[Ecls-list] deftype propagation in the ECL compiler

[Juan Jose Garcia-Ripoll]

On Sun, Nov 2, 2008 at 2:28 PM, Gabriel Dos Reis <gdr at cs.tamu.edu> wrote:
> Hmm, my understanding of DEFTYPE was that, even though the newly
> introduced type specifier, the recursion expansion must terminate.
> Therefore, it would be OK to use a lattice system, since a DEFTYPE
> could not introduce an infinitely branching type.
> Is that not the case?

Maybe you are right. I probably did not read that carefully enough,
but the point is that CONS types are inherently recursive, and
therefore it is not possible to use the scheme we use now in ECL for
the rest of the types.

Juanjo

-- 
Instituto de Física Fundamental, CSIC
c/ Serrano, 113b, Madrid 28009 (Spain)
http://juanjose.garciaripoll.googlepages.com