Add support for C++0x unicode string and character literals, from Craig Topper!

llvm-svn: 136210

8 files changed, 97 insertions, 42 deletions

diff --git a/clang/lib/Sema/SemaChecking.cpp b/clang/lib/Sema/SemaChecking.cpp
index 2e4198b5202..28085ef6eab 100644
--- a/clang/lib/Sema/SemaChecking.cpp
+++ b/clang/lib/Sema/SemaChecking.cpp
@@ -605,7 +605,7 @@ bool Sema::CheckObjCString(Expr *Arg) {
   Arg = Arg->IgnoreParenCasts();
   StringLiteral *Literal = dyn_cast<StringLiteral>(Arg);
 
-  if (!Literal || Literal->isWide()) {
+  if (!Literal || !Literal->isAscii()) {
     Diag(Arg->getLocStart(), diag::err_cfstring_literal_not_string_constant)
       << Arg->getSourceRange();
     return true;
@@ -1805,7 +1805,7 @@ void Sema::CheckFormatString(const StringLiteral *FExpr,
                              bool isPrintf) {
   
   // CHECK: is the format string a wide literal?
-  if (FExpr->isWide()) {
+  if (!FExpr->isAscii()) {
     Diag(FExpr->getLocStart(),
          diag::warn_format_string_is_wide_literal)
     << OrigFormatExpr->getSourceRange();
diff --git a/clang/lib/Sema/SemaDeclAttr.cpp b/clang/lib/Sema/SemaDeclAttr.cpp
index 9e20bc90184..2cbd83a8fd8 100644
--- a/clang/lib/Sema/SemaDeclAttr.cpp
+++ b/clang/lib/Sema/SemaDeclAttr.cpp
@@ -712,7 +712,7 @@ static void handleWeakRefAttr(Sema &S, Decl *D, const AttributeList &Attr) {
     Arg = Arg->IgnoreParenCasts();
     StringLiteral *Str = dyn_cast<StringLiteral>(Arg);
 
-    if (Str == 0 || Str->isWide()) {
+    if (!Str || !Str->isAscii()) {
       S.Diag(Attr.getLoc(), diag::err_attribute_argument_n_not_string)
           << "weakref" << 1;
       return;
@@ -737,7 +737,7 @@ static void handleAliasAttr(Sema &S, Decl *D, const AttributeList &Attr) {
   Arg = Arg->IgnoreParenCasts();
   StringLiteral *Str = dyn_cast<StringLiteral>(Arg);
 
-  if (Str == 0 || Str->isWide()) {
+  if (!Str || !Str->isAscii()) {
     S.Diag(Attr.getLoc(), diag::err_attribute_argument_n_not_string)
       << "alias" << 1;
     return;
@@ -1162,7 +1162,7 @@ static void handleVisibilityAttr(Sema &S, Decl *D, const AttributeList &Attr) {
   Arg = Arg->IgnoreParenCasts();
   StringLiteral *Str = dyn_cast<StringLiteral>(Arg);
 
-  if (Str == 0 || Str->isWide()) {
+  if (!Str || !Str->isAscii()) {
     S.Diag(Attr.getLoc(), diag::err_attribute_argument_n_not_string)
       << "visibility" << 1;
     return;
@@ -2464,7 +2464,7 @@ static void handleCallConvAttr(Sema &S, Decl *D, const AttributeList &Attr) {
   case AttributeList::AT_pcs: {
     Expr *Arg = Attr.getArg(0);
     StringLiteral *Str = dyn_cast<StringLiteral>(Arg);
-    if (Str == 0 || Str->isWide()) {
+    if (!Str || !Str->isAscii()) {
       S.Diag(Attr.getLoc(), diag::err_attribute_argument_n_not_string)
         << "pcs" << 1;
       Attr.setInvalid();
@@ -2519,7 +2519,7 @@ bool Sema::CheckCallingConvAttr(const AttributeList &attr, CallingConv &CC) {
   case AttributeList::AT_pcs: {
     Expr *Arg = attr.getArg(0);
     StringLiteral *Str = dyn_cast<StringLiteral>(Arg);
-    if (Str == 0 || Str->isWide()) {
+    if (!Str || !Str->isAscii()) {
       Diag(attr.getLoc(), diag::err_attribute_argument_n_not_string)
         << "pcs" << 1;
       attr.setInvalid();
@@ -2868,7 +2868,7 @@ static void handleUuidAttr(Sema &S, Decl *D, const AttributeList &Attr) {
 
