debug("local cache will not be used for this feed", $debug_enabled); } if (!$rss) { + foreach ($pluginhost->get_hooks(PluginHost::HOOK_FETCH_FEED) as $plugin) { + $feed_data = $plugin->hook_fetch_feed($feed_data, $fetch_url, $owner_uid, $feed); + } + if (!$feed_data) { - if ($debug_enabled) { - _debug("update_rss_feed: fetching [$fetch_url] (ts: $cache_timestamp/$last_updated_timestamp)"); - } + _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 : $last_article_timestamp); - if ($debug_enabled) { - _debug("update_rss_feed: fetch done."); + global $fetch_curl_used; + + if (!$fetch_curl_used) { + $tmp = @gzdecode($feed_data); + + if ($tmp) $feed_data = $tmp; } + $feed_data = trim($feed_data); + + _debug("fetch done.", $debug_enabled); + + /* if ($feed_data) { + $error = verify_feed_xml($feed_data); + + if ($error) { + _debug("error verifying XML, code: " . $error->code, $debug_enabled); + + if ($error->code == 26) { + _debug("got error 26, trying to decode entities...", $debug_enabled); + + $feed_data = html_entity_decode($feed_data, ENT_COMPAT, 'UTF-8'); + + $error = verify_feed_xml($feed_data); + + if ($error) $feed_data = ''; + } + } + } */ } if (!$feed_data) { global $fetch_last_error; global $fetch_last_error_code; - if ($debug_enabled) { - _debug("update_rss_feed: unable to fetch: $fetch_last_error [$fetch_last_error_code]"); - } + _debug("unable to fetch: $fetch_last_error [$fetch_last_error_code]", $debug_enabled); $error_escaped = ''; // If-Modified-Since if ($fetch_last_error_code != 304) { - $error_escaped = db_escape_string($link, $fetch_last_error); + $error_escaped = db_escape_string($fetch_last_error); } else { - if ($debug_enabled) { - _debug("update_rss_feed: source claims data not modified, nothing to do."); - } + _debug("source claims data not modified, nothing to do.", $debug_enabled); } - db_query($link, + db_query( "UPDATE ttrss_feeds SET last_error = '$error_escaped', last_updated = NOW() WHERE id = '$feed'"); @@ -330,58 +341,41 @@ } } - $pluginhost = new PluginHost($link); - $pluginhost->set_debug($debug_enabled); - $user_plugins = get_pref($link, "_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); + foreach ($pluginhost->get_hooks(PluginHost::HOOK_FEED_FETCHED) as $plugin) { + $feed_data = $plugin->hook_feed_fetched($feed_data, $fetch_url, $owner_uid, $feed); } + // set last update to now so if anything *simplepie* crashes later we won't be + // continuously failing on the same feed + //db_query("UPDATE ttrss_feeds SET last_updated = NOW() WHERE id = '$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); - $feed = db_escape_string($link, $feed); + $feed = db_escape_string($feed); if (!$rss->error()) { // cache data for later if (!$auth_pass && !$auth_login && is_writable(CACHE_DIR . "/simplepie")) { - $rss_data = serialize($rss); $new_rss_hash = sha1($rss_data); - if ($new_rss_hash != $rss_hash) { - if ($debug_enabled) { - _debug("update_rss_feed: saving $cache_filename"); - } - @file_put_contents($cache_filename, serialize($rss)); + if ($new_rss_hash != $rss_hash && count($rss->get_items()) > 0 ) { + _debug("saving $cache_filename", $debug_enabled); + @file_put_contents($cache_filename, $feed_data); } } // We use local pluginhost here because we need to load different per-user feed plugins - $pluginhost->run_hooks($pluginhost::HOOK_FEED_PARSED, "hook_feed_parsed", $rss); + $pluginhost->run_hooks(PluginHost::HOOK_FEED_PARSED, "hook_feed_parsed", $rss); - if ($debug_enabled) { - _debug("update_rss_feed: processing feed data..."); - } + _debug("processing feed