Re: distributed version control

[tarmigan]

On Jan 17, 2008 12:03 PM, tarmigan <address@hidden> wrote:
> On Jan 17, 2008 11:42 AM, John W. Eaton <address@hidden> wrote:
> > On 16-Jan-2008, tarmigan wrote:
> > | On Jan 16, 2008 2:33 PM, tarmigan wrote:
> > | > I don't know the size
> > | > of the repository off the top of my head, but I will check tonight.
> > |
> > | OK, I just did a git-repack to compress the repository, and the entire
> > | octave cvs history is 28M in git.  For reference the checked out files
> > | for octave are 28M and the 3.0.0.tar.gz is 10.7M.
> >
> > Can you tell me precisely how you converted the Octave CVS archive to
> > git?
>
> I used git-cvsimport, with a bunch of flags.  I don't remember the
> options off the top of my head, but I will check tonight.

I think you should be able to do it with

cd octave_git_eval
export CVS_RSH=ssh
git-cvsimport -d :ext:address@hidden:/cvs -k -u -v -m -p -Z,9 octave

for the initial import, as well as getting updates.  The first import
may take a few hours, but the updates are much faster.  I would offer
to post mine, but I did not do the best conversion, so I would not
recommend using mine as a base.

If you decide to use git for real, might make sense to spend some time
making the conversion nice, by specifiying an author file and trying
to get the merges right, for example, which I did not do.  It would be
a good idea to compress that initial conversion extra hard before you
make it public to reduce everyone's disk usage and bandwidth for the
initial import.  You can do this by running 'git-repack -a -d -f
-depth=250 --window=250'  This is how I got the 28M.

> http://www.kernel.org/pub/software/scm/git/docs/git-cvsimport.html

Thanks,
Tarmigan