Import file attachments. Props tellyworth. fixes #5466

git-svn-id: http://svn.automattic.com/wordpress/trunk@6390 1a063a9b-81f0-0310-95a4-ce76da25c4cd

wp-admin/import/wordpress.php

@@ -10,7 +10,8 @@ class WP_Import {
 	var $newauthornames = array ();
 	var $allauthornames = array ();
 	var $j = -1;
-	var $another_pass = false;
+	var $fetch_attachments = false;
+	var $url_remap = array ();
 
 	function header() {
 		echo '<div class="wrap">';
@@ -189,10 +190,17 @@ class WP_Import {
 			$this->users_form($j);
 			echo '</li>';
 		}
+?>	
+</ol>	
+<h2><?php _e('Import Attachments'); ?></h2>
+<p>
+	<input type="checkbox" value="1" name="attachments" id="import-attachments" />
+	<label for="import-attachments"><?php _e('Download and import file attachments') ?></label>
+</p>
 
+<?php
 		echo '<input type="submit" value="Submit">'.'<br />';
 		echo '</form>';
-		echo '</ol>';
 
 	}
 
@@ -333,12 +341,24 @@ class WP_Import {
 			}
 
 			echo '<li>';
-			printf(__('Importing post <i>%s</i>...'), stripslashes($post_title));
 
 			$post_author = $this->checkauthor($post_author); //just so that if a post already exists, new users are not created by checkauthor
 
 			$postdata = compact('post_author', 'post_date', 'post_date_gmt', 'post_content', 'post_title', 'post_excerpt', 'post_status', 'post_name', 'comment_status', 'ping_status', 'post_modified', 'post_modified_gmt', 'guid', 'post_parent', 'menu_order', 'post_type');
+			if ($post_type == 'attachment') {
+				$remote_url = $this->get_tag( $post, 'wp:attachment_url' );
+				if ( !$remote_url )
+					$remote_url = $guid;
+					
+				$comment_post_ID = $post_id = $this->process_attachment($postdata, $remote_url);
+				if ( !$post_id or is_wp_error($post_id) )
+					return $post_id;
+			}
+			else {
+				printf(__('Importing post <i>%s</i>...'), stripslashes($post_title));
 				$comment_post_ID = $post_id = wp_insert_post($postdata);
+			}
+			
 			if ( is_wp_error( $post_id ) )
 				return $post_id;
 
@@ -420,6 +440,79 @@ class WP_Import {
 			$value = stripslashes($value); // add_post_meta() will escape.
 			add_post_meta( $post_id, $key, $value );
 		} }
+		
+		print "</li>\n";
+	}
+	
+	function process_attachment($postdata, $remote_url) {
+		if ($this->fetch_attachments and $remote_url) {
+			printf( __('Importing attachment <i>%s</i>... '), htmlspecialchars($remote_url) );
+			$upload = $this->fetch_remote_file($postdata, $remote_url);
+			if ( is_wp_error($upload) ) {
+				printf( __('Remote file error: %s'), htmlspecialchars($upload->get_error_message()) );
+				return $upload;
+			}
+			else {
+				print '('.size_format(filesize($upload['file'])).')';
+			}
+				
+			$postdata['guid'] = $upload['url'];
+
+			// as per wp-admin/includes/upload.php
+			$post_id = wp_insert_attachment($postdata, $upload['file']);
+			wp_update_attachment_metadata( $post_id, wp_generate_attachment_metadata( $post_id, $upload['file'] ) );
+			return $post_id;
+		}
+		else {
+			printf( __('Skipping attachment <i>%s</i>'), htmlspecialchars($remote_url) );
+		}
+	}
+	
+	function fetch_remote_file($post, $url) {
+		$upload = wp_upload_dir($post['post_date']);
+		
+		// extract the file name and extension from the url
+		$file_name = basename($url);
+
+		// get placeholder file in the upload dir with a unique sanitized filename
+		$upload = wp_upload_bits( $file_name, 0, '', $post['post_date']);
+		if ( $upload['error'] ) {
+			echo $upload['error'];
+			return new WP_Error( 'upload_dir_error', $upload['error'] );
+		}
+		
+		// fetch the remote url and write it to the placeholder file
+		$headers = wp_get_http($url, $upload['file']);
+		
+		// make sure the fetch was successful
+		if ( $headers['response'] != '200' )
+			return new WP_Error( 'import_file_error', __(sprintf('Remote file returned error response %d', intval($headers['response']))) );
+		elseif ( isset($headers['content-length']) && filesize($upload['file']) != $headers['content-length'] )
+			return new WP_Error( 'import_file_error', __('Remote file is incorrect size') );
+			
+		// keep track of the old and new urls so we can substitute them later
+		$this->url_remap[$url] = $upload['url'];
+		// if the remote url is redirected somewhere else, keep track of the destination too
+		if ( $headers['x-final-location'] != $url )
+			$this->url_remap[$headers['x-final-location']] = $upload['url'];
+		
+		return $upload;
+		
+	}
+	
+	// update url references in post bodies to point to the new local files
+	function backfill_attachment_urls() {
+		
+		// make sure we do the longest urls first, in case one is a substring of another
+		function cmpr_strlen($a, $b) {
+			return strlen($b) - strlen($a);
+		}
+		uksort($this->url_remap, 'cmpr_strlen');
+				
+		global $wpdb;
+		foreach ($this->url_remap as $from_url => $to_url) {
+			$wpdb->query( $wpdb->prepare("UPDATE {$wpdb->posts} SET post_content = REPLACE(post_content, '%s', '%s')", $from_url, $to_url) );
+		}
 	}
 	
