Re: Whither delta combiner

On Mon, 2002-02-11 at 14:44, Zack Weinberg wrote:
> I want to stomp on this idea right now: what RCS does is *wrong*. The
> tips of all the branches should be fulltext.

That's fine (well, mostly fine, anyway, until someone expects to be able
to branch large files and make small changes to them without a large
space penalty), and it appears that is what we do now. I just didn't
realize that.

> Otherwise, performance
> rapidly becomes pathological as the branch lengthens and the parent
> tip moves farther away from the branchpoint. This is a major reason
> why people hate working with branches in CVS.

With skip-deltas, performance does not become pathological; the number
of deltas required grows as lg(distance from head to branchpoint) +
lg(distance from branchpoint to tip). That shouldn't be too bad.

> I think the diagram above is confused; I'm not sure what the arrows
> mean. I would recommend deltas on branches be stored something like
> this, where <-- means "retrieve the version on the left by applying a
> delta against the version on the right":

Backwards from what I meant. "1.1 --> 1.2" means "the representation of
1.1 is a delta against the representation of 1.2" in my ASCII art.

I did mistype one "<--" arrow as "--<" by mistake.

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Sat Oct 21 14:37:06 2006