Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Remove as many RelationConcepts from the instance models as possible, by using ModelKind more #2371

Open
JacquesCarette opened this issue May 6, 2021 · 4 comments
Assignees
Labels

Comments

@JacquesCarette
Copy link
Owner

See f23bf7b for example. This is part of adding ModelKind to InstanceModel.

Basically, every instance model in all examples should be looked at, and either

  1. its model should be changed from OthModel to something else (one per PR!)
  2. an issue should be created regarding why it can't be done.
@balacij
Copy link
Collaborator

balacij commented May 12, 2021

Trying to better keep track of them and their similarities:

Things not ticked off are still OthModels.

@balacij
Copy link
Collaborator

balacij commented May 27, 2021

Note to self: Once all OthModels are converted into other things, we can easily implement MayHaveUnit of GenDefns in terms of the ModelKinds instead of having a separate field for it.

@balacij
Copy link
Collaborator

balacij commented Jul 8, 2021

I don't remember where you (@JacquesCarette) mentioned it, but you previously mentioned that ModelKinds may need to be converted into a triple with a custom UID, NP (term), and the original ModelKind, and I think that with the last bunch of models, we will need that. Most of the existing converted ones only needed either a custom UID or term, so we added them to the models (IM/GD/TM) instead of the MK, but some of the last ones appear to need both.

I can see us changing ModelsKinds to something similar to:

type ModelKinds' = (ModelKinds, UID, NP)

or:

data ModelKinds' = MK' {
  _mk    :: ModelKinds,
  _mkUid :: UID,
  _mkNp  :: NP
}

The second variant is probably better.

@JacquesCarette
Copy link
Owner Author

Yes. As I've mentioned elsewhere, 4-tuples (and above): never. 3-tuples? rarely, and need strong justification.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
Development

No branches or pull requests

2 participants