Sanjay wrote:
> Hi All,
>
> Not being able to figure out how are partial classes coded in Python.
>
> Example: Suppose I have a code generator which generates part of a
> business class, where as the custome part is to be written by me. In
> ruby (or C#), I divide the code into two source files. Like this:
I would do this by inheritance if really needed - what isn't working?
That said, you can get this behaviour fairly easy with a metaclass:
class PartialMeta(type):
registry = {}
def __new__(cls,name,bases,dct):
if name in PartialMeta.registry:
cls2=PartialMeta.registry[name]
for k,v in dct.items():
setattr(cls2, k, v)
else:
cls2 = type.__new__(cls,name,bases,dct)
PartialMeta.registry[name] = cls2
return cls2
class PartialClass(object):
__metaclass__=PartialMeta
use:
#generatedperson.py
class Person(PartialClass):
def foo(self): print "foo"
#gandcraftedperson.py
import generatedperson
class Person(PartialClass):
def bar(self): print "bar"
and you should get similar behaviour.
Caveats:
I've used just the name to determine the class involved to be the
same as your Ruby example. However, this means that any class in any
namespace with the name Person and inheriting from PartialClass will be
interpreted as the same - this might not be desirable if some other
library code has a different Person object doing the same thing. Its
easy to change to checking for some property instead - eg. have your
handcrafted class have the line "__extends__=generatedperson.Person" ,
and check for it in the metaclass instead of looking in a name
registry.
Also, if two or more classes define the same name, the last one
evaluated will overwrite the previous one.
--
http://mail.python.org/mailman/listinfo/python-list
