X-Git-Url: https://git.wh0rd.org/?a=blobdiff_plain;f=include%2Frssfuncs.php;h=01a11da6b535183096cb04158c82834575942160;hb=8292d05b7c9723bc91ccdf92f6acc5e46f58ce6c;hp=eaa975a5af04f4b4f09cc8aca6cf732c3e5cf06f;hpb=35cb2b8f67368be40f8f5365a3987bf9f1a5ff60;p=tt-rss.git diff --git a/include/rssfuncs.php b/include/rssfuncs.php index eaa975a5..01a11da6 100644 --- a/include/rssfuncs.php +++ b/include/rssfuncs.php @@ -18,10 +18,10 @@ $count = 0; while ($line = db_fetch_assoc($result)) { - $subscribers = db_escape_string($line["subscribers"]); - $feed_url = db_escape_string($line["feed_url"]); - $title = db_escape_string($line["title"]); - $site_url = db_escape_string($line["site_url"]); + $subscribers = db_escape_string($link, $line["subscribers"]); + $feed_url = db_escape_string($link, $line["feed_url"]); + $title = db_escape_string($link, $line["title"]); + $site_url = db_escape_string($link, $line["site_url"]); $tmp_result = db_query($link, "SELECT subscribers FROM ttrss_feedbrowser_cache WHERE feed_url = '$feed_url'"); @@ -108,9 +108,7 @@ $random_qpart = sql_random_function(); // We search for feed needing update. - $result = db_query($link, "SELECT ttrss_feeds.feed_url,ttrss_feeds.id, ttrss_feeds.owner_uid, - ".SUBSTRING_FOR_DATE."(ttrss_feeds.last_updated,1,19) AS last_updated, - ttrss_feeds.update_interval + $result = db_query($link, "SELECT DISTINCT ttrss_feeds.feed_url FROM ttrss_feeds, ttrss_users, ttrss_user_prefs WHERE @@ -118,44 +116,71 @@ AND ttrss_users.id = ttrss_user_prefs.owner_uid AND ttrss_user_prefs.pref_name = 'DEFAULT_UPDATE_INTERVAL' $login_thresh_qpart $update_limit_qpart - $updstart_thresh_qpart - ORDER BY $random_qpart $query_limit"); + $updstart_thresh_qpart + ORDER BY feed_url $query_limit"); $user_prefs_cache = array(); - if($debug) _debug(sprintf("Scheduled %d feeds to update...\n", db_num_rows($result))); + if($debug) _debug(sprintf("Scheduled %d feeds to update...", db_num_rows($result))); // Here is a little cache magic in order to minimize risk of double feed updates. $feeds_to_update = array(); while ($line = db_fetch_assoc($result)) { - $feeds_to_update[$line['id']] = $line; + array_push($feeds_to_update, db_escape_string($link, $line['feed_url'])); } // We update the feed last update started date before anything else. // There is no lag due to feed contents downloads // It prevent an other process to update the same feed. - $feed_ids = array_keys($feeds_to_update); - if($feed_ids) { + + if(count($feeds_to_update) > 0) { + $feeds_quoted = array(); + + foreach ($feeds_to_update as $feed) { + array_push($feeds_quoted, "'" . db_escape_string($link, $feed) . "'"); + } + db_query($link, sprintf("UPDATE ttrss_feeds SET last_update_started = NOW() - WHERE id IN (%s)", implode(',', $feed_ids))); + WHERE feed_url IN (%s)", implode(',', $feeds_quoted))); } expire_cached_files($debug); expire_lock_files($debug); - // For each feed, we call the feed update function. - while ($line = array_pop($feeds_to_update)) { + $nf = 0; - if($debug) _debug("Feed: " . $line["feed_url"] . ", " . $line["last_updated"]); - - update_rss_feed($link, $line["id"], true); - - sleep(1); // prevent flood (FIXME make this an option?) + // For each feed, we call the feed update function. + foreach ($feeds_to_update as $feed) { + if($debug) _debug("Base feed: $feed"); + + //update_rss_feed($link, $line["id"], true); + + // since we have the data cached, we can deal with other feeds with the same url + + $tmp_result = db_query($link, "SELECT ttrss_feeds.feed_url,ttrss_feeds.id,last_updated + FROM ttrss_feeds, ttrss_users WHERE + ttrss_users.id = ttrss_feeds.owner_uid AND + feed_url = '".db_escape_string($link, $feed)."' AND + ttrss_feeds.update_interval != -1 + $login_thresh_qpart + ORDER BY feed_url $query_limit"); + + if (db_num_rows($tmp_result) > 0) { + while ($tline = db_fetch_assoc($tmp_result)) { + if($debug) _debug(" => " . $tline["last_updated"] . ", " . $tline["id"]); + update_rss_feed($link, $tline["id"], true); + ++$nf; + } + } } + require_once "digest.php"; + // Send feed digests by email if needed. send_headlines_digests($link, $debug); + return $nf; + } // function update_daemon_common // ignore_daemon is not used @@ -163,8 +188,6 @@ $override_url = false) { require_once "lib/simplepie/simplepie.inc"; - require_once "lib/magpierss/rss_fetch.inc"; - require_once 'lib/magpierss/rss_utils.inc'; $debug_enabled = defined('DAEMON_EXTENDED_DEBUG') || $_REQUEST['xdebug']; @@ -173,8 +196,8 @@ } $result = db_query($link, "SELECT id,update_interval,auth_login, - feed_url,auth_pass,cache_images,update_method,last_updated,cache_content, - mark_unread_on_update, owner_uid, update_on_checksum_change, + feed_url,auth_pass,cache_images,last_updated, + mark_unread_on_update, owner_uid, pubsub_state FROM ttrss_feeds WHERE id = '$feed'"); @@ -185,13 +208,10 @@ return false; } - $update_method = db_fetch_result($result, 0, "update_method"); $last_updated = db_fetch_result($result, 0, "last_updated"); $owner_uid = db_fetch_result($result, 0, "owner_uid"); $mark_unread_on_update = sql_bool_to_bool(db_fetch_result($result, 0, "mark_unread_on_update")); - $update_on_checksum_change = sql_bool_to_bool(db_fetch_result($result, - 0, "update_on_checksum_change")); $pubsub_state = db_fetch_result($result, 0, "pubsub_state"); db_query($link, "UPDATE ttrss_feeds SET last_update_started = NOW() @@ -200,106 +220,147 @@ $auth_login = db_fetch_result($result, 0, "auth_login"); $auth_pass = db_fetch_result($result, 0, "auth_pass"); - if ($update_method == 0) - $update_method = DEFAULT_UPDATE_METHOD + 1; + $cache_images = sql_bool_to_bool(db_fetch_result($result, 0, "cache_images")); + $fetch_url = db_fetch_result($result, 0, "feed_url"); + + $feed = db_escape_string($link, $feed); - // 1 - Magpie - // 2 - SimplePie - // 3 - Twitter OAuth + if ($override_url) $fetch_url = $override_url; - if ($update_method == 2) - $use_simplepie = true; - else - $use_simplepie = false; + $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) || $last_updated == '1970-01-01 00:00:00') ? + 30 : get_feed_update_interval($link, $feed) * 60; if ($debug_enabled) { - _debug("update method: $update_method (feed setting: $update_method) (use simplepie: $use_simplepie)\n"); + _debug("update_rss_feed: cache filename: $cache_filename exists: " . file_exists($cache_filename)); + _debug("update_rss_feed: cache age: $cache_age; no cache: $no_cache"); } - if ($update_method == 1) { - $auth_login = urlencode($auth_login); - $auth_pass = urlencode($auth_pass); - } + $cached_feed_data_hash = false; - $cache_images = sql_bool_to_bool(db_fetch_result($result, 0, "cache_images")); - $cache_content = sql_bool_to_bool(db_fetch_result($result, 0, "cache_content")); - $fetch_url = db_fetch_result($result, 0, "feed_url"); + $rss = false; + $rss_hash = false; + $cache_timestamp = file_exists($cache_filename) ? filemtime($cache_filename) : 0; + $last_updated_timestamp = strtotime($last_updated); - $feed = db_escape_string($feed); + if (file_exists($cache_filename) && + is_readable($cache_filename) && + !$auth_login && !$auth_pass && + filemtime($cache_filename) > time() - $cache_age) { - if ($auth_login && $auth_pass ){ - $url_parts = array(); - preg_match("/(^[^:]*):\/\/(.*)/", $fetch_url, $url_parts); + if ($debug_enabled) { + _debug("update_rss_feed: using local cache."); + } - if ($url_parts[1] && $url_parts[2]) { - $fetch_url = $url_parts[1] . "://$auth_login:$auth_pass@" . $url_parts[2]; - } + if ($cache_timestamp > $last_updated_timestamp) { + @$rss_data = file_get_contents($cache_filename); + if ($rss_data) { + $rss_hash = sha1($rss_data); + @$rss = unserialize($rss_data); + } + } else { + if ($debug_enabled) { + _debug("update_rss_feed: local cache valid and older than last_updated, nothing to do."); + } + return; + } } - if ($override_url) - $fetch_url = $override_url; + if (!$rss) { - if ($debug_enabled) { - _debug("update_rss_feed: fetching [$fetch_url]..."); - } + if (!$feed_data) { + if ($debug_enabled) { + _debug("update_rss_feed: fetching [$fetch_url] (ts: $cache_timestamp/$last_updated_timestamp)"); + } - // Ignore cache if new feed or manual update. - $cache_age = (is_null($last_updated) || $last_updated == '1970-01-01 00:00:00') ? - -1 : get_feed_update_interval($link, $feed) * 60; + $feed_data = fetch_file_contents($fetch_url, false, + $auth_login, $auth_pass, false, $no_cache ? 15 : 45, + max($last_updated_timestamp, $cache_timestamp)); - if ($update_method == 1) { + if ($debug_enabled) { + _debug("update_rss_feed: fetch done."); + } - define('MAGPIE_CACHE_AGE', $cache_age); - define('MAGPIE_CACHE_ON', !$no_cache); - define('MAGPIE_FETCH_TIME_OUT', $no_cache ? 15 : 60); - define('MAGPIE_CACHE_DIR', CACHE_DIR . "/magpie"); + } - $rss = @fetch_rss($fetch_url); - } else { - $simplepie_cache_dir = CACHE_DIR . "/simplepie"; + if (!