coq-club AT inria.fr
Subject: The Coq mailing list
List archive
- From: Bruno Barras <Bruno.Barras AT cl.cam.ac.uk>
- To: "Randy Pollack" <rap AT dcs.ed.ac.uk>
- Cc: coq-club AT pauillac.inria.fr, Bruno.Barras AT cl.cam.ac.uk
- Subject: Re: Parameters in "Definitions"?
- Date: Mon, 15 Feb 1999 17:52:31 +0000
Hello Randy,
> As a functional programmer I like to write, e.g.
>
> (1) plus (n,m:nat) : nat = ...
>
> rather than
>
> (2) plus : nat->nat->nat = \n:nat.\n:nat. ...
>
> In Coq, the "Fixpoint" definition lets one write (1) instead of (2).
to avoid redundancy, you could write
Definition plus : nat->nat->nat := [n;m] ...
but I agree this is not as readable as what you would expect.
Anyway, as you have noticed, the Fixpoint notation is more friendly.
So, you can do
Fixpoint plus [n,m:nat]: nat := ...
If your definition is indeed not recursive, it will not generate the
useless fixpoint operator and it behaves like a definition.
> The notation I want almost exists, by using a Fixpoint with a dummy
> boolean argument.
I don't understand why you need this extra argument...
Bruno.
- Parameters in "Definitions"?, Randy Pollack
- Re: Parameters in "Definitions"?, Bruno Barras
- Re: Parameters in "Definitions"?,
Randy Pollack
- Re: Parameters in "Definitions"?, Patrick Loiseleur
- Re: Parameters in "Definitions"?,
Randy Pollack
- Re: Parameters in "Definitions"?, Bruno Barras
Archive powered by MhonArc 2.6.16.