diff --git a/stepmania/src/smpackage/LanguagesDlg.cpp b/stepmania/src/smpackage/LanguagesDlg.cpp
index 95b0f78857..d7f699d966 100644
--- a/stepmania/src/smpackage/LanguagesDlg.cpp
+++ b/stepmania/src/smpackage/LanguagesDlg.cpp
@@ -463,92 +463,96 @@ void LanguagesDlg::OnBnClickedCheckLanguage()
 	RString sFullFile = SpecialDirs::GetDesktopDir() + sFile;
 	file.Open( sFullFile, RageFile::WRITE );
 
-	FOREACH_CONST_Child( &ini1, key )
 	{
-		FOREACH_CONST_Attr( key, value )
+		FOREACH_CONST_Child( &ini1, key )
 		{
-			const RString &sSection = key->m_sName;
-			const RString &sID = value->first;
-			const RString &sBaseLanguage = value->second;
-			RString sCurrentLanguage;
-			ini2.GetValue( sSection, sID, sCurrentLanguage );
-
-			if( sCurrentLanguage.empty() )
-				continue;
-
-			/* Check &FOO;, %{foo}, %-1.2f, %.  Some mismatches here are normal, particularly in
-			 * languages with substantially different word order rules than English. */
+			FOREACH_CONST_Attr( key, value )
 			{
-				RString sRegex = "(&[A-Za-z]+;|%{[A-Za-z]+}|%[+-]?[0-9]*.?[0-9]*[sidf]|%)";
-				vector<RString> asMatchesBase;
-				GetAllMatches( sRegex, sBaseLanguage, asMatchesBase );
+				const RString &sSection = key->m_sName;
+				const RString &sID = value->first;
+				const RString &sBaseLanguage = value->second;
+				RString sCurrentLanguage;
+				ini2.GetValue( sSection, sID, sCurrentLanguage );
 
-				vector<RString> asMatches;
-				GetAllMatches( sRegex, sCurrentLanguage, asMatches );
+				if( sCurrentLanguage.empty() )
+					continue;
 
-				if( asMatchesBase.size() != asMatches.size() ||
-					!std::equal(asMatchesBase.begin(), asMatchesBase.end(), asMatches.begin()) )
+				/* Check &FOO;, %{foo}, %-1.2f, %.  Some mismatches here are normal, particularly in
+				 * languages with substantially different word order rules than English. */
 				{
-					file.PutLine( ssprintf("Entity/substitution mismatch in section [%s] (%s):", sSection.c_str(), sID.c_str()) );
-					file.PutLine( ssprintf("    %s", sCurrentLanguage.c_str()) );
-					file.PutLine( ssprintf("    %s", sBaseLanguage.c_str()) );
+					RString sRegex = "(&[A-Za-z]+;|%{[A-Za-z]+}|%[+-]?[0-9]*.?[0-9]*[sidf]|%)";
+					vector<RString> asMatchesBase;
+					GetAllMatches( sRegex, sBaseLanguage, asMatchesBase );
+
+					vector<RString> asMatches;
+					GetAllMatches( sRegex, sCurrentLanguage, asMatches );
+
+					if( asMatchesBase.size() != asMatches.size() ||
+						!std::equal(asMatchesBase.begin(), asMatchesBase.end(), asMatches.begin()) )
+					{
+						file.PutLine( ssprintf("Entity/substitution mismatch in section [%s] (%s):", sSection.c_str(), sID.c_str()) );
+						file.PutLine( ssprintf("    %s", sCurrentLanguage.c_str()) );
+						file.PutLine( ssprintf("    %s", sBaseLanguage.c_str()) );
+					}
 				}
-			}
 
-			/* Check "foo::bar", "foo\nbar" and double quotes.  Check these independently
-			 * of the above. */
-			{
-				RString sRegex = "(\"|::|\\n)";
-				vector<RString> asMatchesBase;
-				GetAllMatches( sRegex, sBaseLanguage, asMatchesBase );
-
-				vector<RString> asMatches;
-				GetAllMatches( sRegex, sCurrentLanguage, asMatches );
-
-				if( asMatchesBase.size() != asMatches.size() ||
-					!std::equal(asMatchesBase.begin(), asMatchesBase.end(), asMatches.begin()) )
+				/* Check "foo::bar", "foo\nbar" and double quotes.  Check these independently
+				 * of the above. */
 				{
-					file.PutLine( ssprintf("Quote/line break mismatch in section [%s] (%s):", sSection.c_str(), sID.c_str()) );
-					file.PutLine( ssprintf("    %s", sCurrentLanguage.c_str()) );
-					file.PutLine( ssprintf("    %s", sBaseLanguage.c_str()) );
+					RString sRegex = "(\"|::|\\n)";
+					vector<RString> asMatchesBase;
+					GetAllMatches( sRegex, sBaseLanguage, asMatchesBase );
+
+					vector<RString> asMatches;
+					GetAllMatches( sRegex, sCurrentLanguage, asMatches );
+
+					if( asMatchesBase.size() != asMatches.size() ||
+						!std::equal(asMatchesBase.begin(), asMatchesBase.end(), asMatches.begin()) )
+					{
+						file.PutLine( ssprintf("Quote/line break mismatch in section [%s] (%s):", sSection.c_str(), sID.c_str()) );
+						file.PutLine( ssprintf("    %s", sCurrentLanguage.c_str()) );
+						file.PutLine( ssprintf("    %s", sBaseLanguage.c_str()) );
+					}
 				}
 			}
 		}
 	}
 
-	FOREACH_CONST_Child( &ini1, key )
 	{
-		bool bFirst = true;
-		RString sLastSection;
-		RString sLine;
-		FOREACH_CONST_Attr( key, value )
+		FOREACH_CONST_Child( &ini1, key )
 		{
-			const RString &sSection = key->m_sName;
-			const RString &sID = value->first;
-			RString sCurrentLanguage;
-			ini2.GetValue( sSection, sID, sCurrentLanguage );
-			if( !sCurrentLanguage.empty() )
-				continue;
+			bool bFirst = true;
+			RString sLastSection;
+			RString sLine;
+			FOREACH_CONST_Attr( key, value )
+			{
+				const RString &sSection = key->m_sName;
+				const RString &sID = value->first;
+				RString sCurrentLanguage;
+				ini2.GetValue( sSection, sID, sCurrentLanguage );
+				if( !sCurrentLanguage.empty() )
+					continue;
 
-			if( bFirst )
-			{
-				bFirst = false;
-				file.PutLine( ssprintf("Not translated in section [%s]:", sSection.c_str()) );
-			}
-			if( sLine.size() + sID.size() > 100 )
-			{
-				file.PutLine( sLine );
-				sLine = "";
+				if( bFirst )
+				{
+					bFirst = false;
+					file.PutLine( ssprintf("Not translated in section [%s]:", sSection.c_str()) );
+				}
+				if( sLine.size() + sID.size() > 100 )
+				{
+					file.PutLine( sLine );
+					sLine = "";
+				}
+				if( sLine.size() )
+					sLine += ", ";
+				else
+					sLine += "    ";
+
+				sLine += sID;
 			}
 			if( sLine.size() )
-				sLine += ", ";
-			else
-				sLine += "    ";
-
-			sLine += sID;
+				file.PutLine( sLine );
 		}
-		if( sLine.size() )
-			file.PutLine( sLine );
 	}
 
 	file.Close();