$feed_data) { + global $fetch_last_error; + global $fetch_last_error_code; - if (!is_dir($simplepie_cache_dir)) { - mkdir($simplepie_cache_dir); - } + if ($debug_enabled) { + _debug("update_rss_feed: unable to fetch: $fetch_last_error [$fetch_last_error_code]"); + } - $rss = new SimplePie(); - $rss->set_useragent(SELF_USER_AGENT); - $rss->set_timeout($no_cache ? 15 : 60); - $rss->set_feed_url($fetch_url); - $rss->set_output_encoding('UTF-8'); - //$rss->force_feed(true); + $error_escaped = ''; - if ($debug_enabled) { - _debug("feed update interval (sec): " . - get_feed_update_interval($link, $feed)*60); - } + // If-Modified-Since + if ($fetch_last_error_code != 304) { + $error_escaped = db_escape_string($link, $fetch_last_error); + } else { + if ($debug_enabled) { + _debug("update_rss_feed: source claims data not modified, nothing to do."); + } + } - $rss->enable_cache(!$no_cache); + db_query($link, + "UPDATE ttrss_feeds SET last_error = '$error_escaped', + last_updated = NOW() WHERE id = '$feed'"); - if (!$no_cache) { - $rss->set_cache_location($simplepie_cache_dir); - $rss->set_cache_duration($cache_age); + return; } - - $rss->init(); } -// print_r($rss); + $pluginhost = new PluginHost($link); + $pluginhost->set_debug($debug_enabled); + $user_plugins = get_pref($link, "_ENABLED_PLUGINS", $owner_uid); - if ($debug_enabled) { - _debug("update_rss_feed: fetch done, parsing..."); + $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); } - $feed = db_escape_string($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); + $rss->enable_cache(false); - if ($update_method == 2) { - $fetch_ok = !$rss->error(); - } else { - $fetch_ok = !!$rss; + @$rss->init(); } - if ($fetch_ok) { +// print_r($rss); + + $feed = db_escape_string($link, $feed); + + if (!$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)); + } + } + + // 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", $rss); if ($debug_enabled) { _debug("update_rss_feed: processing feed data..."); @@ -313,32 +374,19 @@ $favicon_interval_qpart = "favicon_last_checked < DATE_SUB(NOW(), INTERVAL 12 HOUR)"; } - $result = db_query($link, "SELECT title,icon_url,site_url,owner_uid, + $result = db_query($link, "SELECT title,site_url,owner_uid, (favicon_last_checked IS NULL OR $favicon_interval_qpart) AS favicon_needs_check FROM ttrss_feeds WHERE id = '$feed'"); $registered_title = db_fetch_result($result, 0, "title"); - $orig_icon_url = db_fetch_result($result, 0, "icon_url"); $orig_site_url = db_fetch_result($result, 0, "site_url"); $favicon_needs_check = sql_bool_to_bool(db_fetch_result($result, 0, "favicon_needs_check")); $owner_uid = db_fetch_result($result, 0, "owner_uid"); - if ($use_simplepie) { - $site_url = db_escape_string(trim($rss->get_link())); - } else { - $site_url = db_escape_string(trim($rss->channel["link"])); - } - - // weird, weird Magpie - if (!$use_simplepie) { - if (!$site_url) $site_url = db_escape_string($rss->channel["link_"]); - } - - $site_url = rewrite_relative_url($fetch_url, $site_url); - $site_url = substr($site_url, 0, 250); + $site_url = db_escape_string($link, mb_substr(rewrite_relative_url($fetch_url, $rss->get_link()), 0, 245)); if ($debug_enabled) { _debug("update_rss_feed: checking favicon..."); @@ -353,11 +401,7 @@ if (!$registered_title || $registered_title == "[Unknown]") { - if ($use_simplepie) { - $feed_title = db_escape_string($rss->get_title()); - } else { - $feed_title = db_escape_string($rss->channel["title"]); - } + $feed_title = db_escape_string($link, $rss->get_title()); if ($debug_enabled) { _debug("update_rss_feed: registering title: $feed_title"); @@ -372,21 +416,6 @@ site_url = '$site_url' WHERE id = '$feed'"); } -// print "I: " . $rss->channel["image"]["url"]; - - if (!$use_simplepie) { - $icon_url = db_escape_string(trim($rss->image["url"])); - } else { - $icon_url = db_escape_string(trim($rss->get_image_url())); - } - - $icon_url = rewrite_relative_url($fetch_url, $icon_url); - $icon_url = substr($icon_url, 0, 250); - - if ($icon_url && $orig_icon_url != $icon_url) { - db_query($link, "UPDATE ttrss_feeds SET icon_url = '$icon_url' WHERE id = '$feed'"); - } - if ($debug_enabled) { _debug("update_rss_feed: loading filters & labels..."); } @@ -399,41 +428,11 @@ _debug("update_rss_feed: " . count($filters) . " filters loaded."); } - $filter_plugins = array(); - - if (defined('_ARTICLE_FILTER_PLUGINS')) { - foreach (explode(",", _ARTICLE_FILTER_PLUGINS) as $p) { - $pclass = "filter_" . trim($p); - - if (class_exists($pclass)) { - $plugin = new $pclass($link); - array_push($filter_plugins, $plugin); - } - } - } - - if ($debug_enabled) { - _debug("update_rss_feed: " . count($filter_plugins) . " filter plugins loaded."); - } - - if ($use_simplepie) { - $iterator = $rss->get_items(); - } else { - $iterator = $rss->items; - if (!