Core library bump 6.3.7

[dotasek]

This is a version bump of the org.hl7.fhir.core library from 6.1.2.2 to 6.3.7.

The trivial changes are matching changing method signatures and class changes.

The more difficult changes revolve around the resource validation. HAPI provides its own implementations of IWorkerContext, which are used by core's InstanceValidator to validate resources.

InstanceValidator has changed significantly with regard to logic, particularly regarding code validation:

org.hl7.fhir.validation/src/main/java/org/hl7/fhir/validation/instance/InstanceValidator.java

See comments for discussion of some of these changes.

[heathfrankel]

Hi @dotasek and @jamesagnew . Has there been any progress on this PR.
We are waiting on the resolution of this terminology passthrough validation issue for the AU Sparked Reference Server in readiness for our AU Core Vendor Testing events at the end of the month.
Is there anything we can do to help progress this?

[jamesagnew]

Hi @dotasek and @jamesagnew . Has there been any progress on this PR. We are waiting on the resolution of this terminology passthrough validation issue for the AU Sparked Reference Server in readiness for our AU Core Vendor Testing events at the end of the month. Is there anything we can do to help progress this?

@heathfrankel This is being actively worked on by several people, but it's nontrivial. David has been leading the work, but there are a number of breaking API changes, and at least one breaking functional validation change that we need to account for before we can land this. Hopefully we'll see this land in the next week or 2.

[dotasek]

Hi @dotasek and @jamesagnew . Has there been any progress on this PR. We are waiting on the resolution of this terminology passthrough validation issue for the AU Sparked Reference Server in readiness for our AU Core Vendor Testing events at the end of the month. Is there anything we can do to help progress this?

@heathfrankel I echo what James has said and I would also suggest that any additional code review would be helpful.

[dotasek]

For 4f491e8 there was a change in the underlying test. The history of this test seems to have waffled back and forth between expected responses, so I referred to the original request:

#4461

The test uses patient-with-single-comm-lang-en_US-UNDERSCORE.json, which contains en_US, an invalid language code. It is singular; there are no valid codes included in the coding, and therefore this should always fail, regardless of the value of EnabledValidationForCodingsLogicalAnd

Edit: On reviewing the original ticket, technically, this is a logical ANY applied to a set, not an AND. I feel like it should be renamed.

[tadgh]

LGTM minus 2 minor things