ChuanqiXu added inline comments.
================ Comment at: clang/include/clang/AST/ASTContext.h:1369 } + QualType getDecayedType(QualType T, QualType Decayed) const; ---------------- mizvekov wrote: > ChuanqiXu wrote: > > mizvekov wrote: > > > ChuanqiXu wrote: > > > > Maybe we need a comment for this. The signature looks not straight > > > > forward and I can't relate this to the above comment. > > > I think I wanted to make this an internal variant, since we will probably > > > never need this outside of ASTContext anyway, but there is the friendship > > > situation between Type and ASTContext. Maybe making this private would be > > > for the best. > > Yeah, the signature confused me as well... as long as the second parameter > > is `Decayed` already, why we need to get the decayed type again? I guess > > `Underlying` may be a better name. > Ah I see, but this confusion already existed. See the definition of > DecayedType in Type.h, where you have the `QualType getDecayedType()` member. > This is what this Decayed parameter represents, what we will put in there to > be returned by that method. > > For DecayedType / AdjustedType / AttributedType there exists this confusion > where they have two child nodes, which arbitrarily picking one of them to be > called the `UnderlyingType` could be controversial. You could make the case > that it should be the same one we use for implementing `desugar`, but there > are arguments there for the other case. I see. I got your point. Thanks. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D130308/new/ https://reviews.llvm.org/D130308 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
