Andy Seaborne wrote:
>> There is a (prototype) SPARQL cache that does paging by stripping off
>> LIMIT/OFFSET, issuing the whole (sorted) query and caching the results.
>>  Then, later requests with different LIMIT/OFFSET are sliced out of the
>> cached results.

A couple of links for Rafal:

  // The query cache
  val queryCacheMap = MMap[Query, RS]()

https://github.com/afs/LD-Access/blob/master/src/main/java/org/apache/jena/sparqlcache/SparqlCache.scala
https://github.com/afs/LD-Access/blob/master/src/main/java/org/apache/jena/sparqlcache/RS.scala

So, the cache is a map from Query objects to RS (which do slicing when offset
and/or limit are specified).

Rafal, are you using Java to implement your cache layer?
Is/will it be open source?

Paolo

PS:
You find also another small experiment here:
https://github.com/castagna/sparql-cache
... not as clever as LD-Access.

Reply via email to