- From: Raúl García Castro <rgarcia@fi.upm.es>
- Date: Fri, 25 Jan 2013 16:22:53 +0100
- To: Roger Menday <roger.menday@uk.fujitsu.com>
- CC: Steve Speicher <sspeiche@gmail.com>, Nandana Mihindukulasooriya <nmihindu@fi.upm.es>, "public-ldp-wg@w3.org" <public-ldp-wg@w3.org>
El 24/01/13 16:53, Roger Menday escribió: > >>>> For what it's worth, section 5.2.1 of the LDP spec [2] states that "A >>>> Linked Data Platform Container must also be a conformant Linked Data >>>> Platform Resource." I've always read that as meaning that an LDPC is an >>>> LDPR. >>> >>> >>> In the data model, >>> >>> LDPR: >>> - has a RDF representation >>> >>> LDPC: >>> - has a RDF representation >>> - has a set of reserved properties with their semantics defined by the >>> protocol >>> - contains some protocol data >>> >>> so LDPC is a specialization of LDPR. >>> >>> In the interaction model, >>> >>> GET: >>> LDPR - returns the current state. >>> LDPC - returns the current state. In addition, provides mechanisms to >>> retrieve only part of the state (non-member properties) and provides >>> additional features like paging, ordering based a special property >>> (membership predicate). >>> >>> PUT: >>> LDPR - updates the current state >>> LDPC - Only part of the state may be updated via >>> <containerURL>?non-member-properties. The rest of the state is managed by >>> the server. >>> >>> POST: >>> LDPR - updates it's state by appending new triples ? >>> LDPC - creates new resources >> >> and adds it to the membership >> >>> >>> DELETE: >>> LDPR - deletes itself >>> LDPC - deletes itself and any resources contained by it >>> >>> LDPC and LDPR have different interaction models but I suppose a >>> specialization can have a different interaction model. >>> >>> Best Regards, >>> Nandana >> >> This is simple and clear to me, then again it is the current state of >> the spec (mostly). Nice summary > > It could be that I am reading the spec wrong, but, my understanding is that when a LDPR has only one LDPC, it can be that both are the same resource. In that case where does that leave the HTTP operations in the above - esp. PUT and POST ? Dear all, If an LDPC is an LDPR, then its behaviour should be at least the same as that of an LDPR. Here you have Nandana's summary updated so that an LDPC has the same behaviour as an LDPR with "In addition" clauses. The updates are: .- I have tried to define a common behaviour regarding PUT and POST, which was what Roger mentions in his email. Now, and LDPC behaves like an LDPR. .- Re-reading the HTTP 1.1 spec (http://www.ietf.org/rfc/rfc2616.txt) once you POST to a resource URI, "the posted entity is subordinate to that URI", and the examples given are equivalent to our composition approach. So, I see two options: or the server contains some background information and converts the LDPR into an LDPC (which will still behave as an LDPR) or the server ignores the request (or it fails). I don't see the case of appending the triples as mentioned in ISSUE-45. LDPC is a specialization of LDPR. Representation: LDPR - has an RDF representation LDPC - has a RDF representation - In addition: has a set of reserved properties with their semantics defined by the protocol - In addition: contains some protocol data In the interaction model, GET: LDPR - returns the current state. LDPC - returns the current state. - In addition, provides mechanisms to retrieve only part of the state (non-member properties) - In addition, provides additional features like paging, ordering based a special property (membership predicate) PUT: LDPR - updates the current state LDPC - updates the current state - In addition, restricts the update of members in updates - In addition, only part of the state may be updated via <containerURL>?non-member-properties POST: LDPR - does nothing LDPC - does nothing - In addition, creates a new resource and adds it to the membership DELETE: LDPR - deletes itself LDPC - deletes itself - In addition, any resources contained by it are deleted Kind regards, -- Dr. Raúl García Castro http://delicias.dia.fi.upm.es/~rgarcia/ Ontology Engineering Group Departamento de Inteligencia Artificial Universidad Politécnica de Madrid Campus de Montegancedo, s/n - Boadilla del Monte - 28660 Madrid Phone: +34 91 336 36 70 - Fax: +34 91 352 48 19
Received on Friday, 25 January 2013 15:23:21 UTC