From 3ed191998d4c870d0a39fbe96ecf0b4eba4dc396 Mon Sep 17 00:00:00 2001 From: Theo Chatzimichos Date: Thu, 10 Mar 2011 01:30:01 +0200 Subject: Add wordpress importer --- plugins/wordpress-importer/parsers.php | 601 +++++++++++++++++++++++++++++++++ 1 file changed, 601 insertions(+) create mode 100644 plugins/wordpress-importer/parsers.php (limited to 'plugins/wordpress-importer/parsers.php') diff --git a/plugins/wordpress-importer/parsers.php b/plugins/wordpress-importer/parsers.php new file mode 100644 index 00000000..17247296 --- /dev/null +++ b/plugins/wordpress-importer/parsers.php @@ -0,0 +1,601 @@ +parse( $file ); + + // If SimpleXML succeeds or this is an invalid WXR file then return the results + if ( ! is_wp_error( $result ) || 'SimpleXML_parse_error' != $result->get_error_code() ) + return $result; + } else if ( extension_loaded( 'xml' ) ) { + $parser = new WXR_Parser_XML; + $result = $parser->parse( $file ); + + // If XMLParser succeeds or this is an invalid WXR file then return the results + if ( ! is_wp_error( $result ) || 'XML_parse_error' != $result->get_error_code() ) + return $result; + } + + // We have a malformed XML file, so display the error and fallthrough to regex + if ( isset($result) && defined('IMPORT_DEBUG') && IMPORT_DEBUG ) { + echo '
';
+			if ( 'SimpleXML_parse_error' == $result->get_error_code() ) {
+				foreach  ( $result->get_error_data() as $error )
+					echo $error->line . ':' . $error->column . ' ' . esc_html( $error->message ) . "\n";
+			} else if ( 'XML_parse_error' == $result->get_error_code() ) {
+				$error = $result->get_error_data();
+				echo $error[0] . ':' . $error[1] . ' ' . esc_html( $error[2] );
+			}
+			echo '
'; + echo '

' . __( 'There was an error when reading this WXR file', 'wordpress-importer' ) . '
'; + echo __( 'Details are shown above. The importer will now try again with a different parser...', 'wordpress-importer' ) . '

'; + } + + // use regular expressions if nothing else available or this is bad XML + $parser = new WXR_Parser_Regex; + return $parser->parse( $file ); + } +} + +/** + * WXR Parser that makes use of the SimpleXML PHP extension. + */ +class WXR_Parser_SimpleXML { + function parse( $file ) { + $authors = $posts = $categories = $tags = $terms = array(); + + $internal_errors = libxml_use_internal_errors(true); + $xml = simplexml_load_file( $file ); + // halt if loading produces an error + if ( ! $xml ) + return new WP_Error( 'SimpleXML_parse_error', __( 'There was an error when reading this WXR file', 'wordpress-importer' ), libxml_get_errors() ); + + $wxr_version = $xml->xpath('/rss/channel/wp:wxr_version'); + if ( ! $wxr_version ) + return new WP_Error( 'WXR_parse_error', __( 'This does not appear to be a WXR file, missing/invalid WXR version number', 'wordpress-importer' ) ); + + $wxr_version = (string) trim( $wxr_version[0] ); + // confirm that we are dealing with the correct file format + if ( ! preg_match( '/^\d+\.