On Wed, Jan 14, 2009 at 4:20 PM, Charles Brooking <[email protected]> wrote: > Alexander Klimetschek wrote: > I'm not sure how a different content model would change things - am I > missing something obvious? My intention is just to restrict WebDAV from > modifying certain properties; these are properties the user might > have "permission" to modify, but because WebDAV doesn't allow validation > it could be argued that modifications should only be allowed through an > application-specific interface (eg HTTP/HTML with form validation).
Ok, I see. There are two other validation options that will work without extending the WebDAV code: a) you can put regexp-like value constraints on properties in JCR (via node types) (see section 6.7.16 "Value Constraints" of the JCR spec [1] and node type definition in Jackrabbit [2]), albeit they are quite limited b) use JCR observation listeners to check for validity (but this happens asynchronously after the change was saved) [1] http://www.day.com/specs/jcr/1.0/6.7.16_Value_Constraints.html [2] http://jackrabbit.apache.org/node-types.html > I guess this is life: I spent enough time studying JCR and Jackrabbit, > I already had opinions and knowledge about the REST interface I wanted, > and there didn't seem to be much material for people outside the project > to understand Sling. So I dived-in with plain JAX-RS. > > It has been a good experience so far: JAX-RS annotations mean not too > much code and Jersey supports JSP for view templates. Although maybe it > could be better with Sling - I guess I'll find out when I have time to > take another look at the project. Never mind, that's a good approach, too. Just wanted to let you know about Sling in case you didn't see it yet. And IMO it's really cool ;-) Regards, Alex -- Alexander Klimetschek [email protected]
