Update SPARQL queries to match new ontology descriptions when new cache available

[jyucsiro]

No description provided.

[dr-shorthair]

Some work started here CSIRO-enviro-informatics/loci-integration-api#47

[benjaminleighton]

@dr-shorthair can you comment on https://github.com/CSIRO-enviro-informatics/loci-cache-scripts/blob/master/docker/cache/resources/pre-condition-files/008_dataset_def.sparql does that need changing? I know there are somethings in there that maybe we don't need like the named individual stuff we can change that in another issue but is there anything effected by changes you know of?

[jyucsiro]

@benjaminleighton can we close this?

[dr-shorthair]

I think the query that Ben pointed me at in #49 (comment) might be simplified by removing the

            a owl:NamedIndividual ;

from each insert. From a semantic POV there is no need for this assertion. However, perhaps you guys use this classifier somewhere else in the queries?

[benjaminleighton]

@jyucsiro and @dr-shorthair you are probably both in a better position to decide whether to close this and also to close out CSIRO-enviro-informatics/loci-integration-api#47 at the same time. I believe that 47 is mostly solved through another pull request you made Jonathan? Looking through that pull request here are the things to be considered:

• Addition of a union to ensure that we pickup loci:datasets in addition to dcat:datasets
• Change area lookups to use geox:inCRS
  Additionally there are changes in that pull request to remove transitiveOverlaps but I think we need them at the moment, and to change the ipo to dcterms:isPartOf which might be a nice to have if someone wants to implement it but not necessary.

Is there anything else that changed, structurally, that we need to account for? I'm guessing there aren't any direct dependencies in excelerator on iderdown now either but it might be worth checking with @shaneseaton .