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
diff --git a/include/clang/AST/ASTContext.h b/include/clang/AST/ASTContext.h
index b13028d..56b3e69 100644
--- a/include/clang/AST/ASTContext.h
+++ b/include/clang/AST/ASTContext.h
@@ -198,7 +198,9 @@
QualType VoidTy;
QualType BoolTy;
QualType CharTy;
- QualType WCharTy; // [C++ 3.9.1p5], integer type in C99.
+ QualType WCharTy; // [C++ 3.9.1p5], integer type in C99.
+ QualType Char16Ty; // [C++0x 3.9.1p5], integer type in C99.
+ QualType Char32Ty; // [C++0x 3.9.1p5], integer type in C99.
QualType SignedCharTy, ShortTy, IntTy, LongTy, LongLongTy, Int128Ty;
QualType UnsignedCharTy, UnsignedShortTy, UnsignedIntTy, UnsignedLongTy;
QualType UnsignedLongLongTy, UnsignedInt128Ty;
diff --git a/include/clang/AST/Type.h b/include/clang/AST/Type.h
index 8634bb0..d3f7b5b 100644
--- a/include/clang/AST/Type.h
+++ b/include/clang/AST/Type.h
@@ -567,6 +567,8 @@
Bool, // This is bool and/or _Bool.
Char_U, // This is 'char' for targets where char is unsigned.
UChar, // This is explicitly qualified unsigned char.
+ Char16, // This is 'char16_t' for C++.
+ Char32, // This is 'char32_t' for C++.
UShort,
UInt,
ULong,
diff --git a/include/clang/Basic/TargetInfo.h b/include/clang/Basic/TargetInfo.h
index 537d553..b1f9821 100644
--- a/include/clang/Basic/TargetInfo.h
+++ b/include/clang/Basic/TargetInfo.h
@@ -41,6 +41,8 @@
bool TLSSupported;
unsigned char PointerWidth, PointerAlign;
unsigned char WCharWidth, WCharAlign;
+ unsigned char Char16Width, Char16Align;
+ unsigned char Char32Width, Char32Align;
unsigned char IntWidth, IntAlign;
unsigned char FloatWidth, FloatAlign;
unsigned char DoubleWidth, DoubleAlign;
@@ -77,7 +79,7 @@
};
protected:
IntType SizeType, IntMaxType, UIntMaxType, PtrDiffType, IntPtrType, WCharType,
- Int64Type;
+ Char16Type, Char32Type, Int64Type;
public:
IntType getSizeType() const { return SizeType; }
IntType getIntMaxType() const { return IntMaxType; }
@@ -87,6 +89,8 @@
}
IntType getIntPtrType() const { return IntPtrType; }
IntType getWCharType() const { return WCharType; }
+ IntType getChar16Type() const { return Char16Type; }
+ IntType getChar32Type() const { return Char32Type; }
IntType getInt64Type() const { return Int64Type; }
/// getPointerWidth - Return the width of pointers on this target, for the
@@ -102,13 +106,9 @@
/// target, in bits.
unsigned getBoolWidth(bool isWide = false) const { return 8; } // FIXME
unsigned getBoolAlign(bool isWide = false) const { return 8; } // FIXME
-
- unsigned getCharWidth(bool isWide = false) const {
- return isWide ? getWCharWidth() : 8; // FIXME
- }
- unsigned getCharAlign(bool isWide = false) const {
- return isWide ? getWCharAlign() : 8; // FIXME
- }
+
+ unsigned getCharWidth() const { return 8; } // FIXME
+ unsigned getCharAlign() const { return 8; } // FIXME
/// getShortWidth/Align - Return the size of 'signed short' and
/// 'unsigned short' for this target, in bits.
@@ -130,11 +130,21 @@
unsigned getLongLongWidth() const { return LongLongWidth; }
unsigned getLongLongAlign() const { return LongLongAlign; }
- /// getWcharWidth/Align - Return the size of 'wchar_t' for this target, in
+ /// getWCharWidth/Align - Return the size of 'wchar_t' for this target, in
/// bits.
unsigned getWCharWidth() const { return WCharWidth; }
unsigned getWCharAlign() const { return WCharAlign; }
+ /// getChar16Width/Align - Return the size of 'char16_t' for this target, in
+ /// bits.
+ unsigned getChar16Width() const { return Char16Width; }
+ unsigned getChar16Align() const { return Char16Align; }
+
+ /// getChar32Width/Align - Return the size of 'char32_t' for this target, in
+ /// bits.
+ unsigned getChar32Width() const { return Char32Width; }
+ unsigned getChar32Align() const { return Char32Align; }
+
/// getFloatWidth/Align/Format - Return the size/align/format of 'float'.
unsigned getFloatWidth() const { return FloatWidth; }
unsigned getFloatAlign() const { return FloatAlign; }
diff --git a/include/clang/Frontend/PCHBitCodes.h b/include/clang/Frontend/PCHBitCodes.h
index b615543..a1feec3 100644
--- a/include/clang/Frontend/PCHBitCodes.h
+++ b/include/clang/Frontend/PCHBitCodes.h
@@ -329,7 +329,11 @@
/// \brief The '__int128_t' type.
