pub fn then_all(t1: Tactic, t2: Tactic) -> TacticExpand description
t1 <;> t2: run t1 on the focused goal, then t2 on exactly the goals t1
PRODUCED — not on every goal that happens to be open.
The difference is invisible when <;> is the first branching step, which almost every
textbook use is, and that is why running all_goals here survived a large test suite.
It is not invisible after an earlier step left a goal waiting: split . left <;> assumption must leave the split’s right conjunct alone. When t2 happened to
discharge that goal too, the script succeeded one goal further along than the proof
believed; when it did not, the failure was reported for a goal the source never asked
it to look at. A goal t1 produced is one that was not open before it ran — node
indices only ever grow, so the two sets are told apart exactly.