Reduce the list and add to why on every deletion

list_reduction.py

my_input = [{'why': [u'The Matrix Revolutions'], 'title': u'The Matrix'}, 
{'why': [u'The Matrix Revolutions'], 'title': u'Inception'}, 
{'why': [u'The Matrix'], 'title': u'The Matrix Revolutions'}, 
{'why': [u'The Matrix'], 'title': u'Inception'}, 
{'why': [u'Saving Private Ryan'], 'title': u'The Hurt Locker'},
 {'why': [u'Black Hawk Down'], 'title': u'Saving Private Ryan'}, 
{'why': [u'Black Hawk Down'], 'title': u'The Kingdom'}, 
{'why': [u'Black Hawk Down'], 'title': u'The Hurt Locker'},
 {'why': [u'A Walk To Remember'], 'title': u'The Notebook'}, 
{'why': [u'Cast Away'], 'title': u'Forrest Gump'}, 
{'why': [u'Forrest Gump'], 'title': u'The Notebook'}, 
{'why': [u'Body of Lies'], 'title': u'The Kingdom'}, 
{'why': [u'Unthinkable'], 'title': u'Syriana'}, 
{'why': [u'Lord of War'], 'title': u'Syriana'}, 
{'why': [u'The Terminal'], 'title': u'Forrest Gump'}, 
{'why': [u'Titanic'], 'title': u'The Notebook'}, 
{'why': [u'Catch Me If You Can'], 'title': u'Forrest Gump'}, 
{'why': [u'Syriana'], 'title': u'The Constant Gardener'}, 
{'why': [u'Nothing But the Truth'], 'title': u'Syriana'}, 
{'why': [u'Hotel Rwanda'], 'title': u'The Constant Gardener'}, 
{'why': [u'Hotel Rwanda'], 'title': u'Syriana'}, 
{'why': [u'Inception'], 'title': u'The Matrix'}, 
{'why': [u'Inception'], 'title': u'The Matrix Revolutions'},
 {'why': [u'The Hurt Locker'], 'title': u'Saving Private Ryan'},
 {'why': [u'The Hurt Locker'], 'title': u'The Kingdom'}]

#remove all the duplicate 'title' and append each 'why'
my_output = [{'why': [u'Unthinkable',u'Lord of War',u'Hotel Rwanda'], 'title': u'Syriana'}]

How about the following piece of code?

output = {}
for d in my_input:
    whys = output.setdefault(d['title'], [])
    whys.append(d['title'])

looks ugly, but does the job ;)

a = groupby(
    sorted(
        [(x['title'], x['why'][0]) for x in my_input]),
    key=lambda x: x[0])
my_output = [{'title': x[0], 'why': [y[1] for y in x[1]]} for x in a]

oh, i forgot that you need to

from itertools import groupby

output = {}
for d in my_input:
    whys = output.setdefault(d['title'], [])
    whys.extend(d['why'])

And for the complete example which produces the correct output I'd go for:

output = {}
for d in my_input:
    try:
        unique = output[d['title']]
    except KeyError:
        unique = output[d['title']] = {'why': [], 'title': d['title']}
    unique['why'].extend(d['why'])
output = output.values()