Stream: IG creation
Topic: meta.profile error
Sarah Gaunt (Jan 17 2020 at 02:15):
New error:
Value is 'http://hl7.org/fhir/us/bser/StructureDefinition/BSeR-EarlyChildhoodNutritionFeedbackObservation' but must be 'http://hl7.org/fhir/us/bser/StructureDefinition/BSeR-EarlyChildhoodNutritionObservation'
Getting the above error in my example - this example is supposed to be "BSeR-EarlyChildhoodNutritionFeedbackObservation" but it's telling me it should be for a different profile.
I have two examples for this profile - they both have the same code and very similar values, but only one is erroring.
I'm also getting
The type of element valueCodeableConcept is not known, which is illegal. Valid types at this point are string, boolean
on the same example. That's because it thinks it should conform to the other profile which only allows string and boolen for the value.
Grahame Grieve (Jan 17 2020 at 02:19):
is this one what's in gihtub?
Sarah Gaunt (Jan 17 2020 at 02:22):
It is in github - though github is a super old copy. I'm running locally right now because I'm trying to get the new template working.
But the example hasn't changed: https://github.com/HL7/bser/blob/master/source/resources/observation/observation-BSeR-NutritionFeedbackObs-2-eve-everywoman.xml
This is the other one that isn't failing: https://github.com/HL7/bser/blob/master/source/resources/observation/observation-BSeR-NutritionFeedbackObs-1-eve-everywoman.xml
I'll rebuild the IG as it is in Git and see if it gets the same error.
Sarah Gaunt (Jan 17 2020 at 02:23):
I guess I should really just bite the bullet and switch totally over to the new template for that IG.
Grahame Grieve (Jan 17 2020 at 02:47):
what is in github doesn't build: You must nominate a template - consult the IG Publisher documentation
Sarah Gaunt (Jan 17 2020 at 02:52):
Yeah, I broke it all when I just deleted everything and tried to move over to the new template. There is a template nominated. I'll get back to you when I get it building again.
Sarah Gaunt (Jan 17 2020 at 03:00):
And that's why:
<name value="observation-BSeR-NutritionFeedbackObs-2-eve-everywoman"/> <exampleCanonical value="http://hl7.org/fhir/us/bser/StructureDefinition/BSeR-EarlyChildhoodNutritionObservation"/>
Totally my fault... Guess I shoulda checked that first instead of assuming it was your fault @Grahame Grieve !
Grahame Grieve (Jan 17 2020 at 03:05):
I was hoping that would be the case
Grahame Grieve (Jan 17 2020 at 03:05):
that's probably something that wasn't checked before
Sarah Gaunt (Jan 17 2020 at 03:06):
Nope, it definitely wasn't. In my defense the profile names in this IG are REALLY hard to tell apart!! And so long...
Sarah Gaunt (Jan 17 2020 at 03:07):
Did you fix the issue with the bundle slicing? Don't seem to be seeing those errors anymore... Does that mean I can try removing my meta.profile hack?
Sarah Gaunt (Jan 17 2020 at 03:11):
Actually maybe the budle slicing issues just went away because I just overwrote my test files in git that I was using to force the error...
Grahame Grieve (Jan 17 2020 at 03:11):
I haven't fixed all issues with bundle slicing but I fixed some
Grahame Grieve (Jan 17 2020 at 03:11):
you shouldn't need meta.profile now
Sarah Gaunt (Jan 17 2020 at 03:12):
Ok will test now.
Sarah Gaunt (Jan 17 2020 at 03:12):
So the discriminator should be type/resource?
Grahame Grieve (Jan 17 2020 at 03:13):
the test case looks like this:
<element> <path value="Bundle.entry" /> <slicing> <discriminator> <type value="profile" /> <path value="$this.resource" /> </discriminator> <ordered value="false"/> <rules value="closed" /> </slicing> </element>
Sarah Gaunt (Jan 17 2020 at 03:13):
tx
Grahame Grieve (Jan 17 2020 at 03:13):
but it's simpler and quicker to differentiate by type if you can
Sarah Gaunt (Jan 17 2020 at 03:13):
Ok
Sarah Gaunt (Jan 17 2020 at 03:20):
Looks like that is working now - cool - thanks!
Last updated: Apr 12 2022 at 19:14 UTC
