[Axiom-developer] Re: fixing SPAD

[Martin Rubey]

Gabriel Dos Reis <address@hidden> writes:

>    If you cannot correctly parse what I wrote, then I drop the ball.

Please bear with me, I'm not english native.  Could you just tell me which
sentence I misparsed and maybe rephrase it, so that I can understand?  I read
your mail several times now, I just do not understand it.

Are you saying:

  if A and B or C then ...

should be, but currently isn't allowed in the context of conditional exports?
(I think it is allowed already now.)

On Sat, 19 May 2007, Martin Rubey wrote:

> | > For a given specific case, one might look for a workaround that simulates
> | > type checking at runtime.
> | 
> | I doubt this.
> 
> Please don't. 

I'm still curious how you would do it. I.e., Do you think you can write a
domain constructor Foo(R) that exports Cat1 if R has 2 elements but Cat2 if R
has 3 elements?

> My original statement was an observation about why I believe removing the
> restriction is something that should be considered.

I agree that removing the restriction is important.  But since I have no idea
how long I'll have to wait for it, I wonder whether a workaround exists.
(apart from switching to Aldor)

Many thanks,

Martin