JP

@raghunathsandilya - Thanks for the contribution! Looks great overall! We don't have permissions to publish a jar under the "org.hl7" package, so I updated some package names on this PR...

Hi Preston, We actually used to do it that way but found it was not suited for our purposes. There were two main issues: 1. We deliver the cqf-ruler in...

Hi Preston Thanks for you response. Let me address some of the points you've raised. > Docker's default behavior of docker build … is to cache individual layers unless some...

Hi Preston, > Requiring installation of develpment dependencies on individuals' machines There must be some misunderstanding. That is simply _not the case_. There is a docker image with the build...

Potential dup of https://github.com/cqframework/clinical_quality_language/issues/1243

Whats happening here is that `Medication.medication` is a choice type. It can be either a `CodeableConcept` or a `Reference`. In cases where it's a `Reference`, there's an implied requirement for...

> but is it possible to report it in such a way that it could be differentiated from an explicit dependency on the Medication resource from the CQL? Based on...

> Based on my understanding of data requirements that's not how they are intended to be used. > This interpretation seems to conflict with the discussion here: https://cql.hl7.org/05-languagesemantics.html#artifact-data-requirements > It'd...

Yes, the logic for that analysis is partially implemented internally and is the source for the FHIR data requirements. Here's the package for that. In particular the ElmRequirementsVisitor may be...

@brynrhodes - This requires a change to the ELM spec, possible candidate for CQL 2.0