Ben Collins-Sussman <sussman@newton.ch.collab.net> writes:
> > > No, you include the revision number in which the copy took place in
> > > the property.
> >
> > Oh, very clever. I don't think people thought of that before.
> >
> > > (How you'd get the revision number before you do the commit, I don't
> > > know, but surely that can be addressed somehow.)
> >
> > That does sound a bit tricky. One answer is to let the fs set the
> > properties, in which case it can do whatever magic it has to. Still
> > potentially a wrench in the works, but probably not as bad of a wrench
> > as copy nodes.
>
> Oh! The RA layer is already going to store "wc props" after a commit
> succeeds (Change #3). Isn't this the exact mechanism we would need?
These are properties set by the filesystem, on nodes in the
filesystem. If you think back to the copy nodes discussion, a copy
node would be done instead by a regular node + prop.
Received on Sat Oct 21 14:36:23 2006