data...", $debug_enabled); -// db_query($link, "BEGIN"); +// db_query("BEGIN"); if (DB_TYPE == "pgsql") { $favicon_interval_qpart = "favicon_last_checked < NOW() - INTERVAL '12 hour'"; @@ -389,7 +383,7 @@ $favicon_interval_qpart = "favicon_last_checked < DATE_SUB(NOW(), INTERVAL 12 HOUR)"; } - $result = db_query($link, "SELECT title,site_url,owner_uid, + $result = db_query("SELECT title,site_url,owner_uid,favicon_avg_color, (favicon_last_checked IS NULL OR $favicon_interval_qpart) AS favicon_needs_check FROM ttrss_feeds WHERE id = '$feed'"); @@ -398,70 +392,80 @@ $orig_site_url = db_fetch_result($result, 0, "site_url"); $favicon_needs_check = sql_bool_to_bool(db_fetch_result($result, 0, "favicon_needs_check")); + $favicon_avg_color = db_fetch_result($result, 0, "favicon_avg_color"); $owner_uid = db_fetch_result($result, 0, "owner_uid"); - $site_url = db_escape_string($link, mb_substr(rewrite_relative_url($fetch_url, $rss->get_link()), 0, 245)); + $site_url = db_escape_string(mb_substr(rewrite_relative_url($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 || $force_refetch) { - if ($debug_enabled) { - _debug("update_rss_feed: checking favicon..."); - } - check_feed_favicon($site_url, $feed, $link); + /* terrible hack: if we crash on floicon shit here, we won't check + * the icon avgcolor again (unless the icon got updated) */ + $favicon_file = ICONS_DIR . "/$feed.ico"; + $favicon_modified = @filemtime($favicon_file); + + _debug("checking favicon...", $debug_enabled); - if (file_exists($favicon_file)) { + check_feed_favicon($site_url, $feed); + $favicon_modified_new = @filemtime($favicon_file); + + if ($favicon_modified_new > $favicon_modified) + $favicon_avg_color = ''; + + if (file_exists($favicon_file) && function_exists("imagecreatefromstring") && $favicon_avg_color == '') { require_once "colors.php"; - $favicon_color = db_escape_string($link, + db_query("UPDATE ttrss_feeds SET favicon_avg_color = 'fail' WHERE + id = '$feed'"); + + $favicon_color = db_escape_string( calculate_avg_color($favicon_file)); $favicon_colorstring = ",favicon_avg_color = '".$favicon_color."'"; + } else if ($favicon_avg_color == 'fail') { + _debug("floicon failed on this file, not trying to recalculate avg color", $debug_enabled); } - db_query($link, "UPDATE ttrss_feeds SET favicon_last_checked = NOW() + db_query("UPDATE ttrss_feeds SET favicon_last_checked = NOW() $favicon_colorstring WHERE id = '$feed'"); } if (!$registered_title || $registered_title == "[Unknown]") { - $feed_title = db_escape_string($link, $rss->get_title()); + $feed_title = db_escape_string($rss->get_title()); - if ($debug_enabled) { - _debug("update_rss_feed: registering title: $feed_title"); - } + if ($feed_title) { + _debug("registering title: $feed_title", $debug_enabled); - db_query($link, "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 != $site_url) { - db_query($link, "UPDATE ttrss_feeds SET + db_query("UPDATE ttrss_feeds SET site_url = '$site_url' WHERE id = '$feed'"); } - if ($debug_enabled) { - _debug("update_rss_feed: loading filters & labels..."); - } + _debug("loading filters & labels...", $debug_enabled); - $filters = load_filters($link, $feed, $owner_uid); - $labels = get_all_labels($link, $owner_uid); + $filters = load_filters($feed, $owner_uid); + $labels = get_all_labels($owner_uid); - if ($debug_enabled) { - //print_r($filters); - _debug("update_rss_feed: " . count($filters) . " filters loaded."); - } + _debug("" . count($filters) . " filters loaded.", $debug_enabled); $items = $rss->get_items(); if (!is_array($items)) { - if ($debug_enabled) { - _debug("update_rss_feed: no articles found."); - } + _debug("no articles found.", $debug_enabled); - db_query($link, "UPDATE ttrss_feeds + db_query("UPDATE ttrss_feeds SET last_updated = NOW(), last_error = '' WHERE id = '$feed'"); return; // no articles @@ -469,7 +473,7 @@ if ($pubsub_state != 2 && PUBSUBHUBBUB_ENABLED) { - if ($debug_enabled) _debug("update_rss_feed: checking for PUSH hub..."); + _debug("checking for PUSH hub...", $debug_enabled); $feed_hub_url = false; @@ -482,7 +486,7 @@ } } - if ($debug_enabled) _debug("update_rss_feed: feed hub url: $feed_hub_url"); + _debug("feed hub url: $feed_hub_url", $debug_enabled); if ($feed_hub_url && function_exists('curl_init') && !ini_get("open_basedir")) { @@ -496,17 +500,14 @@ $rc = $s->subscribe($fetch_url); - if ($debug_enabled) - _debug("update_rss_feed: feed hub url found, subscribe request sent."); + _debug("feed hub url found, subscribe request sent.", $debug_enabled); - db_query($link, "UPDATE ttrss_feeds SET pubsub_state = 1 + db_query("UPDATE ttrss_feeds SET pubsub_state = 1 WHERE id = '$feed'"); } } - if ($debug_enabled) { - _debug("update_rss_feed: processing articles..."); - } + _debug("processing articles...", $debug_enabled); foreach ($items as $item) { if ($_REQUEST['xdebug'] == 3) { @@ -517,17 +518,21 @@ if (!$entry_guid) $entry_guid = $item->get_link(); if (!$entry_guid) $entry_guid = make_guid_from_title($item->get_title()); - if ($debug_enabled) { - _debug("update_rss_feed: guid $entry_guid"); - } + _debug("f_guid $entry_guid", $debug_enabled); if (!$entry_guid) continue; $entry_guid = "$owner_uid,$entry_guid"; + $entry_guid_hashed = db_escape_string('SHA1:' . sha1($entry_guid)); + + _debug("guid $entry_guid / $entry_guid_hashed", $debug_enabled); + $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(); @@ -538,18 +543,16 @@ $entry_timestamp_fmt = strftime("%Y/%m/%d %H:%M:%S", $entry_timestamp); - if ($debug_enabled) { - _debug("update_rss_feed: date $entry_timestamp [$entry_timestamp_fmt]"); - } + _debug("date $entry_timestamp [$entry_timestamp_fmt]", $debug_enabled); - $entry_title = html_entity_decode($item->get_title()); +// $entry_title = html_entity_decode($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, $item->get_link()); - if ($debug_enabled) { - _debug("update_rss_feed: title $entry_title"); - _debug("update_rss_feed: link $entry_link"); - } + _debug("title $entry_title", $debug_enabled); + _debug("link $entry_link", $debug_enabled); if (!$entry_title) $entry_title = date("Y-m-d H:i:s", $entry_timestamp);; @@ -557,38 +560,24 @@ if (!$entry_content) $entry_content = $item->get_description(); if ($_REQUEST["xdebug"] == 2) { - print "update_rss_feed: content: "; + print "content: "; print $entry_content; print "\n"; } - $entry_comments = $item->data["comments"]; + $entry_comments = $item->get_comments_url(); + $entry_author = $item->get_author(); - if ($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(); + $entry_guid = db_escape_string(mb_substr($entry_guid, 0, 245)); - $entry_author = db_escape_string($link, $entry_author); - } + $entry_comments = db_escape_string(mb_substr(trim($entry_comments), 0, 245)); + $entry_author = db_escape_string(mb_substr(trim($entry_author), 0, 245)); - $entry_guid = db_escape_string($link, mb_substr($entry_guid, 0, 245)); + $num_comments = (int) $item->get_comments_count(); - $entry_comments = db_escape_string($link, mb_substr($entry_comments, 0, 245)); - $entry_author = db_escape_string($link, mb_substr($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 { - $num_comments = 0; - } - - if ($debug_enabled) { - _debug("update_rss_feed: num_comments: $num_comments"); - _debug("update_rss_feed: looking for tags [1]..."); - } + _debug("author $entry_author", $debug_enabled); + _debug("num_comments: $num_comments", $debug_enabled); + _debug("looking for tags...", $debug_enabled); // parse entries into tags @@ -598,42 +587,26 @@ if (is_array($additional_tags_src)) { foreach ($additional_tags_src as $tobj) { - array_push($additional_tags, $tobj->get_term()); + array_push($additional_tags, $tobj); } } - if ($debug_enabled) { - _debug("update_rss_feed: category tags:"); - print_r($additional_tags); - } - - if ($debug_enabled) { - _debug("update_rss_feed: looking for tags [2]..."); - } - $entry_tags = array_unique($additional_tags); for ($i = 0; $i < count($entry_tags); $i++) $entry_tags[$i] = mb_strtolower($entry_tags[$i], 'utf-8'); - if ($debug_enabled) { - //_debug("update_rss_feed: unfiltered tags found:"); - //print_r($entry_tags); - } + _debug("tags found: " . join(",", $entry_tags), $debug_enabled); - if ($debug_enabled) { - _debug("update_rss_feed: done collecting data."); - } + _debug("done collecting data.", $debug_enabled); // TODO: less memory-hungry implementation - if ($debug_enabled) { - _debug("update_rss_feed: applying plugin filters.."); - } + _debug("applying plugin filters..", $debug_enabled); // FIXME not sure if owner_uid is a good idea here, we may have a base entry without user entry (?) - $result = db_query($link, "SELECT plugin_data,title,content,link,tag_cache,author FROM ttrss_entries, ttrss_user_entries - WHERE ref_id = id AND guid = '".db_escape_string($link, $entry_guid)."' AND owner_uid = $owner_uid"); + $result = db_query("SELECT plugin_data,title,content,link,tag_cache,author FROM ttrss_entries, ttrss_user_entries + WHERE ref_id = id AND (guid = '".db_escape_string($entry_guid)."' OR guid = '$entry_guid_hashed') AND owner_uid = $owner_uid"); if (db_num_rows($result) != 0) { $entry_plugin_data = db_fetch_result($result, 0, "plugin_data"); @@ -657,44 +630,40 @@ "author" => $entry_author, "stored" => $stored_article); - foreach ($pluginhost->get_hooks($pluginhost::HOOK_ARTICLE_FILTER) as $plugin) { + foreach ($pluginhost->get_hooks(PluginHost::HOOK_ARTICLE_FILTER) as $plugin) { $article = $plugin->hook_article_filter($article); } $entry_tags = $article["tags"]; - $entry_guid = db_escape_string($link, $entry_guid); - $entry_title = db_escape_string($link, $article["title"]); - $entry_author = db_escape_string($link, $article["author"]); - $entry_link = db_escape_string($link, $article["link"]); - $entry_plugin_data = db_escape_string($link, $article["plugin_data"]); + $entry_guid = db_escape_string($entry_guid); + $entry_title = db_escape_string($article["title"]); + $entry_author = db_escape_string($article["author"]); + $entry_link = db_escape_string($article["link"]); + $entry_plugin_data = db_escape_string($article["plugin_data"]); $entry_content = $article["content"]; // escaped below - if ($debug_enabled) { - _debug("update_rss_feed: plugin data: $entry_plugin_data"); - } + _debug("plugin data: $entry_plugin_data", $debug_enabled); if ($cache_images && is_writable(CACHE_DIR . '/images')) cache_images($entry_content, $site_url, $debug_enabled); - $entry_content = db_escape_string($link, $entry_content, false); + $entry_content = db_escape_string($entry_content, false); $content_hash = "SHA1:" . sha1($entry_content); - db_query($link, "BEGIN"); + db_query("BEGIN"); - $result = db_query($link, "SELECT