FHIR Chat · Profile name vs title · committers

Stream: committers

Topic: Profile name vs title


view this post on Zulip Grahame Grieve (Feb 20 2018 at 08:58):

The definition of StructureDefinition.name is "A natural language name identifying the structure definition. This name should be usable as an identifier for the module by machine processing applications such as code generation." (The name should be a simple alpha-numeric type name to ensure that it is computable friendly)

The definition of StructureDefinition.title is "A short, descriptive, user-friendly title for the structure definition" (This name does not need to be machine-processing friendly and may contain punctuation, white-space, etc. Applications don't have to use this name, but can always fall back to it. )

view this post on Zulip Grahame Grieve (Feb 20 2018 at 08:59):

many profiles use name the way title is meant to be used. Forth coming tooling in the IG publisher is going to start requiring structure definitions to conform to these rules

view this post on Zulip Ewout Kramer (Feb 20 2018 at 15:09):

@Martijn Harthoorn , we should warn the user about this in Simplifier too!

view this post on Zulip Michel Rutten (Feb 20 2018 at 15:20):

I should update Forge to validate the StructureDefinition.name field.
http://hl7.org/fhir/structuredefinition-definitions.html#StructureDefinition.name
"The name should be a simple alpha-numeric type name to ensure that it is computable friendly."
Does FHIR define an invariant for this rule? If not, should we add one?

(and maybe continue this discussion in another stream?)

view this post on Zulip Lloyd McKenzie (Feb 20 2018 at 17:21):

What are our rules for computable? alpha-numeric, with dashes and underscores?

view this post on Zulip Lloyd McKenzie (Feb 20 2018 at 17:22):

Would it be worthwhile adding a regex to the actual structure definition?

view this post on Zulip Grahame Grieve (Feb 20 2018 at 19:30):

I guess the most likely regex would be xs:token

view this post on Zulip Martijn Harthoorn (Feb 20 2018 at 19:49):

@Ewout, The validator will pick that up, will it not?

view this post on Zulip Grahame Grieve (Feb 20 2018 at 19:54):

not at the moment since there's no regex - it's just narrative

view this post on Zulip Lloyd McKenzie (Feb 20 2018 at 20:06):

That's not actually a regex, but fine with that as the rules.

view this post on Zulip Martijn Harthoorn (Feb 22 2018 at 13:15):

Is this rule going to apply just for StructureDefinition.name or also for other conformance resources, like ValueSet.name?

view this post on Zulip Ewout Kramer (Feb 22 2018 at 15:40):

I think this holds for all!

view this post on Zulip Lloyd McKenzie (Feb 22 2018 at 16:27):

It's going to be painful, but I think the pain is necessary.


Last updated: Apr 12 2022 at 19:14 UTC