Stream: IG creation
Topic: VhDir IG
Brian Postlethwaite (Aug 17 2018 at 10:37):
Also wondering why we are cloning the existing search parameters, and not just indicating that we support those - don't want to redefine them. Or have you just been cleaning up what was already in there?
Brian Postlethwaite (Aug 17 2018 at 10:43):
I'm building locally too, have done the merge, just taking my version of the IG.xml and IG.json so will see how that goes.
Eric Haas (Aug 17 2018 at 10:51):
I have created searchparameters extending the core sp wherever alex indicated an modifier etc in the spreadsheet. where there are no changes I just list the url..and I use the same code when is based on existing . This was a lot more effort than I anticipated. there are 140 searchparameters because of this.
Brian Postlethwaite (Aug 17 2018 at 11:59):
argh. Yuk.
Has to be a better way. Not for today though.
Brian Postlethwaite (Aug 17 2018 at 12:01):
The guide should now build fine as long as you haven't updated to Grahame's updated version.
There are however a few narrative ones that I don't think I quite merged correctly that are appearing in the build errors on the qa.html. Should be pretty easy to pick.
Note: There was one OrgAffil search param that had an ID that was tooo long, so I had to shorten things, should be pretty obvious in the history which one that is.
Brian Postlethwaite (Aug 17 2018 at 12:41):
I'm done for the day on this one. Whoever was working on the payer characteristics valueset/codesystem, its a bit of a mess and needs some love.
Brian Postlethwaite (Aug 17 2018 at 12:44):
Specifically that casing of menthlth vs mentHlth
Brian Postlethwaite (Aug 17 2018 at 12:46):
Adn @Eric Haas , if you get back to it, these come up in the build log, but not in errors, don't know if its effecting anything you were doing.
Liquid Warning: Liquid syntax error (line 209): Unexpected character { in "{{site.data.structuredefinitions.vhdir-practitioner.kind}} == "logical"" in StructureDefinition-vhdir-practitioner.html[0m (55.0123sec)
Dan Chaput (Aug 17 2018 at 13:46):
Dan here. I think I need to just rebase my code. I am not sure I merger correctly. Vocab only.
Dan Chaput (Aug 17 2018 at 13:49):
I'll clean-up payor characteristics.
Alex Kontur (Aug 17 2018 at 13:52):
I think pointing at the core parameter definitions potentially creates versioning issues down the road. It means all the new parameters we're defining are going to be fixed to R4, but any parameters that we're borrowing from core are going to be subject to future changes in definition. Or maybe I'm misunderstanding how things work?
Alex Kontur (Aug 17 2018 at 13:53):
We face a similar question with value sets as well.
Alex Kontur (Aug 17 2018 at 14:43):
Weird formatting issue on the organization profile intro.md: the asterisk for organization.type (1..*) disappears and all of the rest of the text is changed to italics
Eric Haas (Aug 17 2018 at 21:36):
@Brian Postlethwaite I looked at the Liquid Warning and can't see anything obvious right now and since is not a build breaking thingy need to move on. Thanks
Brian Postlethwaite (Aug 17 2018 at 22:19):
Thanks Eric.
Brian Postlethwaite (Aug 17 2018 at 22:21):
Reviewing the QA page http://build.fhir.org/ig/HL7/VhDir/qa.html , there are still issues with the valuesets
- insuranceplantype
- validation
Who's looking into these?
Where did we land with the telecom issue?
(I'll complete the missing couple of examples this morning after breakfast)
Grahame Grieve (Aug 17 2018 at 22:23):
looking at that - you should really fix the names...
Grahame Grieve (Aug 17 2018 at 22:23):
both the value set errors look like wrong code system urls to me
Brian Postlethwaite (Aug 17 2018 at 22:32):
I'm sure that's what they are
Dan Chaput (Aug 17 2018 at 22:36):
I'm working on it.
Dan Chaput (Aug 17 2018 at 22:37):
Just hit insurance plan type, a bunch of validation was missing.
Brian Postlethwaite (Aug 17 2018 at 22:37):
Thanks Dan. I'll get on the examples after my breakfast.
Dan Chaput (Aug 17 2018 at 22:39):
Seems like the terminology server is down
Connect to Terminology Server at http://tx.fhir.org/r4 (26.0012sec)
-tx: Connect to http://tx.fhir.org/r4
Publishing Content Failed: Error sending Http Request: Read timed out (00:01:01.0357sec)
(00:01:01.0358sec)
Use -? to get command line help (00:01:01.0358sec)
Grahame Grieve (Aug 17 2018 at 22:43):
restarting it now
Grahame Grieve (Aug 17 2018 at 22:49):
back
Alex Kontur (Aug 17 2018 at 23:11):
Did consent ever get changed or are we putting it though in the current state?
Dan Chaput (Aug 17 2018 at 23:16):
Quick one, fixed the obvious errors - what does "/f:CodeSystem error Text should not be present" refer to:
C:\fhir\source\resources\codesystem-validation.xml
Path Severity Message
/f:CodeSystem error Text should not be present
CodeSystem/codesystem-validation: CodeSystem warning A resource should have narrative for robust management [text.div.exists()]
Alex Kontur (Aug 17 2018 at 23:30):
Is there a way to change element names in a profile?
