Basic support for C++0x unicode types. Support for literals will follow in an incremental patch

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@75622 91177308-0d34-0410-b5e6-96231b3b80d8

15 files changed, 86 insertions, 4 deletions

diff --git a/lib/AST/ASTContext.cpp b/lib/AST/ASTContext.cpp
index 5de66596e1..9d1e1267e4 100644
--- a/lib/AST/ASTContext.cpp
+++ b/lib/AST/ASTContext.cpp
@@ -170,6 +170,16 @@ void ASTContext::InitBuiltinTypes() {
   else // C99
     WCharTy = getFromTargetType(Target.getWCharType());
 
+  if (LangOpts.CPlusPlus) // C++0x 3.9.1p5, extension for C++
+    InitBuiltinType(Char16Ty,           BuiltinType::Char16);
+  else // C99
+    Char16Ty = getFromTargetType(Target.getChar16Type());
+
+  if (LangOpts.CPlusPlus) // C++0x 3.9.1p5, extension for C++
+    InitBuiltinType(Char32Ty,           BuiltinType::Char32);
+  else // C99
+    Char32Ty = getFromTargetType(Target.getChar32Type());
+
   // Placeholder type for functions.
   InitBuiltinType(OverloadTy,          BuiltinType::Overload);
 
@@ -522,6 +532,14 @@ ASTContext::getTypeInfo(const Type *T) {
       Width = Target.getWCharWidth();
       Align = Target.getWCharAlign();
       break;
+    case BuiltinType::Char16:
+      Width = Target.getChar16Width();
+      Align = Target.getChar16Align();
+      break;
+    case BuiltinType::Char32:
+      Width = Target.getChar32Width();
+      Align = Target.getChar32Align();
+      break;
     case BuiltinType::UShort:
     case BuiltinType::Short:
       Width = Target.getShortWidth();
@@ -2326,6 +2344,12 @@ unsigned ASTContext::getIntegerRank(Type *T) {
   if (T->isSpecificBuiltinType(BuiltinType::WChar))
     T = getFromTargetType(Target.getWCharType()).getTypePtr();
 
+  if (T->isSpecificBuiltinType(BuiltinType::Char16))
+    T = getFromTargetType(Target.getChar16Type()).getTypePtr();
+
+  if (T->isSpecificBuiltinType(BuiltinType::Char32))
+    T = getFromTargetType(Target.getChar32Type()).getTypePtr();
+
   // There are two things which impact the integer rank: the width, and
   // the ordering of builtins.  The builtin ordering is encoded in the
   // bottom three bits; the width is encoded in the bits above that.
diff --git a/lib/AST/Type.cpp b/lib/AST/Type.cpp
index 7a9faac60f..9edb9c046d 100644
--- a/lib/AST/Type.cpp
+++ b/lib/AST/Type.cpp
@@ -1003,6 +1003,8 @@ const char *BuiltinType::getName(const LangOptions &LO) const {
   case Double:            return "double";
   case LongDouble:        return "long double";
   case WChar:             return "wchar_t";
+  case Char16:            return "char16_t";
+  case Char32:            return "char32_t";
   case NullPtr:           return "nullptr_t";
   case Overload:          return "<overloaded function type>";
   case Dependent:         return "<dependent type>";
diff --git a/lib/Basic/TargetInfo.cpp b/lib/Basic/TargetInfo.cpp
index ba7f190408..5b2ffb7d16 100644
--- a/lib/Basic/TargetInfo.cpp
+++ b/lib/Basic/TargetInfo.cpp
@@ -25,6 +25,8 @@ TargetInfo::TargetInfo(const std::string &T) : Triple(T) {
   TLSSupported = true;
   PointerWidth = PointerAlign = 32;
   WCharWidth = WCharAlign = 32;
+  Char16Width = Char16Align = 16;
+  Char32Width = Char32Align = 32;
   IntWidth = IntAlign = 32;
   LongWidth = LongAlign = 32;
   LongLongWidth = LongLongAlign = 64;
@@ -41,6 +43,8 @@ TargetInfo::TargetInfo(const std::string &T) : Triple(T) {
   UIntMaxType = UnsignedLongLong;
   IntPtrType = SignedLong;
   WCharType = SignedInt;
+  Char16Type = UnsignedShort;
+  Char32Type = UnsignedInt;
   Int64Type = SignedLongLong;
   FloatFormat = &llvm::APFloat::IEEEsingle;
   DoubleFormat = &llvm::APFloat::IEEEdouble;
diff --git a/lib/CodeGen/CodeGenTypes.cpp b/lib/CodeGen/CodeGenTypes.cpp
index 6add2c25c7..5ed2b7c2a0 100644
--- a/lib/CodeGen/CodeGenTypes.cpp
+++ b/lib/CodeGen/CodeGenTypes.cpp
@@ -249,6 +249,8 @@ const llvm::Type *CodeGenTypes::ConvertNewType(QualType T) {
     case BuiltinType::LongLong:
     case BuiltinType::ULongLong:
     case BuiltinType::WChar:
+    case BuiltinType::Char16:
+    case BuiltinType::Char32:
       return llvm::IntegerType::get(
         static_cast<unsigned>(Context.getTypeSize(T)));
       
