Fix: Default to UTF-8 for empty blog_charset values

src/wp-includes/default-filters.php

@@ -297,6 +297,8 @@
 // Misc filters.
 add_filter( 'wp_default_autoload_value', 'wp_filter_default_autoload_value_via_option_size', 5, 4 ); // Allow the value to be overridden at the default priority.
 add_filter( 'option_ping_sites', 'privacy_ping_filter' );
+
+add_filter( 'option_blog_charset', '_wp_ensure_blog_charset', 9 );
 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' );

src/wp-includes/functions.php

@@ -7566,7 +7566,13 @@ function get_tag_regex( $tag ) {
  * @return bool Whether the slug represents the UTF-8 encoding.
  */
 function is_utf8_charset( $blog_charset = null ) {
-	return _is_utf8_charset( $blog_charset ?? get_option( 'blog_charset' ) );
+	if ( null === $blog_charset ) {
+		$blog_charset = get_option( 'blog_charset', 'UTF-8' );
+	} elseif ( empty( $blog_charset ) ) {
+		return false;
+	}
+
+	return _is_utf8_charset( $blog_charset );
 }
 
 /**
@@ -7582,6 +7588,10 @@ function is_utf8_charset( $blog_charset = null ) {
  * @return string The canonical form of the charset.
  */
 function _canonical_charset( $charset ) {
+	if ( empty( $charset ) ) {
+		return '';
+	}
+
 	if ( is_utf8_charset( $charset ) ) {
 		return 'UTF-8';
 	}

src/wp-includes/option.php

@@ -3238,3 +3238,15 @@ function wp_autoload_values_to_autoload() {
 
 	return array_intersect( $filtered_values, $autoload_values );
 }
+
+/**
+ * Ensures blog_charset has a value by defaulting to UTF-8 when empty.
+ *
+ * @since 6.8.0
+ *
+ * @param string|null $charset The character set from the option.
+ * @return string The character set, defaulting to UTF-8 when empty.
+ */
+function _wp_ensure_blog_charset( $charset ) {
+	return empty( $charset ) ? 'UTF-8' : $charset;
+}

tests/phpunit/tests/functions/ensureBlogCharset.php

@@ -0,0 +1,91 @@
+<?php
+/**
+ * Tests for the _wp_ensure_blog_charset() function.
+ *
+ * @group functions
+ * @group charset
+ *
+ * @covers ::_wp_ensure_blog_charset
+ */
+class Tests_Functions_EnsureBlogCharset extends WP_UnitTestCase {
+	/**
+	 * Tests that _wp_ensure_blog_charset returns UTF-8 for empty or null values.
+	 *
+	 * @ticket 25693
+	 *
+	 * @dataProvider data_empty_charset_values
+	 *
+	 * @param mixed $empty_charset Empty or null charset value.
+	 */
+	public function test_returns_utf8_for_empty_values( $empty_charset ) {
+		$this->assertSame(
+			'UTF-8',
+			_wp_ensure_blog_charset( $empty_charset ),
+			'Should return UTF-8 when the blog_charset value is empty'
+		);
+	}
+
+	/**
+	 * Tests that _wp_ensure_blog_charset preserves valid charsets.
+	 *
+	 * @ticket 25693
+	 *
+	 * @dataProvider data_valid_charset_values
+	 *
+	 * @param string $valid_charset A valid charset.
+	 */
+	public function test_preserves_valid_charset_values( $valid_charset ) {
+		$this->assertSame(
+			$valid_charset,
+			_wp_ensure_blog_charset( $valid_charset ),
+			'Should preserve the original charset when it is not empty'
+		);
+	}
+
+	/**
+	 * Data provider for empty charset values.
+	 *
+	 * @return array[].
+	 */
+	public static function data_empty_charset_values() {
+		return array(
+			array( null ),
+			array( '' ),
+			array( false ),
+			array( 0 ),
+			array( '0' ),
+		);
+	}
+
+	/**
+	 * Data provider for valid charset values.
+	 *
+	 * @return array[].
+	 */
+	public static function data_valid_charset_values() {
+		return array(
+			array( 'UTF-8' ),
+			array( 'ISO-8859-1' ),
+			array( 'ASCII' ),
+			array( 'Windows-1252' ),
+			array( 'EUC-JP' ),
+		);
+	}
+
+	/**
+	 * Tests the integration of _wp_ensure_blog_charset with the option filter system.
+	 *
+	 * @ticket 25693
+	 */
+	public function test_option_filter_integration() {
+		$original_charset = get_option( 'blog_charset' );
+
+		update_option( 'blog_charset', '' );
+		$this->assertSame( 'UTF-8', get_option( 'blog_charset' ), 'Filter should ensure UTF-8 when blog_charset is empty' );
+
+		update_option( 'blog_charset', 'ISO-8859-1' );
+		$this->assertSame( 'ISO-8859-1', get_option( 'blog_charset' ), 'Filter should preserve valid blog_charset' );
+
+		update_option( 'blog_charset', $original_charset );
+	}
+}

