FHIR Chat · AU IGs on R4 · australia

Stream: australia

Topic: AU IGs on R4

Brett Esler (Jun 26 2019 at 23:02):

Hi all R4 implementation guides are now up for AU Base and AU PD as follows:

• AU Base 3 on R4 : https://build.fhir.org/ig/hl7au/au-fhir-base/ - CI build of content limited as port to R4 continues
• AU PD 2 on R4 : http://build.fhir.org/ig/hl7au/au-fhir-pd/ - CI build using AU Base 3 on R4

Brian Postlethwaite (Jun 27 2019 at 01:52):

Brett and I will be working on these guides for the next week or so to ensure their stability and compatibility with the STU3 version of these guides (not new functionality as yet)

Brian Postlethwaite (Jun 27 2019 at 02:19):

what do we need in the dependency list for the pd base?
(below is what is working for me for the base)

  "dependencyList": [
        {
            "pacakge": "hl7.fhir.au.base",
            "name": "aubase",
            "location": "http://hl7.org.au/fhir/base",
            "version": "current"
        }
  ],

Brian Postlethwaite (Jun 27 2019 at 04:23):

I believe this is what you need to depend on the AU-PD guide

        {
            "pacakge": "hl7.fhir.au.pd",
            "name": "au-pd",
            "location": "http://hl7.org.au/fhir/pd",
            "version": "current"
        }

@Brett Esler is that right?

Brett Esler (Jun 27 2019 at 07:21):

yes that is good - also need something like the following in your ig.xml
<dependsOn id="au-pd">
<uri value="http://hl7.org.au/fhir/pd"/>
<packageId value="hl7.fhir.au.pd"/>
<version value="current"/>
</dependsOn>

Richard Townley-O'Neill (Jul 11 2019 at 01:13):

AUBasePatient does not open in Forge :( . The others are OK.

Brett Esler (Jul 11 2019 at 01:23):

thanks @Richard Townley-O'Neill will investigate...

Grahame Grieve (Jul 11 2019 at 12:19):

I am in Katherine gorge.... an interesting place: I have great telstra signal.... but bandwidth on 4G is measured in bytes/min... I haven't seen that before. Took hours to get Zulip to function....

Brett Esler (Jul 12 2019 at 01:25):

almost disconnected ;)

Brett Esler (Jul 12 2019 at 01:25):

@Richard Townley-O'Neill pull and patient should load now

Brett Esler (Jul 12 2019 at 01:27):

problems with the core extensions conversion to R4 let me know if more issues

Brett Esler (Jul 12 2019 at 01:28):

i.e. was not updated in our subfolder for AU IG

Richard Townley-O'Neill (Jul 12 2019 at 03:37):

Thanks. @Brett Esler Patient is fine.
Now allergyintolerance and medication don't open in Forge.

Brett Esler (Jul 12 2019 at 03:47):

have not done allergy intolerance yet; medication and medicationstatement coming shortly...

Brett Esler (Jul 12 2019 at 04:21):

@Richard Townley-O'Neill believe Medication and MedicationStatement should be good now - this include AU Dosage

Brian Postlethwaite (Jul 22 2019 at 01:35):

ADHA has pointed out that the code system between STU3 and R4 changed, and in R4 au we haven't made a decision on if we should also change the AU terminology to match.
Our R4 valuesets need to be updated to point to these new code system names, but should we change our system value too?
e.g.

{"coding":[{"system":"http://hl7.org/fhir/v2/0203","code":"MC"}]}

=>

{"coding":[{"system":"http://terminology.hl7.org/CodeSystem/v2-0203","code":"MC"}]}

Brett Esler (Jul 22 2019 at 01:47):

So the references to the R4 core published code systems will need the URL change for sure as the references we used on STU3 like http://hl7.org/fhir/v2/0203 don't exist as an R4 version; I am inclined to update the AU 'extension' code system urls also for consistency as the wire protocol is changing anyway... e.g. http://hl7.org.au/fhir/v2/0203 => http://terminology.hl7.org.au/CodeSystem/v2-0203

Brian Postlethwaite (Jul 22 2019 at 01:50):

The odd part is, that this will then make the data in content different beween STU3/R4 for more of the PD content, but do think this is worth while.
@John Carter , @Jared Davison, @Damon Groenveld thoughts on this?

Brian Postlethwaite (Jul 22 2019 at 01:51):

I have been writing a Version Mapper that could be configured into a Facade that could sit in front of an STU3 AuPD server that could handle these changes
(internally we're using it to upgrade all our data)

Brett Esler (Jul 22 2019 at 01:52):

notably there NEEDS to be a change for HPI-I identifier on Practitioner wrt R4 system URL;
changing AU code system URLs would bring changes to all Identifier.type codes wrt 'system' element for PD

Lloyd McKenzie (Jul 22 2019 at 02:29):

Your data won't be terribly interoperable with anyone else's if you don't migrate.

Brian Postlethwaite (Jul 22 2019 at 02:32):

The core values, yes we do need to migrate, however the others that are AU specific terminologies, they just "look" like the core ones for the extended values, except have the .au added into them.
(and the Provider directly has only 2 values in the core where the problem truely exists)

Brian Postlethwaite (Jul 22 2019 at 02:33):

(and they are AU only identifiers too)

Brett Esler (Jul 22 2019 at 03:42):

I think consistency into the future wins i.e. align with R4 style http://hl7.org.au/fhir/v2/0203 => http://terminology.hl7.org.au/CodeSystem/v2-0203 unless there is a REALLY REALLY big issue with that; we are at the stage of the PD project works where this can happen; later on there will be much greater resistance

Jared Davison (Jul 22 2019 at 09:41):

I don't know the rationale why the change was made (it seems like change for the sake of it - but that doesn't really matter). If the fact is that the change has been made, and if this aspect of R4 is normative we should make the change and not entrench the old way. Hopefully wire level changes like this won't continue.

