Jonathan Wakely

I like the direction, but saying implementation-defined means implementations are required to document what strings are used in which places. How about: > where S is a NTBS indicating where...

hmm ... seems a shame to add the member just for this ... but using const_cast would be UB if we had: ``` io_context c; const auto e = c.get_executor();...

Hmm, good point. Maybe this should be discussed in LEWG not just the two of us.

I think something like this: A type `A` meets the proto-allocator requirements if `allocator_traits::rebind_alloc` meets the allocator requirements, where `U` is an object type. [Note: For example, `std::allocator` meets the...

Yes, good point.

I'll probably work on this myself, so assign it to me if you want.

It will need to call the libdfp code for printing decimal types, so can't really live inside libstdc++. All the new code would be defined as inline templates in the...

https://github.com/boostorg/chrono/blob/develop/include/boost/chrono/detail/scan_keyword.hpp has the same license info.

Before: https://godbolt.org/z/en14c9Ps9 After: https://godbolt.org/z/nxPEbfcbf

Hi, GCC dev here. This PR is just completely wrong. It's one thing to insist that users must include `Python.h` before any other headers (I don't like it, and it...