[JENA-1487] Incorrect SPARQL query result set when applying ORDER BY after an empty GROUP BY

• Type: Bug
• Status: Closed
• Priority: Major
• Resolution: Not A Problem
• Affects Version/s: Jena 3.6.0
• Component/s: ARQ
• Assignee: Unassigned
• Reporter: Claus Stadler
• Created: 13/Feb/18 04:01
• Updated: 09/Apr/18 21:57
• Resolved: 13/Feb/18 10:06

Description

Ordering an empty result set must not introduce additional bindings, but this happens with the following example (run on an arbitrary dataset - including an empty one):

./fuseki-server --file=test.ttl /foobar

The Json below is obtained from the Fuski bundle; but I noted this issue first using the ARQ API, where the incorrect result set contains a single BindingProjectNamed instance.

SELECT ?s
WHERE {
  ?s ?p ?o
  FILTER(false)
}
GROUP BY ?s
ORDER BY DESC(COUNT(?o))

{
  "head": {
    "vars": [ "s" ]
  } ,
  "results": {
    "bindings": [
      {
# NOTE THIS EMPTY BINDING HERE!        
      }
    ]
  }
}

The result set is correct without the ORDER BY:

SELECT ?s
WHERE {
  ?s ?p ?o
  FILTER(false)
}
GROUP BY ?s

{
  "head": {
    "vars": [ "s" ]
  } ,
  "results": {
    "bindings": [
# CORRECT      
    ]
  }
}

Issue Links

• is superceded by: JENA-1507 GROUP BY and aggregates when there are no matching of the WHERE pattern.