Preparations for migrating core to use JSON based i18n

LocalisationCache and Language have to take the JSON files into account
in deciding if a language is present or not.

Standardizing language validity checking with isSupportedLanguage
and isValidBuiltInCode.

Co-Authored-By: Niklas Laxström <niklas.laxstrom@gmail.com>
Co-Authored-By: Siebrand Mazeland <siebrand@kitano.nl>
Change-Id: I35bbb3a7a145fc48d14fff620407dff5ecfdd4fc

diff --git a/includes/cache/LocalisationCache.php b/includes/cache/LocalisationCache.php
index d0274a1..8827219 100644 (file)
--- a/includes/cache/LocalisationCache.php
+++ b/includes/cache/LocalisationCache.php
@@ -448,7 +448,7 @@ class LocalisationCache {
 
                # Recache the data if necessary
                if ( !$this->manualRecache && $this->isExpired( $code ) ) {
-                       if ( file_exists( Language::getMessagesFileName( $code ) ) ) {
+                       if ( Language::isSupportedLanguage( $code ) ) {
                                $this->recache( $code );
                        } elseif ( $code === 'en' ) {
                                throw new MWException( 'MessagesEn.php is missing.' );
@@ -680,16 +680,16 @@ class LocalisationCache {
                global $IP;
                wfProfileIn( __METHOD__ );
 
+
+               // This reads in the PHP i18n file with non-messages l10n data
                $fileName = Language::getMessagesFileName( $code );
                if ( !file_exists( $fileName ) ) {
-                       wfProfileOut( __METHOD__ );
-
-                       return false;
+                       $data = array();
+               } else {
+                       $deps[] = new FileDependency( $fileName );
+                       $data = $this->readPHPFile( $fileName, 'core' );
                }
 
-               $deps[] = new FileDependency( $fileName );
-               $data = $this->readPHPFile( $fileName, 'core' );
-
                # Load CLDR plural rules for JavaScript
                $data['pluralRules'] = $this->getPluralRules( $code );
                # And for PHP
@@ -856,10 +856,7 @@ class LocalisationCache {
 
                $codeSequence = array_merge( array( $code ), $coreData['fallbackSequence'] );
 
-               # Load the extension localisations
-               # This is done after the core because we know the fallback sequence now.
-               # But it has a higher precedence for merging so that we can support things
-               # like site-specific message overrides.
+               # Load core messages and the extension localisations.
                wfProfileIn( __METHOD__ . '-extensions' );
                $allData = $initialData;
                foreach ( $wgMessagesDirs as $dirs ) {

diff --git a/languages/Language.php b/languages/Language.php
index fcb0f09..8c68b60 100644 (file)
--- a/languages/Language.php
+++ b/languages/Language.php
@@ -261,7 +261,10 @@ class Language {
         * @since 1.21
         */
        public static function isSupportedLanguage( $code ) {
-               return $code === strtolower( $code ) && is_readable( self::getMessagesFileName( $code ) );
+               return self::isValidBuiltInCode( $code )
+                       && ( is_readable( self::getMessagesFileName( $code ) )
+                               || is_readable( self::getJsonMessagesFileName( $code ) )
+               );
        }
 
        /**
@@ -351,7 +354,7 @@ class Language {
 
        /**
         * Returns true if a language code is of a valid form for the purposes of
-        * internal customisation of MediaWiki, via Messages*.php.
+        * internal customisation of MediaWiki, via Messages*.php or *.json.
         *
         * @param $code string
         *
@@ -910,12 +913,16 @@ class Language {
                        # We do this using a foreach over the codes instead of a directory
                        # loop so that messages files in extensions will work correctly.
                        foreach ( $returnMw as $code => $value ) {
-                               if ( is_readable( self::getMessagesFileName( $code ) ) ) {
+                               if ( is_readable( self::getMessagesFileName( $code ) )
+                                       || is_readable( self::getJsonMessagesFileName( $code ) )
+                               ) {
                                        $namesMwFile[$code] = $names[$code];
                                }
                        }
+
                        return $namesMwFile;
                }
+
                # 'mw' option; default if it's not one of the other two options (all/mwfile)
                return $returnMw;
        }
@@ -4048,10 +4055,7 @@ class Language {
         * @return string $prefix . $mangledCode . $suffix
         */
        public static function getFileName( $prefix = 'Language', $code, $suffix = '.php' ) {
-               // Protect against path traversal
-               if ( !Language::isValidCode( $code )
-                       || strcspn( $code, ":/\\\000" ) !== strlen( $code )
-               ) {
+               if ( !self::isValidBuiltInCode( $code ) ) {
                        throw new MWException( "Invalid language code \"$code\"" );
                }
 
@@ -4086,6 +4090,21 @@ class Language {
                return $file;
        }
 
+       /**
+        * @param $code string
+        * @return string
+        * @since 1.23
+        */
+       public static function getJsonMessagesFileName( $code ) {
+               global $IP;
+
+               if ( !self::isValidBuiltInCode( $code ) ) {
+                       throw new MWException( "Invalid language code \"$code\"" );
+               }
+
+               return "$IP/languages/i18n/$code.json" ;
+       }
+
        /**
         * @param $code string
         * @return string

diff --git a/tests/phpunit/languages/LanguageClassesTestCase.php b/tests/phpunit/languages/LanguageClassesTestCase.php
index 632e037..f93ff7d 100644 (file)
--- a/tests/phpunit/languages/LanguageClassesTestCase.php
+++ b/tests/phpunit/languages/LanguageClassesTestCase.php
@@ -50,7 +50,7 @@ abstract class LanguageClassesTestCase extends MediaWikiTestCase {
                $found = preg_match( '/Language(.+)Test/', get_called_class(), $m );
                if ( $found ) {
                        # Normalize language code since classes uses underscores
-                       $m[1] = str_replace( '_', '-', $m[1] );
+                       $m[1] = strtolower( str_replace( '_', '-', $m[1] ) );
                } else {
                        # Fallback to english language
                        $m[1] = 'en';