\d+$/', $wxr_version ) ) + return new WP_Error( 'WXR_parse_error', __( 'This does not appear to be a WXR file, missing/invalid WXR version number', 'wordpress-importer' ) ); + + $base_url = $xml->xpath('/rss/channel/wp:base_site_url'); + $base_url = (string) trim( $base_url[0] ); + + $namespaces = $xml->getDocNamespaces(); + if ( ! isset( $namespaces['wp'] ) ) + $namespaces['wp'] = 'http://wordpress.org/export/1.1/'; + if ( ! isset( $namespaces['excerpt'] ) ) + $namespaces['excerpt'] = 'http://wordpress.org/export/1.1/excerpt/'; + + // grab authors + foreach ( $xml->xpath('/rss/channel/wp:author') as $author_arr ) { + $a = $author_arr->children( $namespaces['wp'] ); + $login = (string) $a->author_login; + $authors[$login] = array( + 'author_id' => (int) $a->author_id, + 'author_login' => $login, + 'author_email' => (string) $a->author_email, + 'author_display_name' => (string) $a->author_display_name, + 'author_first_name' => (string) $a->author_first_name, + 'author_last_name' => (string) $a->author_last_name + ); + } + + // grab cats, tags and terms + foreach ( $xml->xpath('/rss/channel/wp:category') as $term_arr ) { + $t = $term_arr->children( $namespaces['wp'] ); + $categories[] = array( + 'term_id' => (int) $t->term_id, + 'category_nicename' => (string) $t->category_nicename, + 'category_parent' => (string) $t->category_parent, + 'cat_name' => (string) $t->cat_name, + 'category_description' => (string) $t->category_description + ); + } + + foreach ( $xml->xpath('/rss/channel/wp:tag') as $term_arr ) { + $t = $term_arr->children( $namespaces['wp'] ); + $tags[] = array( + 'term_id' => (int) $t->term_id, + 'tag_slug' => (string) $t->tag_slug, + 'tag_name' => (string) $t->tag_name, + 'tag_description' => (string) $t->tag_description + ); + } + + foreach ( $xml->xpath('/rss/channel/wp:term') as $term_arr ) { + $t = $term_arr->children( $namespaces['wp'] ); + $terms[] = array( + 'term_id' => (int) $t->term_id, + 'term_taxonomy' => (string) $t->term_taxonomy, + 'slug' => (string) $t->term_slug, + 'term_parent' => (string) $t->term_parent, + 'term_name' => (string) $t->term_name, + 'term_description' => (string) $t->term_description + ); + } + + // grab posts + foreach ( $xml->channel->item as $item ) { + $post = array( + 'post_title' => (string) $item->title, + 'guid' => (string) $item->guid, + ); + + $dc = $item->children( 'http://purl.org/dc/elements/1.1/' ); + $post['post_author'] = (string) $dc->creator; + + $content = $item->children( 'http://purl.org/rss/1.0/modules/content/' ); + $excerpt = $item->children( $namespaces['excerpt'] ); + $post['post_content'] = (string) $content->encoded; + $post['post_excerpt'] = (string) $excerpt->encoded; + + $wp = $item->children( $namespaces['wp'] ); + $post['post_id'] = (int) $wp->post_id; + $post['post_date'] = (string) $wp->post_date; + $post['post_date_gmt'] = (string) $wp->post_date_gmt; + $post['comment_status'] = (string) $wp->comment_status; + $post['ping_status'] = (string) $wp->ping_status; + $post['post_name'] = (string) $wp->post_name; + $post['status'] = (string) $wp->status; + $post['post_parent'] = (int) $wp->post_parent; + $post['menu_order'] = (int) $wp->menu_order; + $post['post_type'] = (string) $wp->post_type; + $post['post_password'] = (string) $wp->post_password; + $post['is_sticky'] = (int) $wp->is_sticky; + + foreach ( $item->category as $c ) { + $att = $c->attributes(); + if ( isset( $att['nicename'] ) ) + $post['terms'][] = array( + 'name' => (string) $c, + 'slug' => (string) $att['nicename'], + 'domain' => (string) $att['domain'] + ); + } + + foreach ( $wp->postmeta as $meta ) { + $post['postmeta'][] = array( + 'key' => (string) $meta->meta_key, + 'value' => (string) $meta->meta_value, + ); + } + + foreach ( $wp->comment as $comment ) { + $post['comments'][] = array( + 'comment_id' => (int) $comment->comment_id, + 'comment_author' => (string) $comment->comment_author, + 'comment_author_email' => (string) $comment->comment_author_email, + 