Pythonic update of a `defaultdict`

What is a more "pythonic" version of this code? (in Python 2.x)

from collections import defaultdict
dd = defaultdict(list)
for i in some_list_of_items:
    current_dict = dict_from_an_item(i)
    for (k, v) in current_dict.items():
        dd[k].extend(v)

the dict_from_an_item parses an item and returns a dict that contains not nested lists as the values. Something like this :

{ 'key1': [开发者_JAVA技巧1, 2, 3],
  'key2': [2, 3, 4],
  'key3': [3, 4, 5, 6, 7, 8]}

---

You could spare the current_dict variable. Also k, v is more pythonic than (k, v)

from collections import defaultdict

dd = defaultdict(list)
for i in some_list_of_items:
    for k, v in dict_from_an_item(i).iteritems():
        dd[k].extend(v)

---

if you don't need a defaultdict for performance reasons you could do:

d = {k:v for i in some_list_of_items for k, v in dict_from_an_item(i).iteritems()}