Closed nomeata closed 1 week ago
Mathlib CI status (docs):
nightly-with-mathlib
branch. Try git rebase e0c1afd12d4fc6b0e520774959aed06bf122aba9 --onto e362b50fa95d6823e59dd706803a93c25e888535
. (2024-05-06 08:10:45)What are the chances that this is fixed in the equation compiler so that it doesn't generate terms that violate the unused have linter?
That's plausible; it would entail removing unused have
there. Didn't check yet how hard it would be.
The induction principle generator should then also remove them from the minor premises.
Do you see a case where the user might genuinely want then to be there in the equations? In some cases it maybe might be nice to have them in the goal to intro
them conveniently? But that sounds obscure, generating nicer equation lemmas is probably the better trade off.
otherwise it remains in the equational theorem and may cause the “unused have linter” to trigger. By moving the proof into
decreasing_by
, the equational theorems are unencumbered by termination arguments.see also https://github.com/leanprover/std4/pull/690#issuecomment-2095378609