Paddy O'Loughlin a écrit :
2009/2/20 Bruno Desthuilliers <bruno.42.desthuilli...@websiteburo.invalid>:
Interesting. Why shouldn't you?
I haven't used the property() function
s/function/object/
Nice try, but what I wrote was what I intended to say:
http://docs.python.org/library/functions.html#property
Check by yourself:
>>> import inspect
>>> inspect.isfunction(property)
False
>>> property()
<property object at 0xb7cbc144>
>>> dir(property)
['__class__', '__delattr__', '__delete__', '__doc__', '__get__',
'__getattribute__', '__hash__', '__init__', '__new__', '__reduce__',
'__reduce_ex__', '__repr__', '__set__', '__setattr__', '__str__',
'fdel', 'fget', 'fset']
>>>
The case is that the whole point of using a computed attribute is to perform
some computation on the value. IOW, except for a couple corner cases, only
the accessors should directly access the implementation(s) attributes(s).
And of course, like for any other GoldenRule(tm), it's not meant to be
blindly followed. It's just that most of the times, going thru the accessors
is really what you want - even from within the class code.
Hmm, it doesn't seem to me like it's much of a big deal, for it to
described as anything like a "GoldenRule"
One could say the same for each and any of the usual GoldenRules(tm).
--
http://mail.python.org/mailman/listinfo/python-list
