I want to change the goal from S x = S y
to x = y
. It's like inversion
, but for the goal instead of a hypothesis.
Such a tactic seems legit, because when we have x = y
, we can simply use rewrite
and reflexivity
to prove the goal.
Currently I always find myself using assert (x = y)
to introduce a new subgoal, but it's tedious to write when x
and y
are complex expression.