Stream: implementers
Topic: Error message from FHIR validator
Katie Wheatley (Aug 04 2018 at 06:44):
When running the official FHIR validator against the attached file, I received this error message: ITK003D FAIL ERROR -- DOCUMENT BUNDLE: Must have either extensions or value[x], not both [extension.exists() != value.exists()] * Location: Bundle.entry[1].resource.extension column: 0 ITK-EDIS-TEST-032a.xml
I am not sure what is wrong with the attached file. Please could anyone with knowledge in this area update me on what could be wrong with the file.
Katie Wheatley (Aug 04 2018 at 07:12):
I have received this error message from the official FHIR validator when running it against the attached file: ERROR -- DOCUMENT BUNDLE: None of the codes provided are in the value set https://fhir.nhs.uk/STU3/ValueSet/CareSettingType-1 (https://fhir.nhs.uk/STU3/ValueSet/CareSettingType-1, and a code from this value set is required) (codes = http://snomed.info/sct#893061000000109) * Location: Bundle.entry[1].resource.extension.valueCodeableConcept column: 0. ITK-EDIS-TEST-032a.xml
I am uncertain as to why this error message is being given as I the code I am using is a member of the specified refset. I attach the valueset file: 032a-20180802151411989718_B02C07.xml ValueSet-CareSettingType-1.xml
Lloyd McKenzie (Aug 04 2018 at 13:30):
What version of the validator are you running? You should be using the version from build.fhir.org.
Katie Wheatley (Aug 06 2018 at 09:14):
What version of the validator are you running? You should be using the version from build.fhir.org.
We are using the HAPI FHIR validator at http://hapifhir.io/.
Katie Wheatley (Aug 06 2018 at 09:16):
Regarding the error message I have posted, please could you explain why that error message appeared and whether there is anything wrong with the file I attached which caused the HAPI validator to throw this error message. Thanks.
Apologies, perhaps I should have entitled this thread Error message from the FHIR HAPI validator.
Kevin Mayfield (Aug 06 2018 at 10:06):
Katie, I think the answer is the valueSet refers to SNOMED Refset which can't be resolved.
Kevin Mayfield (Aug 06 2018 at 10:20):
Screen-Shot-2018-08-06-at-11.18.51.png This is ok but the validator doesn't have a way of resolving 999000381000000107
Grahame Grieve (Aug 06 2018 at 11:14):
hmm. where does that snomed code come from?
Kevin Mayfield (Aug 06 2018 at 11:14):
Uk Refset
Grahame Grieve (Aug 06 2018 at 11:15):
hmm should be supported but it's a while since I updated hte UK dirtibution
Kevin Mayfield (Apr 18 2019 at 07:56):
Using r4 validation instead of stu3, was faster and more thorough.
Not sure how to handle the stu3 profiles, I can convert them structurally to r4 but the contents aren't converted (so for AllergyIntollerance.clinicalStatus it's still a code not coding). We don't have r4 profiles.
I'm also noticing a lot issues with profiles themselves (which the r4 validation is discovering).... so I'm thinking: creating r4 profiles with included profile fixes may be the solution?
Grahame Grieve (Apr 18 2019 at 07:58):
there's no framework for auto-upgrading the profiles to describe a different version. that's manual work
Grahame Grieve (Apr 18 2019 at 07:59):
you should be using the r5 validator now - we've moved onto that.
Kevin Mayfield (Apr 18 2019 at 12:18):
I'm coming across a few SNOMED ecl statements which our terminology server rejects as too costly. (e.g. the one here https://fhir.nhs.uk/STU3/ValueSet/CareConnect-MedicationCode-1 ).
Is their another way of asking 'is this code in this valueset' rather than trying to attempting to expand it?
Kevin Mayfield (Apr 18 2019 at 12:18):
I'm doing this in a validationsupport class
Grahame Grieve (Apr 18 2019 at 12:37):
are you asking about the validator? it should be using $validate-code not $expand
Kevin Mayfield (Apr 18 2019 at 13:10):
That would make more sense, to call direct.
At present I'm calling out to a SNOMED term server to answer ecl query (for the ValueSet)
Grahame Grieve (Apr 18 2019 at 13:10):
well, check - use some http traer. I'm sure it calls $validate-code not $expand
Last updated: Apr 12 2022 at 19:14 UTC
