FHIR Chat · Sanity Check: UUID as identifier · implementers

Stream: implementers

Topic: Sanity Check: UUID as identifier


view this post on Zulip Jenni Syed (Jan 30 2019 at 16:48):

We have places where a UUID will be used as an identifier. EG: an id from a different system. The text on identifier.value states: "The value's primary purpose is computational mapping. As a result, it may be normalized for comparison purposes (e.g. removing non-significant whitespace, dashes, etc.) A value formatted for human display can be conveyed using the Rendered Value extension. Identifier.value is to be treated as case sensitive unless knowledge of the Identifier.system allows the processer to be confident that non-case-sensitive processing is safe."

view this post on Zulip Jenni Syed (Jan 30 2019 at 16:49):

However, I assume that the dashes in a UUID are the computable value (and have meaning), and it would be unexpected to trim those out? IE: it should be value=123e4567-e89b-12d3-a456-426655440000

view this post on Zulip Jason Walonoski (Jan 30 2019 at 18:24):

FWIW, in Synthea, we output identifiers with values that are UUIDs in a few places (in addition to the all the Resource.id values that are UUIDs) and most systems we've inserted the data into seem to handle this. And I agree, it should include the dashes.

view this post on Zulip Grahame Grieve (Jan 30 2019 at 19:08):

yes include the dashes. But a typical UUID looks like this:

  "system" : "urn:ietf:rfc:3986",
  "value" : "urn:uuid:a76d9bbf-f293-4fb7-ad4c-2851cac77162"

(see http://hl7.org/fhir/datatypes-examples.html#Identifier)

view this post on Zulip Paul S (Feb 06 2019 at 06:10):

Is there a list of UUIDs I can download? Do they map with XML (CCD) codes? Thanks.

@Grahame Grieve @Jenni Syed @Jason Walonoski

view this post on Zulip Grahame Grieve (Feb 06 2019 at 10:22):

you can look at the naming system resources - they contain some of the UUIDs we defined - if any

view this post on Zulip Paul S (Feb 06 2019 at 15:34):

You said "some of the UUIDs". Is there a repository for "all" not just some UUID? Same question for ResourceType repository?

view this post on Zulip Grahame Grieve (Feb 06 2019 at 20:24):

I'm not sure whether we define any at all, actually. If we define any (or use any) then they might be in NamingSystems.... but mostly we just UUIDs in examples

view this post on Zulip Elliot Silver (Feb 06 2019 at 20:30):

(deleted)

view this post on Zulip Elliot Silver (Feb 06 2019 at 20:37):

(deleted)


Last updated: Apr 12 2022 at 19:14 UTC