tests/phpunit/tests/functions/isUtf8Charset.php

@@ -0,0 +1,123 @@
+<?php
+/**
+ * Tests for the is_utf8_charset() function.
+ *
+ * @group functions
+ * @group charset
+ *
+ * @covers ::is_utf8_charset
+ */
+class Tests_Functions_IsUtf8Charset extends WP_UnitTestCase {
+	/**
+	 * Tests that is_utf8_charset handles null by getting the blog_charset option.
+	 *
+	 * @ticket 25693
+	 */
+	public function test_handles_null_by_getting_option() {
+		$original_charset = get_option( 'blog_charset' );
+
+		update_option( 'blog_charset', 'UTF-8' );
+		$this->assertTrue(
+			is_utf8_charset( null ),
+			'Should return true when null is passed and blog_charset is UTF-8'
+		);
+
+		update_option( 'blog_charset', 'ISO-8859-1' );
+		$this->assertFalse(
+			is_utf8_charset( null ),
+			'Should return false when null is passed and blog_charset is not UTF-8'
+		);
+
+		update_option( 'blog_charset', $original_charset );
+	}
+
+	/**
+	 * Tests that is_utf8_charset returns false for empty values.
+	 *
+	 * @ticket 25693
+	 *
+	 * @dataProvider data_empty_charset_values
+	 *
+	 * @param mixed $empty_charset Empty or null charset value.
+	 */
+	public function test_handles_empty_values( $empty_charset ) {
+		$this->assertFalse(
+			is_utf8_charset( $empty_charset ),
+			'Should return false when empty values are explicitly passed'
+		);
+	}
+
+	/**
+	 * Tests that is_utf8_charset correctly identifies UTF-8 variants.
+	 *
+	 * @ticket 25693
+	 *
+	 * @dataProvider data_utf8_charset_variants
+	 *
+	 * @param string $utf8_charset A UTF-8 charset variant.
+	 */
+	public function test_identifies_utf8_variants( $utf8_charset ) {
+		$this->assertTrue(
+			is_utf8_charset( $utf8_charset ),
+			'Should identify valid UTF-8 charset variants'
+		);
+	}
+
+	/**
+	 * Tests that is_utf8_charset correctly rejects non-UTF-8 charsets.
+	 *
+	 * @ticket 25693
+	 *
+	 * @dataProvider data_non_utf8_charset_values
+	 *
+	 * @param string $non_utf8_charset A non-UTF-8 charset.
+	 */
+	public function test_rejects_non_utf8_charsets( $non_utf8_charset ) {
+		$this->assertFalse(
+			is_utf8_charset( $non_utf8_charset ),
+			'Should reject non-UTF-8 charsets'
+		);
+	}
+
+	/**
+	 * Data provider for empty charset values.
+	 *
+	 * @return array[].
+	 */
+	public static function data_empty_charset_values() {
+		return array(
+			array( '' ),
+			array( false ),
+			array( 0 ),
+			array( '0' ),
+		);
+	}
+
+	/**
+	 * Data provider for UTF-8 charset variants.
+	 *
+	 * @return array[].
+	 */
+	public static function data_utf8_charset_variants() {
+		return array(
+			array( 'UTF-8' ),
+			array( 'utf-8' ),
+			array( 'utf8' ),
+			array( 'UTF8' ),
+		);
+	}
+
+	/**
+	 * Data provider for non-UTF-8 charset values.
+	 *
+	 * @return array[].
+	 */
+	public static function data_non_utf8_charset_values() {
+		return array(
+			array( 'ISO-8859-1' ),
+			array( 'Windows-1252' ),
+			array( 'ASCII' ),
+			array( 'EUC-JP' ),
+		);
+	}
+}