coq-club AT inria.fr
Subject: The Coq mailing list
List archive
- From: Hugo Herbelin <herbelin AT pauillac.inria.fr>
- To: Edsko de Vries <devriese AT cs.tcd.ie>
- Cc: coq-club AT pauillac.inria.fr
- Subject: Re: [Coq-Club] Simple question about tactic notation
- Date: Tue, 17 Mar 2009 12:44:57 +0100
- List-archive: <http://pauillac.inria.fr/pipermail/coq-club/>
Hi,
> Is it possible to define a tactic that at some point calls 'induction n' for
> some natural number 'n'?
You can do it this way:
<<<<<
Ltac OnLastHyp tac :=
match goal with
[ n:_ |-_ ] => tac n || fail 1 "Application on last hypothesis failed"
end.
Tactic Notation "my_induction" integer(n) :=
let ind n := (induction n) in intros until n; OnLastHyp ind.
Inductive even : nat -> Prop :=
| even_zero : even 0
| even_succ : forall n, even n -> even (S (S n)).
Lemma foo : forall (n:nat), even n -> True.
Proof.
my_induction 1.
>>>>>
Passing induction arguments referring to quantified (dependent or non
dependent) hypotheses in Ltac functions was actually not
implemented. This will be made available for further releases.
> In nested Ltac calls to "my_induction'" and "n"), last term evaluation
> failed.
> Error: Variable n should be bound to a term.
>
> BTW There seems to be an extraneous bracket in the error message :)
Oops, not very nice, this will be fixed at once in the bugfix branch.
Hugo Herbelin
- [Coq-Club] Simple question about tactic notation, Edsko de Vries
- Re: [Coq-Club] Simple question about tactic notation,
Thomas Braibant
- Re: [Coq-Club] Simple question about tactic notation,
Samuel E. Moelius III
- Re: [Coq-Club] Simple question about tactic notation,
Edsko de Vries
- Re: [Coq-Club] Simple question about tactic notation,
Samuel E. Moelius III
- Re: [Coq-Club] Simple question about tactic notation,
Samuel E. Moelius III
- Re: [Coq-Club] Simple question about tactic notation, Edsko de Vries
- Re: [Coq-Club] Simple question about tactic notation,
Samuel E. Moelius III
- Re: [Coq-Club] Simple question about tactic notation,
Samuel E. Moelius III
- Re: [Coq-Club] Simple question about tactic notation,
Edsko de Vries
- Re: [Coq-Club] Simple question about tactic notation,
Edsko de Vries
- Re: [Coq-Club] Simple question about tactic notation, Edsko de Vries
- Re: [Coq-Club] Simple question about tactic notation,
Samuel E. Moelius III
- Re: [Coq-Club] Simple question about tactic notation, Hugo Herbelin
- Re: [Coq-Club] Simple question about tactic notation,
Thomas Braibant
Archive powered by MhonArc 2.6.16.