Grahame Grieve (Aug 17 2018 at 23:36):
no
Lloyd McKenzie (Aug 17 2018 at 23:43):
You can add additional aliases to the documentation, but you can't change what goes over the wire. Note that the names used to display content to users aren't governed by FHIR can can be whatever you like - and may well vary from system to system.
Brian Postlethwaite (Aug 18 2018 at 14:32):
Hey Team, the IG is almost ready, there are a couple of examples that need to be added and some broken links and the qa.html looks pretty clean (warnings aside).
I cleaned up all the other issues I could see.
(And I still need to do the boundary contains search parameter)
Brian Postlethwaite (Aug 18 2018 at 14:33):
IG.xml and IG.json are in sync, everything in one is in the other (wrote a tool to check it)
Brian Postlethwaite (Aug 19 2018 at 02:31):
Hey all, The IG is now looking good. No errors, and only 3 broken links.
I am working on the location contains search parameter right now, and that should be it for the guide.
A good solid last minute dash to the finish for everyone. Thanks @Alex Kontur @Daniel Chaput @Eric Haas
Brian Postlethwaite (Aug 19 2018 at 03:39):
Location Contains search parameter defined, and other Location search parameters fixed - others still need work - but that will be post ballot.
Only 1 broken link (to an example that doesn't exist)
Happy to call that ready now.
Grahame Grieve (Aug 19 2018 at 21:04):
so is this good to go now?
Grahame Grieve (Aug 19 2018 at 21:50):
@Brian Postlethwaite I need change notes for this IG
Brian Postlethwaite (Aug 19 2018 at 22:21):
Yes, good to go. There is a change-notes markdown file, i think it was updated, but I'll check.
Grahame Grieve (Aug 19 2018 at 22:24):
also, I see a lot of jekyll errors processing this IG
Brian Postlethwaite (Aug 19 2018 at 22:26):
http://build.fhir.org/ig/HL7/VhDir/source/pages/change-notes.md
But i think these need some updating now.Since the "for comment" ballot in January 2018, the guide has been extensively refined, and incomplete content added. Feedback from the ballot has been I corporates during the updates.
Grahame Grieve (Aug 19 2018 at 22:27):
well, can you commit and update?
Brian Postlethwaite (Aug 19 2018 at 22:27):
The Jekyl stuff are warnings to my knowledge, @eric looked at them and they are ok. If it's the same stuff, they don't effect the output.
Grahame Grieve (Aug 19 2018 at 22:27):
ok
Brian Postlethwaite (Aug 19 2018 at 22:27):
Will do.
Grahame Grieve (Aug 20 2018 at 00:55):
so the md link above doesn't actually have change notes... let me know when it does
Grahame Grieve (Aug 20 2018 at 01:10):
I see one broken link?
Rob Hausam (Aug 20 2018 at 01:11):
I didn't mean to just use GForge - have to start with the getting the list of tasks, though
Brian Postlethwaite (Aug 20 2018 at 01:12):
Can we live with the 1 broken link in the IG?
(its to an example that doesn't exist, and want to resolve it post ballot - rather than remove the link)
Grahame Grieve (Aug 20 2018 at 01:12):
ok
Grahame Grieve (Aug 20 2018 at 01:13):
is it a known issue anywhere?
Brian Postlethwaite (Aug 20 2018 at 01:16):
Not yet I probably should have logged it as a tracker, but only realized in the last few days once all the other issues were cleaned out.
Grahame Grieve (Aug 20 2018 at 01:17):
it's not too late to put a note somewhere in the IG if you want to
Brian Postlethwaite (Aug 20 2018 at 01:17):
Not stressed on this one.
Grahame Grieve (Aug 20 2018 at 01:18):
k
Grahame Grieve (Aug 20 2018 at 01:19):
vhdir is go then
Grahame Grieve (Aug 20 2018 at 01:19):
just need the history notes
Brian Postlethwaite (Aug 20 2018 at 01:19):
APart from the change notes that you're looking for. Doing that now.
Brian Postlethwaite (Aug 20 2018 at 01:39):
Change notes updated, if you're happy with that, good to go.
Brian Postlethwaite (Aug 20 2018 at 01:40):
There is FHIR version 3.4 in lots of resources, should I be updating those to 3.5 now that its ready?
Brian Postlethwaite (Aug 20 2018 at 01:41):
And do you do the replace of all the links from build.fhir.org to 2018Sep?
Grahame Grieve (Aug 20 2018 at 01:43):
where are you seeing 3.4.0?
Grahame Grieve (Aug 20 2018 at 01:44):
hmm plenty of places. I'll check
Brian Postlethwaite (Aug 20 2018 at 01:44):
http://build.fhir.org/ig/HL7/VhDir/CapabilityStatement-vhdir-server.xml.html
(and I expect in the other structure defintions)
I'm happy to do the search and replace.
