On 3/29/07, Rick Yorgason <rick@ldagames.com> wrote:
> Daniel Berlin wrote:
> > You only need to change one define, but it makes it incompatible with
> > subversions not compiled with that value :)
>
> I suppose I could just maintain my own builds of Subversion and
> Tortoise, but I think my team would rather put up with long update times
> and large hard drives, and *I'd* rather the problem be fixed properly,
> with the clients and servers able to handle any window size.
>
> It also sounds like there might be some benefit to researching the value
> of choosing a window size based on file-size, or some other heuristic
> (all funny quotes by Mark Jason Dominus aside).
>
> However, the only way of making this work before the next
> compatibility-breaking release (2.0?) would be to write some code to
> allow the server to translate any delta to a delta with 100Kb windows,
> which would obviously be pretty awkward and probably pretty inefficient.
> So unless I or anybody else decides to do that, then I guess there's
> quite a bit of time for this to get done.
>
> -Rick-
>
Are the delta's just processed by the server or also by the client?
If only the server
uses these deltas (and therefore cares about the window size) then
this could be handled by just bumping the repository format, right?
The gains on larger binary files seems like it could be very
significant. In my testing, I saw xdelta producing deltas several
times smaller than the corresponding subversion rev files. (I believe
that I was properly accounting for the skip-revs.)
Roy
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Fri Mar 30 08:19:46 2007