Index: class-wp-xmlrpc-server.php =================================================================== --- class-wp-xmlrpc-server.php (revision 34138) +++ class-wp-xmlrpc-server.php (working copy) @@ -6115,43 +6115,10 @@ if ( !$pos1 ) return $this->pingback_error( 0, __( 'Is there no link to us?' ) ); - // let's find which post is linked to - // FIXME: does url_to_postid() cover all these cases already? - // if so, then let's use it and drop the old code. - $urltest = parse_url($pagelinkedto); - if ( $post_ID = url_to_postid($pagelinkedto) ) { - // $way - } elseif ( isset( $urltest['path'] ) && preg_match('#p/[0-9]{1,}#', $urltest['path'], $match) ) { - // the path defines the post_ID (archives/p/XXXX) - $blah = explode('/', $match[0]); - $post_ID = (int) $blah[1]; - } elseif ( isset( $urltest['query'] ) && preg_match('#p=[0-9]{1,}#', $urltest['query'], $match) ) { - // the querystring defines the post_ID (?p=XXXX) - $blah = explode('=', $match[0]); - $post_ID = (int) $blah[1]; - } elseif ( isset($urltest['fragment']) ) { - // an #anchor is there, it's either... - if ( intval($urltest['fragment']) ) { - // ...an integer #XXXX (simplest case) - $post_ID = (int) $urltest['fragment']; - } elseif ( preg_match('/post-[0-9]+/',$urltest['fragment']) ) { - // ...a post id in the form 'post-###' - $post_ID = preg_replace('/[^0-9]+/', '', $urltest['fragment']); - } elseif ( is_string($urltest['fragment']) ) { - // ...or a string #title, a little more complicated - $title = preg_replace('/[^a-z0-9]/i', '.', $urltest['fragment']); - $sql = $wpdb->prepare("SELECT ID FROM $wpdb->posts WHERE post_title RLIKE %s", $title ); - if (! ($post_ID = $wpdb->get_var($sql)) ) { - // returning unknown error '0' is better than die()ing - return $this->pingback_error( 0, '' ); - } - } - } else { - // TODO: Attempt to extract a post ID from the given URL - return $this->pingback_error( 33, __('The specified target URL cannot be used as a target. It either doesn’t exist, or it is not a pingback-enabled resource.' ) ); - } - $post_ID = (int) $post_ID; + // let's find which post is linked to. + $post_ID = url_to_postid($pagelinkedto); + $post_ID = intval( $post_ID ); $post = get_post($post_ID); if ( !$post ) // Post_ID not found @@ -6201,69 +6168,55 @@ * @param string $pagelinkedto URL of the page linked to. */ $linea = apply_filters( 'pre_remote_source', $linea, $pagelinkedto ); + /** + * Filter the pingback remote source through kses. + * + * @since + * + * @param array $allowed_tags Tags permitted through the filter, defaults to wp_kses_post. + * @param string $pagelinkedto URL of the page linked to. + */ - // Work around bug in strip_tags(): - $linea = str_replace(']*>/", "\n\n", $linea ); + $allowed_tags = apply_filters( 'filter_pingback_source', wp_kses_allowed_html( 'post' ); + $linea = wp_kses( $linea, $allowed_tags ); - preg_match('|([^<]*?)|is', $linea, $matchtitle); - $title = $matchtitle[1]; - if ( empty( $title ) ) - return $this->pingback_error( 32, __('We cannot find a title on that page.' ) ); + // check if source really links to target + if ( ! strpos( $linea, str_replace( array( 'http://www.', 'http://', 'https://www.', 'https://' ), '', untrailingslashit( preg_replace( '/#.*/', '', $pagelinkedto ) ) ) ) ) { + return $this->pingback_error( 17, __( 'The source URL does not contain a link to the target URL, and so cannot be used as a source.' ) ); + } - $linea = strip_tags( $linea, '' ); // just keep the tag we need + $host = parse_url( $source, PHP_URL_HOST ); + // strip leading www, if any + $host = preg_replace( '/^www\./', '', $host ); - $p = explode( "\n\n", $linea ); - - $preg_target = preg_quote($pagelinkedto, '|'); - - foreach ( $p as $para ) { - if ( strpos($para, $pagelinkedto) !