Endilll wrote: > @Endilll This is my attempt at implementing what you asked for in [#106321 > (comment)](https://github.com/llvm/llvm-project/pull/106321#discussion_r1734379624) > > I'm not a huge fan of this as I think this will be extremely annoying for > downstream forks with very little upside vs just moving the unscoped enum to > be earlier in the `Sema` class declaration (which creates no churn). > > However, if you and @AaronBallman want this version of the PR then we can go > with this.
Yeah, no worries, we've been doing this on-demand for some time now. The biggest downside of unscoped enums, especially nested in classes, is that they are not eligible to be forward declared without including the header they are defined in. So you're following a well-established precedent here. https://github.com/llvm/llvm-project/pull/106453 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