'comment_author_IP' => (string) $comment->comment_author_IP, + 'comment_author_url' => (string) $comment->comment_author_url, + 'comment_date' => (string) $comment->comment_date, + 'comment_date_gmt' => (string) $comment->comment_date_gmt, + 'comment_content' => (string) $comment->comment_content, + 'comment_approved' => (string) $comment->comment_approved, + 'comment_type' => (string) $comment->comment_type, + 'comment_parent' => (string) $comment->comment_parent, + 'comment_user_id' => (int) $comment->comment_user_id, + ); + } + + $posts[] = $post; + } + + return array( + 'authors' => $authors, + 'posts' => $posts, + 'categories' => $categories, + 'tags' => $tags, + 'terms' => $terms, + 'base_url' => $base_url, + 'version' => $wxr_version + ); + } +} + +/** + * WXR Parser that makes use of the XML Parser PHP extension. + */ +class WXR_Parser_XML { + var $wp_tags = array( + 'wp:post_id', 'wp:post_date', 'wp:post_date_gmt', 'wp:comment_status', 'wp:ping_status', + 'wp:status', 'wp:post_name', 'wp:post_parent', 'wp:menu_order', 'wp:post_type', 'wp:post_password', + 'wp:is_sticky', 'wp:term_id', 'wp:category_nicename', 'wp:category_parent', 'wp:cat_name', 'wp:category_description', + 'wp:tag_slug', 'wp:tag_name', 'wp:tag_description', 'wp:term_taxonomy', 'wp:term_parent', + 'wp:term_name', 'wp:term_description', 'wp:author_id', 'wp:author_login', 'wp:author_email', 'wp:author_display_name', + 'wp:author_first_name', 'wp:author_last_name', + ); + var $wp_sub_tags = array( + 'wp:comment_id', 'wp:comment_author', 'wp:comment_author_email', 'wp:comment_author_url', + 'wp:comment_author_IP', 'wp:comment_date', 'wp:comment_date_gmt', 'wp:comment_content', + 'wp:comment_approved', 'wp:comment_type', 'wp:comment_parent', 'wp:comment_user_id', + ); + + function parse( $file ) { + $this->wxr_version = $this->in_post = $this->cdata = $this->data = $this->sub_data = $this->in_tag = $this->in_sub_tag = false; + $this->authors = $this->posts = $this->term = $this->category = $this->tag = array(); + + $xml = xml_parser_create( 'UTF-8' ); + xml_parser_set_option( $xml, XML_OPTION_SKIP_WHITE, 1 ); + xml_parser_set_option( $xml, XML_OPTION_CASE_FOLDING, 0 ); + xml_set_object( $xml, $this ); + xml_set_character_data_handler( $xml, 'cdata' ); + xml_set_element_handler( $xml, 'tag_open', 'tag_close' ); + + if ( ! xml_parse( $xml, file_get_contents( $file ), true ) ) { + $current_line = xml_get_current_line_number( $xml ); + $current_column = xml_get_current_column_number( $xml ); + $error_code = xml_get_error_code( $xml ); + $error_string = xml_error_string( $error_code ); + return new WP_Error( 'XML_parse_error', 'There was an error when reading this WXR file', array( $current_line, $current_column, $error_string ) ); + } + xml_parser_free( $xml ); + + if ( ! preg_match( '/^\d+\.\d+$/', $this->wxr_version ) ) + return new WP_Error( 'WXR_parse_error', __( 'This does not appear to be a WXR file, missing/invalid WXR version number', 'wordpress-importer' ) ); + + return array( + 'authors' => $this->authors, + 'posts' => $this->posts, + 'categories' => $this->category, + 'tags' => $this->tag, + 'terms' => $this->term, + 'base_url' => $this->base_url, + 'version' => $this->wxr_version + ); + } + + function tag_open( $parse, $tag, $attr ) { + if ( in_array( $tag, $this->wp_tags ) ) { + $this->in_tag = substr( $tag, 3 ); + return; + } + + if ( in_array( $tag, $this->wp_sub_tags ) ) { + $this->in_sub_tag = substr( $tag, 3 ); + return; + } + + switch ( $tag ) { + case 'category': + if ( isset($attr['domain'], $attr['nicename']) ) { + $this->sub_data['domain'] = $attr['domain']; + $this->sub_data['slug'] = $attr['nicename']; + } + break; + case 'item': $this->in_post = true; + case 'title': if ( $this->in_post ) $this->in_tag = 'post_title'; break; + case 'guid': $this->in_tag = 'guid'; break; + case 'dc:creator': $this->in_tag = 'post_author'; break; + case 'content:encoded': $this->in_tag = 'post_content'; break; + case 'excerpt:encoded': $this->in_tag = 'post_excerpt'; break; + + case 'wp:term_slug': $this->in_tag = 'slug'; break; + case 'wp:meta_key': $this->in_sub_tag = 'key'; break; + case 'wp:meta_value': $this->in_sub_tag = 'value'; break; + } + } + + function cdata( $parser, $cdata ) { + if ( ! trim( $cdata ) ) + return; + + $this->cdata .= trim( $cdata ); + } + + function tag_close( $parser, $tag ) { + switch ( $tag ) { + case 'wp:comment': + if ( ! empty( $this->sub_data ) ) + $this->data['comments'][] = $this->sub_data; + $this->sub_data = false; + break; + case 'category': + if ( ! empty( $this->sub_data ) ) { + $this->sub_data['name'] = $this->cdata; + $this->data['terms'][] = $this->sub_data; + } + $this->sub_data = false; + break; + case 'wp:postmeta': + if ( ! empty( $this->sub_data ) ) + $this->data['postmeta'][] = $this->sub_data; + $this->sub_data = false; + break; + case 'item': + $this->posts[] = $this->data; + $this->data = false; + break; + case 'wp:category': + case 'wp:tag': + case 'wp:term': + $n = substr( $tag, 3 ); + array_push( $this->$n, $this->data ); + $this->data = false; + break; + case 'wp:author': + if ( ! empty($this->data['author_login']) ) + $this->authors[$this->data['author_login']] = $this->data; + $this->data = false; + break; + case 'wp:base_site_url': + $this->base_url = $this->cdata; + break; + case 'wp:wxr_version': + $this->wxr_version = $this->cdata; + break; + + default: + if ( $this->in_sub_tag ) { + $this->sub_data[$this->in_sub_tag] = ! empty( $this->cdata ) ? $this->cdata : ''; + $this->in_sub_tag = false; + } else if ( $this->in_tag ) { + $this->data[$this->in_tag] = ! empty( $this->cdata ) ? $this->cdata : ''; + $this->in_tag = false; + } + } + + $this->cdata = false; + } +} + +/** + * WXR Parser that uses regular expressions. Fallback for installs without an XML parser. + */ +class WXR_Parser_Regex { + var $authors = array(); + var $posts = array(); + var $categories = array(); + var $tags = array(); + var $terms = array(); + var $base_url = ''; + + function WXR_Parser_Regex() { + $this->__construct(); + } + + function __construct() { + $this->has_gzip = is_callable( 'gzopen' ); + } + + function parse( $file ) { + $wxr_version = $in_post = false; + + $fp = $this->fopen( $file, 'r' ); + if ( $fp ) { + while ( ! $this->feof( $fp ) ) { + $importline = rtrim( $this->fgets( $fp ) ); + + if ( ! $wxr_version && preg_match( '|(\d+\.\d+)|', $importline, $version ) ) + $wxr_version = $version[1]; + + if ( false !== strpos( $importline, '' ) ) { + preg_match( '|(.*?)|is', $importline, $url ); + $this->base_url = $url[1]; + continue; + } + if ( false !== strpos( $importline, '' ) ) { + preg_match( '|(.*?)|is', $importline, $category ); + $this->categories[] = $this->process_category( $category[1] ); + continue; + } + if ( false !== strpos( $importline, '' ) ) { + preg_match( '|(.*?)|is', $importline, $tag ); + $this->tags[] = $this->process_tag( $tag[1] ); + continue; + } + if ( false !