 	// update the post_parent of orphans now that we know the local id's of all parents
@@ -435,8 +528,9 @@ class WP_Import {
 		}
 	}
 
-	function import($id) {
+	function import($id, $fetch_attachments = false) {
 		$this->id = (int) $id;
+		$this->fetch_attachments = (bool) $fetch_attachments;
 
 		$file = get_attached_file($this->id);
 		$this->import_file($file);
@@ -452,6 +546,7 @@ class WP_Import {
 		$this->process_tags();
 		$result = $this->process_posts();
 		$this->backfill_parents();
+		$this->backfill_attachment_urls();
 		wp_defer_term_counting(false);
 		if ( is_wp_error( $result ) )
 			return $result;
@@ -487,7 +582,7 @@ class WP_Import {
 				break;
 			case 2:
 				check_admin_referer('import-wordpress');
-				$result = $this->import( $_GET['id'] );
+				$result = $this->import( $_GET['id'], $_POST['attachments'] );
 				if ( is_wp_error( $result ) )
 					echo $result->get_error_message();
 				break;

wp-admin/includes/file.php

@@ -147,26 +147,7 @@ function wp_handle_upload( &$file, $overrides = false ) {
 	if ( ! ( ( $uploads = wp_upload_dir() ) && false === $uploads['error'] ) )
 		return $upload_error_handler( $file, $uploads['error'] );
 
-	// Increment the file number until we have a unique file to save in $dir. Use $override['unique_filename_callback'] if supplied.
+	$filename = wp_unique_filename( $uploads['path'], $file['name'], $ext, $unique_filename_callback );
-	if ( isset( $unique_filename_callback ) && function_exists( $unique_filename_callback ) ) {
-		$filename = $unique_filename_callback( $uploads['path'], $file['name'] );
-	} else {
-		$number = '';
-		$filename = str_replace( '#', '_', $file['name'] );
-		$filename = str_replace( array( '\\', "'" ), '', $filename );
-		if ( empty( $ext) )
-			$ext = '';
-		else
-			$ext = ".$ext";
-		while ( file_exists( $uploads['path'] . "/$filename" ) ) {
-			if ( '' == "$number$ext" )
-				$filename = $filename . ++$number . $ext;
-			else
-				$filename = str_replace( "$number$ext", ++$number . $ext, $filename );
-		}
-		$filename = str_replace( $ext, '', $filename );
-		$filename = sanitize_title_with_dashes( $filename ) . $ext;
-	}
 
 	// Move the file to the uploads dir
 	$new_file = $uploads['path'] . "/$filename";

wp-includes/functions.php

@@ -531,8 +531,10 @@ function do_enclose( $content, $post_ID ) {
 	}
 }
 
-
+// perform a HTTP HEAD or GET request
-function wp_get_http_headers( $url, $red = 1 ) {
+// if $file_path is a writable filename, this will do a GET request and write the file to that path
+// returns a list of HTTP headers
+function wp_get_http( $url, $file_path = false, $red = 1 ) {
 	global $wp_version;
 	@set_time_limit( 60 );
 
@@ -545,7 +547,12 @@ function wp_get_http_headers( $url, $red = 1 ) {
 	if ( !isset( $parts['port'] ) )
 		$parts['port'] = 80;
 
-	$head = "HEAD $file HTTP/1.1\r\nHOST: $host\r\nUser-Agent: WordPress/" . $wp_version . "\r\n\r\n";
+	if ( $file_path )
+		$request_type = 'GET';
+	else
+		$request_type = 'HEAD';
+		
+	$head = "$request_type $file HTTP/1.1\r\nHOST: $host\r\nUser-Agent: WordPress/" . $wp_version . "\r\n\r\n";
 
 	$fp = @fsockopen( $host, $parts['port'], $err_num, $err_msg, 3 );
 	if ( !$fp )
@@ -555,7 +562,6 @@ function wp_get_http_headers( $url, $red = 1 ) {
 	fputs( $fp, $head );
 	while ( !feof( $fp ) && strpos( $response, "\r\n\r\n" ) == false )
 		$response .= fgets( $fp, 2048 );
-	fclose( $fp );
 	preg_match_all( '/(.*?): (.*)\r/', $response, $matches );
 	$count = count( $matches[1] );
 	for ( $i = 0; $i < $count; $i++ ) {
@@ -567,12 +573,42 @@ function wp_get_http_headers( $url, $red = 1 ) {
 	$headers['response'] = $return[1]; // HTTP response code eg 204, 200, 404
 
