Subject: [Fwd: Re: Error or restriction in "apply"?]
From: Raymond Toy
Date: Sat, 16 Oct 2010 11:46:37 -0400
On 10/16/10 11:32 AM, Leo Butler wrote:
>
> On Sat, 16 Oct 2010, Wilhelm Haager wrote:
>
> < ------------------------ Urspr?ngliche Nachricht -------------------------
> < Betreff: Re: [Maxima] Error or restriction in "apply"?
> < Von: "Wilhelm Haager" <wilhelm.haager at htlstp.ac.at>
> < Datum: Sa, 16.10.2010, 16:03
> < An: "Andrey G. Grozin" <A.G.Grozin at inp.nsk.su>
> < --------------------------------------------------------------------------
> <
> < > On Fri, 15 Oct 2010, Wilhelm Haager wrote:
> < >> In the new version of Maxima (5.22.1) ist is obviously not possible to
> < >> generate complex graphics with draw2d any more.
> < >>
> < >> The command
> < >>
> < >> draw2d(explicit(..), ...<100 objects or options>)
> < >>
> < >> produces the following error message:
> < >>
> < >> Maxima encountered a Lisp error:
> < >> Error in APPLY [or a callee]: Lisps arglist maximum surpassed
> < >> Automatically continuing.
> < >> To enable the Lisp debugger set *debugger-hook* to nil.
> < >>
> < >> Previous versions of maxima (<=5.20.1) did not have any problems!
> <
> < > This depends on which lisp you use to run maxima, not on the version of
> < > maxima. What lisp are you using now? And what lisp you used previously,
> < > when you had no problem?
> < >
> < > Andrey
> < >
> < Thanks to all the answers;
> < I think, it's not a Lisp problem, both versions (5.20.1 working o.k.
> < and 5.22.1 producing the error message) use the same Lisp GCL 2.6.8.
> <
> < As also Ted Woollett noticed, I think it's a draw-problem.
>
> No, the error message tells you the problem is that there
> are too many arguments being passed. This means that it's likely
> to be a GCL configuration issue. This issue has been raised on
> this list in the recent past.
I don't think GCL can be configured to use a larger value for
call-arguments-limit.
I looked a little at this issue. The problem seems to be in
make-scene-2d in draw.lisp. Could be an interpreter issue because if I
compile draw.lisp, the error now happens in append, called from
macsyma-top-level. Not sure what that means.
But I also ran this with cmucl, and as best as I can tell the call to
apply in make-scene-2d is never given anything more than 6 elements long
(for the example with qdraw that was posted).
I don't really know what's going on. The only other thing I noticed is
that the line with apply was changed in May this year (rev 1.70), which
was after 5.20 was released.
Ray