== strpos( $importline, '' ) ) { + preg_match( '|(.*?)|is', $importline, $term ); + $this->terms[] = $this->process_term( $term[1] ); + continue; + } + if ( false !== strpos( $importline, '' ) ) { + preg_match( '|(.*?)|is', $importline, $author ); + $a = $this->process_author( $author[1] ); + $this->authors[$a['author_login']] = $a; + continue; + } + if ( false !== strpos( $importline, '' ) ) { + $post = ''; + $in_post = true; + continue; + } + if ( false !== strpos( $importline, '' ) ) { + $in_post = false; + $this->posts[] = $this->process_post( $post ); + continue; + } + if ( $in_post ) { + $post .= $importline . "\n"; + } + } + + $this->fclose($fp); + } + + if ( ! $wxr_version ) + return new WP_Error( 'WXR_parse_error', __( 'This does not appear to be a WXR file, missing/invalid WXR version number', 'wordpress-importer' ) ); + + return array( + 'authors' => $this->authors, + 'posts' => $this->posts, + 'categories' => $this->categories, + 'tags' => $this->tags, + 'terms' => $this->terms, + 'base_url' => $this->base_url, + 'version' => $wxr_version + ); + } + + function get_tag( $string, $tag ) { + global $wpdb; + preg_match( "|<$tag.*?>(.*?)|is", $string, $return ); + if ( isset( $return[1] ) ) { + $return = preg_replace( '|^$|s', '$1', $return[1] ); + $return = $wpdb->escape( trim( $return ) ); + } else { + $return = ''; + } + return $return; + } + + function process_category( $c ) { + return array( + 'term_id' => $this->get_tag( $c, 'wp:term_id' ), + 'cat_name' => $this->get_tag( $c, 'wp:cat_name' ), + 'category_nicename' => $this->get_tag( $c, 'wp:category_nicename' ), + 'category_parent' => $this->get_tag( $c, 'wp:category_parent' ), + 'category_description' => $this->get_tag( $c, 'wp:category_description' ), + ); + } + + function process_tag( $t ) { + return array( + 'term_id' => $this->get_tag( $t, 'wp:term_id' ), + 'tag_name' => $this->get_tag( $t, 'wp:tag_name' ), + 'tag_slug' => $this->get_tag( $t, 'wp:tag_slug' ), + 'tag_description' => $this->get_tag( $t, 'wp:tag_description' ), + ); + } + + function process_term( $t ) { + return array( + 'term_id' => $this->get_tag( $t, 'wp:term_id' ), + 'term_taxonomy' => $this->get_tag( $t, 'wp:term_taxonomy' ), + 'slug' => $this->get_tag( $t, 'wp:term_slug' ), + 'term_parent' => $this->get_tag( $t, 'wp:term_parent' ), + 'term_name' => $this->get_tag( $t, 'wp:term_name' ), + 'term_description' => $this->get_tag( $t, 'wp:term_description' ), + ); + } + + function process_author( $a ) { + return array( + 'author_id' => $this->get_tag( $a, 'wp:author_id' ), + 'author_login' => $this->get_tag( $a, 'wp:author_login' ), + 'author_email' => $this->get_tag( $a, 'wp:author_email' ), + 'author_display_name' => $this->get_tag( $a, 'wp:author_display_name' ), + 'author_first_name' => $this->get_tag( $a, 'wp:author_first_name' ), + 'author_last_name' => $this->get_tag( $a, 'wp:author_last_name' ), + ); + } + + function process_post( $post ) { + $post_id = $this->get_tag( $post, 'wp:post_id' ); + $post_title = $this->get_tag( $post, 'title' ); + $post_date = $this->get_tag( $post, 'wp:post_date' ); + $post_date_gmt = $this->get_tag( $post, 'wp:post_date_gmt' ); + $comment_status = $this->get_tag( $post, 'wp:comment_status' ); + $ping_status = $this->get_tag( $post, 'wp:ping_status' ); + $status = $this->get_tag( $post, 'wp:status' ); + $post_name = $this->get_tag( $post, 'wp:post_name' ); + $post_parent = $this->get_tag( $post, 'wp:post_parent' ); + $menu_order = $this->get_tag( $post, 'wp:menu_order' ); + $post_type = $this->get_tag( $post, 'wp:post_type' ); + $post_password = $this->get_tag( $post, 'wp:post_password' ); + $is_sticky = $this->get_tag( $post, 'wp:is_sticky' ); + $guid = $this->get_tag( $post, 'guid' ); + $post_author = $this->get_tag( $post, 'dc:creator' ); + + $post_excerpt = $this->get_tag( $post, 'excerpt:encoded' ); + $post_excerpt = preg_replace_callback( '|<(/?[A-Z]+)|', array( &$this, '_normalize_tag' ), $post_excerpt ); + $post_excerpt = str_replace( '
