Created on 2022-06-30.00:00:00 last changed 41 months ago
Proposed resolution:
This wording is relative to N4910.
Modify [ranges.syn], header <ranges> synopsis, as indicated:
[…]
namespace std {
[…]
template<class I, class S, ranges::subrange_kind K>
struct tuple_element<0, ranges::subrange<I, S, K>> {
using type = remove_cv_t<I>;
};
template<class I, class S, ranges::subrange_kind K>
struct tuple_element<1, ranges::subrange<I, S, K>> {
using type = remove_cv_t<S>;
};
template<class I, class S, ranges::subrange_kind K>
struct tuple_element<0, const ranges::subrange<I, S, K>> {
using type = remove_cv_t<I>;
};
template<class I, class S, ranges::subrange_kind K>
struct tuple_element<1, const ranges::subrange<I, S, K>> {
using type = remove_cv_t<S>;
};
}
[…]
[ 2022-07-08; Reflector poll ]
Set priority to 4 after reflector poll. "This is just contrived, especially since lots of the iterator requirements on such a type involve deprecated operations."
std::ranges::subrange<int * volatile>> is weird but valid. The return type (deduced type for auto) of std::ranges::get for this type is int*, because the replacing of that (cv-unqualified) auto drops top-level cv-qualifiers. I think get is doing the right thing, and std::tuple_element_t should be consistent with the return types.
| History | |||
|---|---|---|---|
| Date | User | Action | Args |
| 2022-07-08 20:04:38 | admin | set | messages: + msg12567 |
| 2022-07-03 12:37:52 | admin | set | messages: + msg12548 |
| 2022-06-30 00:00:00 | admin | create | |