FHIR Chat · build failing · committers/notification

Stream: committers/notification

Topic: build failing


view this post on Zulip Rob Hausam (Mar 21 2019 at 05:19):

My local build is now failing with a bunch of errors like this due to failing to validate NDC codes in the medication examples:

     [java] ERROR @ MedicationStatement.contained.code.coding (line 0, col0): medicationstatementexample2:: MedicationStatement.contained.code.coding: The code "50580-506-02" is not valid in the system http://hl7.org/fhir/sid/ndc; The code provided (http://hl7.org/fhir/sid/ndc#50580-506-02) is not  valid in the value set All codes known to the system (src = TerminologyEngine)
     [java] ERROR @ MedicationStatement.medicationReference.contained.code.coding (line 0, col0): medicationstatementexample2:: MedicationStatement.medicationReference.contained.code.coding: The code "50580-506-02" is not valid in the system http://hl7.org/fhir/sid/ndc; The code provided (http://hl7.org/fhir/sid/ndc#50580-506-02) is not  valid in the value set All codes known to the system (src = TerminologyEngine)
     [java] ERROR @ MedicationStatement.contained.code.coding: medicationstatementexample2:: MedicationStatement.contained.code.coding: The code "50580-506-02" is not valid in the system http://hl7.org/fhir/sid/ndc; The code provided (http://hl7.org/fhir/sid/ndc#50580-506-02) is not  valid in the value set All codes known to the system (src = TerminologyEngine)
     [java] ERROR @ MedicationStatement.medicationReference.contained.code.coding: medicationstatementexample2:: MedicationStatement.medicationReference.contained.code.coding: The code "50580-506-02" is not valid in the system http://hl7.org/fhir/sid/ndc; The code provided (http://hl7.org/fhir/sid/ndc#50580-506-02) is not  valid in the value set All codes known to the system (src = TerminologyEngine)
     [java] This was a Full Build                                                      41.945 2579sec 1474MB
     [java] org.hl7.fhir.tools.publisher.ExampleInspector$EValidationFailed: Resource Examples failed instance validation
     [java]     at org.hl7.fhir.tools.publisher.ExampleInspector.summarise(ExampleInspector.java:400)
     [java]     at org.hl7.fhir.tools.publisher.Publisher.validationProcess(Publisher.java:5860)
     [java]     at org.hl7.fhir.tools.publisher.Publisher.execute(Publisher.java:629)
     [java]     at org.hl7.fhir.tools.publisher.Publisher.main(Publisher.java:482)
     [java] FHIR build failure @ Wed, Mar 20, 2019 23:37-0500                          0.847 2580sec 1474MB
     [java] Error: Resource Examples failed instance validation

BUILD FAILED
/Users/rhausam/git-repo/fhir/build.xml:30: The following error occurred while executing this line:
/Users/rhausam/git-repo/fhir/tools/java/org.hl7.fhir.tools.core/build.xml:81: Java returned: 1

view this post on Zulip Grahame Grieve (Mar 21 2019 at 05:24):

well, are they incorrect?

view this post on Zulip Grahame Grieve (Mar 21 2019 at 05:27):

should be 50580-608-02

view this post on Zulip Rob Hausam (Mar 21 2019 at 05:30):

Yes, probably so. 50580-506-02 doesn't exist, but 50580-608-02 does.

view this post on Zulip Rob Hausam (Mar 21 2019 at 05:36):

there are at least several other invalid codes (not sure how many) - which I assume you've seen

view this post on Zulip Grahame Grieve (Mar 21 2019 at 05:50):

all the one code many times

view this post on Zulip Rob Hausam (Mar 21 2019 at 06:10):

I'm seeing several different codes that are invalid, like 0264751010 and 2501-813-16 (there are probably others). 16590-619-30 shows in the NDC browser as "Records marked with (E): This information was removed from publication, because FDA has found inaccuracy in the data submitted by the firm. Data will be released for publication once complete and accurate information is submitted to FDA." and it isn't found on tx.fhir.org (not surprisingly).

view this post on Zulip Rob Hausam (Mar 21 2019 at 06:11):

0069-2587-10 is another not valid one

view this post on Zulip Grahame Grieve (Mar 21 2019 at 20:32):

I think I've committed fixes for all of them. @Melva Peters about 25% of the NDC codes were invalid, and I've and I've gone through and picked the nearest code I could find, but I'm not sure I picked the correct NDC codes

view this post on Zulip Grahame Grieve (Mar 21 2019 at 20:32):

@Rob Hausam the documentation for display : "Use the PACKAGEDESCRIPTION column value from the TSV or Excel distribution file" is entirely useless and wrong

view this post on Zulip Grahame Grieve (Mar 21 2019 at 20:33):

from http://hl7.org/fhir/ndc.html

view this post on Zulip Rob Hausam (Mar 21 2019 at 20:34):

sorry, @Grahame Grieve, which documentation?

view this post on Zulip Rob Hausam (Mar 21 2019 at 20:34):

oh, sorry - see it now

view this post on Zulip Rob Hausam (Mar 21 2019 at 20:38):

yes, that's entirely inadequate, but the PACKAGEDESCRIPTION is (obviously) an important component when you are dealing with a package code
there is no "right" answer to this, but in looking at it over the past few days I think I have a pretty good idea of what an algorithmic approach to generating a useable display string could be

view this post on Zulip Grahame Grieve (Mar 21 2019 at 20:39):

I made one too - trace name + suffix + package description. But that's not great

view this post on Zulip Grahame Grieve (Mar 21 2019 at 20:40):

happy to change mine

view this post on Zulip Melva Peters (Mar 21 2019 at 21:52):

I'm in the middle of fixing as well - I was running a local build today

view this post on Zulip Melva Peters (Mar 21 2019 at 21:52):

Thanks @Grahame Grieve

view this post on Zulip Grahame Grieve (Mar 21 2019 at 22:06):

well, my changes are in and built but feel free to overwrite if you have better code choices


Last updated: Apr 12 2022 at 19:14 UTC