Re: Bug: svn-1.0.4 stops at 65536 revisions

On Wed, Jun 09, 2004 at 09:56:49AM -0500, Wadsworth, Eric (Contractor) wrote:
> > > Because we make use of negative values to indicate throughout the
> > > codebase that the revision number is not valid or that a revision
> > > number argument is not applicable.
>
> Just a thought: You could define invalid to be zero, and start the
> repository with revision 1 instead of zero. For us C coders, it makes sense
> that the first item be number zero, but the rest of the world starts with 1.
> Then you get 4 billion revisions instead of 2 billion.
>
> Like it matters anyway. ;)

It's a little bit late for this now... To change this would cause us to
be really incompatable with existing repos and it's of a limited value.
It gets us more revisions, but we'd already decided 2^31 is more than
enough. By the time anyone has 2^31 revisions, they'll probably be
ready to move to to 64-bit archs, in which case they'll have 2^63. :)

-- 
Ben Reser <ben@reser.org>
http://ben.reser.org
"Conscience is the inner voice which warns us somebody may be looking."
- H.L. Mencken
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org