Stream: terminology / utg
Topic: Issues with value set expansions
Ted Klein (Apr 20 2020 at 21:21):
@Grahame Grieve please look into this. Something is broken with the value set expansions on the UTG publisher build - many of them are not expanding a hierarchy with include.filter.op value="is-a". Details on one that illustrates the problem can be found in the Google Doc at https://docs.google.com/document/d/1hM3TubYKop1XWei7ls5j8O9LxfGUd-LU7gacGnXlX8k/edit?usp=sharing
Grahame Grieve (Apr 21 2020 at 00:07):
fixed next release
Ted Klein (Apr 21 2020 at 17:21):
OK release 81 only half fixed it - now with op set to 'is-a' is shows all the descendants, but not the head code specified. The filter:
Ted Klein (Apr 21 2020 at 17:21):
<filter>
<property value="concept"/>
<op value="descendent-of"/>
<value value="_ServiceDeliveryLocationRoleType"/>
</filter>
Ted Klein (Apr 21 2020 at 17:21):
when the op is changed to 'descendent-of' the expansion contains zero concepts. is-a seems to have them all except the head code
Grahame Grieve (Apr 21 2020 at 20:00):
ok missed descendent-of but can't see how the problem with is-a could be the case, where's an example?
Grahame Grieve (Apr 21 2020 at 20:50):
do we have am email contact set up through HL7 for terminology / UTG issues? e.g. something like terminology-admin@hl7.org?
Grahame Grieve (Apr 23 2020 at 22:13):
@Ted Klein ping on above, and are we done now with the release of .82 IG publisher
Ted Klein (Apr 23 2020 at 22:31):
@Grahame Grieve sorry been heads down all day working on getting rid of the build errors, just noticed you pushed IGP v82. Will download it now and check everything. RE: the email - we do not, but that may be a great idea. Setting up a JIRA workflow and entry with Josh for logging UTG errors, but this is primarily for content issues that need fixing (the old "open issues" stuff mostly). But an email list might be very handy, at least now in the beginning when we are just rolling it out and ironing out problems. Build errors now down to the last couple subtle ones, expect to have those solved by tonight or tomorrow morning US EDT. Let me know when you want to speak - maybe either early today (your time before I go to bed here) or before you are done for the night tomorrow (when I wake up my Friday morning).
Ted Klein (Apr 25 2020 at 11:53):
@Grahame Grieve All the identified items but one have been fixed with Publisher .82; I guess you did not yet look into the additional language display strings on code system concepts (v2 table 163 as the example I went over with you fort some German display strings). Once you get a notion of when you may get around to that please let me know. The item that Lloyd pointed out was that the publisher is inserting <title> as "Definition" for all of the pages generated for Value Sets, but the template is also inserting <title> as "Content Logical". Thus two section headers. Lloyd suggests either you change the publisher to say "Content Logical Definition" and he will remove it from the template, or alternatively you remove it from the Publisher and he will keep it as a mandatory item in the template. But we should not have both. Did I get that correct @Lloyd McKenzie ?
Grahame Grieve (Apr 25 2020 at 11:57):
i thought I had done that
Grahame Grieve (Apr 25 2020 at 11:57):
also, I thought that the IG publisher did not generate title when generating the pages
Grahame Grieve (Apr 25 2020 at 22:36):
@Ted Klein I see language at https://build.fhir.org/ig/HL7/UTG/CodeSystem-v2-0163.html
Grahame Grieve (Apr 26 2020 at 00:40):
well,I really dislike "Content Logical Definition" as a set of words. I suppose we could come up with some thing more obscure if we tried, but we'd have to work at it.
Grahame Grieve (Apr 26 2020 at 00:41):
how about "Logical Definition (CLD)" - CLD is only meaningful to anointed insiders, and for any who are, they'll know "CLD"
Rob Hausam (Apr 26 2020 at 01:12):
We're in the value set context. And is there anything really wrong with just calling it "Definition"? The only reason I can see that we need to add "Logical" is to distinguish it from "Illogical Definition". :)
Grahame Grieve (Apr 26 2020 at 01:39):
"Illogical Definition"
:joy_cat:
Ted Klein (Apr 26 2020 at 02:27):
LOL. I kinda like "illogical definition". To be more serious about it, your suggestion of "Logical Definition (CLD)" is, I think, perfect.
Ted Klein (Apr 26 2020 at 02:28):
Or we can put an Easter egg in it if you hold down some special keys and click it comes up 'Illogical Definition"...sorry this quarantine has me in a very funky mental place these days...
Lloyd McKenzie (Apr 26 2020 at 02:35):
Part of me wants the Illogical definition to be expressed in Z... :)
Last updated: Apr 12 2022 at 19:14 UTC
