Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.lang.python > #35841

Re: father class name

Show all headers | View raw

On Sun, Dec 30, 2012 at 8:18 PM, contro opinion <contropinion@gmail.com> wrote:
> here is my haha  class
> class  haha(object):
>   def  theprint(self):
>     print "i am here"
>
>>>> haha().theprint()
> i am here
>>>> haha(object).theprint()
> Traceback (most recent call last):
>   File "<stdin>", line 1, in <module>
> TypeError: object.__new__() takes no parameters
>
> why   haha(object).theprint()  get wrong output?

The fact that `haha(object)` is textually part of the *declaration*
`class haha(object):` has no bearing on how one instantiates an
instance of the class `haha`.
In the `class` statement, `haha` is being declared to be a subclass of
class `object` (that's what it means for `object` to be in the
parentheses after the class name in a `class` statement; the syntax is
"class <classname>(<base classes>):").

In the first part of the *expression* `haha().theprint()`, you are
using the function-call operator on the `haha` class itself, which has
the effect of instantiating it; since you gave no arguments in the
function call, haha's initializer (i.e. its __init__() method) was
given no arguments. Since you didn't define an __init__() method for
haha, haha inherited the default __init__() method from class
`object`, which takes no arguments, so your call was fine and worked
as expected.

By contrast, in the first part of the *expression*
`haha(object).theprint()`, you passed an argument (namely, `object`).
Since __init__() wasn't expecting any arguments whatsoever, you
therefore got an error.

The parentheses in a `class` statement do NOT signify a function call;
they are part of the syntax of the `class` statement itself.

Cheers,
Chris
--
Note: I'm oversimplifying things a bit for the sake of understandability.

Back to comp.lang.python | Previous | Next | Find similar | Unroll thread

Thread

Re: father class name Chris Rebert <clp2@rebertia.com> - 2012-12-30 23:27 -0800

csiph-web