FHIR Chat · X/FHIRPath for changing cardinality · committers

Stream: committers

Topic: X/FHIRPath for changing cardinality


view this post on Zulip David Pyke (Feb 22 2018 at 18:47):

If one of those who understand XPath and FHIRPath could write the invariants for forcing Consent.patient to have cardinality of 1..1 if Consent.scope is "privacy" I would buy them a beer or two in May...

view this post on Zulip David Pyke (Feb 22 2018 at 18:49):

Or other beverage (or cheesecake) if preferred.

view this post on Zulip Lloyd McKenzie (Feb 22 2018 at 18:57):

XPath: exists(f:patient) or f:scope!='privacy'
FHIRPath: patient.exists() or scope='privacy'

view this post on Zulip Lloyd McKenzie (Feb 22 2018 at 18:57):

Cheapest piece of cheesecake ever ;)

view this post on Zulip Lloyd McKenzie (Feb 22 2018 at 18:57):

We can discuss why privacy is only relevant to patients while I eat... :)

view this post on Zulip David Pyke (Feb 22 2018 at 19:02):

It doesn't but I can clone that for research, ADR and treatment...

view this post on Zulip Lloyd McKenzie (Feb 22 2018 at 20:14):

Will you have extensions that allow the subject to be something else?

view this post on Zulip Brian Postlethwaite (Feb 23 2018 at 10:45):

At the moment patient is 1..1, hoping that means you're relaxing the cardinality for use on others.
But seems weird that if we use it on a different subject then the scope can't be privacy

view this post on Zulip John Moehrke (Feb 23 2018 at 13:38):

I am sure Dave was just being short with his words... he means when the scope is -a patient privacy consent-, not all forms of privacy.

view this post on Zulip Lloyd McKenzie (Feb 23 2018 at 15:53):

Might be good to make the code "pat-privacy" or something then...

view this post on Zulip Lloyd McKenzie (Feb 23 2018 at 15:53):

Or following FHIR convention "patient-privacy"

view this post on Zulip David Pyke (Feb 23 2018 at 17:20):

We are opening patient to be omitted to handle rian's use case.

view this post on Zulip David Pyke (Feb 23 2018 at 17:21):

WE have made patient 0..1 to allow for @Brian Postlethwaite 's use case of submission to a registry. However, at this time, patient is a patient.

view this post on Zulip David Pyke (Feb 23 2018 at 17:22):

I can do that change to the valueset, I'll make the change soon.

view this post on Zulip David Pyke (Feb 23 2018 at 17:24):

At the moment patient is 1..1, hoping that means you're relaxing the cardinality for use on others.
But seems weird that if we use it on a different subject then the scope can't be privacy

I'll change the scope valueset entry to patient-privacy to make it clear. It's now a codableconcept to allow for other scopes.

view this post on Zulip Brian Postlethwaite (Feb 25 2018 at 12:51):

Thanks David, excellent.


Last updated: Apr 12 2022 at 19:14 UTC