>> Waldek writes:
>>> Another question is what we should generate for products,
>>> sums and integrals. The current result is _not_ valid as input...
On Tue, Oct 27, 2009 at 7:15 PM, Martin Rubey wrote:
> why not similar to
>
> (9) -> integrate(f i, i)::INFORM
>
> (9) (integral (fun i) (:: i Symbol))
> Type: InputForm
> just replacing "integral" by "product" or "summation"?
> ...
> the right thing to do is to set a %specialInput property, as in
>
> fspace.spad.pamphlet:596: setProperty(opdiff,SPECIALINPUT,indiff@(List %
> -> InputForm) pretend None)
> liouv.spad.pamphlet:198: setProperty(opint, SPECIALINPUT, inint@(List F
> -> INP) pretend None)
> liouv.spad.pamphlet:199: setProperty(opdint, SPECIALINPUT, indint@(List
> F -> INP) pretend None)
>
Martin,
I have attached a patch 'combfunc.spad.pamphlet' that implements
InputForm as you suggest above.
if F has ConvertibleTo(I) then
setProperty(opdprod, SPECIALINPUT, indprod@(List F -> I) pretend None)
setProperty(opdsum, SPECIALINPUT, indsum@(List F -> I) pretend None)
...
if F has ConvertibleTo(I) then
indprod l ==
convert([convert(_product)$I, convert(summand l)@I, _
convert([convert("="::Symbol)$I, convert(third l)@I, _
convert([convert('SEGMENT)$I, _
convert(fourth l)@I, convert(fourth rest l)@I]$List(I)) _
]$List(I)) _
]$List(I))
indsum l ==
convert([convert(_summation)$I, convert(summand l)@I, _
convert([convert("="::Symbol)$I, convert(third l)@I, _
convert([convert('SEGMENT)$I, _
convert(fourth l)@I, convert(fourth rest l)@I]$List(I))@I _
]$List(I))@I _
]$List(I))@I
With this patch I get:
(1) -> summation(factorial(i+1),i=0..n)::InputForm
(1) (summation (factorial (+ i 1)) (= i (SEGMENT 0 n)))
Type: InputForm
(2) -> interpret %
n
--+
(2) > (i + 1)!
--+
i = 0
Type: Expression(Integer)
(3) -> product(sin x,x=1..n+1)::InputForm
(3) (product (sin x) (= x (SEGMENT 1 (+ n 1))))
Type: InputForm
(4) -> interpret %
n + 1
++-++
(4) | | sin(x)
| |
x = 1
Type: Expression(Integer)
---
However I still have problems with a crash in 'unparse' for example:
(5) -> product(sin x,x=1..n+1)::InputForm
(5) (product (sin x) (= x (SEGMENT 1 (+ n 1))))
Type: InputForm
(6) -> unparse %
debugger invoked on a SIMPLE-CONDITION: break
Type HELP for debugger help, or (SB-EXT:QUIT) to exit from SBCL.
restarts (invokable by number or by possibly-abbreviated name):
0: [CONTINUE] Return from BREAK.
(BREAK "break")
0]
I am up to date with the FriCAS repository:
wsp...@debian:~/fricas-sources/src/algebra$ svn diff
combfunc.spad.pamphlet > ~/algebra-combfunc-1.patch
wsp...@debian:~/fricas-sources/src/algebra$
wsp...@debian:~/fricas-sources/src/algebra$ svn update
At revision 711.
which includes Waldek's recent related patch:
Modified: trunk/src/interp/format.boot
===================================================================
--- trunk/src/interp/format.boot 2009-10-26 22:48:47 UTC (rev 708)
+++ trunk/src/interp/format.boot 2009-10-27 02:44:28 UTC (rev 709)
@@ -461,7 +461,6 @@
$curExpr : local := x
x is ["=", arg1, arg2] or x is ['"=", arg1, arg2] =>
concat(sumOrParen(arg1), '"=", sumOrParen(arg2))
- sumOrParen(x)
sumOrParen(x)
sumOrParen(x) ==
---
Perhaps there is another bug lurking there?
Regards,
Bill Page.
--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"FriCAS - computer algebra system" group.
To post to this group, send email to fricas-devel@googlegroups.com
To unsubscribe from this group, send email to
fricas-devel+unsubscr...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/fricas-devel?hl=en
-~----------~----~----~----~------~----~------~--~---
