gamesh411 wrote: @steakhal Thanks for the feedback on the intention behind using this trait. From this perspective, I see merit in being as conservative as possible with the types we allow in the ProgramState.
> If the issue was that the warning was cryptic, we can continue the discussion > improving the diagnostic for breaking the build in case the trait isn't > specialized for your type. For this case, I can envision a default body for > the trait with a single `static_assert(false)`, and a message stating what to > do in this case. To be honest, my main concern was that my LSP gave me a warning that "polluted" the workspace with an annoying, not trivially actionable warning. I still have to check if this is because of how I use my types with this type trait (aka my fault 😝 ) or is it a consequence of maybe Clangd not being able to pick up the fact that there are these specializations along the way to my eventual usage of them. I'll check that soon and report back. https://github.com/llvm/llvm-project/pull/98150 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
