Open townxelliot opened 7 years ago
Any possibility of an explanation of why a response sometimes contains olo:Slot resources and sometimes doesn't?
Interpretation of returned RDF must always depend upon the properties associated with the resource being requested. olo:Slot
, as with any other property, are present when there is data to include and not when there isn't.
See https://github.com/bbcarchdev/patchwork/issues/3 for the approach we're taking to make this a little clearer, although there will still be cases where the sub-graphs do not include any olo:Slot
properties because there are no related items of the specified type.
The RDF for a proxy URI sometimes contains olo:Slot objects and the expected search results format, and sometimes doesn't.
Compare:
http://acropolis.org.uk/9d6f86627b4c46eca3dda409d611c27e.ttl (no olo:Slot resources) http://acropolis.org.uk/6ec96f5bd2c64232a17d132334f1da75.ttl (olo:Slot resources)
My assumption is that the olo:Slot objects are added by Acropolis and are not part of the original data source that the data came from. If this is the case, the olo:Slot objects should either always be present or always absent.
If the output varies unpredictably, it could make it difficult for a client to figure out how to process the RDF for a proxy URI.