Stream: implementers
Topic: fhir implementation guide vs swagger api
Jefferson (Oct 18 2021 at 12:14):
How much more do you get from the implementation guide vs using a swagger API documentation?
Lloyd McKenzie (Oct 18 2021 at 14:17):
Quite a bit. I'm not a Swagger expert, but my understanding is:
- by default, FHIR APIs support JSON, XML and RDF data. Swagger isn't set to handle that
- Swagger relies on schemas to define data requirements, but schemas don't handle representing slicing, vocabulary bindings, invariants, mustSupport, etc.
- ImplementationGuides are set up to link to underlying specs, so the rendered views automatically link to the core specification, ancestor IGs, etc.
- FHIR profiles have both a snapshot and differential. Swagger isn't really set up to deal with that
- FHIR IGs are set up to ensure mandatory information is available both on-screen and for downloads. E.g. showing version, publisher, copyright, etc. on every page, creating a 'package' file for use with FHIR validation and subsequent publication, etc.
Jefferson (Oct 18 2021 at 15:47):
thanks Lloyd
Last updated: Apr 12 2022 at 19:14 UTC
