HazardyKnusperkeks added a comment. In D131978#3729644 <https://reviews.llvm.org/D131978#3729644>, @rymiel wrote:
> It seems that even with this patch, there is seemingly weird formatting with > the negation in requires clauses, such as: > > template <typename T> > requires !F<T> > int bar(T t); > > This is because the "fake parens" of the unary expression opened is never > closed by the special case clean-up made for requires clauses > However, it turns out this isn't even valid syntax, as it requires > parentheses for disambiguation. Adding those brings back correct indentation > > template <typename T> > requires(!F<T>) > int bar(T t); > > Is it okay to ignore this case, as it wouldn't even compile? Yeah it is okay when invalid code is not formatted "nicely". It wouldn't hurt if we do, but the time can be invested better. We'll give the others some time to react, I will (most likely) push the commit on the weekend. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D131978/new/ https://reviews.llvm.org/D131978 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
