finally a bnode with a uri


Streaming backups via the SPOG SPARQL result format in ARC
The Urban Dictionary describes SPOG as "super pimped out gangsta" or as "a weapon that (...) had a fusion reactor as a power source". Sorry to disappoint you, neither has become part of ARC. Nevertheless, the SPOG I mean is quite powerful, too. It is a constrained SPARQL XML result format from SELECT queries that was proposed by Morten Frederiksen a few months ago. SPOG enables streaming store backups/dumps, and being another RDF serialization, it can be used for streamed loading as well. Support for SPOG was added in the latest revision (2008-07-02) and extends the store and the endpoint components:
  • The store got a dump() method that stream-outputs SPOG from all quads, and a createBackup($path, $alternative_query) method to write a SPOG dump (or custom SPO(G) query result) to a local file
  • The SPARQL endpoint feature list accepts "dump" as a new read operation
  • The SPARQL endpoint accepts "DUMP" as a query type now ("DUMP" also works via the internal query() method)
  • The format detector accepts SPOG XML as an RDF format now, SPARQL+ queries will work fine with LOAD <some-spog-file.srx>. (There is now a dedicated SPOG parser for streaming LOADs.)

These additions should simplify graph exchange and store replication quite a bit.

Morten++ for the idea and an initial implementation.

Documentation - Release Notes

Comments and Trackbacks

0 comments are currently in the approval queue.

Comments are disabled for this post.

Earlier Posts


No Posts found