Python: Dictionaries

A dictionary is a structure in Python for representing a one-way mapping between what we call a key to what we call a value.

For instance, we can create a simple dictionary as follows in Python.

>>> grades = { 'Alice': 97, 'Bob': 85, 'Carrie': 83 }


You can think of this as creating a table, in this case mapping names to test scores.

Alice	→	97
Bob	→	85
Carrie	→	83

In this case, we're using strings as our keys and integers as our values, but Python is very flexible in what can be used for keys and values: We could, for example, have a dictionary mapping points to strings.

You can retrieve a value associated with a key by referencing the dictionary followed by brackets containing the key value. You can also use len to find how many key-value pairs are in the dictionary, the in operator to test whether a key is in the dictionary, and a for loop to iterate through each key in the dictionary.

>>> print grades['Alice']
97
>>> print len(grades)
3
>>> print 'Quincy' in grades
False
>>> for name in grades:
>>>     print name, 'got', grades[name]
Alice got 97
Carrie got 83
Bob got 85


Python doesn't maintain any ordering to the keys, so the for loop could easily go through the keys in a peculiar order.

You can modify the value associated with a key or add a new key-value mapping by assigning something into the dictionary.

>>> grades['Bob'] = 89
>>> grades['Dylan'] = 71
>>> print len(grades)
4


Each key in a dictionary maps to only one value. So when we assign to grades['Bob'], Python notices that 'Bob' already had a value associated with it, and it replaces that association with a new one. When we assign to grades['Dylan'], a new association is added into the dictionary since 'Dylan' wasn't previously associated with a value.

Here is a much more complex example involving a dictionary. It finds which words occur most frequently in a file. The dictionary counts is used to maintain the mapping from each word to the number of times it has been found.

counts = {}                     # We start with an empty mapping
essay = open('essay.txt')
for line in essay:
    for word in line.split():
        if word not in counts:  # This is the first time we have seen word,
            counts[word] = 1    #   so we save that we have seen it once
        else:
            counts[word] += 1   # Otherwise increment the count for word

max_count = 0                   # max_count will be the largest count found
max_words = []                  # max_words is list of words appearing max_count times
for word in counts:
    count = counts[word]
    if count > max_count:       # This word beats out all previous ones
        max_words = [word]
        max_count = count
    elif count == max_count:    # This word ties the previous best
        max_words.append(word)
print max_count, 'times:', ' '.join(max_words)