Extract characteristics for Logicals #250
Conversation
Logical models use special extensions to set their type characteristics. Check for these extensions when extracting keywords. Do not create caret rules for these extensions.
There was a problem hiding this comment.
Everything that's here looks good, but if you run it on the CLI against a logical JSON w/ characteristics, you don't get any of the characteristics in the output FSH. It looks like the Logical class's toFSH() method needs to be updated in SUSHI. Foiled again by toFSH()!
There was a problem hiding this comment.
This looks really good to me. I tested it out with the branch in SUSHI that updates the toFSH function and the Characteristics are there and look good.
One small question I had -- does it make sense to include in either or both of the tests that extensions that are not the characteristics extensions are not included in the characteristics keyword and still have regular rules created?
There was a problem hiding this comment.
This looks good! I like Julia's idea to add a non-context extension to the test JSON to prove it still handles those correctly, but I also recognize that we have other tests that handle normal extensions (and they still pass), so I'm willing to accept this without that change.
I'm guessing we have test cases somewhere that handle extensions on Structure Definitions generically (although I haven't pinpointed exactly which test case covers this), but I don't think we have any tests that test if a Logical has non-characteristics extensions. We only have test cases that have extensions that are all characteristics. Maybe this is testing more than we need to still, but wanted to ask specifically about that. Or maybe I'm just missing something and we have already covered that through something else. |
Fixes #238 and completes task CIMPL-1167.
Logical models use special extensions to set their type characteristics. Check for these extensions when extracting keywords. Do not create caret rules for these extensions.