This is the mail archive of the
cygwin-developers
mailing list for the Cygwin project.
Re: 1.7.2 release?
- From: Andy Koppe <andy dot koppe at gmail dot com>
- To: cygwin-developers at cygwin dot com
- Date: Tue, 26 Jan 2010 18:56:46 +0000
- Subject: Re: 1.7.2 release?
- References: <20100126180213.GZ2402@calimero.vinschen.de>
2010/1/26 Corinna Vinschen:
> - Discussion: Linux or Windows default charsets as Cygwin default charsets
> Âin locales without explicit charset?
>
> ÂMy POV: I'm interested in POSIX/Linux interoperability in the first
> Âplace. ÂNative Windows interop is nice, but second. ÂAnd eventually
> Âthere is a workaround: ÂLANG="foo_BAR.CHARSET".
I fully agree. There's no technical requirement on the Windows side
that "foo_BAR" should mean the ANSI codepage, but there are plenty of
programs on the Unix/Linux side that expect it to imply a particular
charset. So this change should avoid a lot of trouble long-term.
Short-term, it's unfortunate for anyone who has already set their
locale with the ANSI assumption. Hopefully though, there are
relatively few people that have done that, what with UTF-8 being the
default. Also, in most locales the Linux charset is sufficiently
similar to the ANSI codepage that people won't even notice the
difference. And where it does make a big difference, such as in ja_JP,
things don't work too well currently anyway. Finally, as you say, the
answer to any complaints is straightforward.
Andy