Re: 423 Locked

On Thu, 3 Feb 2000, WJCarpenter wrote:
> jamsden> We have to be careful about overloading methods to support
> jamsden> client use cases. This will cause the protocol to bloat and

I thought we were designing a protocol that is used by the client. Why
shouldn't we support it?

IMO, returning an XML body in the 423 response is a good idea. It could
have a DAV:lockdiscovery element.

> jamsden> become complex. It will also reduce interoperability and

The spec is how we fix interoperability issues.

> jamsden> create situations where one client's use cases conflict with
> jamsden> another. Some methods do return failure information that your
> jamsden> client can use.
> 
> You also have to allow for the possibility that it's more expensive
> for a server to answer questions about the details of a LOCK than it
> is to merely answer yes-or-no about a resource being LOCKed.  (E.g., a 
> fixed field in some property table points to a LOCK record stored
> elsewhere.)

While this is true, we are talking about an error case. Personally, I
don't mind taking a few more cycles in an error case. Especially if there
is a possibility that the client will be needing that information anyhow.

Cheers,
-g

-- 
Greg Stein, http://www.lyra.org/

Received on Thursday, 3 February 2000 13:43:03 UTC