Stream: IG creation
Topic: Co-occurence constraints
Grahame Grieve (Sep 03 2020 at 11:07):
I have added a new feature to the IG publisher.
Grahame Grieve (Sep 03 2020 at 11:11):
if you write a constraint on an element that follows this pattern:
(name | name | name).count = 1
(or <= 1)
where all the names in the union are children of the element on which this invariant sits, then the IG publisher will understand that the named elements are a mutually exclusive group. That means that:
- They will be rendered as a mutually exclusive group in the table presentation form
- if you make one required in a profile, the others will automatically be set to max = 0 (that must be true, so it saves you from having to be explicit about it)
- the rendering as a group will collapse out if one of the elements is required
I added this to reduce the comprehension impedence for CCDA guide readers
Grahame Grieve (Sep 03 2020 at 11:12):
Since we pretty much don't do this in the base spec, this is really only relevant for logical models. And honestly, it's probably only really relevant for RIM derived models like CDA, and there's a few other people working on similar v3 derived RMIMs as logical models
Jose Costa Teixeira (Sep 03 2020 at 13:17):
where should we put a reference example for this: IG Guidance or Sample IG?
Jose Costa Teixeira (Sep 03 2020 at 15:47):
@Lloyd McKenzie to document this via an example, where is it best?
Lloyd McKenzie (Sep 03 2020 at 18:22):
Documentation in guidance, example in sampleIg
Last updated: Apr 12 2022 at 19:14 UTC
