Re: [Chicken-users] FFI: wrapping method that returns object

[Sergey Khorev]

> Your wrapper is the correct way of doing this.

Then I have a question ;)) I have created such wrapper, but got
strange access violation in chicken.dll during C_c_pointer_or_null.
My knowledge of Chicken internals are (temporarily) insufficient to
understand what is wrong. It seems that ___abstract is such a reason.
Windows, MSVC6 SP5 and SP6.
#>
class T;
class N
{
  char dummy[128];
public:
  N() {}
  int getX(T&);
};

class T
{
  char dummy[128];
  int x;
public:
  T(int x_) : x(x_) {}
  friend class N;
};
class A
{
  char dummy[128];
public:
  A() {}
  T makeT(int x) { T t(x); return t; }
};

int N::getX(T &t)
{
  return t.x;
}
<#

#>?
class T;
class N
{
public:
  N() {}
  int getX(T &);
};

___abstract class T
{
};

class A
{
public:
  A() {}
};
<#

#>!
T *makeT(A &a, int x)
{
  return new T(a.makeT(x));
}

<#

(require-extension tinyclos)

(define n (make <N>))
(define a (make <A>))
(define t (makeT a 100))
(getX n t) ; got access violation when trying to get pointer to T

> If you want to avoid coding, you might write your own little
> preprocessor using `foreign-parse/spec' (see the manual):
> this macro could (for example) parse the specification, remodel
> the interface and generate foreign declarations again.
> (This is kind of weird, but actually pretty cool).

Definitely, I should try this ;))