PREDEF_TYPE_INT128_ID = 22,
/// \brief The type of 'nullptr'.
- PREDEF_TYPE_NULLPTR_ID = 23
+ PREDEF_TYPE_NULLPTR_ID = 23,
+ /// \brief The C++ 'char16_t' type.
+ PREDEF_TYPE_CHAR16_ID = 24,
+ /// \brief The C++ 'char32_t' type.
+ PREDEF_TYPE_CHAR32_ID = 25
};
/// \brief The number of predefined type IDs that are reserved for
diff --git a/include/clang/Frontend/TypeXML.def b/include/clang/Frontend/TypeXML.def
index 2a78fd9..dd12738 100644
--- a/include/clang/Frontend/TypeXML.def
+++ b/include/clang/Frontend/TypeXML.def
@@ -104,6 +104,8 @@
ENUM_XML(BuiltinType::Double, "double");
ENUM_XML(BuiltinType::LongDouble, "long double");
ENUM_XML(BuiltinType::WChar, "wchar_t");
+ ENUM_XML(BuiltinType::Char16, "char16_t");
+ ENUM_XML(BuiltinType::Char32, "char32_t");
ENUM_XML(BuiltinType::NullPtr, "nullptr_t"); // This is the type of C++0x 'nullptr'.
ENUM_XML(BuiltinType::Overload, "overloaded");
ENUM_XML(BuiltinType::Dependent, "dependent");
diff --git a/include/clang/Parse/DeclSpec.h b/include/clang/Parse/DeclSpec.h
index 0c57915..b6044ff 100644
--- a/include/clang/Parse/DeclSpec.h
+++ b/include/clang/Parse/DeclSpec.h
@@ -68,6 +68,8 @@
TST_void,
TST_char,
TST_wchar, // C++ wchar_t
+ TST_char16, // C++0x char16_t
+ TST_char32, // C++0x char32_t
TST_int,
TST_float,
TST_double,
diff --git a/lib/AST/ASTContext.cpp b/lib/AST/ASTContext.cpp
index 5de6659..9d1e126 100644
--- a/lib/AST/ASTContext.cpp
+++ b/lib/AST/ASTContext.cpp
@@ -170,6 +170,16 @@
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 @@
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 @@
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 7a9faac..9edb9c0 100644
--- a/lib/AST/Type.cpp
+++ b/lib/AST/Type.cpp
@@ -1003,6 +1003,8 @@
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 ba7f190..5b2ffb7 100644
--- a/lib/Basic/TargetInfo.cpp
+++ b/lib/Basic/TargetInfo.cpp
@@ -25,6 +25,8 @@
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 @@
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 6add2c2..5ed2b7c 100644
--- a/lib/CodeGen/CodeGenTypes.cpp
+++ b/lib/CodeGen/CodeGenTypes.cpp
@@ -249,6 +249,8 @@
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 97c26f8..cd0e2ea 100644
--- a/lib/CodeGen/Mangle.cpp
+++ b/lib/CodeGen/Mangle.cpp
@@ -540,8 +540,8 @@
// 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 @@
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 c922b0d..012c6fe 100644
--- a/lib/Frontend/PCHReader.cpp
+++ b/lib/Frontend/PCHReader.cpp
@@ -1984,6 +1984,8 @@
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 57577b7..333bcc9 100644
--- a/lib/Frontend/PCHWriter.cpp
+++ b/lib/Frontend/PCHWriter.cpp
@@ -2005,6 +2005,8 @@
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 c98acc4..a7307c6 100644
--- a/lib/Lex/PPExpressions.cpp
+++ b/lib/Lex/PPExpressions.cpp
@@ -224,8 +224,10 @@
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 8b3b285..e58076e 100644
--- a/lib/Parse/DeclSpec.cpp
+++ b/lib/Parse/DeclSpec.cpp
@@ -159,6 +159,8 @@
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 94855b2..cefd325 100644
--- a/lib/Parse/ParseDecl.cpp
+++ b/lib/Parse/ParseDecl.cpp
@@ -159,6 +159,8 @@
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 @@
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 @@
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 @@
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 @@
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 cd7618f..c2086b9 100644
--- a/lib/Parse/ParseExpr.cpp
+++ b/lib/Parse/ParseExpr.cpp
@@ -739,6 +739,8 @@
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 1c00a8e..68101fc 100644
--- a/lib/Parse/ParseExprCXX.cpp
+++ b/lib/Parse/ParseExprCXX.cpp
@@ -654,6 +654,12 @@
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 02687a2..97f6f52 100644
--- a/lib/Parse/ParseTentative.cpp
+++ b/lib/Parse/ParseTentative.cpp
@@ -681,6 +681,8 @@
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 b7f698e..bee3936 100644
--- a/lib/Sema/SemaOverload.cpp
+++ b/lib/Sema/SemaOverload.cpp
@@ -2792,7 +2792,8 @@
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 7551d93..8195aba 100644
--- a/lib/Sema/SemaType.cpp
+++ b/lib/Sema/SemaType.cpp
@@ -88,6 +88,16 @@
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()) {