[svn.haxx.se] · SVN Dev · SVN Users · SVN Org · TSVN Dev · TSVN Users · Subclipse Dev · Subclipse Users · this month's index

Re: svn: Can't convert string from native encoding to 'UTF-8'

From: Peter Samuelson <peter_at_p12n.org>
Date: 2006-05-02 20:59:25 CEST

[Molle Bestefich]
> # locale
> LANG=
> LC_CTYPE="POSIX"
> LC_NUMERIC="POSIX"
> LC_TIME="POSIX"
> LC_COLLATE="POSIX"
> LC_MONETARY="POSIX"
> LC_MESSAGES="POSIX"
> LC_PAPER="POSIX"
> LC_NAME="POSIX"
> LC_ADDRESS="POSIX"
> LC_TELEPHONE="POSIX"
> LC_MEASUREMENT="POSIX"
> LC_IDENTIFICATION="POSIX"
> LC_ALL=

Yeah, as others have said, this means only ASCII characters are
considered valid, and subversion has no way of knowing what any 8-bit
characters are supposed to represent.

> (Does this mean that if I decide that names in my filesystem are
> Russian, then I'll _have to_ watch Russian error messages from
> Subversion?)

No. That is why the locale has so many LC_* variables. LC_MESSAGES
affects your error messages, LC_CTYPE is used for character encodings.
So you can just set the environment variable LC_CTYPE=ru_RU.KOI8-R but
_not_ set any of the others, and they will continue to default to
"POSIX", aka English with ASCII.

In order to use a particular locale, you have to make sure it is
available on your system. Type 'locale -a' to see a list of these. To
add a locale to your system, use the 'localedef' program, something
like 'localedef -f KOI8-R -i ru_RU ru_RU.KOI8-R'. Or it may be that
gentoo has a frontend like debian's "dpkg-reconfigure locales", which
shows a long menu of possible locales and lets you pick the ones you
want to have available, and which should be the default.

Received on Tue May 2 20:59:54 2006

This is an archived mail posted to the Subversion Dev mailing list.