Test Runner Not Detecting Equal Long Values

[cmoesel]

When running the test CqlArithmeticFunctionsTest:Abs:AbsLong, the test runner reports:

Test CqlArithmeticFunctionsTest:Abs:AbsLong status: fail expected: 1L actual: 1L

My engine is returning the following result:

{
  "resourceType": "Parameters",
  "parameters": [
    {
      "extension": [
        {
          "url": "http://hl7.org/fhir/StructureDefinition/cqf-cqlType",
          "valueString": "System.Long"
        }
      ],
      "name": "return",
      "valueString": "1L"
    }
  ]
}

According to the CQL IG FHIR type mapping, System.Long maps to FHIR.string in R4, so I believe my engine is returning the correct result to represent 1L.

NOTE: I also tried returning it with "valueString": "1" instead of "valueString": "1L", but that also failed.

[brynrhodes]

Agreed, this should just use the string representation (no L): https://jira.hl7.org/browse/FHIR-56227

[bryantaustin13]

Wait. I thought Long was R5 and not R4. I just added skips for trying to test Long with server R4 and earlier. I do see
"The Using CQL With FHIR IG maps System.Long to FHIR.string in R4 and FHIR.integer64 in R5+", but the HAPI server is currently returning "EvaluationError:FHIR R4 does not support Long/Integer64 values" for tests with long.

[bryantaustin13]

Just an observation - shouldn't
{
"resourceType": "Parameters",
"parameters": [

be

{
"resourceType": "Parameters",
"parameter": [

[cmoesel]

@bryantaustin13 - Yes, that should be "parameter": [ -- and indeed my server is returning parameter, so I must have somehow messed it up when entering this issue.

As for Long and R4, I don't think you should automatically skip Long tests in R4 -- because, as you noted, the Using CQL IG contains a mapping for Long to FHIR R4, so that's what should be used. Even if HAPI cannot evaluate CQL Longs, that's a bug in a specific CQL implementation -- and no specific implementation should govern what is and is not a valid test in this framework.