Sorry, there was a short bout of confusion on my part, I forgot that the Rya instance (utilizing the REST endpoint) is already configured with a table prefix (found in the Rya environment.properties file).
That being said, I've managed to write a short Python script (via Requests library) that attempts to POST my RDF files (.owl and .rdf) to the REST endpoint. I'm using "?format=RDFXML" as a URL parameter and "Content-Type: text/plain" in the header as indicated in the manual. Unfortunately, I'm getting a 500 status code. I've captured and attached a screenshot (TomcatRyaLog.PNG) of one of the Tomcat logs. It seems to be saying RDFXML is not a supported RDFFormat, which doesn't make sense. Any thoughts? Thank you so much! Brian From: Boris Pelakh <[email protected]> Sent: Thursday, February 21, 2019 12:44 PM To: Brian Vincent <[email protected]>; [email protected] Subject: Re: Most straight forward approach to loading ontology and data What do you mean by 'table'? Are you referring to the named graph? ________________________________ From: Brian Vincent <[email protected]<mailto:[email protected]>> Sent: Thursday, February 21, 2019 2:09 PM To: Boris Pelakh; [email protected]<mailto:[email protected]> Subject: RE: Most straight forward approach to loading ontology and data So when PUTting to the REST endpoint, how do you specify the table [prefix] you wish to use? I've seen table specified in the other data loading mechanisms, but not the REST endpoint. Thanks again! Brian From: Boris Pelakh <[email protected]<mailto:[email protected]>> Sent: Wednesday, February 20, 2019 7:03 PM To: Brian Vincent <[email protected]<mailto:[email protected]>>; [email protected]<mailto:[email protected]> Subject: Re: Most straight forward approach to loading ontology and data Back when we were loading large volumes of data into the store, we would load ontologies via the REST endpoint, as they tend to be small. We would then convert ABox data into NQuad format (since it's one quad per line) and load them using the Hadoop bulk loader. We scaled up to 2 billion triples that way. Boris ________________________________ From: Brian Vincent <[email protected]<mailto:[email protected]>> Sent: Wednesday, February 20, 2019 6:30 PM To: [email protected]<mailto:[email protected]> Subject: Most straight forward approach to loading ontology and data Hello, In your experience, what has been the most straight forward approach for you and your team for loading ontology files (RDF XML) and data (turtle) into Rya? I ask because the manual lists several approaches (REST endpoint, Bulk load, RDF4J API, and shell). I have several RDFS files (ontology) along with several large (~300MB) Turtle files (ontology and data). Thank you! Brian
