📦 The Collections Module
The collections module in Python provides specialised container datatypes, that are alternatives to Python's general-purpose containers like dicts, lists, sets, and tuples.
These specialised containers are designed to provide additional functionality and can sometimes lead to more efficient code, both in terms of speed and memory usage.
namedtuple()
Use these when you need to access elements by name to make the code more readable and self-documenting. Access is as fast as a normal tuple because namedtuple instances are just as lightweight. For example:
from collections import namedtuple
Point = namedtuple('Point', ['x', 'y'])
pt = Point(1, 2)
print(pt.x, pt.y) # Output: 1 2
deque
Use deque when you need to add or pop elements from both ends of a collection efficiently. It provides faster appends and pops from the left end but is slower than lists for random access:
from collections import deque
dq = deque([1, 2, 3])
dq.appendleft(0) # Add to the left
dq.pop() # Remove from the right
print(dq) # Output: deque([0, 1, 2])
Counter
Counters make counting the occurrence of elements in an iterable, and performing element-wise operations, more efficient than manually using a dictionary. For example:
from collections import Counter
cnt = Counter('abracadabra')
print(cnt) # Output: Counter({'a': 5, 'b': 2, 'r': 2, 'c': 1, 'd': 1})
OrderedDict
Prior to Python 3.7, normal dicts didn’t preserve insertion order. OrderedDict can be useful if your code still relies on earlier Python versions:
from collections import OrderedDict
od = OrderedDict([('a', 1), ('b', 2), ('c', 3)])
od['d'] = 4
print(od) # Output: OrderedDict([('a', 1), ('b', 2), ('c', 3), ('d', 4)])
defaultdict
When you need a dictionary that automatically initialises non-existent keys with a default value, defaultdict can help:
from collections import defaultdict
dd = defaultdict(list)
for i in ['a', 'b', 'a']:
dd[i].append(1)
print(dd) # Output: defaultdict(<class 'list'>, {'a': [1, 1], 'b': [1]})
ChainMap
ChainMap lets you combine multiple dictionaries or mappings into a single view, avoiding potentially costly dictionary merges:
from collections import ChainMap
defaults = {'color': 'red', 'user': 'guest'}
environment = {'user': 'admin', 'path': 'bin'}
cm = ChainMap(environment, defaults)
print(cm['color']) # Output: red
print(cm['user']) # Output: admin