id FROM ttrss_entries - WHERE guid = '$entry_guid'"); + $result = db_query("SELECT id FROM ttrss_entries + WHERE (guid = '$entry_guid' OR guid = '$entry_guid_hashed')"); if (db_num_rows($result) == 0) { - if ($debug_enabled) { - _debug("update_rss_feed: base guid [$entry_guid] not found"); - } + _debug("base guid [$entry_guid] not found", $debug_enabled); // base post entry does not exist, create it - $result = db_query($link, + $result = db_query( "INSERT INTO ttrss_entries (title, guid, @@ -712,7 +681,7 @@ author) VALUES ('$entry_title', - '$entry_guid', + '$entry_guid_hashed', '$entry_link', '$entry_timestamp_fmt', '$entry_content', @@ -736,31 +705,29 @@ $base_entry_id = db_fetch_result($result, 0, "id"); - db_query($link, "UPDATE ttrss_entries SET date_updated = NOW() + db_query("UPDATE ttrss_entries SET date_updated = NOW() WHERE id = '$base_entry_id'"); - $article_labels = get_article_labels($link, $base_entry_id, $owner_uid); + $article_labels = get_article_labels($base_entry_id, $owner_uid); } // now it should exist, if not - bad luck then - $result = db_query($link, "SELECT - id,content_hash,no_orig_date,title,plugin_data, + $result = db_query("SELECT + id,content_hash,no_orig_date,title,plugin_data,guid, ".SUBSTRING_FOR_DATE."(date_updated,1,19) as date_updated, ".SUBSTRING_FOR_DATE."(updated,1,19) as updated, num_comments FROM ttrss_entries - WHERE guid = '$entry_guid'"); + WHERE guid = '$entry_guid' OR guid = '$entry_guid_hashed'"); $entry_ref_id = 0; $entry_int_id = 0; if (db_num_rows($result) == 1) { - if ($debug_enabled) { - _debug("update_rss_feed: base guid [$entry_guid] found, checking for user record"); - } + _debug("base guid found, checking for user record", $debug_enabled); // this will be used below in update handler $orig_content_hash = db_fetch_result($result, 0, "content_hash"); @@ -773,10 +740,18 @@ $ref_id = db_fetch_result($result, 0, "id"); $entry_ref_id = $ref_id; + /* $stored_guid = db_fetch_result($result, 0, "guid"); + if ($stored_guid != $entry_guid_hashed) { + if ($debug_enabled) _debug("upgrading compat guid to hashed one", $debug_enabled); + + db_query("UPDATE ttrss_entries SET guid = '$entry_guid_hashed' WHERE + id = '$ref_id'"); + } */ + // check for user post link to main table // do we allow duplicate posts with same GUID in different feeds? - if (get_pref($link, "ALLOW_DUPLICATE_POSTS", $owner_uid, false)) { + if (get_pref("ALLOW_DUPLICATE_POSTS", $owner_uid, false)) { $dupcheck_qpart = "AND (feed_id = '$feed' OR feed_id IS NULL)"; } else { $dupcheck_qpart = ""; @@ -789,22 +764,20 @@ $entry_tags); if ($debug_enabled) { - _debug("update_rss_feed: article filters: "); + _debug("article filters: ", $debug_enabled); if (count($article_filters) != 0) { print_r($article_filters); } } if (find_article_filter($article_filters, "filter")) { - db_query($link, "COMMIT"); // close transaction in progress + db_query("COMMIT"); // close transaction in progress continue; } $score = calculate_article_score($article_filters); - if ($debug_enabled) { - _debug("update_rss_feed: initial score: $score"); - } + _debug("initial score: $score", $debug_enabled); $query = "SELECT ref_id, int_id FROM ttrss_user_entries WHERE ref_id = '$ref_id' AND owner_uid = '$owner_uid' @@ -812,14 +785,12 @@ // if ($_REQUEST["xdebug"]) print "$query\n"; - $result = db_query($link, $query); + $result = db_query($query); // okay it doesn't exist - create user entry if (db_num_rows($result) == 0) { - if ($debug_enabled) { - _debug("update_rss_feed: user record not found, creating..."); - } + _debug("user record not found, creating...", $debug_enabled); if ($score >= -500 && !find_article_filter($article_filters, 'catchup')) { $unread = 'true'; @@ -845,7 +816,7 @@ if (DB_TYPE == "pgsql" and defined('_NGRAM_TITLE_DUPLICATE_THRESHOLD')) { - $result = db_query($link, "SELECT COUNT(*) AS similar FROM + $result = db_query("SELECT COUNT(*) AS similar FROM ttrss_entries,ttrss_user_entries WHERE ref_id = id AND updated >= NOW() - INTERVAL '7 day' AND similarity(title, '$entry_title') >= "._NGRAM_TITLE_DUPLICATE_THRESHOLD." @@ -853,9 +824,7 @@ $ngram_similar = db_fetch_result($result, 0, "similar"); - if ($debug_enabled) { - _debug("update_rss_feed: N-gram similar results: $ngram_similar"); - } + _debug("N-gram similar results: $ngram_similar", $debug_enabled); if ($ngram_similar > 0) { $unread = 'false'; @@ -865,7 +834,7 @@ $last_marked = ($marked == 'true') ? 'NOW()' : 'NULL'; $last_published = ($published == 'true') ? 'NOW()' : 'NULL'; - $result = db_query($link, + $result = db_query( "INSERT INTO ttrss_user_entries (ref_id, owner_uid, feed_id, unread, last_read, marked, published, score, tag_cache, label_cache, uuid, @@ -877,14 +846,14 @@ if (PUBSUBHUBBUB_HUB && $published == 'true') { $rss_link = get_self_url_prefix() . "/public.php?op=rss&id=-2&key=" . - get_feed_access_key($link, -2, false, $owner_uid); + get_feed_access_key(-2, false, $owner_uid); $p = new Publisher(PUBSUBHUBBUB_HUB); $pubsub_result = $p->publish_update($rss_link); } - $result = db_query($link, + $result = db_query( "SELECT int_id FROM ttrss_user_entries WHERE ref_id = '$ref_id' AND owner_uid = '$owner_uid' AND feed_id = '$feed' LIMIT 1"); @@ -893,17 +862,13 @@ $entry_int_id = db_fetch_result($result, 0, "int_id"); } } else { - if ($debug_enabled) { - _debug("update_rss_feed: user record FOUND"); - } + _debug("user record FOUND", $debug_enabled); $entry_ref_id = db_fetch_result($result, 0, "ref_id"); $entry_int_id = db_fetch_result($result, 0, "int_id"); } - if ($debug_enabled) { - _debug("update_rss_feed: RID: $entry_ref_id, IID: $entry_int_id"); - } + _debug("RID: $entry_ref_id, IID: $entry_int_id", $debug_enabled); $post_needs_update = false; $update_insignificant = false; @@ -923,7 +888,7 @@ $update_insignificant = false; } - if (db_escape_string($link, $orig_title) != $entry_title) { + if (db_escape_string($orig_title) != $entry_title) { $post_needs_update = true; $update_insignificant = false; } @@ -933,12 +898,12 @@ if ($post_needs_update) { if (defined('DAEMON_EXTENDED_DEBUG')) { - _debug("update_rss_feed: post $entry_guid needs update..."); + _debug("post $entry_guid_hashed needs update...", $debug_enabled); } // print ""; - db_query($link, "UPDATE ttrss_entries + db_query("UPDATE ttrss_entries SET title = '$entry_title', content = '$entry_content', content_hash = '$content_hash', updated = '$entry_timestamp_fmt', @@ -948,25 +913,21 @@ if (!$update_insignificant) { if ($mark_unread_on_update) { - db_query($link, "UPDATE ttrss_user_entries + db_query("UPDATE ttrss_user_entries SET last_read = null, unread = true WHERE ref_id = '$ref_id'"); } } } } - db_query($link, "COMMIT"); + db_query("COMMIT"); - if ($debug_enabled) { - _debug("update_rss_feed: assigning labels..."); - } + _debug("assigning labels...", $debug_enabled); - assign_article_to_label_filters($link, $entry_ref_id, $article_filters, + assign_article_to_label_filters($entry_ref_id, $article_filters, $owner_uid, $article_labels); - if ($debug_enabled) { - _debug("update_rss_feed: looking for enclosures..."); - } + _debug("looking for