...
- Can queries be done across triplestores? The original plan was to reference an external URI for bibliographic references (e.g.,
http://da-rdf.library.cornell.edu/individual/b3652730) and not copy any information about the bibliographic reference into our local triplestore. The requirement to sort by title, author, and publication date with you to require that this information be copied into the local triplestore, or that the SPARQL queries are capable of querying across two distinct triplestores.- Drive app data from SOLR instead of triplestore.
- Can search across multiple SOLR indexes?
- Some say yes, some say no, looks like no.
- Can make individual queries and meld results.
- Can add data to my SOLR from multiple triplestores/apps?
- YES, but questions about each app keeping multiple SOLR data in sync
- For example, will library catalog system update my SOLR index when the name of a bib ref gets updated?
- Need some process added to blacklight-cornell catalog that will update my SOLR index
- Need some process to add my triple data into the cornell catalog SOLR index
- Both of these are post Use Case and production implementation issues
- Process:
- adding an item
- copy title, author, pubdate from blacklight-cornell catalog system into my SOLR index
- updates to virtual collection items go into triplestore and into SOLR index
- queries go to my SOLR index
- adding an item
- Why separate SOLR indices? Efficiency of queries, data management (buffer against changes), ease of use by multiple apps
- How does SPARQL handle sorting?
- look at order_by
- How does SPARQL track paginated results? How does SPARQL track tokens for retrieving the next set of search results?
- look at limit and offset
- Missing Triples for an 'object' -- From the SPARQL doc, it appears that if any one of the 'objects' (e.g., title, description, size, visibility) listed in the WHERE of the query is not defined for a virtual collection, then that virtual collection will not be included in the results. I have not tested this. Is there a way around this limitation? I'd like the value for the missing object to be _BLANK_ or nil.
Example from SPARQL doc
Data:
Code Block language none @prefix foaf: <http://xmlns.com/foaf/0.1/> . _:a foaf:name "Johnny Lee Outlaw" . _:a foaf:mbox <mailto:jlow@example.com> . _:b foaf:name "Peter Goodguy" . _:b foaf:mbox <mailto:peter@example.org> . _:c foaf:mbox <mailto:carol@example.org> .
Query:
Code Block language none PREFIX foaf: <http://xmlns.com/foaf/0.1/> SELECT ?name ?mbox WHERE { ?x foaf:name ?name . ?x foaf:mbox ?mbox }Query Result:
name mbox "Johnny Lee Outlaw" <mailto:jlow@example.com> "Peter Goodguy" <mailto:peter@example.org> Question:
- Why isn't mbox=<mailto:carol@example.org> part of the result with name=_BLANK_?
...