     Expr *Arg = Attr.getArg(0);
     StringLiteral *Str = dyn_cast<StringLiteral>(Arg);
-    if (Str == 0 || Str->isWide()) {
+    if (!Str || !Str->isAscii()) {
       S.Diag(Attr.getLoc(), diag::err_attribute_argument_n_not_string)
         << "uuid" << 1;
       return;
diff --git a/clang/lib/Sema/SemaExpr.cpp b/clang/lib/Sema/SemaExpr.cpp
index 4a9b4bcfdf0..dedf7b0d778 100644
--- a/clang/lib/Sema/SemaExpr.cpp
+++ b/clang/lib/Sema/SemaExpr.cpp
@@ -997,11 +997,25 @@ Sema::ActOnStringLiteral(const Token *StringToks, unsigned NumStringToks) {
     StringTokLocs.push_back(StringToks[i].getLocation());
 
   QualType StrTy = Context.CharTy;
-  if (Literal.AnyWide) 
+  if (Literal.isWide())
     StrTy = Context.getWCharType();
+  else if (Literal.isUTF16())
+    StrTy = Context.Char16Ty;
+  else if (Literal.isUTF32())
+    StrTy = Context.Char32Ty;
   else if (Literal.Pascal)
     StrTy = Context.UnsignedCharTy;
 
+  StringLiteral::StringKind Kind = StringLiteral::Ascii;
+  if (Literal.isWide())
+    Kind = StringLiteral::Wide;
+  else if (Literal.isUTF8())
+    Kind = StringLiteral::UTF8;
+  else if (Literal.isUTF16())
+    Kind = StringLiteral::UTF16;
+  else if (Literal.isUTF32())
+    Kind = StringLiteral::UTF32;
+
   // A C++ string literal has a const-qualified element type (C++ 2.13.4p1).
   if (getLangOptions().CPlusPlus || getLangOptions().ConstStrings)
     StrTy.addConst();
@@ -1015,7 +1029,7 @@ Sema::ActOnStringLiteral(const Token *StringToks, unsigned NumStringToks) {
 
   // Pass &StringTokLocs[0], StringTokLocs.size() to factory!
   return Owned(StringLiteral::Create(Context, Literal.GetString(),
-                                     Literal.AnyWide, Literal.Pascal, StrTy,
+                                     Kind, Literal.Pascal, StrTy,
                                      &StringTokLocs[0],
                                      StringTokLocs.size()));
 }
@@ -2412,7 +2426,7 @@ ExprResult Sema::ActOnCharacterConstant(const Token &Tok) {
     return ExprError();
 
   CharLiteralParser Literal(ThisTok.begin(), ThisTok.end(), Tok.getLocation(),
-                            PP);
+                            PP, Tok.getKind());
   if (Literal.hadError())
     return ExprError();
 
@@ -2421,14 +2435,25 @@ ExprResult Sema::ActOnCharacterConstant(const Token &Tok) {
     Ty = Context.IntTy;   // 'x' and L'x' -> int in C.
   else if (Literal.isWide())
     Ty = Context.WCharTy; // L'x' -> wchar_t in C++.
+  else if (Literal.isUTF16())
+    Ty = Context.Char16Ty; // u'x' -> char16_t in C++0x.
+  else if (Literal.isUTF32())
+    Ty = Context.Char32Ty; // U'x' -> char32_t in C++0x.
   else if (Literal.isMultiChar())
     Ty = Context.IntTy;   // 'wxyz' -> int in C++.
   else
     Ty = Context.CharTy;  // 'x' -> char in C++
 
-  return Owned(new (Context) CharacterLiteral(Literal.getValue(),
-                                              Literal.isWide(),
-                                              Ty, Tok.getLocation()));
+  CharacterLiteral::CharacterKind Kind = CharacterLiteral::Ascii;
+  if (Literal.isWide())
+    Kind = CharacterLiteral::Wide;
+  else if (Literal.isUTF16())
+    Kind = CharacterLiteral::UTF16;
+  else if (Literal.isUTF32())
+    Kind = CharacterLiteral::UTF32;
+
+  return Owned(new (Context) CharacterLiteral(Literal.getValue(), Kind, Ty,
+                                              Tok.getLocation()));
 }
 
 ExprResult Sema::ActOnNumericConstant(const Token &Tok) {
@@ -8624,7 +8649,7 @@ static void MakeObjCStringLiteralFixItHint(Sema& SemaRef, QualType DstType,
 
   // Strip off any parens and casts.
   StringLiteral *SL = dyn_cast<StringLiteral>(SrcExpr->IgnoreParenCasts());
-  if (!SL || SL->isWide())
+  if (!SL || !SL->isAscii())
     return;
 
