Random observation while reading code: UnivSubst.subst_univs_constr
(via subst_univs_fn_constr
) does not nothing special for Array
but Vars.subst_instance_constr
does have a case for Array
. Could this be a bug?
To be clear(er): I had expected to find a case for Array
because it is the one(?) kind of term that directly carries universes apart from constants, constructors, inductives, and sorts.
probably buggy yes
we have too many versions of this function
It seems that UnivSubst.subst_univs_constr
is only used when registering hints on arbitrary constr (indirectly via EConstr.Vars.subst_univs_constr
)
https://github.com/coq/coq/pull/14247
Last updated: Sep 09 2024 at 06:02 UTC