You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@subversion.apache.org by Greg Hudson <gh...@MIT.EDU> on 2003/08/10 20:20:39 UTC
Re: svn commit: rev 6696 - in trunk/subversion: libsvn_ra_svn
svnserve
On Sun, 2003-08-10 at 16:08, mbk@tigris.org wrote:
> - svn_boolean_t aborted;
> + svn_boolean_t aborted = TRUE;
That doesn't look right. If svn_ra_svn_drive_editor() is returning
successfully without setting aborted one way or the other, that's a bug;
we shouldn't compensate for it in the caller. (If
svn_ra_svn_drive_editor() returns an error, we never use the value of
aborted.)
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Re: svn commit: rev 6696 - in trunk/subversion: libsvn_ra_svn svnserve
Posted by mark benedetto king <mb...@lowlatency.com>.
On Sun, Aug 10, 2003 at 04:20:39PM -0400, Greg Hudson wrote:
> On Sun, 2003-08-10 at 16:08, mbk@tigris.org wrote:
> > - svn_boolean_t aborted;
> > + svn_boolean_t aborted = TRUE;
>
> That doesn't look right. If svn_ra_svn_drive_editor() is returning
> successfully without setting aborted one way or the other, that's a bug;
> we shouldn't compensate for it in the caller. (If
> svn_ra_svn_drive_editor() returns an error, we never use the value of
> aborted.)
>
Yes, I thought about that as soon as I had actually committed the
change.
The reason why svn_ra_svn_drive_editor is returning success even though
aborted hasn't been set one way or the other is that svn_ra_svn_handle_commands
is eating the error (it's a CMD_ERR thrown from ra_svn_handle_close_edit).
--ben
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org