Some Enumerations and properties are not generated properly

[marcovisserFurore]

Describe the bug Reported by @Rob5045:

Some properties and enumeration types are not generated properly by the code generator. It concerns the following elements:

STU3

Enum Types: (starting with lowercase)

• Sequence.qualityType
• Sequence.repositoryType

Properties: (ending with underscore)

• Claim.Use_
• TestScript.OperationComponent.ContentType_

R4

Enum Types: (starting with lowercase)

• messageheader_response_request
• MolecularSequence.orientationType
• MolecularSequence.qualityType
• MolecularSequence.repositoryType
• MolecularSequence.sequenceType
• MolecularSequence.strandType

Properties: (ending with underscore)

• EffectEvidenceSynthesis.ResultsByExposureComponent.Exposure_
• VerificationResult.Status_

R4B

Enum Types: (starting with lowercase)

• messageheader_response_request
• MolecularSequence.orientationType
• MolecularSequence.qualityType
• MolecularSequence.repositoryType
• MolecularSequence.sequenceType
• MolecularSequence.strandType
• VerificationResult.status

Properties: (ending with underscore)

• Composition.Confidentiality_
• EvidenceVariable.CharacteristicCombination_
• ResearchElementDefinition.VariableType_
• VerificationResult.Status_

R5

Enum Types: (starting with lowercase)

• messageheader_response_request
• MolecularSequence.orientationType
• MolecularSequence.qualityType
• MolecularSequence.repositoryType
• MolecularSequence.sequenceType
• MolecularSequence.strandType
• VerificationResult.status

Properties: (ending with underscore)

• Composition.Confidentiality_
• EvidenceVariable.CharacteristicCombination.GroupMeasure_
• VerificationResult.Status_

Version used:

• FHIR Version: STU3, R4, R4B and R5
• Version: 4.2.1

Changing these elements would introduce breaking changes!

[marcovisserFurore]

After refinement 2022-09-08:

• we should change codegen so we generate the correct types and properties
• the existing incorrect properties and enum will be retained and flagged as obsolete (move them to a separate partial class).

[ewoutkramer]

Note that Gino is updating his latest codegen so it does no longer produce breaking changes in the generated C# code, so we should try to apply this fix to the newest codegen.

[brianpos]

Can you check the codegen branch that I was working on to generate R5 and see if that can be merged into.

[ewoutkramer]

There were two issues here:

• Some HL7 workgroups give their valuesets a ValueSet.name with a lower character (so using lowerCaps), I have made sure we will always use PascalCase.
• Some valuesets have the same names as the property. This is not possible if the enum is defined inside the same class that declares the property. In this case, the generator automatically adds a "_" to the end of the property. I have changed the behaviour of the generator, so it will now throw an Exception if this happens. There is a variable _enunNamesOverride in the generator that you can extend to define a better/more unique name for the enum to avoid these nameclashes. It's too hard to automate it.
• The exception thrown will contain the valueset url that you need to specify in _enumNamesOverride.