Skip site navigation (1)Skip section navigation (2)
Date:      Sun, 1 Nov 2015 17:31:31 +0100
From:      Matthias Apitz <guru@unixarea.de>
To:        Christian Weisgerber <naddy@mips.inka.de>
Cc:        freebsd-questions@freebsd.org
Subject:   Re: tr(1) and LANG=de_DE.UTF-8
Message-ID:  <20151101163131.GB1901@c720-r276659>
In-Reply-To: <slrnn3cf05.1bra.naddy@lorvorc.mips.inka.de>
References:  <20151029103133.GA16882@sh4-5.1blu.de> <slrnn3cf05.1bra.naddy@lorvorc.mips.inka.de>

next in thread | previous in thread | raw e-mail | index | archive | help
El día Sunday, November 01, 2015 a las 04:21:25PM +0000, Christian Weisgerber escribió:

> On 2015-10-29, Matthias Apitz <guru@unixarea.de> wrote:
> 
> > I know that the man page of tr(1) contains a hint about the LANG and
> > environment(7), but would not expect that this means that I can't change
> > a single byte, octal given value, only for the reason that \357 is not a valid
> > Unicode code point.
> >
> > Any ideas/comments on this?
> 
> POSIX jabbers something about "multi-character collating elements",
> which seems related, but I don't understand it.

I think, the truth is (as always) in the man page, which reads from
tr(1):

...
 In the first synopsis form, the characters in string1 are translated
 into the characters in string2 
...

i.e. it speaks about 'characters' and not 'bytes' and the char given by
me was not a valid UTF-8 character.

	matthias

-- 
Matthias Apitz, ✉ guru@unixarea.de, 🌐 http://www.unixarea.de/  ☎ +49-176-38902045



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20151101163131.GB1901>