== false ) { // it exists, but is it a link? - preg_match("|]+?".$preg_target."[^>]*>([^>]+?)|", $para, $context); - - // If the URL isn't in a link context, keep looking - if ( empty($context) ) - continue; - - // We're going to use this fake tag to mark the context in a bit - // the marker is needed in case the link text appears more than once in the paragraph - $excerpt = preg_replace('|\|', '', $para); - - // prevent really long link text - if ( strlen($context[1]) > 100 ) - $context[1] = substr($context[1], 0, 100) . '…'; - - $marker = ''.$context[1].''; // set up our marker - $excerpt= str_replace($context[0], $marker, $excerpt); // swap out the link for our marker - $excerpt = strip_tags($excerpt, ''); // strip all tags but our context marker - $excerpt = trim($excerpt); - $preg_marker = preg_quote($marker, '|'); - $excerpt = preg_replace("|.*?\s(.{0,100}$preg_marker.{0,100})\s.*|s", '$1', $excerpt); - $excerpt = strip_tags($excerpt); // YES, again, to remove the marker wrapper - break; - } + $meta_tags = @get_meta_tags( $source ); + // use meta-author + if ( $meta_tags && is_array( $meta_tags ) && array_key_exists( 'author', $meta_tags ) ) { + $title = $meta_tags['author']; + } elseif ( preg_match( '/(.+)<\/title>/i', $contents, $match ) ) { // use title + $title = trim( $match[1] ); + } else { // or host + // strip leading www, if any + $title = $host; } - if ( empty($context) ) // Link to target not found - return $this->pingback_error( 17, __( 'The source URL does not contain a link to the target URL, and so cannot be used as a source.' ) ); - $pagelinkedfrom = str_replace('&', '&', $pagelinkedfrom); - - $context = '[…] ' . esc_html( $excerpt ) . ' […]'; $pagelinkedfrom = $this->escape( $pagelinkedfrom ); + // generate default text + $context = sprintf( __( 'This post was mentioned on <a href="%s">%s</a>' ), esc_url( $pagelinkedfrom ), $host ); + $comment_post_ID = (int) $post_ID; $comment_author = $title; $comment_author_email = ''; $this->escape($comment_author); $comment_author_url = $pagelinkedfrom; + $comment_content = $context; $this->escape($comment_content); $comment_type = 'pingback'; - $commentdata = compact('comment_post_ID', 'comment_author', 'comment_author_url', 'comment_author_email', 'comment_content', 'comment_type'); - + $commentdata = compact('comment_post_ID', 'comment_author', 'comment_author_url', 'comment_author_email', 'comment_content', 'comment_type', 'linea' ); $comment_ID = wp_new_comment($commentdata); /** Index: comment-functions.php =================================================================== --- comment-functions.php (revision 34138) +++ comment-functions.php (working copy) @@ -1429,6 +1429,12 @@ * @param WP_Comment $comment Comment object. */ do_action( 'wp_insert_comment', $id, $comment ); + + // If there is something extra in $commentdata save it as meta. + $commentmeta = array_diff( $commentdata, $compacted ); + foreach ( $commentmeta as $key => $value ) { + update_comment_meta( $id, $key, $value, true ); + } wp_cache_set( 'last_changed', microtime(), 'comment' ); @@ -1574,7 +1580,10 @@ * @param array $commentdata Comment data. */ $commentdata = apply_filters( 'preprocess_comment', $commentdata ); - + // Unset linea from a pingback after it is made available to preprocessing the pingback. + if ( isset( $commentdata['linea' ) ) { + unset( $commentdata['linea'] ); + } $commentdata['comment_post_ID'] = (int) $commentdata['comment_post_ID']; if ( isset( $commentdata['user_ID'] ) && $prefiltered_user_id !== (int) $commentdata['user_ID'] ) { $commentdata['user_id'] = $commentdata['user_ID'] = (int) $commentdata['user_ID'];