Stream: IG creation
Topic: Error with examples using external codes
Jean Duteau (Nov 17 2020 at 21:09):
I'm getting a number of errors of the following form:
None of the codes provided are in the value set http://hl7.org/fhir/us/davinci-pas/ValueSet/X12278RequestedServiceType (http://hl7.org/fhir/us/davinci-pas/ValueSet/X12278RequestedServiceType), and a code from this value set is required) (codes = http://codesystem.x12.org/005010/1525#HS
I've defined value sets with multiple codes from multiple code systems and one of those is an X12 code system that the validator can't access. Unfortunately with these errors, I can't suppress them. I have that exact message in my ignoreWarnings file but it still appears.
http://build.fhir.org/ig/HL7/davinci-pas/branches/master/qa.html if you want to see the results.
Sarah Gaunt (Nov 17 2020 at 21:17):
Have you tried copying the whole message from http://build.fhir.org/ig/HL7/davinci-pas/branches/master/qa.txt (you might have, I didn't check, but I was having issues until I used that file).
Jean Duteau (Nov 17 2020 at 21:22):
i'll try that. i've had success with just copying from the qa.html
Sarah Gaunt (Nov 17 2020 at 21:24):
I seem to have varying success with just the shorter message from the html, but it seems to consistently work if I copy the whole line including WARNING, etc. from the text file. It does mean you need more lines in your suppression file though because they are profile/instance specific.
Jean Duteau (Nov 17 2020 at 21:25):
you copied the entire line including ERROR and the list where it is found? I haven't done that with any of my other error/warning/info messages.
Jean Duteau (Nov 17 2020 at 21:27):
yeah, that didn't fix it :)
Lloyd McKenzie (Nov 17 2020 at 21:29):
Errors (except for possibly link issues) aren't suppressible
Jean Duteau (Nov 17 2020 at 21:35):
duouh! I think I even knew that.
Sarah Gaunt (Nov 17 2020 at 21:36):
Which is weird - usually those external value set issues are warnings, not errors.
Jean Duteau (Nov 17 2020 at 21:36):
so never mind the "can't suppress" these part of my stream. this seems like a mistake in the publisher validation routine since it can't determine what's in the value set, so it can't really validate the code that I give it.
Lloyd McKenzie (Nov 17 2020 at 21:45):
Right. I think it's a publisher issue. This problem shouldn't be an error
Grahame Grieve (Nov 18 2020 at 10:30):
well, here's a case in point:
Grahame Grieve (Nov 18 2020 at 10:31):
<valueCodeableConcept>
<coding>
<system value="http://codesystem.x12.org/005010/1525"/>
<code value="HS"/>
<display value="Health Services Review"/>
</coding>
</valueCodeableConcept>
Grahame Grieve (Nov 18 2020 at 10:31):
the value set in question says:
Grahame Grieve (Nov 18 2020 at 10:31):
<compose>
<include>
<system value="http://codesystem.x12.org/005010/1365"/>
</include>
<include>
<system value="http://www.ama-assn.org/go/cpt"/>
</include>
<include>
<system value="http://www.cms.gov/Medicare/Coding/HCPCSReleaseCodeSets"/>
</include>
<include>
<system value="http://terminology.hl7.org/CodeSystem/icd9cm"/>
</include>
<include>
<system value="http://www.cms.gov/Medicare/Coding/ICD10"/>
</include>
Grahame Grieve (Nov 18 2020 at 10:32):
so the error sure looks correct to me
Grahame Grieve (Nov 18 2020 at 10:34):
if you point me at a place where the validator is wrong, I'll be happy to look into it
Jean Duteau (Nov 18 2020 at 19:01):
oops, most of those errors were my fault. I think i've got it down to an actual error. Same value set:
<compose>
<include>
<system value="http://codesystem.x12.org/005010/1365"/>
</include>
<include>
<system value="http://www.ama-assn.org/go/cpt"/>
</include>
<include>
<system value="http://www.cms.gov/Medicare/Coding/HCPCSReleaseCodeSets"/>
</include>
<include>
<system value="http://terminology.hl7.org/CodeSystem/icd9cm"/>
</include>
<include>
<system value="http://www.cms.gov/Medicare/Coding/ICD10"/>
</include>
And here is my element:
<code>
<coding>
<system value="http://codesystem.x12.org/005010/1365"/>
<code value="3"/>
<display value="Consultation"/>
</coding>
</code>
ServiceRequest - Referral Request Example
Claim-Referral Authorization Example
are the two example instances that have the error.
Jean Duteau (Nov 18 2020 at 19:12):
Hmm, that's weird. While I get errors in my local build, the CI build has no errors other than a JIRA-Spec-Artifacts error that I don't get on my local build.
Grahame Grieve (Nov 18 2020 at 22:59):
clear your tx-cache?
Patrick Werner (Dec 01 2020 at 16:02):
Lloyd McKenzie said:
Errors (except for possibly link issues) aren't suppressible
The CG IG has several tx related errors due to tx.fhir.org (probably the codes in question are newer than the loinc version on tx.fhir.org)?
Patrick Werner (Dec 01 2020 at 16:03):
As errors arent' filterable, how should we deal with them?
Patrick Werner (Dec 01 2020 at 16:13):
If i try the validation of these LOINC manually (Postman) it seems to work:
http://tx.fhir.org/r4/CodeSystem/$lookup?system=http://loinc.org&code=92822-6works but throws errors during IG build:
The code "92822-6" is not valid in the system http://loinc.org; The code provided (http://loinc.org#92822-6) is not valid in the value set 'All codes known to the system' (from http://tx.fhir.org/r4) for 'http://loinc.org#92822-6'
Rob Hausam (Dec 01 2020 at 16:28):
I agree that the $lookup is working correctly on the server. Is this possibly a cache issue?
Patrick Werner (Dec 01 2020 at 16:33):
hmm the error is reported by the CI build which should reset the tx cache every build?
Rob Hausam (Dec 01 2020 at 16:37):
Actually, I don't think the CI build automatically resets the cache. If a file for the code system is checked in for txcache I believe it will use it. This also returns 'true': http://tx.fhir.org/r4/ValueSet/$validate-code?url=http://loinc.org/vs&system=http://loinc.org&code=92822-6
Lloyd McKenzie (Dec 01 2020 at 16:46):
Grahame manually purges the cache before official publication, but the CI build generally uses the cache - for performance reasons and to keep the load on tx.fhir.org reasonable.
Rob Hausam (Dec 01 2020 at 16:48):
Yes.
Eric Haas (Dec 01 2020 at 17:06):
so is the autobuild build using a cache each time? is there a way to clear it prior to the next build?
Patrick Werner (Dec 01 2020 at 17:26):
Rob Hausam said:
Actually, I don't think the CI build automatically resets the cache. If a file for the code system is checked in for txcache I believe it will use it. This also returns 'true':
http://tx.fhir.org/r4/ValueSet/$validate-code?url=http://loinc.org/vs&system=http://loinc.org&code=92822-6
oh! Thanks for the info. Will try to reset the cache locally.
Grahame Grieve (Dec 01 2020 at 20:39):
I am considering asking everyone to remove txcache from git. the teminology system is more robust than it used to be, and the cache is starting to be a negative value. I have some internal metric goals to meet before doing this
Patrick Werner (Dec 02 2020 at 09:48):
Ah, so there isn't a reused server-side cache, but the tx cache folder is also used by the CI build?
Last updated: Apr 12 2022 at 19:14 UTC
