pygccxml-development

[pygccxml-development] Future freeze

[Roman Y. <rom...@gm...>]

Hi. I would like to freeze Py++ & pygccxml code for introducing new features.
The only exception to this is FT feature. I think Py++ accumulated
enough changes and we should release them.

I ask you to test the source code and to report any bug you have. Also another
very important task is documentation. Please read the documentation and report
any problem you have with it( spelling errors, not good enough, not
clear enough ... ).
It could be nice if you say what area of Py++ is still undocumented
and should be.

Thank you.

-- 
Roman Yakovenko
C++ Python language binding
http://www.language-binding.net/

Re: [pygccxml-development] Future freeze

[Allen B. <al...@vr...>]

Roman Yakovenko wrote:

>Hi. I would like to freeze Py++ & pygccxml code for introducing new features.
>The only exception to this is FT feature. I think Py++ accumulated
>enough changes and we should release them.
>
>I ask you to test the source code and to report any bug you have. Also another
>very important task is documentation. Please read the documentation and report
>any problem you have with it( spelling errors, not good enough, not
>clear enough ... ).
>It could be nice if you say what area of Py++ is still undocumented
>and should be.
>
>Thank you.
>
>  
>
Roman:

Will this version of Py++ have a helper method that can completely 
finalize a class?

-Allen

Re: [pygccxml-development] Future freeze

[Roman Y. <rom...@gm...>]

On 10/10/06, Allen Bierbaum <al...@vr...> wrote:
> Will this version of Py++ have a helper method that can completely
> finalize a class?

Do you mean to not create wrapper for other cases than virtual and
pure virtual member
functions? If so than the answer is not, it will not contain such feature.


-- 
Roman Yakovenko
C++ Python language binding
http://www.language-binding.net/

Re: [pygccxml-development] Future freeze

[Allen B. <al...@vr...>]

Roman Yakovenko wrote:

> On 10/10/06, Allen Bierbaum <al...@vr...> wrote:
>
>> Will this version of Py++ have a helper method that can completely
>> finalize a class?
>
>
> Do you mean to not create wrapper for other cases than virtual and
> pure virtual member
> functions? If so than the answer is not, it will not contain such 
> feature.
>
>
I mean don't create a wrapper at all.

-Allen

Re: [pygccxml-development] Future freeze

[Roman Y. <rom...@gm...>]

On 10/10/06, Allen Bierbaum <al...@vr...> wrote:
> Roman Yakovenko wrote:
>
> > On 10/10/06, Allen Bierbaum <al...@vr...> wrote:
> >
> >> Will this version of Py++ have a helper method that can completely
> >> finalize a class?
> >
> >
> > Do you mean to not create wrapper for other cases than virtual and
> > pure virtual member
> > functions? If so than the answer is not, it will not contain such
> > feature.
> >
> >
> I mean don't create a wrapper at all.

No.

-- 
Roman Yakovenko
C++ Python language binding
http://www.language-binding.net/

Re: [pygccxml-development] Future freeze

[Allen B. <al...@vr...>]

Ok.

Well, then I have a method in goodies that tries to finalize a class.  
It looks like this:

-------------------------------------
def finalize(cls, finalize_pure_virtuals=True):
   if isinstance(cls, pd.mdecl_wrapper_t):
      for x in cls:
         finalize(x)
   else:
      matcher = pd.virtuality_type_matcher( pd.VIRTUALITY_TYPES.VIRTUAL )
      if finalize_pure_virtuals:
         matcher = matcher | pd.virtuality_type_matcher( 
pd.VIRTUALITY_TYPES.PURE_VIRTUAL)
      members = cls.decls( matcher,
                           decl_type=pd.member_calldef_t,
                           allow_empty=True)
      members.set_virtuality( pd.VIRTUALITY_TYPES.NOT_VIRTUAL )
--------------------------------------

This gets rid of all the virtuals but there are still some things left 
over that make Py++ create a wrapper.  Any hints at what other things 
will make Py++ create wrappers?

-Allen

Re: [pygccxml-development] Future freeze

[Roman Y. <rom...@gm...>]

On 10/10/06, Allen Bierbaum <al...@vr...> wrote:
>Any hints at what other things
> will make Py++ create wrappers?

http://svn.sourceforge.net/viewvc/pygccxml/pyplusplus_dev/pyplusplus/decl_wrappers/class_wrapper.py?view=markup

Take a look on is_wrapper_needed method

-- 
Roman Yakovenko
C++ Python language binding
http://www.language-binding.net/

Re: [pygccxml-development] Future freeze

[Allen B. <al...@vr...>]

Roman Yakovenko wrote:

> On 10/10/06, Allen Bierbaum <al...@vr...> wrote:
>
>> Any hints at what other things
>> will make Py++ create wrappers?
>
>
> http://svn.sourceforge.net/viewvc/pygccxml/pyplusplus_dev/pyplusplus/decl_wrappers/class_wrapper.py?view=markup 
>
>
> Take a look on is_wrapper_needed method
>
Thanks.  That is a very helpful method. :)

I understand most of the reasons in there, but I don't understand any of 
the member variable reasons.  Why does a bitfield, T*, T&, or array 
member variable require a wrapper to be created?

-Allen

Re: [pygccxml-development] Future freeze

[Roman Y. <rom...@gm...>]

On 10/10/06, Allen Bierbaum <al...@vr...> wrote:
> Roman Yakovenko wrote:
>
> > On 10/10/06, Allen Bierbaum <al...@vr...> wrote:
> >
> >> Any hints at what other things
> >> will make Py++ create wrappers?
> >
> >
> > http://svn.sourceforge.net/viewvc/pygccxml/pyplusplus_dev/pyplusplus/decl_wrappers/class_wrapper.py?view=markup
> >
> >
> > Take a look on is_wrapper_needed method
> >
> Thanks.  That is a very helpful method. :)
>
> I understand most of the reasons in there, but I don't understand any of
> the member variable reasons.  Why does a bitfield, T*, T&, or array
> member variable require a wrapper to be created?

bitfield - you can not take reference to the bit, get\set functions are needed
T*, & - you should instruct Boost.Python how to manage life time of
the variable - you
    have to suppy call policies
array - boost.python does not supoprt arrays, so there is a need to
construct instance
of the class that exposes array to Python.


-- 
Roman Yakovenko
C++ Python language binding
http://www.language-binding.net/