Although the gitosis docs give little info about the back-end mechanisms, I think gitosis is doing some of what I'm babbling about when I say things like "bare repo" and "push isn't the opposite of pull".
If you push directly to a central repo, you can confuse yourself and the repo pretty quickly. Very bad: changes can't quite take, but don't really go away either. The glylib repos used for push and pull are bare repos, only encrypted content. No one ever touches the actual repo except the admin, and it is never pushed to. I think gitosis making it look like a push to central, but it isn't directly.
I will have some minor changes to push soon...
:-) Lachele
--
B. Lachele Foley, PhD '92,'02
Assistant Research Scientist
Complex Carbohydrate Research Center, UGA
706-542-0263
lfoley.ccrc.uga.edu
----- Original Message -----
From: Lachele Foley
[mailto:lfoley.ccrc.uga.edu]
To: AMBER Developers Mailing List
[mailto:amber-developers.ambermd.org]
Sent: Tue, 11 May 2010 18:43:33
-0400
Subject: Re: [AMBER-Developers] New Patch Procedures...
> PS: Not sure if this is in your thoughts already, but... I think you can
> use a sort of branching and have different people/groups push to different
> branches. Then, you can pull to the main repo from each branch separately.
> That might be a big hassle or the best thing ever, and I have no idea which
> because I'm as new to this as all of you, if not newer. I have in mind
> something like the working from home sync. Here's a page on that.
>
> https://git.wiki.kernel.org/index.php/GitFaq#How_would_I_use_.22git_push.22_to_sync_out_of_a_host_that_I_cannot_pull_from.3F
>
> ...of course, this doesn't use bare clones in the middle. Not sure what
> layer of annoyance that would add.
>
> :-) Lachele
> --
> B. Lachele Foley, PhD '92,'02
> Assistant Research Scientist
> Complex Carbohydrate Research Center, UGA
> 706-542-0263
> lfoley.ccrc.uga.edu
>
>
> ----- Original Message -----
> From: Lachele Foley
> [mailto:lfoley.ccrc.uga.edu]
> To: AMBER Developers Mailing List
> [mailto:amber-developers.ambermd.org]
> Sent: Tue, 11 May 2010 18:03:42
> -0400
> Subject: Re: [AMBER-Developers] New Patch Procedures...
>
>
> > > I don't understand. If you have "them set up separately" how did you do
> > it?
> > > How do keep them (two?) in sync with each other? What makes it "safer
> and
> > > saner"? Basically, I don't understand how you have things set up for
> > > glylib.
> >
> > I sort-of copied the idea from here:
> >
> >
> http://www.kernel.org/pub/software/scm/git/docs/user-manual.html#public-repositories
> >
> > See the diagram just a short scroll down.
> >
> > The only difference is that instead of "your public repo" I have one
> central
> > "deposit repo," and I pull from there to the main repo. I chose the
> slight
> > alteration because not everyone has a computer that has public access from
> > which I can pull -- and to limit the number of places I had to pull from.
> > Our web designer set up the web site differently, so it's not the only
> way.
> >
> > Honestly, when I set it up I was thoroughly confused and groping around,
> but
> > what that means is that if something really screwy gets pushed to the
> > "depository" (as in makes the repo behave badly -- not just a coding
> error),
> > people can still clone from the public repo without cloning the screwy
> bits.
> >
> > If curious, here's our docs:
> > http://128.192.9.183/pmwiki/pmwiki.php/Main/GIT
> > (the IP address not for public sharing, please)
> > Also, read those docs with the thought "produced on a crisis-to-crisis
> > basis" in mind.
> > Feel free to steal/share any of the text/docs if you like them.
> >
> > > Both the "master" branch and the "amber11" branch are definitely there,
> > but
> > > maybe there is some additional incantation needed to get them both --
> does
> > > "git clone" only get the master branch? Can any git guru help Jason out
> > here?
> >
> > I am *NOT* a guru, but... Are they different? I think you have to pull
> > from both, and say where they're being pulled to, and then merge them
> > together if you pull to different places. But, I can't recall off the top
> > of my head.
> >
> > :-) Lachele
> > --
> > B. Lachele Foley, PhD '92,'02
> > Assistant Research Scientist
> > Complex Carbohydrate Research Center, UGA
> > 706-542-0263
> > lfoley.ccrc.uga.edu
> >
> >
> > ----- Original Message -----
> > From: case
> > [mailto:case.biomaps.rutgers.edu]
> > To: AMBER Developers Mailing List
> > [mailto:amber-developers.ambermd.org]
> > Sent: Tue, 11 May 2010 17:43:08
> > -0400
> > Subject: Re: [AMBER-Developers] New Patch Procedures...
> >
> >
> > > On Tue, May 11, 2010, Lachele Foley wrote:
> > >
> > > > > How will we know when commit access is open? (Or, rather, I should
> > say
> > > 'push' access).
> > >
> > > Email.
> > >
> > > >
> > > > heh... push is not the opposite of pull. (learned the hard way trying
> to
> > > > work from home)
> > > >
> > > > Will we push and pull from the same bare repo? I have them set up
> > > > separately for glylib. It seemed saner and safer that way, but YMMV.
> > >
> > > I don't understand. If you have "them set up separately" how did you do
> > it?
> > > How do keep them (two?) in sync with each other? What makes it "safer
> and
> > > saner"? Basically, I don't understand how you have things set up for
> > > glylib.
> > >
> > > ....dac
> > >
> > >
> > > _______________________________________________
> > > AMBER-Developers mailing list
> > > AMBER-Developers.ambermd.org
> > > http://lists.ambermd.org/mailman/listinfo/amber-developers
> > >
> >
> > _______________________________________________
> > AMBER-Developers mailing list
> > AMBER-Developers.ambermd.org
> > http://lists.ambermd.org/mailman/listinfo/amber-developers
> >
>
> _______________________________________________
> AMBER-Developers mailing list
> AMBER-Developers.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber-developers
>
_______________________________________________
AMBER-Developers mailing list
AMBER-Developers.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber-developers
Received on Wed May 12 2010 - 06:30:03 PDT