Brian Postlethwaite (Aug 20 2018 at 01:59):
Updated 3.4.0 to 3.5.0 in the IG now.
Brian Postlethwaite (Aug 20 2018 at 02:02):
There is a new warning appearing in the build too, is this something to do with the new Tx server?
ValueSet http://terminology.hl7.org/ValueSet/v2-2.7-0360 not found
Brian Postlethwaite (Aug 20 2018 at 02:03):
From my perspective, VhDir IG is ready to go (provided the links to build.fhir.org are updated)
Grahame Grieve (Aug 21 2018 at 02:52):
@Brian Postlethwaite let me know ASAP if this shouldn't be v0.2.0 of VHDir
Brian Postlethwaite (Aug 21 2018 at 02:55):
This is fine. First for comment ballot was 0.1.0
Brian Postlethwaite (Aug 21 2018 at 02:56):
Is that too inconsistent with other IGs?
Grahame Grieve (Aug 21 2018 at 02:58):
no that's what they're all going to be unless someone presents a compelling reason - 0.1.0 or 0.2.0 if they're a repeat
Brian Postlethwaite (Jun 03 2019 at 07:56):
@Grahame Grieve , any chance you could take a 2 second look at the current status of this IG to see if there is anything missing from a publishing standpoint? I'm trying to get the final approved trackers applied, and have done the QA I could.
https://build.fhir.org/ig/HL7/VhDir/qa.html
There is one external link that's failing, and a couple of examples too.
Including one weird warning:
The display "English (United States)" is not a valid display for the code
{urn:ietf:bcp:47}en-US - should be one of ["English (Region=United States)"]
And also a code that is in the core spec, but doesn't seem to be resolving for some reason
Unknown Code http://terminology.hl7.org/CodeSystem/verificationresult-communication-method#owner in http://terminology.hl7.org/CodeSystem/verificationresult-communication-method
Brian Postlethwaite (Jun 03 2019 at 08:49):
And while you're in there, I updated all the Canonical URIs to be consistent, can you take a check to see if you're happy with them?
Grahame Grieve (Jun 04 2019 at 04:13):
what is http://build.fhir.org/ig/HL7/bulk-data-export supposed to reference?
Grahame Grieve (Jun 04 2019 at 04:16):
I don't see "owner" in the specified code system?
Grahame Grieve (Jun 04 2019 at 04:17):
the canonicals look ok
Brian Postlethwaite (Jun 04 2019 at 04:54):
That is the current CI build of the bulk data IG
Brian Postlethwaite (Jun 04 2019 at 04:55):
It revolves happily. If there was a snapshot version, I'd happily reference that.
Brian Postlethwaite (Jun 04 2019 at 04:58):
Yup, overlooked that terminology one, too close to it.
How about the language one?
The TxLog actually showed the right display too.
Grahame Grieve (Jun 04 2019 at 05:28):
canonicals don't resolve on build.fhir.org because that isn't the canonical
Grahame Grieve (Jun 04 2019 at 05:28):
language - I'm not sure but we can let that one go....
Brian Postlethwaite (Jun 04 2019 at 05:37):
Its not a canonical, its just a guidance link to get to the Bulk Data IG.
Grahame Grieve (Jun 04 2019 at 05:38):
well, then shouldn't it hvae a .html?
Brian Postlethwaite (Jun 04 2019 at 05:39):
If I make it directly reference the index.html will that go away then?
Grahame Grieve (Jun 04 2019 at 05:46):
might. it will at least work
Grahame Grieve (Jun 04 2019 at 05:46):
but you're not supposed to reference build.fhir.org so it will still complain
Brian Postlethwaite (Jun 04 2019 at 05:54):
Is there a snapshot of the BulkData guide anywhere I can reference that isn't in the CI build?
Grahame Grieve (Jun 04 2019 at 05:55):
http://hl7.org/fhir/us/bulkdata/2019May/
Grahame Grieve (Jun 04 2019 at 05:55):
you'll have to add it as a dependency - hl7.fhir.us.bulkdata # 0.1.0
Brian Postlethwaite (Jun 04 2019 at 07:48):
Even though not directly referencing any fhir resources?
Grahame Grieve (Jun 04 2019 at 13:31):
yes. Maybe at some stage I'll do something different but for now that is the only way that the I GPublisher can know it's a valid reference
Brian Postlethwaite (Jun 04 2019 at 13:32):
Ok, I've updated the guide again, and cleared out as many warnings and info messages as I can.
From where you sit does that look Ok for publishing? - while I go through all the other admin steps.
Grahame Grieve (Jun 04 2019 at 13:33):
yes it's good to go ( @Lynn Laakso )
Brian Postlethwaite (Jun 05 2019 at 00:17):
Thanks, before you do, I'll make sure its been through all the final admin hurdles through the appropriate channels, publication request, ballot withdrawls etc.
Lynn Laakso (Jun 19 2020 at 19:07):
I'll watch for it
Last updated: Apr 12 2022 at 19:14 UTC
