FHIR Chat · Extensions when _summary=true? · implementers

Stream: implementers

Topic: Extensions when _summary=true?


view this post on Zulip Lloyd McKenzie (Mar 08 2018 at 19:08):

Is it prohibited to include any extensions when someone has asked for _summary=true view of a resource?

view this post on Zulip Brian Postlethwaite (Mar 08 2018 at 21:22):

Many of the servers (mine included) violate this is a few places, specifically in the conformance statement to include the security urls

view this post on Zulip Michel Rutten (Mar 08 2018 at 21:32):

Would such extension definitions then need to express isSummary = true, to be explicit and computable?

view this post on Zulip Lloyd McKenzie (Mar 08 2018 at 21:37):

If we allow it at all, then yes allowing extensions to override isSummary would be appropriate/necessary

view this post on Zulip Michel Rutten (Mar 08 2018 at 21:47):

According to this table: http://build.fhir.org/elementdefinition.html#interpretation, the isSummary property is prohibited on root elements, including extension definitions. And child elements cannot override isSummary, including extension element within a profile. So I guess we're not allowing this right now.

view this post on Zulip Lloyd McKenzie (Mar 08 2018 at 23:14):

I agree it's not technically allowed at present. I guess the more apt question is "should it be"? As Brian noted, current convention is ignoring the rule and I expect there will be other cases where rule violation is expected/necessary. That's a pretty strong argument for fixing the rule - especially before we lock it in stone.

view this post on Zulip Michel Rutten (Mar 09 2018 at 12:14):

Indeed, hence my Socratic question ;p

view this post on Zulip Lloyd McKenzie (Mar 09 2018 at 14:41):

@Grahame Grieve - thoughts on this?

view this post on Zulip Grahame Grieve (Mar 09 2018 at 20:21):

Not sure. It's not about the type - it's about DomainResource.extension. It's definitely necessary to return some extensions when _summary=true but I do that as a special rule.

view this post on Zulip Grahame Grieve (Mar 09 2018 at 20:22):

I'm not sure that it's necessary to make it computable but even if we did, I don't think that solving it with extensions overriding the summary on their definition is right either

view this post on Zulip Lloyd McKenzie (Mar 09 2018 at 20:45):

Well, right now returning those extensions is non-conformant, so we need to expose something that allows defining whether they're allowed or not.

view this post on Zulip Grahame Grieve (Mar 09 2018 at 23:33):

please explain why it's non conforment

view this post on Zulip Lloyd McKenzie (Mar 10 2018 at 00:27):

You're right, it's not. But the current rules essentially mean that systems are free to totally ignore what's marked as "summary" and send whatever they like - is that really what we want?

view this post on Zulip Grahame Grieve (Mar 10 2018 at 09:07):

not sure

view this post on Zulip Lloyd McKenzie (Mar 10 2018 at 15:57):

GF#15742

view this post on Zulip Michael Lawley (Mar 17 2018 at 02:59):

I would have thought must understand extensions would need to be included?


Last updated: Apr 12 2022 at 19:14 UTC