FHIR Chat · Logical Model element types · implementers

Stream: implementers

Topic: Logical Model element types


view this post on Zulip Jose Costa Teixeira (Jun 03 2020 at 07:38):

When working with Logical Models, we want to indicate "this is a number, this is text". We can use physical data types but those are specific to FHIR, which does not seem very kosher..

view this post on Zulip Jose Costa Teixeira (Jun 03 2020 at 07:38):

Any ideas on this? What are others doing there?
Before, I removed elementDefinition.type altogether - which makes the Publisher react a bit, but I don't mind.

view this post on Zulip Oliver Egger (Jun 03 2020 at 08:13):

Jose Costa Teixeira said:

When working with Logical Models, we want to indicate "this is a number, this is text". We can use physical data types but those are specific to FHIR, which does not seem very kosher..

why not use FHIR primitive types ? the spec calls for that for logical models: https://www.hl7.org/fhir/structuredefinition.html#logical

The same definition structure can also be used to define any arbitrary structures that are a directed acyclic graph with typed nodes, where the primitive types are those defined by the FHIR specification.

view this post on Zulip Jose Costa Teixeira (Jun 03 2020 at 08:27):

My understanding is that the primitive types are also on the physical layer - if string means "maximum 1024 characters" if we have "markdown" instead of "formatted text", this seems a physical implementation

view this post on Zulip Jose Costa Teixeira (Jun 03 2020 at 08:27):

(I may be wrong in this assumption)

view this post on Zulip Jose Costa Teixeira (Jun 03 2020 at 08:29):

BTW,
http://build.fhir.org/types.html
the link to Logical Models does not point to anywhere - and I think some documentation on Logical Models would be good

view this post on Zulip Jose Costa Teixeira (Jun 03 2020 at 19:56):

Any recommendation on what to use as types for Logical Models? Are Primitive types Logical ? or Physical (they seem physical to me)

view this post on Zulip Jose Costa Teixeira (Jun 05 2020 at 10:55):

@Grahame Grieve ?

view this post on Zulip Grahame Grieve (Jun 05 2020 at 21:32):

I'm not sure what you're asking. But I think that answer might lie in the fact that logical models cannot define their own primitive types

view this post on Zulip Giorgio Cangioli (Jun 05 2020 at 21:44):

@Jose Costa Teixeira i defined abstract data types for the IPS logical model representing the EN 17269 IPS data set (now ISO 27269)

view this post on Zulip Jose Costa Teixeira (Jun 06 2020 at 05:33):

I think this is what I need as well


Last updated: Apr 12 2022 at 19:14 UTC