Date: Wed, 09 Nov 2005 10:01:50 -0600 (CST) From: Sergey Babkin <babkin@verizon.net> To: John Baldwin <jhb@freebsd.org>, Pawel Jakub Dawidek <pjd@freebsd.org> Cc: Craig Rodrigues <rodrigc@freebsd.org>, cvs-src@freebsd.org, Poul-Henning Kamp <phk@phk.freebsd.dk>, src-committers@freebsd.org, cvs-all@freebsd.org Subject: Re: Re: cvs commit: src/sys/kern vfs_mount.c Message-ID: <16835761.1131552110894.JavaMail.root@vms074.mailsrvcs.net>
next in thread | raw e-mail | index | archive | help
>From: John Baldwin <jhb@freebsd.org> >On Wednesday 09 November 2005 09:30 am, Pawel Jakub Dawidek wrote: >> On Wed, Nov 09, 2005 at 03:08:39PM +0100, Poul-Henning Kamp wrote: >> +> In message <20051109140419.GD15580@garage.freebsd.pl>, Pawel Jakub >> Dawidek writes: +> >> +> >> +> >In current world order you can decide based on LC_ALL/LANG, etc. >> +> >> +> Read my lips: printf(9) and panic(9) are _never_ going to respect >> +> any environment variable. >> >> Printf(9) and panic(9) are for administrator, not for user. >> Anyway, let's stop here, as we're going nowhere. > >If you are going to provide l10n of these messages, you'll need to do it in >userland somehow. However, if you return strings that task becomes a bit >harder. I still prefer returning error data, not presentation, and letting >something else (userland) take care of formatting error data and presenting >it to the user. Maybe it would make sense to split the error code and any extra data related to the error (for example, the pathname that caused it or any extra explanation). The error code would go into errno while only the extra data would go into the string error. This extra data won't need to be translaed anyway. -SB
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?16835761.1131552110894.JavaMail.root>