A. Jiang

The issue seems previously discovered in #1901.

IMHO tags can be helpful when analyzing errors reported with Godbolt.

Remaining unqualified `_Construct_in_place` and `_Destroy_in_place` calls after #4430 are all in containers and `valarray`.

The associated constraints of `_Type_eraser` are being removed by #4133. I think we can investigate the errors after merging that PR.

The remaining issue (without my mistake in and accidental fix) should only be about the `const` which is made required by LWG-3636. It seems that we can reform `_Formattable_with` like...

Currently `constexpr std::vector v{};` compiles in release mode (`_ITERATOR_DEBUG_LEVEL == 0`) but not in debug mode, which is surprisingly inconsitent. Such inconsistency can't be fixed as long as the proxy...

I'm afraid that we can't do this at this moment due to ABI compatibility, see #1036. (Perhaps the reason why the standard doesn't require this is keeping ABI compatibility.)

Off-topic (but related to [test failures](https://github.com/microsoft/STL/pull/3928/checks?check_run_id=21056419304)): Resolved It seems that we should use `static_cast(_Count)` here: https://github.com/microsoft/STL/blob/202e3820d6390505855ce9a58a4bcee4f28feacd/stl/inc/xutility#L4753 Also we should make the `size` variables of type `size_t` in these files: -...

> For now I'm not sure whether it is UB, or a valid case. Per [[alg.min.max]](https://eel.is/c++draft/alg.min.max), it seems that processing NaN is UB for ranges::meow_element (due to `indirect_strict_weak_order`) but not...

> > > For now I'm not sure whether it is UB, or a valid case. > > > > > > Per [[alg.min.max]](https://eel.is/c++draft/alg.min.max), it seems that processing NaN is...