FHIR Chat · Forge issue for display & saving of extensions? · tooling

Stream: tooling

Topic: Forge issue for display & saving of extensions?


view this post on Zulip Morten Ernebjerg (Mar 09 2021 at 13:14):

I was trying to view the STU1 IPS MedicationStatement profile in Forge and noticed something strange. Somehow, the data-absent-reason extension on the effective[x] element was being displayed at the top level.

Screenshot-2021-03-09-at-14.01.24.png

In StructureDefinition.differential, this extension is declared like this:

{
    "id": "MedicationStatement.effective[x].extension",
    "path": "MedicationStatement.effective[x].extension",
    "short": "effective[x] absence reason",
    "definition": "Provides a reason why the effectiveTime is missing.",
    "max": "1",
    "type": [{
        "code": "Extension",
        "profile": [
            "http://hl7.org/fhir/StructureDefinition/data-absent-reason"
         ]
     }],
    "mustSupport": true
}

What is more, if I made a change and saved it (I have saving snapshots option on), the extension would completely disappear from the StructureDefinition (snapshot & diff)! @Ward Weistra is this a known issue?

view this post on Zulip Matthijs van der Wielen (Mar 10 2021 at 12:01):

Hi @Morten Ernebjerg ,
Thank you for pointing this out. When a StructureDefinition contains an extension on a variant (e.g. Observation.value[x]), this extension is not correctly processed by the SnapshotGenerator. It is moved to the observation's root 'extension' element definitions, while it would be expected to remain in the Observation.value[x].extension element definitions. Forge previously did not support extensions on variants; that's why this problem probably never popped up before.
We first need to resolve this issue in the .NET SDK: https://github.com/FirelyTeam/firely-net-sdk/issues/1619. After this is implemented we expect this to be implemented in Forge in one of the upcoming sprints. We can keep you posted on the progress if you like?

view this post on Zulip Morten Ernebjerg (Mar 10 2021 at 16:14):

@Matthijs van der Wielen Thanks for the update, would be great if you could keep me posted!


Last updated: Apr 12 2022 at 19:14 UTC