Access to C functions from a python class using self ?

[Giorgi Lekishvili]

Giorgi Lekishvili wrote:

> Hi!
> I guess I should try this:
> PyObject* PyObject_GetAttr(PyObject *o, PyObject *attr_name)
>
> See http://www.python.org/doc/current/api/object.html

Thomas is right, GetAttrString is better.

>
>
> On the other hand, my opinion is that, you'd better use either swig in a way
> that make all stuff in C/C++ first and then wrap ot to Python, or consider
> Pyrex:
> http://www.cosc.canterbury.ac.nz/~greg/python/Pyrex/
>
> christophe grimault wrote:
>
> > Hi there,
> >
> > I've been looking to 'ext & emb python', swig, cxx, ... for quite some
> > time now, and I still can't figure out how to do the kind of stuff
> > explained below...
> >
> > I define a class in python, let's say this simple class, just for the
> > purpose of discussion :
> >
> > class Vector:
> >     def __init__(self, x1, x2, x3):
> >         self.x1, self.x2, self.x3 = x1, x2, x3
> >
> >     def sum(self):
> >         return self.x1 + self.x2 + self.x3
> >
> > Now, I find that sum is slow, and I want to speedup things by replacing
> > sum by a call to C.
> > Looking at the manual, I can write something like:
> >
> > static PyObject *sum( PyObject *self, PyObject *args )
> >
> > and  the manual tells me how to access elements passed in args, but does
> > not tell how to access elements (in fact attributes) of self. How can I
> > retrieve self.x1, self.x2, and self.x3 as defined in the python class
> > from *self passed to the C function ?
> >
> > The goal for me is to create classes quickly in Python. These classes
> > have, say, a constructor and 5 to 30 methods. The object itself is quite
> > complex (lots of attributes of different types) and only one or two
> > method need very high speed and thus, must call C.
> >
> > And i want to avoid passing all the attributes needed by those C calls
> > by increasing the *args !
> >
> > Any help would be appreciated !
> >
> > Christophe
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20021230/c0bf023f/attachment.html>