Which methods implement the buffer interface in Python?
I have a custom class with a serialize
method, and I want to be able to write this class directly to files and have the return value of the serialize
method get written, in Python 2.6. (I'm not trying to pickle my objects, this is something totally different.) For example:
class Foo(object):
def serialize(self):
return "Hello World!"
__str__ = serialize
foo = Foo()
f = open("foo.dat", "wb")
f.write(foo)
However, when I run this code, I get the following exception
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
TypeError: argument 1 must be convertible to a buffer, not Foo
Okay, so I need my Foo
class to implement the buffer
interface. I even see in the buffer documentation that it says: "An example user of the buffer interface is the file object’s write() method. Any object that can export a series of bytes through the buffer interface can be written to a file."
So apparently I can do what I want, but the docs don't actually say which methods I need to implement in order to have implemented the buffer interface. I've tried implementing __str__
, __unicode__
, __len__
, and 开发者_如何学Ceven __sizeof__
. I've implemented __getitem__
, __setitem__
, and __delitem__
, accepting both int
and slice
arguments. I've even tried implementing the deprecated __getslice__
, __setslice__
, and __delslice__
methods just to be safe. No matter what I try, I still get exactly the same exception.
For reference, here are the methods of the buffer
builtin class:
>>> dir(buffer)
['__add__', '__class__', '__cmp__', '__delattr__', '__delitem__', '__delslice__',
'__doc__', '__format__', '__getattribute__', '__getitem__', '__getslice__',
'__hash__', '__init__', '__len__', '__mul__', '__new__', '__reduce__',
'__reduce_ex__', '__repr__', '__rmul__', '__setattr__', '__setitem__',
'__setslice__', '__sizeof__', '__str__', '__subclasshook__']
I'd like to avoid implementing all of them one by one, and I'd especially like to find the documentation on exactly which methods are necessary.
Is this something that can only be implemented in C extension classes? Or am I missing something obvious?
Peculiarly, there are no special methods that a pure-Python coded class can directly implement to support the buffer
interface -- that would have been PEP 298, but it was withdrawn.
I'm afraid you'll have to use some explicit attribute or method (or a built-in like str
which internally calls a special method) to pass your class's instances to file.write
&c:-(.
The documentation says,
Python objects implemented in C can export a group of functions called the “buffer interface.”
Your Foo class is not implemented in C, is it...?
精彩评论