Re: My observations and experience with cvs2svn.py on Windows

On Tue, 2004-02-17 at 12:04, kfogel@collab.net wrote:

> Ben's description of Subversion tagging applies to the ideal case
> only: giving a tag name to one exact revision tree. Anything more
> complex than that will take up more space, proportional to the
> complexity ("mixedness") of the tagged tree.

Sure... I was just saying that my impression is that cvs2svn.py
generally creates tags that have more "mixedness" than what normally
would have resulted if svn had been used from the very beginning. I
think it's just a side-effect of the hard problem cvs2svn.py is solving
via heuristics.

>
> Bern, thanks for the excellent mail. I've saved it, along with some
> other recent cvs2svn mails. I'm finishing issue #1510 right now, but
> will come back to these issues. Mike Pilato or others may jump on
> them sooner, time permitting.

His mail is excellent cvs2svn feedback, you're right. Sorry, Bern, I
didn't mean to sound hostile.

The problem is that this is the third email I've seen where somebody
confuses the efficiency of cvs2svn.py with the efficiency of subversion
itself. It's because a hot-button issue for me. I guess I'm being
silly. :-)

---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@subversion.tigris.org
For additional commands, e-mail: users-help@subversion.tigris.org
Received on Tue Feb 17 20:17:05 2004