On Wed, Dec 20, 2000 at 01:52:40PM -0600, Karl Fogel wrote:
> Greg Stein <gstein@lyra.org> writes:
> > > But you
> > > don't, of course -- the target ancestry is implied by the ancestry of
> > > the parent in which the target is being replace.
> >
> > Well, I'm not quite so sure that I agree... the target's ancestry is *not*
> > implied by the parent. Consider the following operation:
> >
> > $ svn update some/dir
> > [ a time period elapses ]
> > $ svn update some/dir/foo.c
> > P some/dir/foo.c # this updates foo.c to v7; dir is ???
> >
> > Now, let's say that we attempt to copy something over the top of foo.c. If
> > we tell the server that we're replacing "foo.c, v6", then the server is
> > going to bark at us. It will complain because it doesn't know if you're
> > attempting to copy over v6 or v7 of foo.c.
>
> I see what you're saying. I'm not sure this is a problem that needs
> to be solved by changing the editor interface, though; I think there
> may be enough information there to detect this case anyway.
If you can find it, then great :-) I wasn't able to.
> My apologies for having to defer so many issues till "tomorrow" (two
> days I've done that now). There's a lot of other noise going on over
> here right now, and technically I don't even work on Wednesdays :-),
> so at a certain point I just punt until the next day, heh.
Not a problem. I thought you were going to cut out a couple hours ago :-) I
was surprised when you kept emailing. Just can't help yourself, eh? :-)
> I think we can get all this stuff resolved during the day tomororw,
> though, and get back to coding. Ben and I are trying to organize
> these various threads into a list of specific issues and tackle them
> one by one.
I hear ya. I've rapid-fired a number of items that came up, which then blew
up into a several more issues... I've tried to revise the subject line for
the emails to create thread specific to each problem at hand. (I find if
several issues are in a single mail, then one gets lost if/when somebody
concentrates on 1 of N issues).
> > And a separate topic/point: just what is the version of the directory in the
> > above case? I'd think it is somewhere between v6 and v7. Note that it isn't
> > v6 because foo.c is at v7. And it isn't v7 because we only fetched foo.c
> > (there may be other v7 changes in the directory that we *didn't* fetch).
>
> Yeah, that's probably a meaningless question :-). The directory
> records its version as being v6 until you do "update .", but you could
> end up with every file in the directory being at some version other
> than v6 before that happens.
Okay. Just wanted to verify what I was thinking there. Also, we may need to
take this into account during various operations.
Cheers,
-g
--
Greg Stein, http://www.lyra.org/
Received on Sat Oct 21 14:36:17 2006