', '
', $post_excerpt ); + $post_excerpt = str_replace( '
', '
', $post_excerpt ); + + $post_content = $this->get_tag( $post, 'content:encoded' ); + $post_content = preg_replace_callback( '|<(/?[A-Z]+)|', array( &$this, '_normalize_tag' ), $post_content ); + $post_content = str_replace( '
', '
', $post_content ); + $post_content = str_replace( '
', '
', $post_content ); + + $postdata = compact( 'post_id', 'post_author', 'post_date', 'post_date_gmt', 'post_content', 'post_excerpt', + 'post_title', 'status', 'post_name', 'comment_status', 'ping_status', 'guid', 'post_parent', + 'menu_order', 'post_type', 'post_password', 'is_sticky' + ); + + preg_match_all( '|(.+?)|is', $post, $terms, PREG_SET_ORDER ); + foreach ( $terms as $t ) { + $post_terms[] = array( + 'slug' => $t[2], + 'domain' => $t[1], + 'name' => str_replace( array( '' ), '', $t[3] ), + ); + } + if ( ! empty( $post_terms ) ) $postdata['terms'] = $post_terms; + + preg_match_all( '|(.+?)|is', $post, $comments ); + $comments = $comments[1]; + if ( $comments ) { + foreach ( $comments as $comment ) { + $post_comments[] = array( + 'comment_id' => $this->get_tag( $comment, 'wp:comment_id' ), + 'comment_author' => $this->get_tag( $comment, 'wp:comment_author' ), + 'comment_author_email' => $this->get_tag( $comment, 'wp:comment_author_email' ), + 'comment_author_IP' => $this->get_tag( $comment, 'wp:comment_author_IP' ), + 'comment_author_url' => $this->get_tag( $comment, 'wp:comment_author_url' ), + 'comment_date' => $this->get_tag( $comment, 'wp:comment_date' ), + 'comment_date_gmt' => $this->get_tag( $comment, 'wp:comment_date_gmt' ), + 'comment_content' => $this->get_tag( $comment, 'wp:comment_content' ), + 'comment_approved' => $this->get_tag( $comment, 'wp:comment_approved' ), + 'comment_type' => $this->get_tag( $comment, 'wp:comment_type' ), + 'comment_parent' => $this->get_tag( $comment, 'wp:comment_parent' ), + ); + } + } + if ( ! empty( $post_comments ) ) $postdata['comments'] = $post_comments; + + preg_match_all( '|(.+?)|is', $post, $postmeta ); + $postmeta = $postmeta[1]; + if ( $postmeta ) { + foreach ( $postmeta as $p ) { + $post_postmeta[] = array( + 'key' => $this->get_tag( $p, 'wp:meta_key' ), + 'value' => $this->get_tag( $p, 'wp:meta_value' ), + ); + } + } + if ( ! empty( $post_postmeta ) ) $postdata['postmeta'] = $post_postmeta; + + return $postdata; + } + + function _normalize_tag( $matches ) { + return '<' . strtolower( $matches[1] ); + } + + function fopen( $filename, $mode = 'r' ) { + if ( $this->has_gzip ) + return gzopen( $filename, $mode ); + return fopen( $filename, $mode ); + } + + function feof( $fp ) { + if ( $this->has_gzip ) + return gzeof( $fp ); + return feof( $fp ); + } + + function fgets( $fp, $len = 8192 ) { + if ( $this->has_gzip ) + return gzgets( $fp, $len ); + return fgets( $fp, $len ); + } + + function fclose( $fp ) { + if ( $this->has_gzip ) + return gzclose( $fp ); + return fclose( $fp ); + } +} -- cgit v1.2.3-65-gdbad