Skip to Content.
Sympa Menu

coq-club - Re: [Coq-Club] Program fixpoints with measure

coq-club AT inria.fr

Subject: The Coq mailing list

List archive

Re: [Coq-Club] Program fixpoints with measure


chronological Thread 
  • From: Michael Day <mikeday AT yeslogic.com>
  • To: Julien Forest <forest AT ensiie.fr>
  • Cc: coq-club AT inria.fr
  • Subject: Re: [Coq-Club] Program fixpoints with measure
  • Date: Sun, 28 Mar 2010 10:42:15 +1100

Hi Julien,

you can try to use Function instead of Program Fixpoint. The two
feature are not exactly equivalent. In particular, Function will give
you a fixpoint equation for you function (aka a proof that $forall x, f
x = body_of_f$) and an induction principle for f which will (hopefully)
help you to reason about f.

Is it possible to use Function to write functions that are not directly structurally recursive? For example, a function that takes a list whose length decreases in recursive calls, without necessarily being a subterm of the original list?

Cheers,

Michael

--
Print XML with Prince!
http://www.princexml.com



Archive powered by MhonArc 2.6.16.

Top of Page