Found a minor typo in the comment introducing <=-tactics, which has = instead of <=
Hit an error on the first decompose call, whose hypothesis is not named in the same way on my machine (8.9.0) and in the source file. (This may be related to the rename_depth setting that we play with just before, but I didn’t find a depth setting that makes the naming match: either the boolean part is more detailed, or the arithmetic part is less detailed than in the Demo name.)
Thanks! Indeed the demo file is broken. I will fix this soon.
It should read like this:
(* Let us have really big names. *)
Ltac rename_depth ::= constr:(5).
!intros.
onAllHyps move_up_types.
(* decompose and revert all new hyps *)
decompose [ex and] h_ex_and_neq_true_andb_false_true_and_le_w_w_eq_w_x ;!; revertHyp.
...