diff --git a/lib/CodeGen/Mangle.cpp b/lib/CodeGen/Mangle.cpp
index 97c26f87b1..cd0e2eae7d 100644
--- a/lib/CodeGen/Mangle.cpp
+++ b/lib/CodeGen/Mangle.cpp
@@ -540,8 +540,8 @@ void CXXNameMangler::mangleType(const BuiltinType *T) {
   // UNSUPPORTED:    ::= De # IEEE 754r decimal floating point (128 bits)
   // UNSUPPORTED:    ::= Df # IEEE 754r decimal floating point (32 bits)
   // UNSUPPORTED:    ::= Dh # IEEE 754r half-precision floating point (16 bits)
-  // UNSUPPORTED:    ::= Di # char32_t
-  // UNSUPPORTED:    ::= Ds # char16_t
+  //                 ::= Di # char32_t
+  //                 ::= Ds # char16_t
   //                 ::= u <source-name>    # vendor extended type
   // From our point of view, std::nullptr_t is a builtin, but as far as mangling
   // is concerned, it's a type called std::nullptr_t.
@@ -557,6 +557,8 @@ void CXXNameMangler::mangleType(const BuiltinType *T) {
   case BuiltinType::UInt128: Out << 'o'; break;
   case BuiltinType::SChar: Out << 'a'; break;
   case BuiltinType::WChar: Out << 'w'; break;
+  case BuiltinType::Char16: Out << "Ds"; break;
+  case BuiltinType::Char32: Out << "Di"; break;
   case BuiltinType::Short: Out << 's'; break;
   case BuiltinType::Int: Out << 'i'; break;
   case BuiltinType::Long: Out << 'l'; break;
diff --git a/lib/Frontend/PCHReader.cpp b/lib/Frontend/PCHReader.cpp
index c922b0d6db..012c6fecfe 100644
--- a/lib/Frontend/PCHReader.cpp
+++ b/lib/Frontend/PCHReader.cpp
@@ -1984,6 +1984,8 @@ QualType PCHReader::GetType(pch::TypeID ID) {
     case pch::PREDEF_TYPE_OVERLOAD_ID:   T = Context->OverloadTy;         break;
     case pch::PREDEF_TYPE_DEPENDENT_ID:  T = Context->DependentTy;        break;
     case pch::PREDEF_TYPE_NULLPTR_ID:    T = Context->NullPtrTy;          break;
+    case pch::PREDEF_TYPE_CHAR16_ID:     T = Context->Char16Ty;           break;
+    case pch::PREDEF_TYPE_CHAR32_ID:     T = Context->Char32Ty;           break;
     }
 
     assert(!T.isNull() && "Unknown predefined type");
diff --git a/lib/Frontend/PCHWriter.cpp b/lib/Frontend/PCHWriter.cpp
index 57577b7dfa..333bcc9893 100644
--- a/lib/Frontend/PCHWriter.cpp
+++ b/lib/Frontend/PCHWriter.cpp
@@ -2005,6 +2005,8 @@ void PCHWriter::AddTypeRef(QualType T, RecordData &Record) {
     case BuiltinType::Double:     ID = pch::PREDEF_TYPE_DOUBLE_ID;     break;
     case BuiltinType::LongDouble: ID = pch::PREDEF_TYPE_LONGDOUBLE_ID; break;
     case BuiltinType::NullPtr:    ID = pch::PREDEF_TYPE_NULLPTR_ID;    break;
+    case BuiltinType::Char16:     ID = pch::PREDEF_TYPE_CHAR16_ID;     break;
+    case BuiltinType::Char32:     ID = pch::PREDEF_TYPE_CHAR32_ID;     break;
     case BuiltinType::Overload:   ID = pch::PREDEF_TYPE_OVERLOAD_ID;   break;
     case BuiltinType::Dependent:  ID = pch::PREDEF_TYPE_DEPENDENT_ID;  break;
     case BuiltinType::UndeducedAuto:
diff --git a/lib/Lex/PPExpressions.cpp b/lib/Lex/PPExpressions.cpp
index c98acc4deb..a7307c6b56 100644
--- a/lib/Lex/PPExpressions.cpp
+++ b/lib/Lex/PPExpressions.cpp
@@ -224,8 +224,10 @@ static bool EvaluateValue(PPValue &Result, Token &PeekTok, DefinedTracker &DT,
     unsigned NumBits;
     if (Literal.isMultiChar())
       NumBits = TI.getIntWidth();
+    else if (Literal.isWide())
+      NumBits = TI.getWCharWidth();
     else
-      NumBits = TI.getCharWidth(Literal.isWide());
+      NumBits = TI.getCharWidth();
 