$iterator || !is_array($iterator)) $iterator = $rss->entries; - if (!$iterator || !is_array($iterator)) $iterator = $rss; - } - - if (!is_array($iterator)) { - /* db_query($link, "UPDATE ttrss_feeds - SET last_error = 'Parse error: can\'t find any articles.' - WHERE id = '$feed'"); */ - - // clear any errors and mark feed as updated if fetched okay - // even if it's blank + $items = $rss->get_items(); + if (!is_array($items)) { if ($debug_enabled) { - _debug("update_rss_feed: entry iterator is not an array, no articles?"); + _debug("update_rss_feed: no articles found."); } db_query($link, "UPDATE ttrss_feeds @@ -447,34 +446,13 @@ if ($debug_enabled) _debug("update_rss_feed: checking for PUSH hub..."); $feed_hub_url = false; - if ($use_simplepie) { - $links = $rss->get_links('hub'); - - if ($links && is_array($links)) { - foreach ($links as $l) { - $feed_hub_url = $l; - break; - } - } - - } else { - $atom = $rss->channel['atom']; - if ($atom) { - if ($atom['link@rel'] == 'hub') { - $feed_hub_url = $atom['link@href']; - } + $links = $rss->get_links('hub'); - if (!$feed_hub_url && $atom['link#'] > 1) { - for ($i = 2; $i <= $atom['link#']; $i++) { - if ($atom["link#$i@rel"] == 'hub') { - $feed_hub_url = $atom["link#$i@href"]; - break; - } - } - } - } else { - $feed_hub_url = $rss->channel['link_hub']; + if ($links && is_array($links)) { + foreach ($links as $l) { + $feed_hub_url = $l; + break; } } @@ -504,33 +482,14 @@ _debug("update_rss_feed: processing articles..."); } - foreach ($iterator as $item) { - if ($_REQUEST['xdebug'] == 2) { + foreach ($items as $item) { + if ($_REQUEST['xdebug'] == 3) { print_r($item); } - if ($use_simplepie) { - $entry_guid = $item->get_id(); - if (!$entry_guid) $entry_guid = $item->get_link(); - if (!$entry_guid) $entry_guid = make_guid_from_title($item->get_title()); - - } else { - - $entry_guid = $item["id"]; - - if (!$entry_guid) $entry_guid = $item["guid"]; - if (!$entry_guid) $entry_guid = $item["about"]; - if (!$entry_guid) $entry_guid = $item["link"]; - if (!$entry_guid) $entry_guid = make_guid_from_title($item["title"]); - } - - if ($cache_content) { - $entry_guid = "ccache:$entry_guid"; - } - - if ($auth_login || $auth_pass) { - $entry_guid = "auth,$owner_uid:$entry_guid"; - } + $entry_guid = $item->get_id(); + if (!$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 $entry_guid"); @@ -538,23 +497,13 @@ if (!$entry_guid) continue; - $entry_timestamp = ""; - - if ($use_simplepie) { - $entry_timestamp = strtotime($item->get_date()); - } else { - $rss_2_date = $item['pubdate']; - $rss_1_date = $item['dc']['date']; - $atom_date = $item['issued']; - if (!$atom_date) $atom_date = $item['updated']; + $entry_guid = "$owner_uid,$entry_guid"; - if ($atom_date != "") $entry_timestamp = parse_w3cdtf($atom_date); - if ($rss_1_date != "") $entry_timestamp = parse_w3cdtf($rss_1_date); - if ($rss_2_date != "") $entry_timestamp = strtotime($rss_2_date); + $entry_timestamp = ""; - } + $entry_timestamp = strtotime($item->get_date()); - if ($entry_timestamp == "" || $entry_timestamp == -1 || !$entry_timestamp) { + if ($entry_timestamp == -1 || !$entry_timestamp || $entry_timestamp > time()) { $entry_timestamp = time(); $no_orig_date = 'true'; } else { @@ -567,21 +516,9 @@ _debug("update_rss_feed: date $entry_timestamp [$entry_timestamp_fmt]"); } - if ($use_simplepie) { - $entry_title = $item->get_title(); - } else { - $entry_title = trim(strip_tags($item["title"])); - } - - if ($use_simplepie) { - $entry_link = $item->get_link(); - } else { - // strange Magpie workaround - $entry_link = $item["link_"]; - if (!$entry_link) $entry_link = $item["link"]; - } + $entry_title = $item->get_title(); - $entry_link = rewrite_relative_url($site_url, $entry_link); + $entry_link = rewrite_relative_url($site_url, $item->get_link()); if ($debug_enabled) { _debug("update_rss_feed: title $entry_title"); @@ -590,40 +527,8 @@ if (!$entry_title) $entry_title = date("Y-m-d H:i:s", $entry_timestamp);; - $entry_link = strip_tags($entry_link); - - if ($use_simplepie) { - $entry_content = $item->get_content(); - if (!$entry_content) $entry_content = $item->get_description(); - } else { - $entry_content = $item["content:escaped"]; - - if (!