Data Science

Defining a ChainMap

A ChainMap is a data structure that allows you to treat multiple dictionaries as one dictionary
In other words, a ChainMap is an updatable view over many dicts
However, it behaves just like a normal dict
The use cases include the following:
1. Searching through multiple dictionaries quickly
2. Making changes to multiple dictionaries quickly
3. Finding unique keys across multiple dictionaries quickly
  - This also can refer to providing a chain of default values

Example of Use-Case 1

>>> toys = {'Blocks': 30, 'Monopoly': 20}
>>> computers = {'iMac': 1000, 'iPad': 800, 'PC': 400}
>>> clothing = {'Jeans': 40, 'T-Shirt': 10}

>>> from collections import ChainMap
>>> inventory = ChainMap(toys, computers, clothing)

>>> print(inventory['Jeans'])
40
>>> inventory.get('Mario Bros.')
None

Example of Use-Case 2

>>> toys = {'Blocks': 30, 'Monopoly': 20}
>>> computers = {'iMac': 1000, 'iPad': 800, 'PC': 400}
>>> clothing = {'Jeans': 40, 'T-Shirt': 10}

>>> from collections import ChainMap
>>> inventory = ChainMap(toys, computers, clothing)

>>> inventory.pop('Blocks')
200
>>> print(toys)
{'Monopoly': 20}

>>> clothing['Jeans'] = 0
>>> print(inventory['Jeans'])
0

Example of Use-Case 3

>>> toys = {'Blocks': 30, 'Monopoly': 20}
>>> computers = {'iMac': 1000, 'iPad': 800, 'PC': 400}
>>> clothing = {'Jeans': 40, 'T-Shirt': 10}

>>> from collections import ChainMap
>>> inventory = ChainMap(toys, computers, clothing)

>>> unique_items = list(inventory.keys())
>>> print(unique_items)
['Jeans', 'T-Shirt', 'iMac', 'iPad', 'PC', 'Blocks', 'Monopoly']

References

Defaultdict

Namedtuple