We could certainly fork the Schram to experiment with extensions.
On Jul 11, 2018, at 15:10, Karima Rafes <karima.rafes@gmail.com<mailto:karima.rafes@gmail.com>> wrote:
A need also very close...
At the next extended, we need to think to insert a place in the XML for the errors' messages of different SPARQL services for federated queries, with a mode verbore ;)
It's maybe a good idea to start to prepare a new version of SPARQL results...
Bye
Karima
2018-07-11 15:49 GMT+02:00 Bijan Parsia <bijan.parsia@manchester.ac.uk<mailto:bijan.parsia@manchester.ac.uk>>:
Other hacks:
1) Wrap the result in a custom format and strip it off before passing the results to a results processor
Pro: pretty easy and clean
Con: Clients which are unaware of this will bork hard
2) Dork the schema to allow for the meta data and dork the schema location to point to your dorked schema
Pro: Lets you use the format of choice. Should work with most processors even validating ones
Con: If the process makes assumptions about the PSVI you could get borkage.
I think link with datauri is likely to be the most robust if a bit unpleasant for e.g. inspecting the results by hand.
Received on Wednesday, 11 July 2018 14:13:04 UTC