2023-05-30 Architecture Meeting Minutes

Public Page

Date: May 30, 2023

Agenda

• Network/Welcome

• Antitrust Statement

  • This meeting is subject to the terms of our anti-trust statement shown here.  In addition, this meeting may be recorded. 

• Review Meeting Minutes

• Review work completed

Meeting Minutes

• Antitrust Accepted

• Meeting Minutes Reviewed Accepted

• We have committees that are working on new Code List that will be coming up through the process.

• Review Changes made since last meeting.

  • Build problems.

  • Changes to definitions

  • Changes to instance documents of the new parts invoices

  • Deleted Address Type, Com Phone, e com, email preferred communications type

  • Changes property names com qualifier to phone Num Type and address to IBM

  • Changed the type to format email and then in code list I actually renamed Communication Qualifier

  • Windows can’t build the bundles, GitHub kicks off under Linux and ran all bundles, work around is to do local workstation build.

    • The problem is it makes schemas in both cases, but the schemas don't validate and because of the the references that are bundler has written into the bundled files don't resolve and that happened to me and to Mike when we ran the clone project down to our Windows computers, it didn't happen for Andy on his Linux machine and then it doesn't happen on the framework that runs up in GitHub itself.

  • Andy worked on the bundler, it works but the files are much bigger and you can tell there is a different algorithm for bundling.

    • Updated JSON Schema Ref parser

    • The big change was the way we reference schemas and using the ID

    • from a coding standpoint, there's there's a bunch of changes, you know that are related to the code because it's a new library, but from a kind of a, you know, fundamentally how do we do our schemas standpoint, how do we reference and everything.

    • So you could see on the left the bundle used to have an ID which was just the file name and the new library which I don't actually know what I should call it.

    • The hyper JSON schema library

  • Mike found a UTube video discussing JSON Schemas and the tool and in short he says we did everything wrong.

  • To assign ID and reference the ID is outside of bundling is a best practice and the reason it is is cause you could look at the screen you have right here and know that the Schema who's opening bracket is Line 267 is what's being referenced by what you see on 268, whereas in the existing scheme you have to keep paging up to work your way back through the Jason path.

    • The question is, does XML spy understand Jason Schema and know about the concept of the IDs and the references? Or does it only know about file systems and network and you know directory structures and whatnot?

    • Is there a way of adding these schemas into quote unquote, the system you know into you know the validation for XML Spy, then there's probably a way to do this and that how it would know, but I I don't have that particular tool and to be honest, I haven't loaded it up into oxygen yet either. But this is the way that the validation works for that validator HIV you have to Add all the schemas to it.

     

Great Work everyone

Up Next

• Network/Welcome

• Antitrust

• Review Meeting Minutes

• Review Work Changes

Action items

Decisions

Reference

Participants

• Paulette Reed

• Dan Webster

• Andy Bober

• Jeff Shroder

• Paul Barry

• Mike Hastings

• Phil Martinez

Participants in the meetings are noted for your information.  If you have questions on the committee’s activities, please contact a recent attendee. Architecture Committee