Ok, I now reuse Sema::RecordDeclSetTy. As for putting a bit into FieldDecl as opposed to the UnusedPrivateField SetVector: It seems like FieldDecl is carefully crafted for size. Just adding another bit for this use case doesn't seem right.
Please find the new version attached. On Wed, May 9, 2012 at 5:41 PM, Jordan Rose <[email protected]> wrote: > > On May 9, 2012, at 19:24, Daniel Jasper wrote: > > >> - IsRecordFullyDefined should probably use SmallPtrSetImpl, to avoid > specifying a size in the function prototype. (Not that it /really/ matters, > but...) > >> > > But all the methods in SmallPtrSetImpl are protected!? > > Oops, sorry. I didn't actually check the docs; I just assumed it was like > SmallVector and SmallVectorImpl. I suppose a typedef, then, is the next > best thing, to avoid having to specify the count in two different places.
clangpatch
Description: Binary data
_______________________________________________ cfe-commits mailing list [email protected] http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits
