[clang]

- Sema::isSimpleTypeSpecifier return true for _Bool in c99 (currently returns false for _Bool, regardless of C dialect). (Fixes #72203)
- move simple type decision code into shared location (IdentifierInfo)
- replace the logic with a check for simple types and a proper check for a valid keyword in the appropriate dialect
- change all call sites to match the above new API

Author: Carl Peto

clang/include/clang/Basic/IdentifierTable.h

@@ -427,6 +427,10 @@ class alignas(IdentifierInfoAlignment) IdentifierInfo {
   /// language.
   bool isCPlusPlusKeyword(const LangOptions &LangOpts) const;
 
+  /// Return true if this token is a simple type specifier
+  /// in the specified language.
+  bool isSimpleTypeSpecifier(const LangOptions &LangOpts) const;
+
   /// Get and set FETokenInfo. The language front-end is allowed to associate
   /// arbitrary metadata with this token.
   void *getFETokenInfo() const { return FETokenInfo; }

clang/include/clang/Sema/Sema.h

@@ -2636,7 +2636,7 @@ class Sema final {
 
   void DiagnoseUseOfUnimplementedSelectors();
 
-  bool isSimpleTypeSpecifier(tok::TokenKind Kind) const;
+  bool isSimpleTypeSpecifier(const IdentifierInfo &II) const;
 
   ParsedType getTypeName(const IdentifierInfo &II, SourceLocation NameLoc,
                          Scope *S, CXXScopeSpec *SS = nullptr,

clang/lib/Basic/IdentifierTable.cpp

@@ -419,6 +419,45 @@ StringRef IdentifierInfo::deuglifiedName() const {
   return Name;
 }
 
+/// Determine whether the token kind starts a simple-type-specifier.
+bool IdentifierInfo::isSimpleTypeSpecifier(const LangOptions &LangOpts) const {
+  auto Kind = getTokenID();
+
+  switch (Kind) {
+  case tok::kw_short:
+  case tok::kw_long:
+  case tok::kw___int64:
+  case tok::kw___int128:
+  case tok::kw_signed:
+  case tok::kw_unsigned:
+  case tok::kw_void:
+  case tok::kw_char:
+  case tok::kw_int:
+  case tok::kw_half:
+  case tok::kw_float:
+  case tok::kw_double:
+  case tok::kw___bf16:
+  case tok::kw__Float16:
+  case tok::kw___float128:
+  case tok::kw_wchar_t:
+  case tok::kw_bool:
+  case tok::kw___underlying_type:
+  case tok::kw___auto_type:
+  case tok::kw__Bool:
+  case tok::annot_typename:
+  case tok::kw_char16_t:
+  case tok::kw_char32_t:
+  case tok::kw_typeof:
+  case tok::annot_decltype:
+  case tok::kw_decltype:
+  case tok::kw_char8_t:
+    return isKeyword(LangOpts);
+
+  default:
+    return false;
+  }
+}
+
 tok::PPKeywordKind IdentifierInfo::getPPKeywordID() const {
   // We use a perfect hash function here involving the length of the keyword,
   // the first and third character.  For preprocessor ID's there are no

clang/lib/Parse/ParseExpr.cpp

@@ -1597,7 +1597,7 @@ ExprResult Parser::ParseCastExpression(CastParseKind ParseKind,
       if (TryAnnotateTypeOrScopeToken())
         return ExprError();
 
-      if (!Actions.isSimpleTypeSpecifier(Tok.getKind()))
+      if (!Tok.getIdentifierInfo() || !Actions.isSimpleTypeSpecifier(*Tok.getIdentifierInfo()))
         // We are trying to parse a simple-type-specifier but might not get such
         // a token after error recovery.
         return ExprError();

clang/lib/Parse/ParseObjc.cpp

@@ -2971,7 +2971,7 @@ bool Parser::ParseObjCXXMessageReceiver(bool &IsExpr, void *&TypeOrExpr) {
                   tok::annot_cxxscope))
     TryAnnotateTypeOrScopeToken();
 
-  if (!Actions.isSimpleTypeSpecifier(Tok.getKind())) {
+  if (!Tok.getIdentifierInfo() || !Actions.isSimpleTypeSpecifier(*Tok.getIdentifierInfo())) {
     //   objc-receiver:
     //     expression
     // Make sure any typos in the receiver are corrected or diagnosed, so that

clang/lib/Sema/SemaDecl.cpp

@@ -128,52 +128,8 @@ class TypeNameValidatorCCC final : public CorrectionCandidateCallback {
 } // end anonymous namespace
 
 /// Determine whether the token kind starts a simple-type-specifier.
-bool Sema::isSimpleTypeSpecifier(tok::TokenKind Kind) const {
-  switch (Kind) {
-  // FIXME: Take into account the current language when deciding whether a
-  // token kind is a valid type specifier
-  case tok::kw_short:
-  case tok::kw_long:
-  case tok::kw___int64:
-  case tok::kw___int128:
-  case tok::kw_signed:
-  case tok::kw_unsigned:
-  case tok::kw_void:
-  case tok::kw_char:
-  case tok::kw_int:
-  case tok::kw_half:
-  case tok::kw_float:
-  case tok::kw_double:
-  case tok::kw___bf16:
-  case tok::kw__Float16:
-  case tok::kw___float128:
-  case tok::kw___ibm128:
-  case tok::kw_wchar_t:
-  case tok::kw_bool:
-  case tok::kw__Accum:
-  case tok::kw__Fract:
-  case tok::kw__Sat:
-#define TRANSFORM_TYPE_TRAIT_DEF(_, Trait) case tok::kw___##Trait:
-#include "clang/Basic/TransformTypeTraits.def"
-  case tok::kw___auto_type:
-    return true;
-
-  case tok::annot_typename:
-  case tok::kw_char16_t:
-  case tok::kw_char32_t:
-  case tok::kw_typeof:
-  case tok::annot_decltype:
-  case tok::kw_decltype:
-    return getLangOpts().CPlusPlus;
-
-  case tok::kw_char8_t:
-    return getLangOpts().Char8;
-
-  default:
-    break;
-  }
-
-  return false;
+bool Sema::isSimpleTypeSpecifier(const IdentifierInfo &II) const {
+  return II.isSimpleTypeSpecifier(getLangOpts());
 }
 
 namespace {