Eric Niebler
Eric Niebler
> This may be evidence of a class of iterators "between" `Bidirectional` and `RandomAccess` I see it as evidence that cyclic iterators are fundamentally broken.
UPDATE from San Diego: we should also reconsider `std::ranges::reverse` and `std::ranges::reverse_copy`.
I don't agree with Casey's interpretation. Or, if that was his interpretation from the beginning, that wasn't clear to me. We should hunt down the places where it matters and...
> It seems more consistent to me to have the coroutine frame automatically destroyed once the coroutine runs to completion in all code-paths regardless of whether the coroutine is executing...
Thanks for the patch. I currently traveling to the C++ standardization committee meeting in Kona. I'll look at this as time allows.
would this solve the problem? https://godbolt.org/z/j3djoY1Ys