This is your friendly reminder that we are meeting *today*, in about 90
minutes. See you there!

Tom.

On 9/24/24 11:54 AM, Tom Honermann via SG16 wrote:
>
> SG16 will hold a meeting on Wednesday, September 25th, at 19:30 UTC
> (timezone conversion
> <https://www.timeanddate.com/worldclock/converter.html?iso=20240925T193000&p1=1440&p2=tz_pdt&p3=tz_mdt&p4=tz_cdt&p5=tz_edt&p6=tz_cest>).
>
> The agenda follows.
>
> * P2319R1: Prevent path presentation problems
> <https://wg21.link/p2319r1>.
> * P2019R7: Thread attributes <https://wg21.link/p2019r7>.
>
> P2319R0 was reviewed during the 2024-09-11 SG16 meeting, but
> discussion was not exhausted. The following polls were taken to gauge
> the emerging consensus at the time.
>
> * Poll 1: P2319R0: The string() member function of
> std::filesystem::path should be deprecated.
> o Attendance: 9
> o
> SF
> F
> N
> A
> SA
> 2
> 4
> 1
> 0
> 0
>
> o Consensus
> * Poll 2: P2319R0: The proposed system_string() member function
> should be added to std::filesystem::path.
> o Attendance: 9
> o
> SF
> F
> N
> A
> SA
> 0
> 2
> 4
> 1
> 0
>
> o No consensus
>
> Victor has since published a new P2319R1 revision that adjusts the
> proposal based on the discussion and poll results. The new revision
> proposes deprecating the string() and generic_string() member
> functions without replacement. We'll discuss the changes and consider
> whether to forward the paper or request additional changes.
>
> P2019R6 was on the agenda for the last meeting but we ran out of time
> to discuss it (again). Corentin has an updated revision anyway! The
> new revision removes previously proposed support for thread names
> expressed in char8_t. P2019 has not previously been reviewed by SG16.
> The only concern for SG16 to weigh in on is the handling of thread
> names. See the "Threads have a name" and "Constraints" sections for
> background information and the proposed wording for the
> thread::name_hint class. As proposed, a thread name may be provided as
> a NTBS in the execution encoding via std::basic_string_view<char>;
> fully portable code will be constrained to characters from the /basic
> literal character set/. No requirements for handling of transcoding
> issues or ill-formed encoded names are specified.
>
> Tom.
>
>