FHIR Chat · GF#15883 - search parameter type · fhir/infrastructure-wg

Stream: fhir/infrastructure-wg

Topic: GF#15883 - search parameter type


view this post on Zulip Grahame Grieve (Jul 22 2018 at 22:16):

GF#15883 creates a problem - if the Questionniare.item.definition type is canonical(ElementDefinition) then the search parameter is a reference, and the target type of that is... not a resource....

view this post on Zulip Bryn Rhodes (Jul 22 2018 at 22:38):

Yeah, I thought that's why we left item.definition as a uri, because it could potentially point to an ElementDefinition. Could we have a two delements: definition: canonical(StructureDefinition) and element: String?

view this post on Zulip Grahame Grieve (Jul 22 2018 at 23:59):

well, we discussed the consequences of making the change, to canonical(ElementDefinition) and we made them all of them... except this one

view this post on Zulip Lloyd McKenzie (Jul 23 2018 at 03:22):

We'd talked about it a little, but you're right that we never landed it. The question was how you declare that you want to allow to point to sub-elements within a resource rather than just the resource alone. Could we say canonical(StructureDefinition,StructureDefinition#) as the convention for saying you can point to either the full structure definition or an element within it?

view this post on Zulip Grahame Grieve (Jul 23 2018 at 03:45):

that's not necessary nor does it deal with the problem, I think

view this post on Zulip Lloyd McKenzie (Jul 23 2018 at 05:59):

You'd probably want two search parameters - one of type Reference that matched on structure defintiion (regardless of element) and one of type uri that matched on a specific element

view this post on Zulip Grahame Grieve (Jul 23 2018 at 06:02):

better to leave it as just uri then

view this post on Zulip Grahame Grieve (Jul 25 2018 at 04:35):

so where have we landed with this one? nowhere yet... but we need to. For the moment, I have rolled it back to uri...

view this post on Zulip Lloyd McKenzie (Jul 25 2018 at 12:37):

Can we have a URI that uses the conventions of a canonical with a target element?

view this post on Zulip Grahame Grieve (Jul 25 2018 at 12:44):

well, we can; we defined it. what isn't clear is the definition of the search parmaeter

view this post on Zulip Lloyd McKenzie (Jul 25 2018 at 12:51):

The desired search parameter behavior would ideally be to allow exact match, to allow match on URL only, URL + version, or url + element without version. It's that last one I don't think we can do right now with a "standard" search parameter and available qualifiers.

view this post on Zulip Lloyd McKenzie (Jul 25 2018 at 12:51):

Though perhaps you could with both an "above" and a "below"


Last updated: Apr 12 2022 at 19:14 UTC