extension modules in C for python, check for korrekt Object

[Torsten Mohr]

Hi,

i write a set of extension modules for python in C.
I also use some new types that i defined there.

At the moment i write some methods that expect some
of these types as parameters. I now want to check
that the right types were given as parameters.

How can i do that?

PyArg_ParseTupl e only checks that an object was
given as parameter, but not its type.

Do i need to check tp_name in the objects type?
Thanks for hints,
Torsten.

[Greg Ewing]

Torsten Mohr wrote:

At the moment i write some methods that expect some
of these types as parameters. I now want to check
that the right types were given as parameters.

Use PyObject_IsInst ance().

--
Greg Ewing, Computer Science Dept,
University of Canterbury,
Christchurch, New Zealand
http://www.cosc.canterbury.ac.nz/~greg

[Heiko Wundram]

Am Mittwoch, 2. Juni 2004 05:26 schrieb Greg Ewing:

Use PyObject_IsInst ance().

Or PyObject_TypeCh eck()... All documented in the Python C/API documentation on
python.org.

Heiko.

[Michael Hudson]

Torsten Mohr <tm***@s.netic. de> writes:

Hi,

i write a set of extension modules for python in C.
I also use some new types that i defined there.

At the moment i write some methods that expect some
of these types as parameters. I now want to check
that the right types were given as parameters.

How can i do that?

PyArg_ParseTupl e only checks that an object was
given as parameter, but not its type.
It can check type, using the "O!" incantation.

If you're asking "how do I get hold of the type object I defined in
some other extension module?", then that's a good question. One
approach is to use "CObjects":

http://docs.python.org/ext/using-cobjects.html
Do i need to check tp_name in the objects type?

This sounds fragile :-)

Cheers,
mwh

--
Of the four project development variables - scope, cost, time and
quality - quality isn't really a free variable. The only possible
values are "excellent" and "insanely excellent", depending on
whether lives are at stake. -- Kent Beck, XP Explained

[Steve Menard]

Torsten Mohr wrote:

Hi,

i write a set of extension modules for python in C.
I also use some new types that i defined there.

At the moment i write some methods that expect some
of these types as parameters. I now want to check
that the right types were given as parameters.

How can i do that?

PyArg_ParseTupl e only checks that an object was
given as parameter, but not its type.

Do i need to check tp_name in the objects type?
Thanks for hints,
Torsten.

The PyArg_ParseTupl e method can do type checking for you, as long as
your requirements are pretty simple.

if you method accepts only one argument, and that argument must be of a
specific type, the following will work :

PyObject* some_method(PyO bject* self, PyObject* args)
{
PyObject* value;

if (! PyArg_ParseTupl e(args, "O!", &typeStruct, &value))
{
// Exception ahs already been set byt the call ... so
// only return NULL
return NULL;
}
...
}

typeStruct is a reference to the PyObejctType structure that defines the
type. either the one you defined or those already available through python.

Steve