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

Re: Checking Invalidity of UTF8 filename, is that work properly?

From: Ulrich Eckhardt <eckhardt_at_satorlaser.com>
Date: 2004-04-28 09:25:05 CEST

Hojin Choi wrote:
> $ svn add HHHH.txt
> svn: Can't recode string
> (above HHHH is really a korean character file name)
>
> And I checked the source code where the message is created. and it is found
> from UTF8 handling file. Maybe it is not UTF8 conforming filename.
> But that's not a utf8. It's euc-kr..
>
> What if? Can subversion accept only UTF8?
> Because of that reason, character encoding conversion must be occurred
> automatically.
> But the conversion is not smart yet, or subversion treat all non-ascii file
> name as a utf8 one.

You're really close to the solution. ;)
The point is that Subversion uses UTF-8 internally for all paths. Anything you
give it is converted to UTF-8 according to the current locale(!). Now, if
your locale says it is UTF-8 but your encoding is really EUC, this conversion
fails. Luckily, as it might as well produce garbage.

So, the only thing to do is to set the locale correctly. LC_CTYPE is the env.
variable IIRC, but check the manpages.

Uli

---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@subversion.tigris.org
For additional commands, e-mail: users-help@subversion.tigris.org
Received on Wed Apr 28 09:26:08 2004

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

This site is subject to the Apache Privacy Policy and the Apache Public Forum Archive Policy.