-
Notifications
You must be signed in to change notification settings - Fork 345
20190218 Community Meeting Minutes
Time: 10:00-11:00 UTC (11:00-12:00 CET, 05:00-06:00 EST, 21:00-22:00 AEDT) Location: Join Zoom Meeting https://zoom.us/j/573322038
See https://youtu.be/dcUSUVSP7z8 for the recording of this meeting
- Figure out whether how to have finer-grained control over the "syntactic" checks: Seems to be more difficult then we thought. Will be discussed later in this meeting and it will be part of 1.3 and not 1.2.
- Marco created a Azure Pipeline which is triggered by every new alpha package of the FHIR .NET API. It will then download the lastest development version of Vonk and restore it with the new alpha package of the API. After that, we check whether Vonk is still building and whether all the unittest still succeed. With this pipeline we test the API indirectly with an extra set of unittests.
- In a week or so, Brian will pull the WebApi into a separate repository. And delete the WebApi out of the source of the .FHIR NET API.
- Other action items are owned by Ewout and the status will be discussed in the next Community meeting.
Some last parts of the pull request of Brian (typeref.profile and typeref.targetProfile) are merged now into the 1.2 version. The only thing left is a pull request of Michel. Because we cannot pull everything of that PR, we have to merge code bit by bit. This is almost done. Furthermore we have to shift the FhirPath code from R4 to DSTU2 base and then we can create a release 1.2-beta.
We have to ship release 1.2-beta soon, because the R4 release now on Nuget is certainly not the correct one.
George managed to get all the unittests working again. You can see the progress on common-code-separation. The Resource class is used everywhere and it would be nice if we can Resource version independent. We can only do that if we remove invariant constraints out of Resource. Perhaps we can offer this functionality as an extension method, so it is not part anymore of the Resource class. Brain suggests to ask the community (Zulip) whether they use this functionality or not. Maybe Brain is the only one using it.
When we choose for a special enum value for unknown enums, that will cause a change to the generator. We have to think about it this a bit more.
There is PR on the shelve for that. Marco will start on that on that next week.
We already decided not to support .NET 4.x for our FHIR API, unless lstratman can convince us to do so. We have not heard from him and this community meeting was the second meeting where he could have had his voice heard. We will remove this item from the agenda.
Work in progress at the moment. FP is not normative yet.
Not so easy as we thought it would be. We have to think about this a bit longer. Perhaps we could use validation definitions which you can switch on or off. This needs some more time to think about it. It will not go in to version 1.2, but perhaps in 1.3.
- For the 1.3 release, we will have to figure out whether we need finer-grained control over the "syntactic"
- Brian will pull WebApi into a separate repository.
- Ewout to set up a targeted meeting to discuss these different interfaces for WebApi
- Ewout to schedule a meeting to show proposed changes to the internals of the validator [DELAYED until further notice]