$entry_content) $entry_content = $item["content:encoded"]; - if (!$entry_content && is_array($entry_content)) $entry_content = $item["content"]["encoded"]; - if (!$entry_content) $entry_content = $item["content"]; - - if (is_array($entry_content)) $entry_content = $entry_content[0]; - - // Magpie bugs are getting ridiculous - if (trim($entry_content) == "Array") $entry_content = false; - - if (!$entry_content) $entry_content = $item["atom_content"]; - if (!$entry_content) $entry_content = $item["summary"]; - - if (!$entry_content || - strlen($entry_content) < strlen($item["description"])) { - $entry_content = $item["description"]; - }; - - // WTF - if (is_array($entry_content)) { - $entry_content = $entry_content["encoded"]; - if (!$entry_content) $entry_content = $entry_content["escaped"]; - } - } - - if ($cache_images && is_writable(CACHE_DIR . '/images')) - $entry_content = cache_images($entry_content, $site_url, $debug_enabled); + $entry_content = $item->get_content(); + if (!$entry_content) $entry_content = $item->get_description(); if ($_REQUEST["xdebug"] == 2) { print "update_rss_feed: content: "; @@ -631,61 +536,31 @@ print "\n"; } - $entry_cached_content = ""; - - if ($use_simplepie) { - $entry_comments = strip_tags($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(); - - $entry_author = db_escape_string($entry_author); - } - } else { - $entry_comments = strip_tags($item["comments"]); - - $entry_author = db_escape_string(strip_tags($item['dc']['creator'])); - - if ($item['author']) { - - if (is_array($item['author'])) { + $entry_comments = $item->data["comments"]; - if (!$entry_author) { - $entry_author = db_escape_string(strip_tags($item['author']['name'])); - } - - if (!$entry_author) { - $entry_author = db_escape_string(strip_tags($item['author']['email'])); - } - } + if ($item->get_author()) { + $entry_author_item = $item->get_author(); + $entry_author = $entry_author_item->get_name(); + if (!$entry_author) $entry_author = $entry_author_item->get_email(); - if (!$entry_author) { - $entry_author = db_escape_string(strip_tags($item['author'])); - } - } + $entry_author = db_escape_string($link, $entry_author); } - if (preg_match('/^[\t\n\r ]*$/', $entry_author)) $entry_author = ''; + $entry_guid = db_escape_string($link, mb_substr($entry_guid, 0, 245)); - $entry_guid = db_escape_string(strip_tags($entry_guid)); - $entry_guid = mb_substr($entry_guid, 0, 250); - - $result = db_query($link, "SELECT id FROM ttrss_entries - WHERE guid = '$entry_guid'"); + $entry_comments = db_escape_string($link, mb_substr($entry_comments, 0, 245)); + $entry_author = db_escape_string($link, mb_substr($entry_author, 0, 245)); - $entry_comments = mb_substr(db_escape_string($entry_comments), 0, 250); - $entry_author = mb_substr($entry_author, 0, 250); + $num_comments = $item->get_item_tags('http://purl.org/rss/1.0/modules/slash/', 'comments'); - if ($use_simplepie) { - $num_comments = 0; #FIXME# + if (is_array($num_comments) && is_array($num_comments[0])) { + $num_comments = (int) $num_comments[0]["data"]; } else { - $num_comments = db_escape_string($item["slash"]["comments"]); + $num_comments = 0; } - if (!$num_comments) $num_comments = 0; - if ($debug_enabled) { + _debug("update_rss_feed: num_comments: $num_comments"); _debug("update_rss_feed: looking for tags [1]..."); } @@ -693,76 +568,24 @@ $additional_tags = array(); - if ($use_simplepie) { - - $additional_tags_src = $item->get_categories(); - - if (is_array($additional_tags_src)) { - foreach ($additional_tags_src as $tobj) { - array_push($additional_tags, $tobj->get_term()); - } - } - - if ($debug_enabled) { - _debug("update_rss_feed: category tags:"); - print_r($additional_tags); - } - - } else { - - $t_ctr = $item['category#']; + $additional_tags_src = $item->get_categories(); - if ($t_ctr == 0) { - $additional_tags = array(); - } else if ($t_ctr > 0) { - $additional_tags = array($item['category']); - - if ($item['category@term']) { - array_push($additional_tags, $item['category@term']); - } - - for ($i = 0; $i <= $t_ctr; $i++ ) { - if ($item["category#$i"]) { - array_push($additional_tags, $item["category#$i"]); - } - - if ($item["category#$i@term"]) { - array_push($additional_tags, $item["category#$i@term"]); - } - } + if (is_array($additional_tags_src)) { + foreach ($additional_tags_src as $tobj) { + array_push($additional_tags, $tobj->get_term()); } + } - // parse elements - - $t_ctr = $item['dc']['subject#']; - - if ($t_ctr > 0) { - array_push($additional_tags, $item['dc']['subject']); - - for ($i = 0; $i <= $t_ctr; $i++ ) { - if ($item['dc']["subject#$i"]) { - array_push($additional_tags, $item['dc']["subject#$i"]); - } - } - } + if ($debug_enabled) { + _debug("update_rss_feed: category tags:"); + print_r($additional_tags); } if ($debug_enabled) { _debug("update_rss_feed: looking for tags [2]..."); } - /* taaaags */ - // , // - - $entry_tags = null; - - preg_match_all("/([^<]+)<\/a>/i", - $entry_content, $entry_tags); - - $entry_tags = $entry_tags[1]; - - $entry_tags = array_merge($entry_tags, $additional_tags); - $entry_tags = array_unique($entry_tags); + $entry_tags = array_unique($additional_tags); for ($i = 0; $i < count($entry_tags); $i++) $entry_tags[$i] = mb_strtolower($entry_tags[$i], 'utf-8'); @@ -773,58 +596,74 @@ } if ($debug_enabled) { - _debug("update_rss_feed: done collecting data [TITLE:$entry_title]"); + _debug("update_rss_feed: done collecting data."); } // TODO: less memory-hungry implementation - if (count($filter_plugins) > 0) { - if ($debug_enabled) { - _debug("update_rss_feed: applying plugin filters..."); - } - $article = array("owner_uid" => $owner_uid, - "title" => $entry_title, - "content" => $entry_content, - "link" => $entry_link, - "tags" => $entry_tags, - "author" => $entry_author); + if ($debug_enabled) { + _debug("update_rss_feed: applying plugin filters.."); + } - foreach ($filter_plugins as $plugin) { - $article = $plugin->filter_article($article); - } + // 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 = $owner_uid"); + + if (db_num_rows($result) != 0) { + $entry_plugin_data = db_fetch_result($result, 0, "plugin_data"); + $stored_article = array("title" => db_fetch_result($result, 0, "title"), + "content" => db_fetch_result($result, 0, "content"), + "link" => db_fetch_result($result, 0, "link"), + "tags" => explode(",", db_fetch_result($result, 0, "tag_cache")), + "author" => db_fetch_result($result, 0, "author")); + } else { + $entry_plugin_data = ""; + $stored_article = array(); + } - $entry_title = $article["title"]; - $entry_content = $article["content"]; - $entry_tags = $article["tags"]; - $entry_author = $article["author"]; + $article = array("owner_uid" => $owner_uid, // read only + "guid" => $entry_guid, // read only + "title" => $entry_title, + "content" => $entry_content, + "link" => $entry_link, + "tags" => $entry_tags, + "plugin_data" => $entry_plugin_data, + "author" => $entry_author, + "stored" => $stored_article); + + foreach ($pluginhost->get_hooks($pluginhost::HOOK_ARTICLE_FILTER) as $plugin) { + $article = $plugin->hook_article_filter($article); } - $entry_content = db_escape_string($entry_content, false); - $entry_title = db_escape_string($entry_title); - $entry_author = db_escape_string($entry_author); - $entry_link = db_escape_string($entry_link); + $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_content = $article["content"]; // escaped below - $content_hash = "SHA1:" . sha1(strip_tags($entry_content)); - db_query($link, "BEGIN"); + if ($debug_enabled) { + _debug("update_rss_feed: plugin data: $entry_plugin_data"); + } - if (db_num_rows($result) == 0) { + if ($cache_images && is_writable(CACHE_DIR . '/images')) + cache_images($entry_content, $site_url, $debug_enabled); - if ($debug_enabled) { - _debug("update_rss_feed: base guid not found"); - } + $entry_content = db_escape_string($link, $entry_content, false); - if ($cache_content) { - if ($debug_enabled) { - _debug("update_rss_feed: caching content (initial)..."); - } + $content_hash = "SHA1:" . sha1($entry_content); + + db_query($link, "BEGIN"); - $entry_cached_content = cache_content($link, $entry_link, $auth_login, $auth_pass); + $result = db_query($link, "SELECT id FROM ttrss_entries + WHERE guid = '$entry_guid'"); - if ($cache_images && is_writable(CACHE_DIR . '/images')) - $entry_cached_content = cache_images($entry_cached_content, $site_url, $debug_enabled); + if (db_num_rows($result) == 0) { - $entry_cached_content = db_escape_string($entry_cached_content, false); + if ($debug_enabled) { + _debug("update_rss_feed: base guid [$entry_guid] not found"); } // base post entry does not exist, create it @@ -843,6 +682,7 @@ date_entered, comments, num_comments, + plugin_data, author) VALUES ('$entry_title', @@ -851,12 +691,13 @@ '$entry_timestamp_fmt', '$entry_content', '$content_hash', - '$entry_cached_content', + '', $no_orig_date, NOW(), - NOW(), + '$date_feed_processed', '$entry_comments', '$num_comments', + '$entry_plugin_data', '$entry_author')"); $article_labels = array(); @@ -878,10 +719,10 @@ // now it should exist, if not - bad luck then $result = db_query($link, "SELECT - id,content_hash,no_orig_date,title, + id,content_hash,no_orig_date,title,plugin_data, ".SUBSTRING_FOR_DATE."