Re: [musl] [PATCH 0/2] Support printing localized RADIXCHAR

[Markus Wichmann <nullplan@gmx.net>]

Am Sat, Dec 16, 2023 at 06:10:37PM -0500 schrieb Rich Felker:
> This is a topic that's been controversial. I have always been against
> having variable radix character, but I've also been seeking input from
> users who want localized output whether the lack of this functionality
> is a serious problem that needs revisiting.
>

Speaking as a German, if output is mostly translated (because
LC_MESSAGES is at the discretion of gettext) but numbers have a '.' as
radix and are output with a precision of 3, it is disorienting and
requires re-reading every time. Because in the German locale, '.' is a
valid thousands separator (space and inch mark also being acceptable),
and the thousands grouping is just 3.

And if they're output with a different precision, it is still jarring at
least.

> Last time it was discussed, I believe my position was that, if we do
> this, it needs to be a 1-bit setting, where a locale necessarily has
> either '.' or ',' as the radix. No other values actually appear in
> real-world conventions, and on other implementations such as glibc,
> the allowance for arbitrary characters allows doing some ~nasty~ stuff
> with output and input processing. For example, you could define the
> radix character to be '1' or something that makes conversions fail to
> round-trip.
>

That is reasonable. Overgeneralization usually leads to bad code or
unforseen outcomes. This is also the case for file names, with just
recently a bug report being introduced against POSIX to disallow control
characters there. (Imagine someone setting '\r' as radix point).

Ciao,
Markus