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

Re: [PATCH] readable error messages on non-ASCII systems

From: Daniel Shahaf <d.s_at_daniel.shahaf.name>
Date: Wed, 12 May 2010 07:59:07 +0300 (Jerusalem Daylight Time)

Greg Ames wrote on Tue, 11 May 2010 at 19:36 -0400:
> The error messages are in the native code page to start with, so running
> them through a UTF-8 -> native conversion doesn't do anything helpful.
>
...
> Index: subversion/libsvn_subr/cmdline.c
> ===================================================================
> --- subversion/libsvn_subr/cmdline.c (revision 943316)
> +++ subversion/libsvn_subr/cmdline.c (working copy)
> @@ -318,24 +318,15 @@
> svn_error_t *
> svn_cmdline_fputs(const char *string, FILE* stream, apr_pool_t *pool)
> {
> - svn_error_t *err;
> - const char *out;
> + /* "string" is native. do not try to convert from UTF-8 */

The doc string of this function (see subversion/include/svn_cmdline.h)
specifically promises that it'll do conversion from UTF-8. We cannot make
it unconditionally do the opposite.

(Perhaps with suitable #ifdef's we could do it; or perhaps your problem
can be fixed elsewhere (e.g., the error-printing code).)

Is your issue only with the encoding of error messages? Or with the the
encoding of all svn output?
Received on 2010-05-12 06:58:53 CEST

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