This meeting will start in ~5 minutes!

Tom.

SG16 will hold a meeting today/tomorrow, Wednesday, May 28th, at 19:30 UTC (timezone conversion).

If you need a .ics file to import into your calendar, you can download it here.

The agenda follows.

• P3671R0: Clarifying the interaction of the literal and execution encodings.
• P3677R0: Preserving LC_CTYPE at program start for UTF-8 locales.
  

This will be yet another celebration of Corentin and his prolific paper writing abilities!

P3671R0 seeks to make the C++ standard more explicit with regard to expected behavior (or lack thereof) when either the ordinary/wide literal encoding or the corresponding encoding of the execution (wide-)character set includes characters in its repertoire that are either not representable or are represented differently in the other encoding. The proposal adds non-normative notes to the standard that clarify that mojibake may result in unexpected/undefined behavior.

P3677R0 targets both C (WG14) and C++ (WG21). It seeks to do for C and C++ what PEP-538 did for Python; to preserve the environment locale encoding when it is UTF-8 while also preserving the historical behavior that C and C++ programs start as if an implicit call to setlocale(LC_ALL, "C") was made. Basically, it says that, on a system that has a "C.utf8" locale, if the environment locale the program is started in uses UTF-8 (e.g., "en_US.utf8"), then the behavior is as if an implicit call to setlocale(LC_ALL, "C.utf8") was performed instead.

Tom.