Changeset 47638 for trunk

Timestamp:

04/29/2020 03:38:43 PM (6 years ago)

Author:

whyisjake

Message:

Formatting: Expand sanitize_file_name to have better support for utf8 characters.

Props: xknown, peterwilsoncc.

Location:

trunk

Files:

• src/wp-includes/formatting.php (modified) (2 diffs)
• tests/phpunit/tests/formatting/SanitizeFileName.php (modified) (1 diff)

trunk/src/wp-includes/formatting.php

@@ -2006 +2006 @@
         $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.
@@ -2015 +2033 @@
          */
         $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 );

trunk/tests/phpunit/tests/formatting/SanitizeFileName.php

@@ -69 +69 @@
                 $this->assertEquals( 'no-extension', sanitize_file_name( '_.no-extension' ) );
         }
+
+        /**
+         * @dataProvider data_wp_filenames
+         */
+        function test_replaces_invalid_utf8_characters( $input, $expected ) {
+                $this->assertEquals( $expected, sanitize_file_name( $input ) );
+        }
+
+        function data_wp_filenames() {
+                return array(
+                        [ urldecode( '%B1myfile.png' ), 'myfile.png' ],
+                        [ urldecode( '%B1myfile' ), 'myfile' ],
+                        [ 'demo bar.png', 'demo-bar.png' ],
+                        [ 'demo' . json_decode( '"\u00a0"' ) . 'bar.png', 'demo-bar.png' ],
+                );
+        }
 }