Problem with SwigValueWrapper

[Robert Lupton]

This one I do think is a bug rather than a "feature";  the non-use of  
SwigValueWrapper when it's needed, and the ignoring of a %feature
("valuewrapper").  This is swig 1.3.40 and -python -c++.

// -*- c++ -*-
%module hoo

%inline %{
     template<typename Matrix>
     class MatrixBase {
     private:
        MatrixBase() {}
     };

     class Matrix { };

     template<typename T>
     class TMatrix { };
%}

//%template(MMatrix) MatrixBase<Matrix>;   // makes no difference
//%feature("valuewrapper") MatrixBase<Matrix>; // makes no difference

%typemap(in) MatrixBase<Matrix>() { }

%inline %{
#if 1
    void size(MatrixBase<Matrix> mat); // fails
#else
    void size(MatrixBase<TMatrix<int> > mat); // OK
#endif
%}

With "#if 1" (i.e. void size(MatrixBase<Matrix> mat);) the generated  
code uses:

SWIGINTERN PyObject *_wrap_size(PyObject *SWIGUNUSEDPARM(self),  
PyObject *args) {
   PyObject *resultobj = 0;
   MatrixBase< Matrix > arg1 ;

and fails to compile due to MatrixBase's private constructor.  With  
"#if 0" the code
uses SwigValueWrapper and all is well:

SWIGINTERN PyObject *_wrap_size(PyObject *SWIGUNUSEDPARM(self),  
PyObject *args) {
   PyObject *resultobj = 0;
   SwigValueWrapper< MatrixBase< TMatrix< int > > > arg1 ;

I'm actually most concerned about %feature("valuewrapper")  
MatrixBase<Matrix>; being apparently ignored  as I have a class that  
should have a private default constructor, but in fact has one that  
just crashes...

                                        R


------------------------------------------------------------------------------
Come build with us! The BlackBerry(R) Developer Conference in SF, CA
is the only developer event you need to attend this year. Jumpstart your
developing skills, take BlackBerry mobile applications to market and stay
ahead of the curve. Join us from November 9 - 12, 2009. Register now!
http://p.sf.net/sfu/devconference
_______________________________________________
Swig-user mailing list
Swig-user@...
https://lists.sourceforge.net/lists/listinfo/swig-user

[wsfulton]

Robert Lupton the Good wrote:

> This one I do think is a bug rather than a "feature";  the non-use of  
> SwigValueWrapper when it's needed, and the ignoring of a %feature
> ("valuewrapper").  This is swig 1.3.40 and -python -c++.
>
> // -*- c++ -*-
> %module hoo
>
> %inline %{
>      template<typename Matrix>
>      class MatrixBase {
>      private:
>         MatrixBase() {}
>      };
>
>      class Matrix { };
>
>      template<typename T>
>      class TMatrix { };
> %}
>
> //%template(MMatrix) MatrixBase<Matrix>;   // makes no difference
> //%feature("valuewrapper") MatrixBase<Matrix>; // makes no difference
>
> %typemap(in) MatrixBase<Matrix>() { }
>
> %inline %{
> #if 1
>     void size(MatrixBase<Matrix> mat); // fails
> #else
>     void size(MatrixBase<TMatrix<int> > mat); // OK
> #endif
> %}
>
> With "#if 1" (i.e. void size(MatrixBase<Matrix> mat);) the generated  
> code uses:
>
> SWIGINTERN PyObject *_wrap_size(PyObject *SWIGUNUSEDPARM(self),  
> PyObject *args) {
>    PyObject *resultobj = 0;
>    MatrixBase< Matrix > arg1 ;
>
> and fails to compile due to MatrixBase's private constructor.  With  
> "#if 0" the code
> uses SwigValueWrapper and all is well:
>
> SWIGINTERN PyObject *_wrap_size(PyObject *SWIGUNUSEDPARM(self),  
> PyObject *args) {
>    PyObject *resultobj = 0;
>    SwigValueWrapper< MatrixBase< TMatrix< int > > > arg1 ;
>
> I'm actually most concerned about %feature("valuewrapper")  
> MatrixBase<Matrix>; being apparently ignored  as I have a class that  
> should have a private default constructor, but in fact has one that  
> just crashes...
>

There is some strange bug here due to symbol clashes between the
template parameter 'Matrix' and the class 'Matrix'. If you change

      template<typename Matrix> ...

to

      template<typename M> ...

you can work around the problem. Alternatively, you need your feature
declarations before the class that you want to attach them to, so use this:

      %feature("valuewrapper") MatrixBase;

before MatrixBase is defined.

William

------------------------------------------------------------------------------
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
trial. Simplify your report design, integration and deployment - and focus on
what you do best, core application coding. Discover what's new with
Crystal Reports now.  http://p.sf.net/sfu/bobj-july
_______________________________________________
Swig-user mailing list
Swig-user@...
https://lists.sourceforge.net/lists/listinfo/swig-user

[Sören Freudiger-2]

Hi
The wrapper module of SWIG 1.3.40 s not 100% ruby 1.9.1 compatible since
they changed

rb_intern(char *)

and
ID GC_VALUE::hash_id = rb_intern("hash");

does not compile any more...

see:
http://sourceforge.net/tracker/index.php?func=detail&aid=2859614&group_id=16
45&atid=101645

Is there any patch or workaround available yet?

Best,
SirAnn



------------------------------------------------------------------------------
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
trial. Simplify your report design, integration and deployment - and focus on
what you do best, core application coding. Discover what's new with
Crystal Reports now.  http://p.sf.net/sfu/bobj-july
_______________________________________________
Swig-user mailing list
Swig-user@...
https://lists.sourceforge.net/lists/listinfo/swig-user