Python: update a dict by object

When you have a dict and a non-iterable object you'd like to update the dict from, you can do this:

class A:
    def __init__(self):
        self.fields_names_list = [..attribute names here..]

    def __iter__(self):
        for name in self.fields_names_list:
            yield name, getattr(self, name, None)

And voila, it works:

a = A()
a.foo = 'bar'
b = dict()
b.update(a)
a
# {'foo': 'bar'}

Update: my solution is for legacy code, in general, it's better to use @rhymes solution from the comments.

Comments

[rhymes]

Hi Maria, neat trick, you made me think about it and I found a few alternatives if you don't care about whitelisting:

>>> class A:
...     def __init__(self):
...         self.a = 3
...     def __iter__(self):
...         for name in self.__dict__:
...             yield name, getattr(self, name)
...
>>> a = A()
>>>
>>> a.b = 4
>>>
>>> dict(a)
{'a': 3, 'b': 4}

and even easier one without __iter__:

>>> class B:
...     def __init__(self):
...         self.a = 3
...
>>> b = B()
>>> b.c = 5
>>> dict(b)
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
TypeError: 'B' object is not iterable
>>> vars(b)
{'a': 3, 'c': 5}
>>> b.__dict__
{'a': 3, 'c': 5}

None is wrong, it depends on how much control you want to have and the last one works also on objects without __iter__. Keep in mind that vars works with the object with iter as well.

[Maria Boldyreva]

Hi. In general case, your solution is better and cleaner than mine. I was so excited to have found the solution (it was a work task) that I forgot to mention, that the object is a huge legacy object and I needed just a few of its attributes and .update(obj) is a legacy line I have no control over. I'll update the post soon, thanks for the comment.

[rhymes]

Ahah you see, then your solution is more apt than mine ;-)

I just gave you two other options that are useful in case you cannot modify the source code of the class.