(date_updated,1,19) as date_updated, ".SUBSTRING_FOR_DATE."(updated,1,19) as updated, - num_comments, cached_content + num_comments FROM ttrss_entries WHERE guid = '$entry_guid'"); @@ -892,16 +733,16 @@ if (db_num_rows($result) == 1) { if ($debug_enabled) { - _debug("update_rss_feed: base guid found, checking for user record"); + _debug("update_rss_feed: base guid [$entry_guid] found, checking for user record"); } // this will be used below in update handler $orig_content_hash = db_fetch_result($result, 0, "content_hash"); $orig_title = db_fetch_result($result, 0, "title"); $orig_num_comments = db_fetch_result($result, 0, "num_comments"); - $orig_cached_content = trim(db_fetch_result($result, 0, "cached_content")); $orig_date_updated = strtotime(db_fetch_result($result, 0, "date_updated")); + $orig_plugin_data = db_fetch_result($result, 0, "plugin_data"); $ref_id = db_fetch_result($result, 0, "id"); $entry_ref_id = $ref_id; @@ -995,12 +836,17 @@ } } + $last_marked = ($marked == 'true') ? 'NOW()' : 'NULL'; + $last_published = ($published == 'true') ? 'NOW()' : 'NULL'; + $result = db_query($link, "INSERT INTO ttrss_user_entries (ref_id, owner_uid, feed_id, unread, last_read, marked, - published, score, tag_cache, label_cache, uuid) + published, score, tag_cache, label_cache, uuid, + last_marked, last_published) VALUES ('$ref_id', '$owner_uid', '$feed', $unread, - $last_read_qpart, $marked, $published, '$score', '', '', '')"); + $last_read_qpart, $marked, $published, '$score', '', '', + '', $last_marked, $last_published)"); if (PUBSUBHUBBUB_HUB && $published == 'true') { $rss_link = get_self_url_prefix() . @@ -1035,40 +881,23 @@ $post_needs_update = false; $update_insignificant = false; - $cached_content_needs_update = false; if ($orig_num_comments != $num_comments) { $post_needs_update = true; $update_insignificant = true; } - if ($content_hash != $orig_content_hash) { + if ($entry_plugin_data != $orig_plugin_data) { $post_needs_update = true; - $update_insignificant = false; - $cached_content_needs_update = true; + $update_insignificant = true; } - if ($cache_content) { - if ($debug_enabled) { - _debug("update_rss_feed: caching content because original checksum changed..."); - } - - $entry_cached_content = cache_content($link, $entry_link, $auth_login, $auth_pass); - - if ($entry_cached_content) { - if ($cache_images && is_writable(CACHE_DIR . '/images')) - $entry_cached_content = cache_images($entry_cached_content, $site_url, $debug_enabled); - - $entry_cached_content = db_escape_string($entry_cached_content, false); - $post_needs_update = true; - } else { - $entry_cached_content = db_escape_string($orig_cached_content); - } - } else { - $entry_cached_content = db_escape_string($orig_cached_content); + if ($content_hash != $orig_content_hash) { + $post_needs_update = true; + $update_insignificant = false; } - if (db_escape_string($orig_title) != $entry_title) { + if (db_escape_string($link, $orig_title) != $entry_title) { $post_needs_update = true; $update_insignificant = false; } @@ -1086,19 +915,15 @@ db_query($link, "UPDATE ttrss_entries SET title = '$entry_title', content = '$entry_content', content_hash = '$content_hash', - cached_content = '$entry_cached_content', updated = '$entry_timestamp_fmt', - num_comments = '$num_comments' + num_comments = '$num_comments', + plugin_data = '$entry_plugin_data' WHERE id = '$ref_id'"); if (!$update_insignificant) { if ($mark_unread_on_update) { db_query($link, "UPDATE ttrss_user_entries SET last_read = null, unread = true WHERE ref_id = '$ref_id'"); - } else if ($update_on_checksum_change) { - db_query($link, "UPDATE ttrss_user_entries - SET last_read = null WHERE ref_id = '$ref_id' - AND unread = false"); } } } @@ -1121,67 +946,16 @@ $enclosures = array(); - if ($use_simplepie) { - $encs = $item->get_enclosures(); - - if (is_array($encs)) { - foreach ($encs as $e) { - $e_item = array( - $e->link, $e->type, $e->length); - - array_push($enclosures, $e_item); - } - } - - } else { - // - - $e_ctr = $item['enclosure#']; - - if ($e_ctr > 0) { - $e_item = array($item['enclosure@url'], - $item['enclosure@type'], - $item['enclosure@length']); - - array_push($enclosures, $e_item); - - for ($i = 