Dictionaries

Another basic data type which we will instroduce is the dictionary, or short, dict.

Similar to lists, dictionaries contain values. In contrast to lists, where all elements are in a specific order, there are two types of elements in dictionaries: key and value. Exactly one value exists for every key. (I don't want to confuse you right at the beginning, but just so you know -- an empty list can also be a value for a key.)

You use a dictionary when each piece of data has an individual name, but you want to work with the data as one variable.

There is a dictionary with 3 keys, and each one of them has a value:

>>> me = {'name': 'Marketa', 'city': 'Prague', 'numbers': [20, 8]}

When you print the dictionary you may find that your keys and values are in a different order. Dictionaries don't have a fixed order of elements. They just assign values to keys.

You can get values from the dictionary similar as from lists, but instead of an index, you have to use the key.

>>> me['name']
'Marketa'

If you try to access a non-existent key, Python won't like it:

>>> me['age']
Traceback (most recent call last):
  File "<stdin>", line 1, in &lt;module&gt;
KeyError: 'age'

You can change the values of keys:

>>> me['numbers'] = [3, 7, 42]
>>> me
{'name': 'Marketa', 'city': 'Prague', 'numbers': [20, 8, 42]}

... or add keys and values:

>>> me['language'] = 'Python'
>>> me
{'name': 'Marketa', 'city': 'Prague', 'numbers': [20, 8, 42], 'language': 'Python'}

... or delete keys and values using the del command (also the same as for lists):

>>> del me['numbers']
>>> me
{'name': 'Marketa', 'city': 'Prague', 'language': 'Python'}

Lookup table

A use of dictionaries other than data clustering is the so-called lookup table. It stores values of same type.

This is useful for example with phone book. For every name there is one phone number. Other examples are dictionaries with properties of food, or word translations.

phones = {
    'Mary': '153 85283',
    'Theresa': '237 26505',
    'Paul': '385 11223',
    'Michael': '491 88047',
}

colours = {
    'pear': 'green',
    'apple': 'red',
    'melon': 'green',
    'plum': 'purple',
    'radish': 'red',
    'cabbage': 'green',
    'carrot': 'orange',
}

Iteration

When you loop over a dictionary using for, you will get only keys:

>>> func_descript = {'len': 'length', 'str': 'string', 'dict': 'dictionary'}
>>> for key in func_descript:
...     print(key)
str
dict
len

If you want to access the values, you will have to use the method values:

>>> for value in func_descript.values():
...     print(value)
string
dictionary
length

But in most cases, you will need both -- keys and values. For this purpose, dictionaries have the method items.

>>> for key, value func_descript.items():
...     print('{}: {}'.format(key, value))
str: string
dict: dictionary
len: length

In a for loop, you can't add keys to dictionary nor delete them:

>>> for key in func_descript:
...     del func_descript[key]
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
RuntimeError: dictionary changed size during iteration

But you can change values for already existing keys.

How to create a dictionary

Dictionaries can be created in two ways. The first way uses square brackets []. The other way is by using the keyword dict. This works similar to strings, integer or list, so it will convert some specific objects to a dictionary.

A dictionary has very specific structure -- numbers or simple lists can't be converted into a dictionary. But we can convert a dictionary into another dictionary. This new dictionary won't be in any way connected to the old one.

colour_riped = dict(colours)
for key in colour_riped:
    colour_riped[key] = 'blackish-brownish-' + colour_riped[key]
print(colours['apple'])
print(colour_riped['apple'])

We can also convert a list which contains tuples with pairs (which work as ke and value) into a dictionary:

data = [(1, 'one'), (2, 'two'), (3, 'three')]
number_names = dict(data)

And that's all that we can convert into a dictionary.

As a bonus function, dict can also work with named arguments. Each argument's name will be a key and the argument itself will be the value:

func_descript = dict(len='length', str='string', dict='dictionary')
print(func_descript['len'])

And that's all for now

If you would like to know all the tricks about dictionaries you can look at (and also print) this cheatsheet.

A complete description can be found here in the Python documentation.