Re: ra_replay: Uncovering a latent design failure

On 01/11/2013 02:23 PM, C. Michael Pilato wrote:
> Actually, I think I'll take a slightly differently approach here. Rather
> than have the client hit the "me resource" (which represents the whole
> repository), I'll have it hit the revision resource associated with the
> single revision that's being replayed.
>
> .../!svn/rev/REVNUM
>
> That way, requests to replay revisions which don't exist will 404 (as we'd
> expect), plus we won't have to marshal the revision number in the request
> body -- just the filtering path. (In fact, in the future, we may decide to
> allow multiple filtering paths here.)

Fixed in r1432357.

-- 
C. Michael Pilato <cmpilato_at_collab.net>
CollabNet   <>   www.collab.net   <>   Enterprise Cloud Development