Ben Collins-Sussman <sussman@collab.net> writes:
> On Mar 23, 2005, at 12:47 PM, Philip Martin wrote:
> >>
> >> How would you implement svn info wcpath without it?
> >
> > I guess your assuming "svn info wcpath" must display all that lock
> > info. Perhaps "svn info wcpath" should only display the lock token,
> > as the only thing cached in the wc, and there would be some flag (-u?)
> > to get more info from the repository.
>
> This is certainly a reasonable position to take...
... and one I've advocated myself.
> ... but I think we were following a precedent. The entries file
> already caches repository data (last-changed-rev, last-author,
> last-changed-date), thus making 'svn info wc' and 'svn info URL'
> show nearly identical output. So it seemed plausible to cache extra
> lock fields as well, so that 'svn info wc' and 'svn info URL' would
> continue to show the same output.
As I pointed out to Ben just yesterday, there are problems with even
having last-changed-rev, last-author, and last-changed-date cached in
the working copy. I mean, the latter two of those things are
completely mutable as properties of the first.
This continues to be a concern for locking stuffs, too, where a
working copy might claim that a lock is held on a given resource, but
in truth, that lock could have been broken a week ago and the working
copy just be out of date.
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Fri Mar 25 04:09:12 2005