Paolino wrote:
> class NotInitializedObjects(type):
> def __init__(cls,*_):
> realInit=cls.__init__
> def __newInit__(self,*pos,**key):
> def _init():
> realInit(self,*pos,**key)
> self._init=_init
> cls.__init__=__newInit__
> def __getattribute__(self,attr):
> def getter(attr):
> return object.__getattribute__(self,attr)
> if '_init' in getter('__dict__'):
> getter('_init')()
> del self._init
> return getter(attr)
> cls.__getattribute__=__getattribute__
>
A lighter solution can be overriding __getattr__.
This will produce more object-like behaving instances even when not
initialized, aka you can call methods and access class attributes
without triggering the init (not very useful)
class NotInitializedObjects(type):
def __init__(cls,*_):
realInit=cls.__init__
def __newInit__(self,*pos,**key):
def _init():
realInit(self,*pos,**key)
self._init=_init
cls.__init__=__newInit__
def __getattr__(self,attr):
if hasattr(self,'_init'):
self._init()
del self._init
if hasattr(self,attr):
return getattr(self,attr)
raise AttributeError
cls.__getattr__=__getattr__
### Test with previous testing code
A cleaner solution is decoupling the intensive calculation attributes
from __init__ and use descriptors for them.But this is impossible if
/the/ instance value is the intensive one to be calculated.
Ciao Paolino
___________________________________
Aggiungi la toolbar di Yahoo! Search sul tuo Browser, e'gratis!
http://it.toolbar.yahoo.com
--
http://mail.python.org/mailman/listinfo/python-list
