You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@subversion.apache.org by Alexis Huxley <ah...@gmx.net> on 2002/07/15 19:53:54 UTC
commit of out-of-date WC with mod'd dir props fails - why?
Hi
(I check the issues, and #561 seems sort of the opposite of this,
but definitely different.)
I'm keeping some web pages under SVN. I've a tree like this:
/repository ____ hiking _____ wornersattel ____ images
\ \____ wallberg ________ images
\
\__ software __...
I check out 'wornersattel' and leave it to get older.
Meanwhile I check out 'software' into a totally separate WC directory,
and make some modifications and commit them.
Then I go back to 'wornersattel' and make some edits and some
'svn:ignore' property changes and then commit them. I get this:
svn_error: #21047 : <Working copy is not up-to-date>
Commit failed (details follow):
svn_error: #21047 : <Working copy is not up-to-date>
Cannot commit propchanges for directory '/amd/dione/diskb/home/alexis/doc/html/alexis-hp/UNSTABLE/hiking/wornersattel/images'
Yes, it is right; the WC is *not* up to date, but it is not up to
date in the same way that any simple edit in a sibling directory
would create out-of-dated-ness, but in *that* situation, a commit is
still possible.
I guess there is some logic to this, but I cannot see it. Could
someone explain please? Thanks.
Alexis
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Re: commit of out-of-date WC with mod'd dir props fails - why?
Posted by Ben Collins-Sussman <su...@collab.net>.
Alexis Huxley <ah...@gmx.net> writes:
> > > I guess there is some logic to this, but I cannot see it. Could
> > > someone explain please? Thanks.
> >
> > Propchanges on directories can only be committed if the directory is
> > at HEAD. This is currently our "one inflexibility" in the way we
> > allow mixed-revision working copies.
> >
> > For an explanation, read /doc/user/manual/dirversioning.texi. In
> > particular, read the "overeager directory" problem.
>
> Ah ... Damn ... I should start searching closer to home, thanks
> Ben.
>
> Out of curiousity, is this likely to be the way it stays, or
> is a solution - even if a very long way off - likely to be implemented?
Well, it's a complicated philosophical issue. I'm sure this isn't
permanent. I just don't know what the answer is yet. :-)
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Re: commit of out-of-date WC with mod'd dir props fails - why?
Posted by Alexis Huxley <ah...@gmx.net>.
> > I guess there is some logic to this, but I cannot see it. Could
> > someone explain please? Thanks.
>
> Propchanges on directories can only be committed if the directory is
> at HEAD. This is currently our "one inflexibility" in the way we
> allow mixed-revision working copies.
>
> For an explanation, read /doc/user/manual/dirversioning.texi. In
> particular, read the "overeager directory" problem.
Ah ... Damn ... I should start searching closer to home, thanks
Ben.
Out of curiousity, is this likely to be the way it stays, or
is a solution - even if a very long way off - likely to be implemented?
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Re: commit of out-of-date WC with mod'd dir props fails - why?
Posted by Ben Collins-Sussman <su...@collab.net>.
Alexis Huxley <ah...@gmx.net> writes:
> svn_error: #21047 : <Working copy is not up-to-date>
> Cannot commit propchanges for directory '/amd/dione/diskb/home/alexis/doc/html/alexis-hp/UNSTABLE/hiking/wornersattel/images'
>
> Yes, it is right; the WC is *not* up to date, but it is not up to
> date in the same way that any simple edit in a sibling directory
> would create out-of-dated-ness, but in *that* situation, a commit is
> still possible.
>
> I guess there is some logic to this, but I cannot see it. Could
> someone explain please? Thanks.
Propchanges on directories can only be committed if the directory is
at HEAD. This is currently our "one inflexibility" in the way we
allow mixed-revision working copies.
For an explanation, read /doc/user/manual/dirversioning.texi. In
particular, read the "overeager directory" problem.
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org