- From: Jim Amsden <jamsden@us.ibm.com>
- Date: Sun, 17 Dec 2000 10:21:45 -0500
- To: ietf-dav-versioning@w3.org
- Message-ID: <OFEF00E735.E6BD8F4D-ON852569B8.0052E007@raleigh.ibm.com>
Greg, Glad to see we're hitting some our your issues. A third model: the client does a CHECKOUT to get a working resource, then does a bunch of PUT/PROPPATCH to that working resource. When the client is done, it does a CHECKIN. <jra>This was the model Tim was referring to, he just left out the checkout. The checkout can be done either when the client gets the resource to indicate an intent to change and/or reserve the resource, or when ready to do the updates on the server. The protocol can support either policy.</jra> >... > Suggested that there be no branching in core, no labels in core, but that > there should be a version history available in core. Um. Why? At the moment, I'm not seeing a need for these in Subversion. I'd like to understand the need for these in the core. <jra>The purpose of the version history resource is just to provide a place to get information about the versioned resource, and to provide a resource the server can iterate over in a report/list. This is in case there are no version-controlled resources on a version and it becomes "lost".</jra> Couldn't you do this with a property report, fetching the DAV:activity-set for each version resource identified by DAV:version-set of the version history resource? <jra>We could, but this is likely to be a common operation, especially just before merging an activity into a a workspace.</jra>
Received on Sunday, 17 December 2000 10:21:55 UTC