From fd21a9346c776d3c4397999dfb9a6128ee8f432c Mon Sep 17 00:00:00 2001
From: whyisjake <whyisjake@git.wordpress.org>
Date: Wed, 29 Apr 2020 15:40:07 +0000
Subject: [PATCH] Formatting: Expand `sanitize_file_name` to have better
 support for utf8 characters.

Props: xknown, peterwilsoncc.

Built from https://develop.svn.wordpress.org/trunk@47638


git-svn-id: http://core.svn.wordpress.org/trunk@47413 1a063a9b-81f0-0310-95a4-ce76da25c4cd
---
 wp-includes/formatting.php | 19 ++++++++++++++++++-
 wp-includes/version.php    |  2 +-
 2 files changed, 19 insertions(+), 2 deletions(-)

diff --git a/wp-includes/formatting.php b/wp-includes/formatting.php
index 7d8f9a9a64..1b5b2ed512 100644
--- a/wp-includes/formatting.php
+++ b/wp-includes/formatting.php
@@ -2005,6 +2005,24 @@ function remove_accents( $string ) {
 function sanitize_file_name( $filename ) {
 	$filename_raw  = $filename;
 	$special_chars = array( '?', '[', ']', '/', '\\', '=', '<', '>', ':', ';', ',', "'", '"', '&', '$', '#', '*', '(', ')', '|', '~', '`', '!', '{', '}', '%', '+', chr( 0 ) );
+
+	// Check for support for utf8 in the installed PCRE library once and store the result in a static.
+	static $utf8_pcre = null;
+	if ( ! isset( $utf8_pcre ) ) {
+		// phpcs:ignore WordPress.PHP.NoSilencedErrors.Discouraged
+		$utf8_pcre = @preg_match( '/^./u', 'a' );
+	}
+
+	if ( ! seems_utf8( $filename ) ) {
+		$_ext     = pathinfo( $filename, PATHINFO_EXTENSION );
+		$_name    = pathinfo( $filename, PATHINFO_FILENAME );
+		$filename = sanitize_title_with_dashes( $_name ) . '.' . $_ext;
+	}
+
+	if ( $utf8_pcre ) {
+		$filename = preg_replace( "#\x{00a0}#siu", ' ', $filename );
+	}
+
 	/**
 	 * Filters the list of characters to remove from a filename.
 	 *
@@ -2014,7 +2032,6 @@ function sanitize_file_name( $filename ) {
 	 * @param string   $filename_raw  The original filename to be sanitized.
 	 */
 	$special_chars = apply_filters( 'sanitize_file_name_chars', $special_chars, $filename_raw );
-	$filename      = preg_replace( "#\x{00a0}#siu", ' ', $filename );
 	$filename      = str_replace( $special_chars, '', $filename );
 	$filename      = str_replace( array( '%20', '+' ), '-', $filename );
 	$filename      = preg_replace( '/[\r\n\t -]+/', '-', $filename );
diff --git a/wp-includes/version.php b/wp-includes/version.php
index e6fc51d75e..ae568573f7 100644
--- a/wp-includes/version.php
+++ b/wp-includes/version.php
@@ -13,7 +13,7 @@
  *
  * @global string $wp_version
  */
-$wp_version = '5.5-alpha-47637';
+$wp_version = '5.5-alpha-47638';
 
 /**
  * Holds the WordPress DB revision, increments when changes are made to the WordPress DB schema.