   Hint = FixItHint::CreateInsertion(SL->getLocStart(), "@");
diff --git a/clang/lib/Sema/SemaExprCXX.cpp b/clang/lib/Sema/SemaExprCXX.cpp
index 94a5bafa7c7..1812510942e 100644
--- a/clang/lib/Sema/SemaExprCXX.cpp
+++ b/clang/lib/Sema/SemaExprCXX.cpp
@@ -2041,12 +2041,20 @@ Sema::IsStringLiteralToNonConstPointerConversion(Expr *From, QualType ToType) {
           = ToPtrType->getPointeeType()->getAs<BuiltinType>()) {
         // This conversion is considered only when there is an
         // explicit appropriate pointer target type (C++ 4.2p2).
-        if (!ToPtrType->getPointeeType().hasQualifiers() &&
-            ((StrLit->isWide() && ToPointeeType->isWideCharType()) ||
-             (!StrLit->isWide() &&
-              (ToPointeeType->getKind() == BuiltinType::Char_U ||
-               ToPointeeType->getKind() == BuiltinType::Char_S))))
-          return true;
+        if (!ToPtrType->getPointeeType().hasQualifiers()) {
+          switch (StrLit->getKind()) {
+            case StringLiteral::UTF8:
+            case StringLiteral::UTF16:
+            case StringLiteral::UTF32:
+              // We don't allow UTF literals to be implicitly converted
+              break;
+            case StringLiteral::Ascii:
+              return (ToPointeeType->getKind() == BuiltinType::Char_U ||
+                      ToPointeeType->getKind() == BuiltinType::Char_S);
+            case StringLiteral::Wide:
+              return ToPointeeType->isWideCharType();
+          }
+        }
       }
 
   return false;
diff --git a/clang/lib/Sema/SemaExprObjC.cpp b/clang/lib/Sema/SemaExprObjC.cpp
index fccea7c0e1b..e88726b6d2a 100644
--- a/clang/lib/Sema/SemaExprObjC.cpp
+++ b/clang/lib/Sema/SemaExprObjC.cpp
@@ -47,8 +47,8 @@ ExprResult Sema::ParseObjCStringLiteral(SourceLocation *AtLocs,
     for (unsigned i = 0; i != NumStrings; ++i) {
       S = Strings[i];
 
-      // ObjC strings can't be wide.
-      if (S->isWide()) {
+      // ObjC strings can't be wide or UTF.
+      if (!S->isAscii()) {
         Diag(S->getLocStart(), diag::err_cfstring_literal_not_string_constant)
           << S->getSourceRange();
         return true;
@@ -64,7 +64,7 @@ ExprResult Sema::ParseObjCStringLiteral(SourceLocation *AtLocs,
     // Create the aggregate string with the appropriate content and location
     // information.
     S = StringLiteral::Create(Context, StrBuf,
-                              /*Wide=*/false, /*Pascal=*/false,
+                              StringLiteral::Ascii, /*Pascal=*/false,
                               Context.getPointerType(Context.CharTy),
                               &StrLocs[0], StrLocs.size());
   }
diff --git a/clang/lib/Sema/SemaInit.cpp b/clang/lib/Sema/SemaInit.cpp
index adf88c62ccd..c406ad98405 100644
--- a/clang/lib/Sema/SemaInit.cpp
+++ b/clang/lib/Sema/SemaInit.cpp
@@ -49,20 +49,30 @@ static Expr *IsStringInit(Expr *Init, const ArrayType *AT,
   if (SL == 0) return 0;
 
   QualType ElemTy = Context.getCanonicalType(AT->getElementType());
-  // char array can be initialized with a narrow string.
-  // Only allow char x[] = "foo";  not char x[] = L"foo";
-  if (!SL->isWide())
+
+  switch (SL->getKind()) {
+  case StringLiteral::Ascii:
+  case StringLiteral::UTF8:
+    // char array can be initialized with a narrow string.
+    // Only allow char x[] = "foo";  not char x[] = L"foo";
     return ElemTy->isCharType() ? Init : 0;
+  case StringLiteral::UTF16:
+    return ElemTy->isChar16Type() ? Init : 0;
+  case StringLiteral::UTF32:
+    return ElemTy->isChar32Type() ? Init : 0;
+  case StringLiteral::Wide:
+    // wchar_t array can be initialized with a wide string: C99 6.7.8p15 (with
+    // correction from DR343): "An array with element type compatible with a
+    // qualified or unqualified version of wchar_t may be initialized by a wide
+    // string literal, optionally enclosed in braces."
+    if (Context.typesAreCompatible(Context.getWCharType(),
+                                   ElemTy.getUnqualifiedType()))
+      return Init;
 
