Barry Revzin

> Annoying bug to troubleshoot if your invokable is not regular. Regular is a requirement on the invocable for `transform` and `for_each`.

The specific reason for the change in behavior from `array` to `vector`: ```cpp auto in = std::array{1, 2, 3}; auto out = in | views::transform([](int i){ return std::vector{i}; } |...

> > Regular is a requirement on the invocable for `transform` and `for_each`. > > Any observable behaviour will be depended upon ;). This isn't a Hyrum's Law thing, you're...

Yeah, range-v3 doesn't implement [P2415](https://wg21.link/p2415) yet.

Just for info, [`{fmt}` does](https://github.com/fmtlib/fmt/blob/d5e9166f54ed71f92eb6ba189342f706fe97469e/include/fmt/core.h#L210-L217): ```cpp #ifndef FMT_BEGIN_NAMESPACE #define FMT_BEGIN_NAMESPACE namespace fmt { inline namespace v9 { #define FMT_END_NAMESPACE } } #endif ``` And then uses those two macros exclusively....

I guess this means I should also revisit #1472

> > and removing the partially-formed default constructors that I removed in that paper > > This will break code needlessly (IMO). I would prefer that part of the change...

Is this really a projection issue? This is just one example of part of a broader problem of passing overload sets into function templates. I feel like this either requires...

> > or a language feature to perform overload resolution during template deduction time to bind which overload you want > > Like https://wg21.link/P1167? No, that's about class templates? What...

> I meant the part titled "Deduction guides for function templates". I don't see how that's relevant either.