enclosures...", $debug_enabled); // enclosures @@ -983,28 +944,28 @@ } if ($debug_enabled) { - _debug("update_rss_feed: article enclosures:"); + _debug("article enclosures:", $debug_enabled); print_r($enclosures); } - db_query($link, "BEGIN"); + db_query("BEGIN"); foreach ($enclosures as $enc) { - $enc_url = db_escape_string($link, $enc[0]); - $enc_type = db_escape_string($link, $enc[1]); - $enc_dur = db_escape_string($link, $enc[2]); + $enc_url = db_escape_string($enc[0]); + $enc_type = db_escape_string($enc[1]); + $enc_dur = db_escape_string($enc[2]); - $result = db_query($link, "SELECT id FROM ttrss_enclosures + $result = db_query("SELECT id FROM ttrss_enclosures WHERE content_url = '$enc_url' AND post_id = '$entry_ref_id'"); if (db_num_rows($result) == 0) { - db_query($link, "INSERT INTO ttrss_enclosures + db_query("INSERT INTO ttrss_enclosures (content_url, content_type, title, duration, post_id) VALUES ('$enc_url', '$enc_type', '', '$enc_dur', '$entry_ref_id')"); } } - db_query($link, "COMMIT"); + db_query("COMMIT"); // check for manual tags (we have to do it here since they're loaded from filters) @@ -1023,7 +984,7 @@ // Skip boring tags - $boring_tags = trim_array(explode(",", mb_strtolower(get_pref($link, + $boring_tags = trim_array(explode(",", mb_strtolower(get_pref( 'BLACKLISTED_TAGS', $owner_uid, ''), 'utf-8'))); $filtered_tags = array(); @@ -1040,7 +1001,7 @@ $filtered_tags = array_unique($filtered_tags); if ($debug_enabled) { - _debug("update_rss_feed: filtered article tags:"); + _debug("filtered article tags:", $debug_enabled); print_r($filtered_tags); } @@ -1048,22 +1009,22 @@ if (count($filtered_tags) > 0) { - db_query($link, "BEGIN"); + db_query("BEGIN"); foreach ($filtered_tags as $tag) { $tag = sanitize_tag($tag); - $tag = db_escape_string($link, $tag); + $tag = db_escape_string($tag); if (!tag_is_valid($tag)) continue; - $result = db_query($link, "SELECT id FROM ttrss_tags + $result = db_query("SELECT id FROM ttrss_tags WHERE tag_name = '$tag' AND post_int_id = '$entry_int_id' AND owner_uid = '$owner_uid' LIMIT 1"); if ($result && db_num_rows($result) == 0) { - db_query($link, "INSERT INTO ttrss_tags + db_query("INSERT INTO ttrss_tags (owner_uid,tag_name,post_int_id) VALUES ('$owner_uid','$tag', '$entry_int_id')"); } @@ -1075,73 +1036,60 @@ $tags_to_cache = array_unique($tags_to_cache); - $tags_str = db_escape_string($link, join(",", $tags_to_cache)); + $tags_str = db_escape_string(join(",", $tags_to_cache)); - db_query($link, "UPDATE ttrss_user_entries + db_query("UPDATE ttrss_user_entries SET tag_cache = '$tags_str' WHERE ref_id = '$entry_ref_id' AND owner_uid = $owner_uid"); - db_query($link, "COMMIT"); + db_query("COMMIT"); } - if (get_pref($link, "AUTO_ASSIGN_LABELS", $owner_uid, false)) { - if ($debug_enabled) { - _debug("update_rss_feed: auto-assigning labels..."); - } + if (get_pref("AUTO_ASSIGN_LABELS", $owner_uid, false)) { + _debug("auto-assigning labels...", $debug_enabled); foreach ($labels as $label) { $caption = preg_quote($label["caption"]); if ($caption && preg_match("/\b$caption\b/i", "$tags_str " . strip_tags($entry_content) . " $entry_title")) { if (!labels_contains_caption($article_labels, $caption)) { - label_add_article($link, $entry_ref_id, $caption, $owner_uid); + label_add_article($entry_ref_id, $caption, $owner_uid); } } } } - if ($debug_enabled) { - _debug("update_rss_feed: article processed"); - } + _debug("article processed", $debug_enabled); } if (!