-  // wchar_t array can be initialized with a wide string: C99 6.7.8p15 (with
-  // correction from DR343): "An array with element type compatible with a
-  // qualified or unqualified version of wchar_t may be initialized by a wide
-  // string literal, optionally enclosed in braces."
-  if (Context.typesAreCompatible(Context.getWCharType(),
-                                 ElemTy.getUnqualifiedType()))
-    return Init;
+    return 0;
+  }
 
-  return 0;
+  llvm_unreachable("missed a StringLiteral kind?");
 }
 
 static Expr *IsStringInit(Expr *init, QualType declType, ASTContext &Context) {
diff --git a/clang/lib/Sema/SemaStmt.cpp b/clang/lib/Sema/SemaStmt.cpp
index 0fd3f039823..56161ed9b45 100644
--- a/clang/lib/Sema/SemaStmt.cpp
+++ b/clang/lib/Sema/SemaStmt.cpp
@@ -1952,13 +1952,13 @@ StmtResult Sema::ActOnAsmStmt(SourceLocation AsmLoc, bool IsSimple,
   SmallVector<TargetInfo::ConstraintInfo, 4> OutputConstraintInfos;
 
   // The parser verifies that there is a string literal here.
-  if (AsmString->isWide())
+  if (!AsmString->isAscii())
     return StmtError(Diag(AsmString->getLocStart(),diag::err_asm_wide_character)
       << AsmString->getSourceRange());
 
   for (unsigned i = 0; i != NumOutputs; i++) {
     StringLiteral *Literal = Constraints[i];
-    if (Literal->isWide())
+    if (!Literal->isAscii())
       return StmtError(Diag(Literal->getLocStart(),diag::err_asm_wide_character)
         << Literal->getSourceRange());
 
@@ -1987,7 +1987,7 @@ StmtResult Sema::ActOnAsmStmt(SourceLocation AsmLoc, bool IsSimple,
 
   for (unsigned i = NumOutputs, e = NumOutputs + NumInputs; i != e; i++) {
     StringLiteral *Literal = Constraints[i];
-    if (Literal->isWide())
+    if (!Literal->isAscii())
       return StmtError(Diag(Literal->getLocStart(),diag::err_asm_wide_character)
         << Literal->getSourceRange());
 
@@ -2034,7 +2034,7 @@ StmtResult Sema::ActOnAsmStmt(SourceLocation AsmLoc, bool IsSimple,
   // Check that the clobbers are valid.
   for (unsigned i = 0; i != NumClobbers; i++) {
     StringLiteral *Literal = Clobbers[i];
-    if (Literal->isWide())
+    if (!Literal->isAscii())
       return StmtError(Diag(Literal->getLocStart(),diag::err_asm_wide_character)
         << Literal->getSourceRange());
 
diff --git a/clang/lib/Sema/SemaTemplate.cpp b/clang/lib/Sema/SemaTemplate.cpp
index ceab7e93ac6..006017f5a47 100644
--- a/clang/lib/Sema/SemaTemplate.cpp
+++ b/clang/lib/Sema/SemaTemplate.cpp
@@ -4131,10 +4131,22 @@ Sema::BuildExpressionFromIntegralTemplateArgument(const TemplateArgument &Arg,
   assert(Arg.getKind() == TemplateArgument::Integral &&
          "Operation is only valid for integral template arguments");
   QualType T = Arg.getIntegralType();
-  if (T->isCharType() || T->isWideCharType())
+  if (T->isAnyCharacterType()) {
+    CharacterLiteral::CharacterKind Kind;
+    if (T->isWideCharType())
+      Kind = CharacterLiteral::Wide;
+    else if (T->isChar16Type())
+      Kind = CharacterLiteral::UTF16;
+    else if (T->isChar32Type())
+      Kind = CharacterLiteral::UTF32;
+    else
+      Kind = CharacterLiteral::Ascii;
+
     return Owned(new (Context) CharacterLiteral(
-                                             Arg.getAsIntegral()->getZExtValue(),
-                                             T->isWideCharType(), T, Loc));
+                                            Arg.getAsIntegral()->getZExtValue(),
+                                            Kind, T, Loc));
+  }
+
   if (T->isBooleanType())
     return Owned(new (Context) CXXBoolLiteralExpr(
                                             Arg.getAsIntegral()->getBoolValue(),