Look what I found

@error It's an utter pain in the arse when writing a compiler too, and even worse is when goto gets involved.

@mikee I actually tried it. It works just fine to my surprise. But the clangd formatter gone mad:

@error @_the_cloud Only in the original example, not in the first two. I'd set it to 3 and chuckle too. I'm an FPGA guy, so I always think about else/default.

@error gloop, arfle, barfle says it all.

@error ha! Thanks for testing!

@error it's beautiful

@error reminds me of Duff's Device for abitrary length byte copy using the switch as an index into an architecture-appropriate unrolled loop https://en.wikipedia.org/wiki/Duff%27s_device

@vaporeon_ please don't

@error I don't think I ever would have tried to pass that off, myself, but I sort of admire the fact that, even though it's language abuse, it produces a predictable outcome. C has a lot of undefined behavior, but somehow, this isn't.

I think I tried to use Duff's Device in a project once, but my colleagues objected and made me change it. Or maybe I just explained Duff's Device to them and they were horrified. Like, "I don't know if I can trust that guy anymore" horrified. It was a long time ago.

This is interesting but it's not really saving a ton of time or code space. I think Duff's Device is probably more defensible.

@error @TomDuff, call your office...

@mmu_man @error Never heard of these before; thank you!

@WesternInfidels It produces warning, so I personally won't use it. But it looks beautiful

@error reminds me of duff’s device https://en.wikipedia.org/wiki/Duff%27s_device

@ewhac @error Insane. No one in his right mind calls functions arfle and barfle.

@error time to see who at my workplace I can aggravate the most

@error yes

@error this is devious and i love it

@error musl does this in its printf implementation: https://git.musl-libc.org/cgit/musl/tree/src/stdio/vfprintf.c#n597

@error wow it does not even -Wextra the nested labels

@gloriouscow @error the important part is the boundaries tbh