Basically, doing in a class method
def(self, a, b, c):
self.a = a
self.b = b
self.c = c
sounds stupid. With next instance I'll loose a, b, c, so I have to save
then to a variable, "self." prefix is generally proposed way. But
it's not surprising a gets to self.a, right? Actually, I thought about
the *args tuple and even **kwargs, but I thought this will make the core
even less readable. Thinking of it now, you'd probably do
self.__dict__.update(kwargs), right? Hmm, but does it assign to self or not?
I mean, does it equivalent to `a = 1' or `self.a = 1' ? The latter seem to
be true, right?
Try it and see:
py> def update1(obj, **kwargs): ... obj.__dict__.update(kwargs) ... py> def update2(obj, a, b, c): ... obj.a, obj.b, obj.c = a, b, c ... py> class C(object): ... pass ... py> c = C() py> update1(c, a=1, b=2, c=3) py> c.a, c.b, c.c (1, 2, 3) py> c = C() py> update2(c, a=1, b=2, c=3) py> c.a, c.b, c.c (1, 2, 3) py> c = C() py> update1(c, 1, 2, 3) Traceback (most recent call last): File "<interactive input>", line 1, in ? TypeError: update1() takes exactly 1 argument (4 given) py> c = C() py> update2(c, 1, 2, 3) py> c.a, c.b, c.c (1, 2, 3)
Note however that unless you do something like
def update(obj, **kwargs):
assert list(kwargs) == 'a b c'.split()
...then the update with the **kwargs will take any parameters (not just a, b and c). Don't know if that matters to you.
STeVe -- http://mail.python.org/mailman/listinfo/python-list
