On Mon, Aug 02, 2010 at 02:18:45AM +0300, Daniel Shahaf wrote:
Why not
err2 = svn_dirent_get_absolute();
if (err2)
return svn_error_compose_create(err2, err);
?
Oh, right! Thanks for the cluestick :)
On Mon, Aug 02, 2010 at 10:10:18AM -0500, Peter Samuelson wrote:
[s...@apache.org]
+ switch (kind)
+{
+case svn_node_none:
+ SVN_ERR(write_config(fs, pool));
+ break;
+case svn_node_dir:
+ return svn_error_return(svn_error_createf(SVN_ERR_FS_GENERAL,
-Original Message-
From: s...@apache.org [mailto:s...@apache.org]
Sent: vrijdag 30 juli 2010 16:54
To: comm...@subversion.apache.org
Subject: svn commit: r980811 - in
/subversion/trunk/subversion/libsvn_fs_fs: fs.h fs_fs.c
Author: stsp
Date: Fri Jul 30 14:54:26 2010
New
On Sun, Aug 01, 2010 at 10:39:04AM +0200, Bert Huijben wrote:
+ if (APR_STATUS_IS_ENOENT(err-apr_err))
+{
+ /* 1.6.0 to 1.6.11 did not copy the configuration file
during
+ * hotcopy. So if we're hotcopying a repository which
has been
+
-Original Message-
From: 'Stefan Sperling' [mailto:s...@elego.de]
Sent: zondag 1 augustus 2010 12:48
To: Bert Huijben
Cc: dev@subversion.apache.org
Subject: Re: svn commit: r980811 - in
/subversion/trunk/subversion/libsvn_fs_fs: fs.h fs_fs.c
On Sun, Aug 01, 2010 at 10:39:04AM
-Original Message-
From: s...@apache.org [mailto:s...@apache.org]
Sent: vrijdag 30 juli 2010 16:54
To: comm...@subversion.apache.org
Subject: svn commit: r980811 - in
/subversion/trunk/subversion/libsvn_fs_fs: fs.h fs_fs.c
Author: stsp
Date: Fri Jul 30 14:54:26 2010
New
On Sun, Aug 01, 2010 at 03:50:49PM +0200, Bert Huijben wrote:
Yes, her you do, but before that there are a few SVN_ERR() statements that
can return WITHOUT wrapping the error.
+ SVN_ERR(svn_dirent_get_absolute(src_abspath, src_path,
pool));
+
On Sun, Aug 01, 2010 at 03:58:14PM +0200, Bert Huijben wrote:
I don't think we should remove the '/db' part here.
At the repos layer the '/db' doesn't make sense, but this is an
implementation on the 'fs' layer and there the 'db' does make sense.
(And if we decide to remove /db here, we
-Original Message-
From: 'Stefan Sperling' [mailto:s...@elego.de]
Sent: zondag 1 augustus 2010 22:50
To: Bert Huijben
Cc: dev@subversion.apache.org
Subject: Re: svn commit: r980811 - in
/subversion/trunk/subversion/libsvn_fs_fs: fs.h fs_fs.c
On Sun, Aug 01, 2010 at 03:50:49PM
On Sun, Aug 01, 2010 at 10:56:47PM +0200, Bert Huijben wrote:
Looking a bit further: Do we really need to take the absolute path for the
error message? (What do we in similar code paths?)
I would like Subversion to print helpful error message.
Though the path need not be absolute, but see
-Original Message-
From: 'Stefan Sperling' [mailto:s...@elego.de]
Sent: zondag 1 augustus 2010 23:09
To: Bert Huijben
Cc: dev@subversion.apache.org
Subject: Re: svn commit: r980811 - in
/subversion/trunk/subversion/libsvn_fs_fs: fs.h fs_fs.c
On Sun, Aug 01, 2010 at 10:56:47PM
On Sun, Aug 01, 2010 at 11:21:02PM +0200, Bert Huijben wrote:
But you've also said that you didn't want the dirname() hack to strip
off the trailing 'db'.
I'm not against doing this.. I just think we should fix the other messages
too, if we decide this is the right thing to do.
Where
-Original Message-
From: 'Stefan Sperling' [mailto:s...@elego.de]
Sent: zondag 1 augustus 2010 23:09
To: Bert Huijben
Cc: dev@subversion.apache.org
Subject: Re: svn commit: r980811 - in
/subversion/trunk/subversion/libsvn_fs_fs: fs.h fs_fs.c
On Sun, Aug 01, 2010 at 10:56:47PM
'Stefan Sperling' wrote on Sun, Aug 01, 2010 at 22:50:25 +0200:
On Sun, Aug 01, 2010 at 03:50:49PM +0200, Bert Huijben wrote:
Yes, her you do, but before that there are a few SVN_ERR() statements that
can return WITHOUT wrapping the error.
+
14 matches
Mail list logo