     // Set the width.
     llvm::APSInt Val(NumBits);
diff --git a/lib/Parse/DeclSpec.cpp b/lib/Parse/DeclSpec.cpp
index 8b3b2851c1..e58076ee5a 100644
--- a/lib/Parse/DeclSpec.cpp
+++ b/lib/Parse/DeclSpec.cpp
@@ -159,6 +159,8 @@ const char *DeclSpec::getSpecifierName(DeclSpec::TST T) {
   case DeclSpec::TST_void:        return "void";
   case DeclSpec::TST_char:        return "char";
   case DeclSpec::TST_wchar:       return "wchar_t";
+  case DeclSpec::TST_char16:      return "char16_t";
+  case DeclSpec::TST_char32:      return "char32_t";
   case DeclSpec::TST_int:         return "int";
   case DeclSpec::TST_float:       return "float";
   case DeclSpec::TST_double:      return "double";
diff --git a/lib/Parse/ParseDecl.cpp b/lib/Parse/ParseDecl.cpp
index 94855b24f5..cefd325248 100644
--- a/lib/Parse/ParseDecl.cpp
+++ b/lib/Parse/ParseDecl.cpp
@@ -159,6 +159,8 @@ AttributeList *Parser::ParseAttributes(SourceLocation *EndLoc) {
             break;
           case tok::kw_char:
           case tok::kw_wchar_t:
+          case tok::kw_char16_t:
+          case tok::kw_char32_t:
           case tok::kw_bool:
           case tok::kw_short:
           case tok::kw_int:
@@ -999,6 +1001,12 @@ void Parser::ParseDeclarationSpecifiers(DeclSpec &DS,
     case tok::kw_wchar_t:
       isInvalid = DS.SetTypeSpecType(DeclSpec::TST_wchar, Loc, PrevSpec);
       break;
+    case tok::kw_char16_t:
+      isInvalid = DS.SetTypeSpecType(DeclSpec::TST_char16, Loc, PrevSpec);
+      break;
+    case tok::kw_char32_t:
+      isInvalid = DS.SetTypeSpecType(DeclSpec::TST_char32, Loc, PrevSpec);
+      break;
     case tok::kw_bool:
     case tok::kw__Bool:
       isInvalid = DS.SetTypeSpecType(DeclSpec::TST_bool, Loc, PrevSpec);
@@ -1226,6 +1234,12 @@ bool Parser::ParseOptionalTypeSpecifier(DeclSpec &DS, int& isInvalid,
   case tok::kw_wchar_t:
     isInvalid = DS.SetTypeSpecType(DeclSpec::TST_wchar, Loc, PrevSpec);
     break;
+  case tok::kw_char16_t:
+    isInvalid = DS.SetTypeSpecType(DeclSpec::TST_char16, Loc, PrevSpec);
+    break;
+  case tok::kw_char32_t:
+    isInvalid = DS.SetTypeSpecType(DeclSpec::TST_char32, Loc, PrevSpec);
+    break;
   case tok::kw_bool:
   case tok::kw__Bool:
     isInvalid = DS.SetTypeSpecType(DeclSpec::TST_bool, Loc, PrevSpec);
@@ -1714,6 +1728,8 @@ bool Parser::isTypeSpecifierQualifier() {
   case tok::kw_void:
   case tok::kw_char:
   case tok::kw_wchar_t:
+  case tok::kw_char16_t:
+  case tok::kw_char32_t:
   case tok::kw_int:
   case tok::kw_float:
   case tok::kw_double:
@@ -1802,6 +1818,9 @@ bool Parser::isDeclarationSpecifier() {
   case tok::kw_void:
   case tok::kw_char:
   case tok::kw_wchar_t:
+  case tok::kw_char16_t:
+  case tok::kw_char32_t:
+
   case tok::kw_int:
   case tok::kw_float:
   case tok::kw_double:
diff --git a/lib/Parse/ParseExpr.cpp b/lib/Parse/ParseExpr.cpp
index cd7618f665..c2086b9e1a 100644
--- a/lib/Parse/ParseExpr.cpp
+++ b/lib/Parse/ParseExpr.cpp
@@ -739,6 +739,8 @@ Parser::OwningExprResult Parser::ParseCastExpression(bool isUnaryExpression,
 
