On 04/28/2014 08:32 AM, jiazhaoconga wrote:
> The reason that 'plus (f,g) == x+->f(x)+g(x)' fails in interpreter may
> have nothing to do with type system, but because of interpreter can
> not handle closure:
You are not totally right, it seems.
(4) -> a(x: Integer):(Integer->Integer) == (y:Integer):Integer +-> y+x
Function declaration a : Integer -> (Integer -> Integer) has been
added to workspace.
Type:
Void
(5) -> a 4
Compiling function a with type Integer -> (Integer -> Integer)
(5)
theMap
#<FUNCTION (LAMBDA (#:G663 |envArg|) :IN |*1;a;1;frame1|)
{10048D621B}>
,
655
Type: (Integer ->
Integer)
(6) -> (a 4) 7
> For example, the 'adder' function:
> adder x == y +-> y + x
>
> It totally works in Spad form,
You mean without giving any type?
> but in interpreter:
>
> (2) -> q := adder 3
> Compiling function adder with type PositiveInteger ->
> AnonymousFunction
>
> (2) y +-> y + #1
> Type: AnonymousFunction
> (3) -> q 4
>
> (3) #1 + 4
> Type: Polynomial(Integer)
>
> The '#1' should be the captured value of 'x', thus 3.
>
> So, interpreter can not handle captured variable, aka closure,
> this is a bug, am I right?
Right. That smells like a bug. The #1 notation for formal parameters has
been removed from FriCAS. Maybe the following makes it more apparent,
what the problem might be.
(7) -> b(x: Any):(Any->Any) == (y:Any):Any +-> y+x
Function declaration b : Any -> (Any -> Any) has been added to
workspace.
Type:
Void
(9) -> (b 5) 8
(9) G665 + 8
Type:
Polynomial(Integer)
To be honest, I would have liked that the definition of b is rejected,
since the type Any does not export a + function, so y+x is actually not
possible.
Ralf
--
You received this message because you are subscribed to the Google Groups
"FriCAS - computer algebra system" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
To post to this group, send email to [email protected].
Visit this group at http://groups.google.com/group/fricas-devel.
For more options, visit https://groups.google.com/d/optout.