-
Notifications
You must be signed in to change notification settings - Fork 12
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
Rectification Campaign II: Restrictions, etc. #230
Conversation
Roadmap:
|
…non-restriction type" This reverts commit f4df7db.
4de2302
to
fcc741e
Compare
(* This is not backwards, FYI. *) | ||
List.rev @@ | ||
(* ~~This is not backwards, FYI.~~ *) | ||
(* NARRATOR VOICE: it was backwards. *) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@ecavallo 😆 😎
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I should be forbidden to use List.rev
... I get so owned by it.
This comment has been minimized.
This comment has been minimized.
b155909
to
3d24751
Compare
This should work:
|
I think I will actually merge this now that it is cleaned up; the PR is getting too big. |
Removing box modality, because the cure was worse than the disease. We should wait until we can implement time warps 😉
Preliminary work toward Find a way to limit the pain of restriction pretypes (or remove them) #231
resolves bug in path type coercion? #228
resolves composition in path types #222
resolves variables getting reversed in HITs #234 (?)
resolves Add neutral coe #212
resolves coercing in Code-like type...? #237
resolves "Tried to evaluate non-dimension term ... as dimension" #240
resolves Find a way to implement strict composition #238
very basic
auto
tactic (will become better as I refactor the refiner)disabling strict composition for now, because what we had was wrong, and I am not yet decided on how to make it right
Huge thanks to @ecavallo for exercising all the bugs fixed in this PR.