On Sun, May 17, 2009 at 09:02:11AM -0700, webpost_at_tigris.org wrote:
> Hello all, I'm kinda at an impass right now trying to figure out the best way to proceed.
>
> I would like to customize some software for my own purposes that is
> hosted by google code, and I am not a developer of that software.
>
> I am trying to figure out the best way to "fork" their code and make
> my own modifications. More importantly, I would like to be able to
> sync up to the upstream code without major effort.
>
> I thought that, looking at the handbook, a vendor branch would be the
> way to go, but I only have read-only access to the repository, and it
> seems that svn import needs that.
What do you mean? svn import needs what? Write access?
I think you misunderstood -- you create your own new repository and
import their code into your own repository. That way, you don't need
write access to their repository.
> So, is there a svn way to do what I am looking for? I'd really like to
> avoid having to create a patch, revert changes, update, apply the
> patch, and reconcile every time I want to update the with the
> upstream.
Managing a vendor branch as described in the book is done to avoid
all this work, yes.
You should also take a look at svn-load, which is a replacement
for svn_load_dirs.pl: http://free.linux.hp.com/~dannf/svn-load/
Because svn_load_dirs has no license we may stop distributing it in
the future.
You could also give mercurial patch queues a look, especially if you
intend to send patches to the upstream project. It's a slightly different
approach to the vendor-branching problem:
http://hgbook.red-bean.com/read/managing-change-with-mercurial-queues.html
Stefan
Received on 2009-05-17 19:41:15 CEST