Ticket #36917: 36917.diff

src/wp-includes/wp-db.php

@@ -735 +735 @@
          */
         public function init_charset() {
                 if ( function_exists('is_multisite') && is_multisite() ) {
-                        $this->charset = 'utf8';
+                        $charset = 'utf8';
                         if ( defined( 'DB_COLLATE' ) && DB_COLLATE ) {
-                                $this->collate = DB_COLLATE;
+                                $collate = DB_COLLATE;
                         } else {
-                                $this->collate = 'utf8_general_ci';
+                                $collate = 'utf8_general_ci';
                         }
                 } elseif ( defined( 'DB_COLLATE' ) ) {
-                        $this->collate = DB_COLLATE;
+                        $collate = DB_COLLATE;
                 }
 
                 if ( defined( 'DB_CHARSET' ) ) {
-                        $this->charset = DB_CHARSET;
+                        $charset = DB_CHARSET;
                 }
 
+                list( $this->charset, $this->collate ) = $this->determine_charset( $charset, $collate );
+        }
+
+        /**
+         * Given a charset and collation, determine the best charset and collation to use.
+         *
+         * For example, when able, utf8mb4 should be used instead of utf8.
+         *
+         * @since 4.6.0
+         *
+         * @param  string $charset The character set to check.
+         * @param  string $collate The collation to check.
+         */
+        public function determine_charset( $charset, $collate ) {
                 if ( ( $this->use_mysqli && ! ( $this->dbh instanceof mysqli ) ) || empty( $this->dbh ) ) {
-                        return;
+                        return array( $charset, $collate );
                 }
 
-                if ( 'utf8' === $this->charset && $this->has_cap( 'utf8mb4' ) ) {
-                        $this->charset = 'utf8mb4';
+                if ( 'utf8' === $charset && $this->has_cap( 'utf8mb4' ) ) {
+                        $charset = 'utf8mb4';
                 }
 
-                if ( 'utf8mb4' === $this->charset ) {
+                if ( 'utf8mb4' === $charset ) {
                         // _general_ is outdated, so we can upgrade it to _unicode_, instead.
-                        if ( ! $this->collate || 'utf8_general_ci' === $this->collate ) {
-                                $this->collate = 'utf8mb4_unicode_ci';
+                        if ( ! $collate || 'utf8_general_ci' === $collate ) {
+                                $collate = 'utf8mb4_unicode_ci';
                         } else {
-                                $this->collate = str_replace( 'utf8_', 'utf8mb4_', $this->collate );
+                                $collate = str_replace( 'utf8_', 'utf8mb4_', $collate );
                         }
                 }
 
                 // _unicode_520_ is a better collation, we should use that when it's available.
-                if ( $this->has_cap( 'utf8mb4_520' ) && 'utf8mb4_unicode_ci' === $this->collate ) {
-                        $this->collate = 'utf8mb4_unicode_520_ci';
+                if ( $this->has_cap( 'utf8mb4_520' ) && 'utf8mb4_unicode_ci' === $collate ) {
+                        $collate = 'utf8mb4_unicode_520_ci';
                 }
+
+                return array( $charset, $collate );
         }
 
         /**

tests/phpunit/tests/db.php

@@ -955 +955 @@
 
                 $wpdb->check_connection();
         }
+
+        /**
+         * @ticket 36917
+         */
+        function test_charset_not_determined_when_disconnected() {
+                global $wpdb;
+
+                $charset = 'utf8';
+                $collate = 'this_isnt_a_collation';
+
+                $wpdb->close();
+
+                $result = $wpdb->determine_charset( $charset, $collate );
+
+                $this->assertSame( array( $charset, $collate ), $result );
+
+                $wpdb->check_connection();
+        }
 }