You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@subversion.apache.org by David Glasser <gl...@mit.edu> on 2006/10/27 20:57:35 UTC
Re: svn commit: r22140 - in trunk/subversion: svnsync tests/cmdline tests/cmdline/svnsync_tests_data
On 10/27/06, rooneg@tigris.org <ro...@tigris.org> wrote:
> * subversion/svnsync/main.c
> (copy_revprops): Add a new sync parameter, which makes us remove
> revprops in the destination revision that don't exist in the source
> revision.
> (do_initialize): Pass FALSE as the sync argument of copy_revprops.
> (do_synchornize): Pass TRUE as the sync argument of copy_revprops.
> (do_copy_revprops): Pass FALSE as the sync argument of copy_revprops.
Hmm. Why shouldn't do_copy_revprops delete revprops that get deleted
in the source?
(Also, typo in log message "do_synchronize".)
--dave
--
David Glasser | glasser@mit.edu | http://www.davidglasser.net/
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Re: svn commit: r22140 - in trunk/subversion: svnsync tests/cmdline tests/cmdline/svnsync_tests_data
Posted by Garrett Rooney <ro...@electricjellyfish.net>.
(Added cc of dev@subversion.t.o, I assume David wanted it to go there...)
On 10/27/06, David Glasser <gl...@mit.edu> wrote:
> On 10/27/06, Garrett Rooney <ro...@electricjellyfish.net> wrote:
> > Well, I figured that the user might have added revprops in the
> > meantime and would be annoyed if they went away, especially since
> > we've called the command 'copy-revprops', not 'sync-revprops', and
> > 'copy' doesn't imply 'deletes some stuff' to me.
>
> Hmm, good point.
>
> The situation I was envisioning is something like "somebody wrote a
> script that accidently set svn:logg on all the revisions, we synced
> it, the upstream repo went back and deleted them, we should be able to
> get those changes too".
In retrospect, I wish I'd called the damn command 'sync-revprops', so
we could avoid all of this crap, yeah ;-)
> But I guess you're thinking about "user puts extra mirror-specific
> info on the mirror".
>
> Maybe a new flag for the subcommand?
Yeah, I'd like to see that. I didn't do it as part of this change
because it wouldn't be backportable, but I wouldn't object to one
being added later.
-garrett
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Re: svn commit: r22140 - in trunk/subversion: svnsync tests/cmdline tests/cmdline/svnsync_tests_data
Posted by Garrett Rooney <ro...@electricjellyfish.net>.
On 10/27/06, David Glasser <gl...@mit.edu> wrote:
> On 10/27/06, rooneg@tigris.org <ro...@tigris.org> wrote:
> > * subversion/svnsync/main.c
> > (copy_revprops): Add a new sync parameter, which makes us remove
> > revprops in the destination revision that don't exist in the source
> > revision.
> > (do_initialize): Pass FALSE as the sync argument of copy_revprops.
> > (do_synchornize): Pass TRUE as the sync argument of copy_revprops.
> > (do_copy_revprops): Pass FALSE as the sync argument of copy_revprops.
>
> Hmm. Why shouldn't do_copy_revprops delete revprops that get deleted
> in the source?
Well, I figured that the user might have added revprops in the
meantime and would be annoyed if they went away, especially since
we've called the command 'copy-revprops', not 'sync-revprops', and
'copy' doesn't imply 'deletes some stuff' to me.
> (Also, typo in log message "do_synchronize".)
Oops. Will fix.
-garrett
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org