This is the mail archive of the
libc-alpha@sourceware.org
mailing list for the glibc project.
Re: [PATCH] Fix Charset comment in fi_FI, fi_FI at euro
- From: keld at keldix dot com
- To: Marko Myllynen <myllynen at redhat dot com>
- Cc: GNU C Library <libc-alpha at sourceware dot org>, libc-locales at sourceware dot org
- Date: Tue, 26 Nov 2013 19:11:26 +0100
- Subject: Re: [PATCH] Fix Charset comment in fi_FI, fi_FI at euro
- Authentication-results: sourceware.org; auth=none
- References: <5294D4F3 dot 40006 at redhat dot com> <20131126172545 dot GA11818 at www5 dot open-std dot org> <5294DC9E dot 201 at redhat dot com>
Hi again
Well, libc locales are able to sort the full ucs charset, so that is not an argument to
exclude iso-8859-15. Finland uses Euro, so that excludes iso-8859-1.
I do not think we have specs that specifies use of citation marks etc, which could be
characters outside 8859-15 for Finnish.
The Charset comment is meant to give charsets for which a locale would compile
with localedef without fatal errors. It can be used to generate
all the locale/charset combinations that a locale is valid in.
Best regards
keld
On Tue, Nov 26, 2013 at 07:38:38PM +0200, Marko Myllynen wrote:
> Hi,
>
> ANGSTROM SIGN is specified in the standard to sort it alongside with å/Å.
>
> There is also a recent proposal to extend the auxiliary character set
> (that's the closest translation I came up with, it's
> "lisämerkkivalikoima" in Finnish in the proposal) with characters that
> might be needed when processing texts in the Finnish context and the
> proposal contains several non-ISO-8859-15 characters, see
> http://www.kotoistus.fi/kannanotot_html/avoimet/lisamerkisto-2013.
>
> So I think it's better not to suggest using ISO-8859-15 any more.
>
> Thanks,
>
> On 2013-11-26 19:25, keld@keldix.com wrote:
> > It is not necessary to specify UTF-8, all locales support UTF-8.
> > The question is whether the locale can be used with iso-8859-15.
> > More than one charset may be specified with the charset comment.
> > If you like you could specify both UTF-8 and iso-8859-15,
> > with utf-8 first, if that is the preference.
> >
> > What is ANGSTROM used for? As an alternative to Å in the sorting?
> > And other places?
> >
> > Best regards
> > keld
> >
> > On Tue, Nov 26, 2013 at 07:05:55PM +0200, Marko Myllynen wrote:
> >> Hi,
> >>
> >> please see the patch below to fix the Charset comment in fi_FI and
> >> fi_FI@euro. ISO-8859-1 is certainly incorrect with Euro and since SFS-EN
> >> 13710 (which we will hopefully support later) specifies a
> >> non-ISO-8859-15 character (ANGSTROM SIGN) it seems best to use UTF-8.
> >>
> >> 2013-11-26 Marko Myllynen <myllynen@redhat.com>
> >>
> >> * locales/fi_FI: Fix Charset comment.
> >> * locales/fi_FI@euro: Likewise.
> >>
> >> diff --git a/localedata/locales/fi_FI b/localedata/locales/fi_FI
> >> index f779964..6ea9301 100644
> >> --- a/localedata/locales/fi_FI
> >> +++ b/localedata/locales/fi_FI
> >> @@ -16,7 +16,7 @@ comment_char %
> >> % Date: 1996-10-15
> >> % Application: general
> >> % Users: general
> >> -% Charset: ISO-8859-1
> >> +% Charset: UTF-8
> >> % Distribution and use is free, also
> >> % for commercial purposes.
> >> %
> >> diff --git a/localedata/locales/fi_FI@euro b/localedata/locales/fi_FI@euro
> >> index 76f78ee..ef7f8cd 100644
> >> --- a/localedata/locales/fi_FI@euro
> >> +++ b/localedata/locales/fi_FI@euro
> >> @@ -8,7 +8,7 @@ comment_char %
> >> % Date: 2000-08-20
> >> % Application: general
> >> % Users: general
> >> -% Charset: ISO-8859-15
> >> +% Charset: UTF-8
> >> % Distribution and use is free, also
> >> % for commercial purposes.
> >>
> >>
> >> --
> >> Marko Myllynen
>
>
> --
> Marko Myllynen