coq-club AT inria.fr
Subject: The Coq mailing list
List archive
- From: AUGER Cedric <Cedric.Auger AT lri.fr>
- To: S3 <scubed2 AT gmail.com>
- Cc: Pierre Casteran <pierre.casteran AT labri.fr>, AUGER Cedric <Cedric.Auger AT lri.fr>, coq-club AT inria.fr
- Subject: Re: [Coq-Club] Prove inductive type implies false in cycle
- Date: Wed, 3 Aug 2011 10:50:48 +0200
Le Tue, 02 Aug 2011 23:04:43 -0700,
S3
<scubed2 AT gmail.com>
a écrit :
>
> Thanks. That's neat. I didn't realize you can use
> left and right on things other than \/. That makes
> for a very short proof.
>
Ok, so you should be interested by this generalization:
split applies to 1-constructor inductives, that is for instance:
* A /\ B
* a = b
* True
* point
(assuming you have defined point as "Record point := {x:ℝ;y:ℝ}.")
* Acc
* Stream
left and right apply to 2-constructor inductives
(left for the first and right for the second)
constructor [n] apply to [m]-constructor inductives (where m≥n)
so that:
* left and split can be replaced by "constructor 1"
* right can be replaced by "constructor 2"
when there is need of some parameter, you can use the
"[constructor-tactic] with [arg(s)]".
So that "exists x." can be replaced by "split with x."
there are also existential versions of these tactics which allow you
to omit the "with" part:
econstructor, eleft, eright, esplit
so that "eexists." can be replaced with "esplit."
- [Coq-Club] Prove inductive type implies false in cycle, S3
- [Coq-Club] Prove inductive type implies false in cycle,
S3
- Re: [Coq-Club] Prove inductive type implies false in cycle, Brian Huffman
- Message not available
- Re: [Coq-Club] Prove inductive type implies false in cycle,
Pierre Casteran
- Re: [Coq-Club] Prove inductive type implies false in cycle,
S3
- Re: [Coq-Club] Prove inductive type implies false in cycle, Jean-Francois Monin
- Re: [Coq-Club] Prove inductive type implies false in cycle, AUGER Cedric
- Re: [Coq-Club] Prove inductive type implies false in cycle,
S3
- Re: [Coq-Club] Prove inductive type implies false in cycle, AUGER Cedric
- [Coq-Club] Prove inductive type implies false in cycle,
S3
Archive powered by MhonArc 2.6.16.