   case tok::kw_char:
   case tok::kw_wchar_t:
+  case tok::kw_char16_t:
+  case tok::kw_char32_t:
   case tok::kw_bool:
   case tok::kw_short:
   case tok::kw_int:
diff --git a/lib/Parse/ParseExprCXX.cpp b/lib/Parse/ParseExprCXX.cpp
index 1c00a8eabe..68101fcdf4 100644
--- a/lib/Parse/ParseExprCXX.cpp
+++ b/lib/Parse/ParseExprCXX.cpp
@@ -654,6 +654,12 @@ void Parser::ParseCXXSimpleTypeSpecifier(DeclSpec &DS) {
   case tok::kw_wchar_t:
     DS.SetTypeSpecType(DeclSpec::TST_wchar, Loc, PrevSpec);
     break;
+  case tok::kw_char16_t:
+    DS.SetTypeSpecType(DeclSpec::TST_char16, Loc, PrevSpec);
+    break;
+  case tok::kw_char32_t:
+    DS.SetTypeSpecType(DeclSpec::TST_char32, Loc, PrevSpec);
+    break;
   case tok::kw_bool:
     DS.SetTypeSpecType(DeclSpec::TST_bool, Loc, PrevSpec);
     break;
diff --git a/lib/Parse/ParseTentative.cpp b/lib/Parse/ParseTentative.cpp
index 02687a216c..97f6f526d4 100644
--- a/lib/Parse/ParseTentative.cpp
+++ b/lib/Parse/ParseTentative.cpp
@@ -681,6 +681,8 @@ Parser::TPResult Parser::isCXXDeclarationSpecifier() {
 
   case tok::kw_char:
   case tok::kw_wchar_t:
+  case tok::kw_char16_t:
+  case tok::kw_char32_t:
   case tok::kw_bool:
   case tok::kw_short:
   case tok::kw_int:
diff --git a/lib/Sema/SemaOverload.cpp b/lib/Sema/SemaOverload.cpp
index b7f698eb0b..bee3936048 100644
--- a/lib/Sema/SemaOverload.cpp
+++ b/lib/Sema/SemaOverload.cpp
@@ -2792,7 +2792,8 @@ Sema::AddBuiltinOperatorCandidates(OverloadedOperatorKind Op,
                  LastPromotedArithmeticType = 16;
   const unsigned NumArithmeticTypes = 16;
   QualType ArithmeticTypes[NumArithmeticTypes] = {
-    Context.BoolTy, Context.CharTy, Context.WCharTy,
+    Context.BoolTy, Context.CharTy, Context.WCharTy, 
+//    Context.Char16Ty, Context.Char32Ty, 
     Context.SignedCharTy, Context.ShortTy,
     Context.UnsignedCharTy, Context.UnsignedShortTy,
     Context.IntTy, Context.LongTy, Context.LongLongTy,
diff --git a/lib/Sema/SemaType.cpp b/lib/Sema/SemaType.cpp
index 7551d93c66..8195aba967 100644
--- a/lib/Sema/SemaType.cpp
+++ b/lib/Sema/SemaType.cpp
@@ -88,6 +88,16 @@ QualType Sema::ConvertDeclSpecToType(const DeclSpec &DS,
       Result = Context.getUnsignedWCharType();
     }
     break;
+  case DeclSpec::TST_char16:
+      assert(DS.getTypeSpecSign() == DeclSpec::TSS_unspecified &&
+        "Unknown TSS value");
+      Result = Context.Char16Ty;
+    break;
+  case DeclSpec::TST_char32:
+      assert(DS.getTypeSpecSign() == DeclSpec::TSS_unspecified &&
+        "Unknown TSS value");
+      Result = Context.Char32Ty;
+    break;
   case DeclSpec::TST_unspecified:
     // "<proto1,proto2>" is an objc qualified ID with a missing id.
     if (DeclSpec::ProtocolQualifierListTy PQ = DS.getProtocolQualifiers()) {