On Thu, Jun 17, 2021 at 10:57 PM Peter Brett via SG16 <sg16_at_[hidden]>
wrote:

> Hi all,
>
> The current proposed resolution for LWG3565 (https://wg21.link/LWG3565)
> involves transcoding from the locale encoding to UTF-8. This makes me a
> little uncomfortable.
>

Can you clarify what makes you uncomfortable?


>
> Is it possible instead to say that, if the string literal encoding is
> UTF-8, then the effective locale is _as if_ the specified or global
> locale was modified by replacing the associated codeset with UTF-8?
>
> So, the following code:
>
> std::locale l1("Russian.1251");
> auto s = std::format(l1, "День недели: {:L}", std::chrono::Monday);
>
> Would behave as if replaced by:
>
> std::locale l1("Russian.1251");
> std::locale l2(l1, std::locale("Russian.UTF-8"), locale::time);
> auto s = std::format(l2, "День недели: {:L}", std::chrono::Monday);
>
> This would permit an implementation that has UTF-8 locale data available
> to use it directly, rather than being required to use the 1251 codeset
> locale data and transcode in order to conform to the standard.
>

"associated codeset with UTF-8" is not really a thing.
The ".UTF-8" locales merely exist by convention on some platforms

There is no spec that says that

* Russian.1251 is not UTF-8
* Russian.1251.UTF-8 exists
* Russian.1251 and Russian.1251.UTF-8 only differ by encoding if both exist

Transcoding is therefore more generally applicable.

Note that I have my own reservations about this issue, namely how much
effort are we willing to put
into mending a system that only works for a narrow subset of cultures,
languages and circumstances?
That being said, even if that issue amounts to putting duct tape over a
giant crack in the wall,
It also doesn't hurt.
It is undoubtedly more correct than the status quo and it might make the
life of our windows users a bit less painful
as a stopgap solution


>
> Peter
>
> P.S. How would one go about writing a locale object that customizes
> chrono formatting with std::format? Does anyone have a code sample?
> --
> SG16 mailing list
> SG16_at_[hidden]
> https://lists.isocpp.org/mailman/listinfo.cgi/sg16
>