[IPython-dev] [IPython-user] Porting SVN history to Launchpad - help needed.

[Brian Granger]

One thing you might try is posting a question on the launchpad
"answers" section.  They were pretty good about answering a question
when I asked.  They might even be able to help or give good direction
- we can't be the first to have faced this.

Brian

On Sat, May 31, 2008 at 10:04 PM, Fernando Perez <fperez.net at gmail.com> wrote:
> Hi all,
>
> A couple of months ago we started the launchpad/bzr experiment, and
> that has been very successful.  I don't think there's any doubt that
> we'll stick with bzr for the time being, nobody feels like going back
> to SVN.
>
> To get things started,  Ville created a clean bzr branch without any
> SVN history, partly because the scipy server at the time kept timing
> out and the svn import into launchpad wasn't working.  Now that import
> does function, and one can check out the bzr version of SVN trunk via
>
> bzr branch lp:~vcs-imports/ipython/main/ ipython-svn
>
> The main ipython dev branch (our new 'trunk') can be obtained with
>
> bzr branch lp:ipython
>
> The problem I'm having is how to merge the latter into the former
> while preserving the full  history.  Because the bzr trunk was
> initialized without any SVN history, every attempt I've made so far
> has failed.  I've tried bzr merge and rebase with various flags,
> without any success.  I'm not even sure it's possible to do this
> cleanly, I'm afraid.
>
> The SVN trunk has several years of history that I'm not willing to
> lose (I already made the mistake of losing all  project history in the
> CVS->SVN transition, I won't repeat that).  I'd love to hear from
> anyone who might know how to make this merge, if it is possible.  If
> it can't be made with bzr merge tools, I guess we can try to produce a
> set of bundles and apply them one by one or as a lump change with one
> gigantic commit message that preserves the other log...
>
> We're probably going to lose history no matter what, because of the
> 'branch folding' problem I mentioned earlier.  The current bzr 'trunk'
> has a really funky log that shows these, and hopefully we won't make
> these mistakes with bzr in the future.  But I'd like to minimize the
> losses, so I'd love if a bzr guru amongst us knows the magic trick.
>
> Cheers,
>
> f
> _______________________________________________
> IPython-user mailing list
> IPython-user at scipy.org
> http://lists.ipython.scipy.org/mailman/listinfo/ipython-user
>