Stream: IG creation
Topic: Logical Models with no cardinality specified
Giorgio Cangioli (Feb 08 2019 at 11:44):
I'm trying to represent the IPS EN 17269 data set as FHIR logical models.
This dataset doesn't specify the cardinality for the defined concepts. I tried to compile using the same "configuration" we use for the IPS FHIR profiles, but i noticed that the build finally fails if the cardinality are not specified, where instead all is (more or less) good if cardinality are given.
I can suppose that the failure is due to the fact that with the used page template there is the attempt to create a table with the cardinalities ….
have you any suggestion on how to overcome this problem ? is there any page template that i can use instead for this kind of logical Models ?
Grahame Grieve (Feb 08 2019 at 12:02):
cardinality is 0..* then
Giorgio Cangioli (Feb 08 2019 at 13:19):
Telling the truth the model would allow to determine the min cardinality;
but (0 or 1) …* is different from saying that the max cardinality is not specified.
If for an element of a data set the cardinality is set to 0...* , someone could always complain saying that for that element the cardinality should be 0..1 and not 0...*....
Grahame Grieve (Feb 08 2019 at 15:26):
but if you say 0..1, someone could always complain.... and 0..* is the outer limit...
Giorgio Cangioli (Feb 08 2019 at 17:20):
sure :-) ... But my intention is to represent as FHIR logical model the current IPS data set as documented in the EN 17269 standard .
Giorgio Cangioli (Feb 08 2019 at 17:21):
So , since in this standard the max cardinality is not defined, I don't want to add in the model details that are not part of the original model as the max cardinality..
Giorgio Cangioli (Feb 08 2019 at 17:22):
and if i put tin the model max = * is not the same of omitting it
Giorgio Cangioli (Feb 08 2019 at 17:23):
.. however my question was how to handle the absence of (max) cardinality with the IG builder :-)
John Moehrke (Feb 08 2019 at 17:41):
and if i put tin the model max = * is not the same of omitting it
I don't agree... sees to me * is explicitly saying it is undefined.
Lloyd McKenzie (Feb 08 2019 at 17:41):
It's not allowed by the methodology. The only reason the cardinality on 'min' and 'max' is allowed to be 0 is for differentials that aren't wanting to tighten the cardinality.
Lloyd McKenzie (Feb 08 2019 at 17:41):
If you want to have an element with unspecified cardinality, the only thing you can do is 0..* for everything.
Last updated: Apr 12 2022 at 19:14 UTC
