I have to disagree here. There are some problems with
Macsyma arrays, but the commercial product provided some
functions to mitigate that. I have some code that uses
macsyma arrays, in particular lindstedt and some multivariable
taylor series code. The biggest problem with arrays, is getting
them to act as symbols, and I don't think arrays created
with make_array does that. However make_array is not supported
in the commercial macsyma either. I will try to get a chance to
look at my code, and see what it would take to not use array,
and get back to the list.
Dan Stanger
Barton Willis wrote:
> I thank Robert Dodier for the clear explanation of Maxima arrays. A few
> opinions
>
> (1) Arrays in Macsyma are just plain broken; we shouldn't preserve
> compatibility with them.
>
> (2) Setting use_fast_arrrays to true in Maxima already breaks
> compatibility with commercial Macsyma.
>
> (3) I'd make use_fast_arrays : true the default and prevent a user from
> resetting it. I don't think porting code
> that uses array(name, dim, ...) to the make_array scheme should be all
> that much work. I never used
> Macsyma arrays; they couldn't be used as function arguments, so I never
> trusted them.
>
> (4) make_array doesn't honor the element-type
>
> (C1) use_fast_arrays : true$
> (C2) a : make_array('fixnum,10)$
> (C3) a[0] : x;
> (D3) x
> (C4) a[1] : 4.5;
> (D4) 4.5
> (C5) a;
> (D5) #($x 4.5 0 0 0 0 0 0 0 0)
>
> This need to be cleaned up.
>
> Barton
>
> _______________________________________________
> Maxima mailing list
> Maxima@www.math.utexas.edu
> http://www.math.utexas.edu/mailman/listinfo/maxima