Brian Postlethwaite (Jul 22 2019 at 12:56):

The reason the change happened was that the terminology is being used outside fhir, so removing it from the base of all HL7 defined code systems made sense, and just prior to R4 going normative was the last opportunity to do it.
The only other reason for me to not change the other au terminology over to this is that the old one was soo close to the other, and in part why we missed the change.
We have to change the core system values, but don't have to for the AU systems.

Brian Postlethwaite (Jul 22 2019 at 12:56):

(hope that all made sense)

Brett Esler (Jul 23 2019 at 01:56):

ok will be changing today

Brett Esler (Jul 24 2019 at 07:29):

@Grahame Grieve is this error part of snapshot ids issues or something else; only occurs for complex-type extension definitions with child extension http://build.fhir.org/ig/hl7au/au-fhir-base/branches/master/build.log
was building under SNAPSHOT 42 of ig-builder but not now on v23 in CI build

Brett Esler (Jul 24 2019 at 07:32):

apologies to those looking for AU Base on R4; it is currently offline so this publishing issue can be assessed

Brett Esler (Jul 24 2019 at 11:40):

the extension structuredefinition-au-receivingapplication is causing the crash

Grahame Grieve (Jul 26 2019 at 21:18):

which error?

Brett Esler (Jul 29 2019 at 13:48):

java.lang.NumberFormatException: For input string: "0][url='http://hl"
at java.base/java.lang.NumberFormatException.forInputString(NumberFormatException.java:68)
at java.base/java.lang.Integer.parseInt(Integer.java:658)
at java.base/java.lang.Integer.parseInt(Integer.java:776)
at org.hl7.fhir.utilities.Utilities.fhirPathToXPath(Utilities.java:1100)
at org.hl7.fhir.r5.utils.OperationOutcomeUtilities.convertToIssue(OperationOutcomeUtilities.java:46)
at org.hl7.fhir.igtools.renderers.ValidationPresenter.generate(ValidationPresenter.java:142)
at org.hl7.fhir.igtools.publisher.Publisher.createIg(Publisher.java:743)
at org.hl7.fhir.igtools.publisher.Publisher.execute(Publisher.java:612)
at org.hl7.fhir.igtools.publisher.Publisher.main(Publisher.java:6208)
Exception in thread "main" java.lang.NullPointerException
at org.hl7.fhir.igtools.publisher.Publisher.main(Publisher.java:6218)

Brett Esler (Jul 29 2019 at 13:49):

added back the first 3 references in ig.xml (extensions) which seem to cause the issuse

Brett Esler (Jul 29 2019 at 13:59):

i

Brett Esler (Jul 29 2019 at 13:59):

http://build.fhir.org/ig/hl7au/au-fhir-base/branches/master/build.log

Grahame Grieve (Jul 29 2019 at 22:06):

the right error would be :

Grahame Grieve (Jul 29 2019 at 22:06):

The FHIRPath expression 'Endpoint/example0: Endpoint.extension[0][url='http://hl7.org.au/fhir/StructureDefinition/au-receivingfacility']' is not valid

Brett Esler (Jul 30 2019 at 02:45):

thanks @Grahame Grieve CI is building now with example references to extensions removed

Brett Esler (Jul 30 2019 at 03:28):

Hi @Grahame Grieve I can't get these examples to work and can;t work out why; practitionerrole-example4.xml is causing the fail and /au-assigningauthority is the referenced extension

Brett Esler (Jul 30 2019 at 03:30):

if I try and add a fixed Extension.url element defintion into the au-assigningauthority definiton I get a warning about mismatch in snapshot genration; not sure if that is related to the example generating an error, fixd Extension.url is not defined in au-assigningauthority extension at the moment to avoid the warning

Brett Esler (Jul 30 2019 at 03:31):

@Brian Postlethwaite any ideas?

Grahame Grieve (Jul 30 2019 at 03:38):

I might have to debug

Brett Esler (Jul 30 2019 at 13:18):

Not sure what the errors mean... I am getting a warning for each of the complex extensions like
Error in snapshot generation: Differential for http://hl7.org.au/fhir/StructureDefinition/au-receivingapplication with id: Extension.value[x] has an element that is not marked with a snapshot match

Brett Esler (Jul 30 2019 at 13:19):

then get ublishing Content Failed: The FHIRPath expression 'Endpoint/example0: Endpoint.extension[0][url='http://hl7.org.au/fhir/StructureDefinition/au-receivingfacility']' is not valid (00:01:12.0566sec)

Brett Esler (Jul 30 2019 at 13:21):

not sure what the FHIRPath expression is referring to... if remove the complex extensions from the examples then will not crash

Brett Esler (Jul 30 2019 at 13:21):

other simple extensions are fine

Grahame Grieve (Jul 30 2019 at 22:31):

I'm not seeing "Extension.value[x] has an element that is not marked with a snapshot match"

Brett Esler (Jul 30 2019 at 23:23):

it is around line 62 of http://build.fhir.org/ig/hl7au/au-fhir-base/branches/master/build.log there are 3 one for each complex extension

Brett Esler (Jul 30 2019 at 23:24):

full line is Error in snapshot generation: Differential for http://hl7.org.au/fhir/StructureDefinition/au-receivingapplication with id: Extension.value[x] has an element that is not marked with a snapshot match

Grahame Grieve (Jul 30 2019 at 23:52):

oh the log

Grahame Grieve (Jul 31 2019 at 01:03):

fixed next release - sometime today

Brett Esler (Jul 31 2019 at 02:01):

thanks will keep an eye out

Last updated: Apr 12 2022 at 19:14 UTC