Patch #1671450: add a section about subclassing builtin types to the

"extending and embedding" tutorial.
 (backport from rev. 54150)
This commit is contained in:
Georg Brandl 2007-03-06 10:02:59 +00:00
parent 0ea891603d
commit 1d56c2ff6a
3 changed files with 190 additions and 1 deletions

View File

@ -489,7 +489,6 @@ this?
garbage collection, there are calls that can be made to ``untrack''
the object from garbage collection, however, these calls are
advanced and not covered here.}
\item
\end{itemize}
@ -930,6 +929,102 @@ That's pretty much it. If we had written custom \member{tp_alloc} or
collection. Most extensions will use the versions automatically
provided.
\subsection{Subclassing other types}
It is possible to create new extension types that are derived from existing
types. It is easiest to inherit from the built in types, since an extension
can easily use the \class{PyTypeObject} it needs. It can be difficult to
share these \class{PyTypeObject} structures between extension modules.
In this example we will create a \class{Shoddy} type that inherits from
the builtin \class{list} type. The new type will be completely compatible
with regular lists, but will have an additional \method{increment()} method
that increases an internal counter.
\begin{verbatim}
>>> import shoddy
>>> s = shoddy.Shoddy(range(3))
>>> s.extend(s)
>>> print len(s)
6
>>> print s.increment()
1
>>> print s.increment()
2
\end{verbatim}
\verbatiminput{shoddy.c}
As you can see, the source code closely resembles the \class{Noddy} examples in previous
sections. We will break down the main differences between them.
\begin{verbatim}
typedef struct {
PyListObject list;
int state;
} Shoddy;
\end{verbatim}
The primary difference for derived type objects is that the base type's
object structure must be the first value. The base type will already
include the \cfunction{PyObject_HEAD} at the beginning of its structure.
When a Python object is a \class{Shoddy} instance, its \var{PyObject*} pointer
can be safely cast to both \var{PyListObject*} and \var{Shoddy*}.
\begin{verbatim}
static int
Shoddy_init(Shoddy *self, PyObject *args, PyObject *kwds)
{
if (PyList_Type.tp_init((PyObject *)self, args, kwds) < 0)
return -1;
self->state = 0;
return 0;
}
\end{verbatim}
In the \member{__init__} method for our type, we can see how to call through
to the \member{__init__} method of the base type.
This pattern is important when writing a type with custom \member{new} and
\member{dealloc} methods. The \member{new} method should not actually create the
memory for the object with \member{tp_alloc}, that will be handled by
the base class when calling its \member{tp_new}.
When filling out the \cfunction{PyTypeObject} for the \class{Shoddy} type,
you see a slot for \cfunction{tp_base}. Due to cross platform compiler
issues, you can't fill that field directly with the \cfunction{PyList_Type};
it can be done later in the module's \cfunction{init} function.
\begin{verbatim}
PyMODINIT_FUNC
initshoddy(void)
{
PyObject *m;
ShoddyType.tp_base = &PyList_Type;
if (PyType_Ready(&ShoddyType) < 0)
return;
m = Py_InitModule3("shoddy", NULL, "Shoddy module");
if (m == NULL)
return;
Py_INCREF(&ShoddyType);
PyModule_AddObject(m, "Shoddy", (PyObject *) &ShoddyType);
}
\end{verbatim}
Before calling \cfunction{PyType_Ready}, the type structure must have the
\member{tp_base} slot filled in. When we are deriving a new type, it is
not necessary to fill out the \member{tp_alloc} slot with
\cfunction{PyType_GenericNew} -- the allocate function from the base type
will be inherited.
After that, calling \cfunction{PyType_Ready} and adding the type object
to the module is the same as with the basic \class{Noddy} examples.
\section{Type Methods
\label{dnt-type-methods}}

91
Doc/ext/shoddy.c Normal file
View File

@ -0,0 +1,91 @@
#include <Python.h>
typedef struct {
PyListObject list;
int state;
} Shoddy;
static PyObject *
Shoddy_increment(Shoddy *self, PyObject *unused)
{
self->state++;
return PyInt_FromLong(self->state);
}
static PyMethodDef Shoddy_methods[] = {
{"increment", (PyCFunction)Shoddy_increment, METH_NOARGS,
PyDoc_STR("increment state counter")},
{NULL, NULL},
};
static int
Shoddy_init(Shoddy *self, PyObject *args, PyObject *kwds)
{
if (PyList_Type.tp_init((PyObject *)self, args, kwds) < 0)
return -1;
self->state = 0;
return 0;
}
static PyTypeObject ShoddyType = {
PyObject_HEAD_INIT(NULL)
0, /* ob_size */
"shoddy.Shoddy", /* tp_name */
sizeof(Shoddy), /* tp_basicsize */
0, /* tp_itemsize */
0, /* tp_dealloc */
0, /* tp_print */
0, /* tp_getattr */
0, /* tp_setattr */
0, /* tp_compare */
0, /* tp_repr */
0, /* tp_as_number */
0, /* tp_as_sequence */
0, /* tp_as_mapping */
0, /* tp_hash */
0, /* tp_call */
0, /* tp_str */
0, /* tp_getattro */
0, /* tp_setattro */
0, /* tp_as_buffer */
Py_TPFLAGS_DEFAULT |
Py_TPFLAGS_BASETYPE, /* tp_flags */
0, /* tp_doc */
0, /* tp_traverse */
0, /* tp_clear */
0, /* tp_richcompare */
0, /* tp_weaklistoffset */
0, /* tp_iter */
0, /* tp_iternext */
Shoddy_methods, /* tp_methods */
0, /* tp_members */
0, /* tp_getset */
0, /* tp_base */
0, /* tp_dict */
0, /* tp_descr_get */
0, /* tp_descr_set */
0, /* tp_dictoffset */
(initproc)Shoddy_init, /* tp_init */
0, /* tp_alloc */
0, /* tp_new */
};
PyMODINIT_FUNC
initshoddy(void)
{
PyObject *m;
ShoddyType.tp_base = &PyList_Type;
if (PyType_Ready(&ShoddyType) < 0)
return;
m = Py_InitModule3("shoddy", NULL, "Shoddy module");
if (m == NULL)
return;
Py_INCREF(&ShoddyType);
PyModule_AddObject(m, "Shoddy", (PyObject *) &ShoddyType);
}

View File

@ -457,6 +457,9 @@ Tests
Documentation
-------------
- Patch #1671450: add a section about subclassing builtin types to the
"extending and embedding" tutorial.
- Bug #1629125: fix wrong data type (int -> Py_ssize_t) in PyDict_Next
docs.