coq-club AT inria.fr
Subject: The Coq mailing list
List archive
- From: Nils Anders Danielsson <nad AT chalmers.se>
- To: Brandon Moore <brandon_m_moore AT yahoo.com>
- Cc: coq-club <coq-club AT inria.fr>
- Subject: Re: [Coq-Club] mixed induction and coinduction
- Date: Thu, 30 Aug 2012 11:36:02 +0200
On 2012-08-29 23:55, Brandon Moore wrote:
A partially-coinductive definition in Agda behaves exactly as I would
like, but I haven't been able to define something equivalent in Coq.
data Proof : Rel term where
Prim : {x y : term} -> Axiom x y -> Proof x y
Subst : (v : term){x y : term} -> Proof x y -> Proof (subst v x) (subst v
y)
Trans : (y : term){x z : term} -> Proof x y -> ∞ (Proof y z) -> Proof x z
Perhaps you could use Keiko and Tarmo's trick:
Mixed induction-coinduction at work for Coq
Nakata and Uustalu
http://www.cs.ioc.ee/~keiko/papers/Coq2.pdf
--
/NAD
- [Coq-Club] mixed induction and coinduction, Brandon Moore, 08/29/2012
- Re: [Coq-Club] mixed induction and coinduction, Nils Anders Danielsson, 08/30/2012
Archive powered by MHonArc 2.6.18.