FHIR Chat · References - Missing invariant? · implementers

Stream: implementers

Topic: References - Missing invariant?


view this post on Zulip Richard Kavanagh (Dec 18 2017 at 14:04):

Looking at the reference definition here: http://build.fhir.org/references.html
Is there an invariant missing to state that there should not be both a "reference" and an "identifier" ?

view this post on Zulip John Moehrke (Dec 18 2017 at 14:08):

why not both? as long as they are identifying the exact same thing...

view this post on Zulip Richard Kavanagh (Dec 18 2017 at 14:23):

But what if they don't - which one do you go with?

view this post on Zulip Lloyd McKenzie (Dec 18 2017 at 15:45):

It's totally ok to have both. Typically if a system understands Reference, it won't pay much attention to identifier, but there may be some downstream systems that can't deal with the reference and will need the identifier

view this post on Zulip Lloyd McKenzie (Dec 18 2017 at 15:46):

It would be an error if the logical business identifier reference and the FHIR resource reference don't resolve to the same artifact though (for whatever business mechanism Reference.identifier is resolvable)

view this post on Zulip Grahame Grieve (Dec 18 2017 at 18:11):

actually, this is dicussed:

view this post on Zulip Grahame Grieve (Dec 18 2017 at 18:12):

When both an identifier and a literal reference are provided, the literal reference is preferred. Applications processing the resource are allowed - but not required - to check that the identifier matches the literal reference, if they understand how to resolve the logical reference.


Last updated: Apr 12 2022 at 19:14 UTC