FHIR Chat · Servers requiring multiple parameters · Covid-19 Response

Stream: Covid-19 Response

Topic: Servers requiring multiple parameters


view this post on Zulip Michael Donnelly (May 13 2020 at 16:49):

There's something I think we need to put in the spec, but I'm not sure how to word it. Hopefully someone else can point at a comparable example.

I agree completely that supporting search parameters for date, measure, and location is essential. Servers need to have the option to require multiple of these. For example, unless someone is searching by ID, we are probably going to require a measure and at least one of location and period.

view this post on Zulip Michael Donnelly (May 13 2020 at 16:49):

How can we express this in an IG?

view this post on Zulip Abbie Watson (May 13 2020 at 16:54):

That seems to be an issue of push vs pull architecture. The group has brainstormed and agreed in prior threads that both usecases are feasible and of interest. But for the purposes of an IG, it helps to pick one (or both) to be explicit. It's common to provide a sequence diagram in IGs. So, I'm thinking a) define the usecase describing a client with pull architecture, b) provide a sequence diagram, c) document API calls.

view this post on Zulip John Moehrke (May 13 2020 at 16:55):

We are working toward the specifics of valid grouping of search parameters. There is a mechaism that US-Core used

view this post on Zulip John Moehrke (May 13 2020 at 16:55):

worst case we just say it narratively.

view this post on Zulip John Moehrke (May 13 2020 at 16:56):

so, write an issue with the suggested grouping of parameters you think are good.

view this post on Zulip John Moehrke (May 13 2020 at 16:56):

Abigail. Push is a different transaction from Pull. so there is different transport requirements/expectations

view this post on Zulip Abbie Watson (May 13 2020 at 16:58):

I agree. I'm saying that the sequence diagrams often capture in their examples when multiple search parameters or headers are required; and at which step.

view this post on Zulip Michael Donnelly (May 13 2020 at 17:07):

John Moehrke said:

so, write an issue with the suggested grouping of parameters you think are good.

https://github.com/AudaciousInquiry/fhir-saner/issues/74

view this post on Zulip Eric Haas (May 13 2020 at 21:58):

John Moehrke said:

We are working toward the specifics of valid grouping of search parameters. There is a mechaism that US-Core used

... That involves an extension on the CapabilityStatement and some generated text


Last updated: Apr 12 2022 at 19:14 UTC