You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@subversion.apache.org by John Szakmeister <jo...@szakmeister.net> on 2005/02/23 03:30:26 UTC
Re: svn commit: r13123 - in trunk/subversion: clients/cmdline include libsvn_client
On Tuesday 22 February 2005 18:21, sussman@tigris.org wrote:
> Author: sussman
[snip]
> - /* Note: we have to be paranoid about checking that these are
> - valid, since svn_wc_entry() doesn't fill them in if they
> - aren't in the entries file. */
> -
> - if (entry->name && strcmp (entry->name, SVN_WC_ENTRY_THIS_DIR))
> - SVN_ERR (svn_cmdline_printf (pool, _("Name: %s\n"), entry->name));
> + /* ### remove this someday: it's only here for cmdline output
> + compatibility with svn 1.1 and older. */
> + SVN_ERR (svn_cmdline_printf (pool, _("Name: %s\n"),
> + svn_path_basename(target, pool)));
This change broke switch tests 11 and 13 (at least on my build). In these
2 test cases, svn info was being run on a directory, and hence, the
entry->name was "". Therefore, the "Name:" was never printed before.
With this change, it's always printing the "Name:" line.
The attached patch works against ra_local. I haven't tested it against
the other layers.
-John
Re: svn commit: r13123 - in trunk/subversion: clients/cmdline include libsvn_client
Posted by Ben Collins-Sussman <su...@collab.net>.
On Feb 22, 2005, at 9:46 PM, John Szakmeister wrote:
>
> I was so confused by your comment... then I realized I sent the wrong
> patch! The switch patch was something I did a while ago to prevent
> switching files to dirs and vice versa.
Ha!
>
> Here's the real patch... all 1 line of it! :-)
>
Commit away.
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Re: svn commit: r13123 - in trunk/subversion: clients/cmdline include libsvn_client
Posted by John Szakmeister <jo...@szakmeister.net>.
On Tuesday 22 February 2005 22:40, Ben Collins-Sussman wrote:
> On Feb 22, 2005, at 9:30 PM, John Szakmeister wrote:
> > This change broke switch tests 11 and 13 (at least on my build). In
> > these
> > 2 test cases, svn info was being run on a directory, and hence, the
> > entry->name was "". Therefore, the "Name:" was never printed before.
> > With this change, it's always printing the "Name:" line.
> >
> > The attached patch works against ra_local. I haven't tested it
> > against the other layers.
>
> Why such a complex patch to the switch code? Why not just make the new
> 'svn info' printing code behave like the old code? Why not just have
> the code not print an empty basename?
I was so confused by your comment... then I realized I sent the wrong
patch! The switch patch was something I did a while ago to prevent
switching files to dirs and vice versa.
Here's the real patch... all 1 line of it! :-)
-John
Re: svn commit: r13123 - in trunk/subversion: clients/cmdline include libsvn_client
Posted by Ben Collins-Sussman <su...@collab.net>.
On Feb 22, 2005, at 9:30 PM, John Szakmeister wrote:
>
> This change broke switch tests 11 and 13 (at least on my build). In
> these
> 2 test cases, svn info was being run on a directory, and hence, the
> entry->name was "". Therefore, the "Name:" was never printed before.
> With this change, it's always printing the "Name:" line.
>
> The attached patch works against ra_local. I haven't tested it against
> the other layers.
Why such a complex patch to the switch code? Why not just make the new
'svn info' printing code behave like the old code? Why not just have
the code not print an empty basename?
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org