AaronBallman wrote:
> > One question I have is whether we can also add a helper function for
> > emitting these diagnostics.
>
> I was thinking about that too. My first idea was that if we can rely on the
> order of records in tablegen influencing the order of diagnostic IDs, then
> one idea is to have some `CompatDiag()` function that can be called as
>
> ```c++
> CompatDiag(loc, getLangOpts().CPlusPlusXY, diag::compat_cxx_xy_foo)
> ```
>
> where `CompatDiag()` is basically defined as
>
> ```c++
> auto Sema::CompatDiag(SourceLocation Loc, bool Std, unsigned DiagId) {
> return Diag(Loc, Std ? DiagId : DiagId + 1);
> ```
>
> but I’m not sure we can rely on that always doing the right thing.
Yeah, I'm not certain that will always work (but it could be fine). I was
thinking more along the lines of having tablegen emit something to help with it.
> > the helper function handles translating that to the getLangOpts() dance
> > with the correct diagnostic.
>
> That would be a bit more involved I think as we’d have to actually modify the
> diagnostics generator, but I think it _should_ be possible; I haven’t looked
> into that yet though.
+1 but it could be a follow-up.
https://github.com/llvm/llvm-project/pull/132129
_______________________________________________
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
