FHIR Chat · capabilitystatement shortcomings · implementers

Stream: implementers

Topic: capabilitystatement shortcomings


view this post on Zulip Jens Villadsen (Nov 11 2020 at 09:00):

Hi all - I think I've found a shortcomming (yes, it may have others, but keep on reading) in the capabilitystatement (more specifically STU3). In my setup, I have multiple profiles (4 to be exact)on the Appointment resource type. The business rules differs dependent on which profile a resource conforms to and resources cannot conform to more than one of the profiles (as the stated profile conformity is bound to third party integration behind the scene). My issue (its a minor issue) is that I cannot cleary express what HTTP operations are allowed on the different profiles in the capability statement as these differ. For some of the profiles, only POST and PATCH are allowed for editing (where PUT is allowed for others) - but this level of expression is AFAIK not supported by capability statements. Is this beyond what should be supported by the capabilitystatement?

view this post on Zulip Lloyd McKenzie (Nov 11 2020 at 14:38):

Yes. CapabilityStatement only allows you to define what's done on a per resource bases (e.g. Observation). You can't say "this is what I support for lab" vs. "this is what I support for vitals"

view this post on Zulip Lloyd McKenzie (Nov 11 2020 at 14:38):

(At least not without using custom extensions)

view this post on Zulip Jens Villadsen (Nov 11 2020 at 15:27):

extensions ... you're saying someting there ... and I guess that the standard don't have any standard extensions for expressing this?

view this post on Zulip Lloyd McKenzie (Nov 11 2020 at 16:36):

Correct. There's no standard way to say something like this right now.

view this post on Zulip Jens Villadsen (Nov 12 2020 at 10:02):

what about http://build.fhir.org/extension-capabilitystatement-expectation.html - couldn't that apply?

view this post on Zulip Jens Villadsen (Nov 12 2020 at 10:02):

Btw: @Lloyd McKenzie image.png

view this post on Zulip Jens Villadsen (Nov 12 2020 at 10:02):

[WG Name] ... is that a new WG with the name WG Name :rolling_on_the_floor_laughing:

view this post on Zulip Lloyd McKenzie (Nov 12 2020 at 14:08):

@Jens Villadsen Can you submit a technical correction Jira on the 'WG Name' issue?

The extension allows you to define SHOULD/SHALL/MAY in an implementation guide. It doesn't allow you to define different sets of behavior on the basis of different profiles.

view this post on Zulip Jens Villadsen (Nov 12 2020 at 18:46):

@Lloyd McKenzie JIRA: https://jira.hl7.org/browse/FHIR-29686

view this post on Zulip Jens Villadsen (Nov 12 2020 at 18:46):

I'll guess I'll have to draft my own extension then

view this post on Zulip Lloyd McKenzie (Nov 12 2020 at 18:57):

The base profile declaration should be the union of the capabilities

view this post on Zulip Jens Villadsen (Nov 13 2020 at 16:09):

The union does not really reflect the capabilities when the possibilities are dependant on the profiles used


Last updated: Apr 12 2022 at 19:14 UTC