0; $i <= $e_ctr; $i++ ) { - - if ($item["enclosure#$i@url"]) { - $e_item = array($item["enclosure#$i@url"], - $item["enclosure#$i@type"], - $item["enclosure#$i@length"]); - array_push($enclosures, $e_item); - } - } - } - - // - // can there be many of those? yes -fox - - $m_ctr = $item['media']['content#']; - - if ($m_ctr > 0) { - $e_item = array($item['media']['content@url'], - $item['media']['content@medium'], - $item['media']['content@length']); + $encs = $item->get_enclosures(); + if (is_array($encs)) { + foreach ($encs as $e) { + $e_item = array( + $e->link, $e->type, $e->length); array_push($enclosures, $e_item); - - for ($i = 0; $i <= $m_ctr; $i++ ) { - - if ($item["media"]["content#$i@url"]) { - $e_item = array($item["media"]["content#$i@url"], - $item["media"]["content#$i@medium"], - $item["media"]["content#$i@length"]); - array_push($enclosures, $e_item); - } - } - } } - if ($debug_enabled) { _debug("update_rss_feed: article enclosures:"); print_r($enclosures); @@ -1190,9 +964,9 @@ db_query($link, "BEGIN"); foreach ($enclosures as $enc) { - $enc_url = db_escape_string($enc[0]); - $enc_type = db_escape_string($enc[1]); - $enc_dur = db_escape_string($enc[2]); + $enc_url = db_escape_string($link, $enc[0]); + $enc_type = db_escape_string($link, $enc[1]); + $enc_dur = db_escape_string($link, $enc[2]); $result = db_query($link, "SELECT id FROM ttrss_enclosures WHERE content_url = '$enc_url' AND post_id = '$entry_ref_id'"); @@ -1253,7 +1027,7 @@ foreach ($filtered_tags as $tag) { $tag = sanitize_tag($tag); - $tag = db_escape_string($tag); + $tag = db_escape_string($link, $tag); if (!tag_is_valid($tag)) continue; @@ -1275,7 +1049,7 @@ $tags_to_cache = array_unique($tags_to_cache); - $tags_str = db_escape_string(join(",", $tags_to_cache)); + $tags_str = db_escape_string($link, join(",", $tags_to_cache)); db_query($link, "UPDATE ttrss_user_entries SET tag_cache = '$tags_str' WHERE ref_id = '$entry_ref_id' @@ -1290,9 +1064,9 @@ } foreach ($labels as $label) { - $caption = $label["caption"]; + $caption = preg_quote($label["caption"]); - if (preg_match("/\b$caption\b/i", "$tags_str " . strip_tags($entry_content) . " $entry_title")) { + if ($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, $owner_uid); } @@ -1325,26 +1099,18 @@ } else { - if ($use_simplepie) { - $error_msg = mb_substr($rss->error(), 0, 250); - } else { - $error_msg = mb_substr(magpie_error(), 0, 250); - } + $error_msg = db_escape_string($link, mb_substr($rss->error(), 0, 245)); if ($debug_enabled) { _debug("update_rss_feed: error fetching feed: $error_msg"); } - $error_msg = db_escape_string($error_msg); - db_query($link, "UPDATE ttrss_feeds SET last_error = '$error_msg', last_updated = NOW() WHERE id = '$feed'"); } - if ($use_simplepie) { - unset($rss); - } + unset($rss); if ($debug_enabled) { _debug("update_rss_feed: done"); @@ -1392,7 +1158,7 @@ $node = $doc->getElementsByTagName('body')->item(0); - return $doc->saveXML($node, LIBXML_NOEMPTYTAG); + return $doc->saveXML($node); } function expire_lock_files($debug) { @@ -1417,7 +1183,7 @@ } function expire_cached_files($debug) { - foreach (array("magpie", "simplepie", "images", "export") as $dir) { + foreach (array("simplepie", "images", "export") as $dir) { $cache_dir = CACHE_DIR . "/$dir"; if ($debug) _debug("Expiring $cache_dir"); @@ -1467,11 +1233,13 @@ foreach ($filters as $filter) { $match_any_rule = $filter["match_any_rule"]; + $inverse = $filter["inverse"]; $filter_match = false; foreach ($filter["rules"] as $rule) { $match = false; $reg_exp = $rule["reg_exp"]; + $rule_inverse = $rule["inverse"]; if (!$reg_exp) continue; @@ -1504,6 +1272,8 @@ break; } + if ($rule_inverse) $match = !$match; + if ($match_any_rule) { if ($match) { $filter_match = true; @@ -1517,6 +1287,8 @@ } } + if ($inverse) $filter_match = !$filter_match; + if ($filter_match) { foreach ($filter["actions"] AS $action) { array_push($matches, $action); @@ -1578,24 +1350,10 @@ } } - function cache_content($link, $url, $login, $pass) { - - $content = fetch_file_contents($url, $login, $pass); - - if ($content) { - $doc = new DOMDocument(); - @$doc->loadHTML($content); - $xpath = new DOMXPath($doc); - - $node = $doc->getElementsByTagName('body')->item(0); - - if ($node) { - $content = $doc->saveXML($node, LIBXML_NOEMPTYTAG); + function make_guid_from_title($title) { + return preg_replace("/[ \"\',.:;]/", "-", + mb_strtolower(strip_tags($title), 'utf-8')); + } - return $content; - } - } - return ""; - } ?>