Normalize the UTF-8 and ISO-8859-1 charset strings stored in blog_charset to make them friendlier with PHP functions that accept a charset such as htmlspecialchars().

fixes #23688


git-svn-id: http://core.svn.wordpress.org/trunk@24510 1a063a9b-81f0-0310-95a4-ce76da25c4cd

wp-includes/default-filters.php

@@ -179,6 +179,7 @@ add_filter( 'the_author',         'ent2ncr',      8 );
 // Misc filters
 add_filter( 'option_ping_sites',        'privacy_ping_filter'                 );
 add_filter( 'option_blog_charset',      '_wp_specialchars'                    ); // IMPORTANT: This must not be wp_specialchars() or esc_html() or it'll cause an infinite loop
+add_filter( 'option_blog_charset',      '_canonical_charset'                  );
 add_filter( 'option_home',              '_config_wp_home'                     );
 add_filter( 'option_siteurl',           '_config_wp_siteurl'                  );
 add_filter( 'tiny_mce_before_init',     '_mce_set_direction'                  );

wp-includes/functions.php

@@ -3998,3 +3998,25 @@ function get_tag_regex( $tag ) {
 		return;
 	return sprintf( '<%1$s[^<]*(?:>[\s\S]*<\/%1$s>|\s*\/>)', tag_escape( $tag ) );
 }
+
+/**
+ * Return a canonical form of the provided charset appropriate for passing to PHP
+ * functions such as htmlspecialchars() and charset html attributes.
+ *
+ * @link http://core.trac.wordpress.org/ticket/23688
+ * @since 3.6.0
+ *
+ * @param string A charset name
+ * @return string The canonical form of the charset
+ */
+function _canonical_charset( $charset ) {
+	if ( 'UTF-8' === $charset || 'utf-8' === $charset || 'utf8' === $charset ||
+		'UTF8' === $charset )
+		return 'UTF-8';
+
+	if ( 'ISO-8859-1' === $charset || 'iso-8859-1' === $charset ||
+		'iso8859-1' === $charset || 'ISO8859-1' === $charset )
+		return 'ISO-8859-1';
+
+	return $charset;
+}