`RAND` operator should produce a `xsd:double`

[romaintailhurat]

According to the SPARQL 1.1 spec, the RAND operator should produce a xsd:double between 0 and 1.

But it's not what I'm getting from the API. For example, with a query like:

PREFIX dc: <http://purl.org/dc/elements/1.1/>
PREFIX rdfs: <http://www.w3.org/2000/01/rdf-schema#>
select ?disorder ?random ?label where {
    ?disorder dc:type "disorder" .
    ?disorder rdfs:label ?label
    FILTER langMatches(lang(?label), "fr")
    BIND(RAND() as ?random)
} order by ?random limit 1000

I'm getting this kind of payload:

{
  "disorder": {
    "type": "uri",
    "value": "http://snomed.info/id/4448006"
  },
  "label": {
    "type": "literal",
    "value": "céphalée allergique",
    "xml:lang": "fr"
  },
  "random": {
    "datatype": "http://www.w3.org/2001/XMLSchema#int",
    "type": "literal",
    "value": "-576459726194747256"
  }
}

The type of the random variable being xsd:int instead of xsd:double and the value is way out of the spec bounds.

[hannahbast]

@romaintailhurat Thanks for pointing this out to us + a PR is already under way. Did this just occur to you or is the range and datatype important for you application?

[romaintailhurat]

Nothing important for my current use case, i was just testing this operator on two endpoints: a qlever instance and a GraphDB one, i noticed this subtle difference.

[romaintailhurat]

And thank you for the quick fix!