This meeting is starting now-ish.

Tom.

On 10/22/25 12:26 AM, Tom Honermann via SG16 wrote:

SG16 will hold a meeting today, Wednesday, October 22nd, at 19:30 UTC (timezone conversion).

If you need a .ics file to import into your calendar, well, you're on your own as I am no longer able to provide a public URL from which to download one.

For those that subscribe to the WG21 shared calendar, please note that public subscriptions are no longer available and you likely need to re-subscribe to a new CalDAV URL. See the recent email sent to the "all" mailing list or this wiki page.

Now for the good stuff. The agenda for this meeting includes, you guessed it, more NB comment processing!

US 5-018 argues for adopting P3657 (A Grammar for Whitespace Characters) for which a fresh R1 revision is hot off the press. The paper offers no functional changes; its goal is to provide a more rigorous definition of what constitutes whitespace (e.g., comments) and whitespace characters (space, tab, form feed, etc...) and how they are treated during lexing. In doing so, the intent is to provide a more solid foundation for future work along the lines of Corentin's P2348 (Whitespaces Wording Revamp).

We briefly began discussion on US 189-304 during our last meeting. This comment asserts that the system_encoded_string() member function ([fs.path.native.obs]) and its generic sibling added to std::filesystem::path by P2319R5 (Prevent path presentation problems) are misnamed because 1) the C++ standard does not have a definition for "system encoding", and 2) colloquial use of that term would lead readers to an unintended encoding. On Windows, according to existing implementations, the intended encoding is determined by querying the AreFileApisANSI() Win32 function. Or maybe not. At least one implementation takes the choose your own adventure route and yields a different answer if the current thread has a thread-local locale and that locale uses a UTF-8 encoding. The proposed change includes a few options to consider. No matter our preferences, we should at least offer a recommendation to do something about the mysterious note that is [fs.path.type.cvt]p(2.1).

Tom.