[Maxima-bugs] [ maxima-Bugs-663873 ] taylor / fixes

[SourceForge.net <no...@so...>]

Bugs item #663873, was opened at 2003-01-07 18:08
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=104933&aid=663873&group_id=4933

Category: None
Group: None
Status: Open
Resolution: None
Priority: 5
Submitted By: Martin Rubey (kratt5)
Assigned to: Nobody/Anonymous (nobody)
Summary: taylor / fixes

Initial Comment:
1. an obvious fix is  
 
1129c1129,1131 
<        (T (break "Undefined limit product ~A * ~A in 
lim-times lim1 lim2"))) 
--- 
> ;; kratt5 
> ;;     (T (break "Undefined limit product ~A * ~A in 
lim-times lim1 lim2"))) 
>        (T (break "Undefined limit product ~A * ~A in 
lim-times" lim1 lim2))) 
 
which corrects a faulty error message... 
 
2. taylor gives the following     
     
(C1) taylor(sum(k,k,0,m),m,0,3);      
      
(D1)/T/                            k + . . .      
     
which is of course rubbish, because the limit depends on m.     
     
A fix is the following:     
2225c2227,2231    
<       ((memq (caar e) '(%SUM %PRODUCT)) (tsprsum    
(cadr e) (cddr e) (caar e)))    
---    
> ;; kratt5    
> ;;    ((memq (caar e) '(%SUM %PRODUCT)) (tsprsum    
(cadr e) (cddr e) (caar e)))    
>       ((and (memq (caar e) '(%SUM %PRODUCT))    
>             (mfreel (cddr e) tvars))    
>        (tsprsum (cadr e) (cddr e) (caar e)))    
    
but I'm not sure whether this is the best way to do it. This    
just checks whether the limits contain a tvar, if so, do the    
general diff-expand algorithm. However: care should be   
taken that no other case of the cond applies, I do not really   
understand  the cond clause  
  
       ((or (eq (caar e) '%AT)  
             (do ((l (mapcar 'car tlist) (cdr l)))  
                 ((null l) t)  
                 (or (free e (car l)) (return ()))))  
         (newsym e))  
  
which comes two lines later. 
-------------------------------------------------------------      
      
Maxima version: 5.9.0rc3      
Maxima build date: 21:0 11/26/2002      
host type: i686-pc-linux-gnu      
lisp-implementation-type: Kyoto Common Lisp      
lisp-implementation-version: GCL-2-5.0      
      
-------------------------------------------------------------      
The above information is also available from the Maxima      
function build_info().      

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=104933&aid=663873&group_id=4933