$last_updated) { - if ($debug_enabled) { - _debug("update_rss_feed: new feed, catching it up..."); - } - catchup_feed($link, $feed, false, $owner_uid); + _debug("new feed, catching it up...", $debug_enabled); + catchup_feed($feed, false, $owner_uid); } - if ($debug_enabled) { - _debug("purging feed..."); - } + _debug("purging feed...", $debug_enabled); - purge_feed($link, $feed, 0, $debug_enabled); + purge_feed($feed, 0, $debug_enabled); - db_query($link, "UPDATE ttrss_feeds + db_query("UPDATE ttrss_feeds SET last_updated = NOW(), last_error = '' WHERE id = '$feed'"); -// db_query($link, "COMMIT"); +// db_query("COMMIT"); } else { - $error_msg = db_escape_string($link, mb_substr($rss->error(), 0, 245)); + $error_msg = db_escape_string(mb_substr($rss->error(), 0, 245)); - if ($debug_enabled) { - _debug("update_rss_feed: error fetching feed: $error_msg"); - } + _debug("error fetching feed: $error_msg", $debug_enabled); - db_query($link, + db_query( "UPDATE ttrss_feeds SET last_error = '$error_msg', last_updated = NOW() WHERE id = '$feed'"); } unset($rss); - if ($debug_enabled) { - _debug("update_rss_feed: done"); - } - + _debug("done", $debug_enabled); } function cache_images($html, $site_url, $debug) { @@ -1187,8 +1135,21 @@ return $doc->saveXML($node); } + function expire_error_log($debug) { + if ($debug) _debug("Removing old error log entries..."); + + if (DB_TYPE == "pgsql") { + db_query("DELETE FROM ttrss_error_log + WHERE created_at < NOW() - INTERVAL '7 days'"); + } else { + db_query("DELETE FROM ttrss_error_log + WHERE created_at < DATE_SUB(NOW(), INTERVAL 7 DAY)"); + } + + } + function expire_lock_files($debug) { - if ($debug) _debug("Removing old lock files..."); + //if ($debug) _debug("Removing old lock files..."); $num_deleted = 0; @@ -1197,7 +1158,7 @@ if ($files) { foreach ($files as $file) { - if (!file_is_locked($file) && time() - filemtime($file) > 86400*2) { + if (!file_is_locked(basename($file)) && time() - filemtime($file) > 86400*2) { unlink($file); ++$num_deleted; } @@ -1205,14 +1166,14 @@ } } - if ($debug) _debug("Removed $num_deleted files."); + if ($debug) _debug("Removed $num_deleted old lock files."); } function expire_cached_files($debug) { foreach (array("simplepie", "images", "export", "upload") as $dir) { $cache_dir = CACHE_DIR . "/$dir"; - if ($debug) _debug("Expiring $cache_dir"); +// if ($debug) _debug("Expiring $cache_dir"); $num_deleted = 0; @@ -1230,7 +1191,7 @@ } } - if ($debug) _debug("Removed $num_deleted files."); + if ($debug) _debug("$cache_dir: removed $num_deleted files."); } } @@ -1264,7 +1225,7 @@ foreach ($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) @@ -1293,8 +1254,12 @@ $match = @preg_match("/$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)) { + $match = true; + break; + } + } break; } @@ -1369,11 +1334,11 @@ return false; } - function assign_article_to_label_filters($link, $id, $filters, $owner_uid, $article_labels) { + function assign_article_to_label_filters($id, $filters, $owner_uid, $article_labels) { foreach ($filters as $f) { if ($f["type"] == "label") { if (!labels_contains_caption($article_labels, $f["param"])) { - label_add_article($link, $id, $f["param"], $owner_uid); + label_add_article($id, $f["param"], $owner_uid); } } } @@ -1384,5 +1349,26 @@ mb_strtolower(strip_tags($title), 'utf-8')); } + /* 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($debug); + expire_lock_files($debug); + expire_error_log($debug); + + $count = update_feedbrowser_cache(); + _debug("Feedbrowser updated, $count feeds processed."); + + purge_orphans( true); + $rc = cleanup_tags( 14, 50000); + + _debug("Cleaned $rc cached tags."); + } ?>