Lecture 10

Dictionaries

MCS 260 Fall 2020
David Dumas

Reminders

• Work on:
  • Project 1 (due Friday, 6pm central)
  • Worksheet 4
  • Quiz 4
• Docstrings now required for all functions & files
• Come to Tue/Thu discussion ready to share screen and code (if at all possible)

Dictionaries

Lists and tuples are sequences: they store an ordered collection of values that can be retrieved by index (a nonnegative integer).

A dictionary or dict in Python is an unordered collection of key:value pairs. Values can be retrieved using the associated key, similar to indexing a list.

The values in a dictionary can be of any type, but there are some restrictions on the keys.

Dictionaries are mutable.

Example of syntax for working with dictionaries:

        >>> # define a new dict
        >>> tuition = { "UIC": 10584, 
        ...             "Stanford": 50703, 
        ...             "Harvard": 46340 }
        >>> # Access an item
        >>> tuition["UIC"]
        10584
        >>> # Add or change an item
        >>> tuition["PSU"] = 18454
        >>> tuition
        {'UIC': 10584, 
         'Stanford': 50703, 
         'Harvard': 46340, 
         'PSU': 18454}
        >>> # Remove an item
        >>> del tuition["Harvard"]
        >>> tuition
        {'UIC': 10584, 'Stanford': 50703, 'PSU': 18454}
    

Mixed types are ok for keys or values.

        d = { 1: "fish", "two": "fish", "x": [7,6,5] }
    

Methods:

>>> d.keys()   # All keys (like range(len(L)))
dict_keys([1, 'two', 'x'])
>>> d.items()  # All key-value pairs (like enumerate(L))
dict_items([(1, 'fish'), ('two', 'fish'), ('x', [7, 6, 5])])
>>> d.values() # All values
dict_values(['fish', 'fish', [7, 6, 5]])
    

dict_keys, dict_items, dict_values types behave a lot like list, and can be converted to a list with $\texttt{list()}$.

Membership testing

Membership in a dictionary means being a key!

        >>> d = { 1: "fish", "two": "fish", "x": [7,6,5] }
        >>> "fish" in d
        False
        >>> 1 in d
        True
    

Forgetting this is a very common source of programming errors.

Other languages

Python dicts are examples of associative arrays, also known as maps.

In other languages with a built-in associative array type, the type is often called map or Map (e.g. in C++, Java, Go)

The rules (allowable keys, type heterogeneity, etc.) vary by language.

Iteration over dicts

dicts are iterable, but iterate over the keys.

        for k in d:  # loop over keys
            print(k,"is one of the keys")

        for k in d:  # loop over keys (index to get value)
            print("Key",k,"has value",d[k])
            
        for k,v in d.items(): # loop over keys,value pairs
            print("Key",k,"has value",v)
    

It is common for the values in a dict to be dicts themselves. This is the usual way to make a collection of labeled data indexed by a key.

    schooldata = {
        "UIC": {
            "fullname": "University of Illinois at Chicago",
            "tuition": 10584,
            "undergrad_students": 21641, 
            },
        "Stanford": {
            "fullname": "University of Illinois at Chicago",
            "tuition": 50703,
            "undergrad_students": 7083
            },
        "Harvard": {
            "fullname": "Harvard University",
            "tuition": 46340,
            "undergrad_students": 6755
            }
        }
    
    for shortname,data in schooldata.items():
        print(shortname,"has annual tuition",data["tuition"])
    

Dictionaries as rules

    pr_replacements = {
        "accident": "unplanned event",
        "escape": "departure",
        "laser-sharks": "fish"
    }
    original = "an accident involving the escape of laser-sharks"
    words = original.split()  # [ "an", "accident", ... ]
    for w in words:
        if w in pr_replacements:
            w = pr_replacements[w]
        print(w,end=" ")
    print()
    

        an unplanned event involving the departure of fish
    

Dictionary comprehensions

Analogous to list comprehensions, but using $\texttt{ \{ key:value for name in iterable ... \} }$.

    >>> words = [ "Chicago", "cat", "cinemas" ]
    >>> word_data = { w: { "length": len(w), 
    ...                    "all lower": w==w.lower() } 
    ...               for w in words }
    >>> word_data
    {'Chicago': {'length': 7, 'all lower': False},
     'cat': {'length': 3, 'all lower': True},
     'cinemas': {'length': 7, 'all lower': True}
    }