[Gcl-devel] Re: [Maxima] Re: Special variables!

[Camm Maguire]

Camm Maguire <address@hidden> writes:

> BTW, down to 2 issues with maxima and 2.7.0.  Anyone know the
> (apparently result sorting) routine that is responsible for the last
> two?
> 
> Take care,
> 

The issue here is the following from algsys.lisp:

(defun condensesolnl (tempsolnl)
  (let (solnl)
    (mapl #'(lambda (q) (or (subsetl (cdr q) (car q))
                            (setq solnl (cons (car q) solnl))))
          (sort tempsolnl (function (lambda (a b) (> (length a)
                                                     (length b))))))
    solnl))

I think you need stable-sort here.  GCL has a sort inliner which
avoids the pathological qsort behavior on previously sorted lists with
a little equal item reshuffling.

Take care,


> 
> ********************** Problem 70 ***************
> Input:
> block([fpprec : 35], ev(ans, x : 1, bfloat) - ev(ans, x : 0, bfloat))
> 
> 
> Result:
> 4.5994649519620995822199830432826167B-1
> 
> This differed from the expected result:
> 5.779160182042402B-1
> 
> 93/94 tests passed.
> The following 1 problem failed: (70)
> Running tests in rtest8.mac: 
> ********************** Problem 20 ***************
> Input:
> algsys([f1, f2], [x, y])
> 
> 
> Result:
>        1              1             1        1
> [[x = -------, y = - -------], [x = - -, y = - -], 
>       sqrt(3)      sqrt(3)          3        3
>                                           1            1
>                                 [x = - -------, y = -------], [x = 1, y = 1]]
>                                        sqrt(3)      sqrt(3)
> 
> This differed from the expected result:
>       - 1           1              1           - 1          - 1      - 1
> [[x = -------, y = -------], [x = -------, y = -------], [x = ---, y = ---], 
>       sqrt(3)    sqrt(3)        sqrt(3)      sqrt(3)         3        3
>                                                               [x = 1, y = 1]]
> 
> ********************** Problem 24 ***************
> Input:
> solve(%, [x, y])
> 
> 
> Result:
> [[x = 2, y = 2], [x = - 1.733751846381093, y = - 0.15356757100196963], 
> [x = - 0.52025943886520076 %i - 0.13312403573587062, 
> y = 3.6080032218702871 %i + 0.076783785237877766], 
> [x = 0.52025943886520076 %i - 0.13312403573587062, 
> y = 0.076783785237877766 - 3.6080032218702871 %i]]
> 
> This differed from the expected result:
> [[x = 2, y = 2], [x = 0.52025943886520076 %i - 0.13312403573587059, 
> y = 0.076783785237877766 - 3.6080032218702871 %i], 
> [x = - 0.13312403573587059 - 0.52025943886520076 %i, 
> y = 0.076783785237877766 + 3.6080032218702871 %i], 
> [x = - 1.733751846381093, y = - 0.1535675710019696]]
> 
> 48/50 tests passed.
> 
> 
> > Ray
> > 
> > _______________________________________________
> > Maxima mailing list
> > address@hidden
> > http://www.math.utexas.edu/mailman/listinfo/maxima
> > 
> > 
> > 
> 
> -- 
> Camm Maguire                                          address@hidden
> ==========================================================================
> "The earth is but one country, and mankind its citizens."  --  Baha'u'llah
> 
> _______________________________________________
> Maxima mailing list
> address@hidden
> http://www.math.utexas.edu/mailman/listinfo/maxima
> 
> 
> 

-- 
Camm Maguire                                            address@hidden
==========================================================================
"The earth is but one country, and mankind its citizens."  --  Baha'u'llah