 		$code = $headers['response'];
-		if ( ( '302' == $code || '301' == $code ) && isset( $headers['location'] ) )
+		if ( ( '302' == $code || '301' == $code ) && isset( $headers['location'] ) ) {
-				return wp_get_http_headers( $headers['location'], ++$red );
+				fclose($fp);
+				return wp_get_http_headers( $headers['location'], $get, ++$red );
+		}
 	
+	// make a note of the final location, so the caller can tell if we were redirected or not
+	$headers['x-final-location'] = $url;
+
+	// HEAD request only
+	if ( !$file_path ) {
+		fclose($fp);
 		return $headers;
 	}
 	
+	// GET request - fetch and write it to the supplied filename
+	$content_length = $headers['content-length'];
+	$got_bytes = 0;
+	$out_fp = fopen($file_path, 'w');
+	while ( !feof($fp) ) {
+		$buf = fread( $fp, 4096 );
+		fwrite( $out_fp, $buf );
+		$got_bytes += strlen($buf);
+		// don't read past the content-length
+		if ($content_length and $got_bytes >= $content_length)
+			break;
+	}
+	
+	fclose($out_fp);
+	fclose($fp);
+	return $headers;
+}
+
+function wp_get_http_headers( $url ) {
+	return wp_get_http( $url, false );
+}
+
 
 function is_new_day() {
 	global $day, $previousday;
@@ -992,7 +1028,7 @@ function wp_mkdir_p( $target ) {
 
 
 // Returns an array containing the current upload directory's path and url, or an error message.
-function wp_upload_dir() {
+function wp_upload_dir( $time = NULL ) {
 	$siteurl = get_option( 'siteurl' );
 	//prepend ABSPATH to $dir and $siteurl to $url if they're not already there
 	$path = str_replace( ABSPATH, '', trim( get_option( 'upload_path' ) ) );
@@ -1009,6 +1045,7 @@ function wp_upload_dir() {
 
 	if ( get_option( 'uploads_use_yearmonth_folders' ) ) {
 		// Generate the yearly and monthly dirs
+		if ( !$time )
 			$time = current_time( 'mysql' );
 		$y = substr( $time, 0, 4 );
 		$m = substr( $time, 5, 2 );
@@ -1026,7 +1063,35 @@ function wp_upload_dir() {
 	return apply_filters( 'upload_dir', $uploads );
 }
 
-function wp_upload_bits( $name, $deprecated, $bits ) {
+// return a filename that is sanitized and unique for the given directory
+function wp_unique_filename( $dir, $name, $ext, $unique_filename_callback = NULL ) {
+	
+	// Increment the file number until we have a unique file to save in $dir. Use $override['unique_filename_callback'] if supplied.
+	if ( $unique_filename_callback && function_exists( $unique_filename_callback ) ) {
+		$filename = $unique_filename_callback( $dir, $name );
+	} else {
+		$number = '';
+		$filename = str_replace( '#', '_', $name );
+		$filename = str_replace( array( '\\', "'" ), '', $filename );
+		if ( empty( $ext) )
+			$ext = '';
+		else
+			$ext = ".$ext";
+		$filename = $filename . $ext;
+		while ( file_exists( $dir . "/$filename" ) ) {
+			if ( '' == "$number$ext" )
+				$filename = $filename . ++$number . $ext;
+			else
+				$filename = str_replace( "$number$ext", ++$number . $ext, $filename );
+		}
+		$filename = str_replace( $ext, '', $filename );
+		$filename = sanitize_title_with_dashes( $filename ) . $ext;
+	}
+	
+	return $filename;
+}
+
+function wp_upload_bits( $name, $deprecated, $bits, $time = NULL ) {
 	if ( empty( $name ) )
 		return array( 'error' => __( "Empty filename" ) );
 
@@ -1034,25 +1099,16 @@ function wp_upload_bits( $name, $deprecated, $bits ) {
 	if ( !$wp_filetype['ext'] )
 		return array( 'error' => __( "Invalid file type" ) );
 
-	$upload = wp_upload_dir();
+	$upload = wp_upload_dir( $time );
 
 	if ( $upload['error'] !== false )
 		return $upload;
 
-	$number = '';
 	$filename = $name;
 	$path_parts = pathinfo( $filename );
 	$ext = $path_parts['extension'];
-	if ( empty( $ext ) )
+	
-		$ext = '';
+	$filename = wp_unique_filename( $upload['path'], $path_parts['basename'], $ext );
-	else
-		$ext = ".$ext";
-	while ( file_exists( $upload['path'] . "/$filename" ) ) {
-		if ( '' == "$number$ext" )
-			$filename = $filename . ++$number . $ext;
-		else
-			$filename = str_replace( "$number$ext", ++$number . $ext, $filename );
-	}
 
 	$new_file = $upload['path'] . "/$filename";
 	if ( ! wp_mkdir_p( dirname( $new_file ) ) ) {