Python Inverse Zip

Zip

zip is a builtin that groups iterables together. Imagine you have a list of names of users and a list of user ids:

username = ['alice', 'bob', 'carol']
user_id = [1111, 2222, 1337]

users = zip(username, userid)
for name, id in users:
  print(f'I am {name} and my id is {id}')

# Less elegant but equivalent
for i in range(min(len(username), len(user_id))):
  name, id = username[i], user_id[i]
  print(f'I am {name} and my id is {id}')

The data is segregated and we want to make it sane to work with. Maybe the data came from a columnar-esque datastore.

Inverse

But what’s the opposite? How do you unzip?

Apparently you use zip and tuple unpacking.

username, user_id = zip(*users)

How does this even work? What is this wizardry?

It’s actually not magic or special-case handling at all. It’s combining tuple unpacking (*args) with zip, to reverse the operation.

users = [('alice', 111), ('bob', 222), ('carol', 1337)]

# *users expands to zip(('alice', 111), ('bob', 222), ('carol', 1337))
# It passes in 3 arguments, each an iterable of 2 items.
# Zip grabs the first item of each and collects them together
username, user+id = zip(*users)

Well, would you look at that.

Put more succinctly:

In other words lambda x: zip(*x) is self-inverse.