RE: Moving files within the repository

> This hash thing seems nice at first look but I don't know how this could
> solve the "move and modify" problem.

That problem exists the way things are now. What it comes down to is that by
checking hashes, we have one more way to reduce the size of the repository.
*No* functionality would be lost. And that can only be a good thing, right?

Also, if we're using hashes, we might be able to stop storing a duplicate of
each file in the .svn\text-base directory. If we use a hash that tells us
which part of the file has changed, we could request just that part of the
file from the repository, and then do the diff against that. For people with
large working copies, having each file stored in 3 places (The working copy,
the working copy's .svn\text-base directory, and in the repository) really
begins to hurt, especially when one of those could be eliminated.

Anyone think that I should ask these questions on the dev list? Maybe a
long-term feature request on the issue list? Or am I getting way ahead of
myself?

-Greg Martyn

---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@subversion.tigris.org
For additional commands, e-mail: users-help@subversion.tigris.org
Received on Sun Feb 1 23:09:14 2004