8. Inheritance

In this chapter we look at a larger example using object oriented programming and learn about the very useful OOP feature of inheritance.

8.1. Composition

By now, you have seen several examples of composition. One of the first examples was using a method invocation as part of an expression. Another example is the nested structure of statements; you can put an if statement within a while loop, within another if statement, and so on.

Having seen this pattern, and having learned about lists and objects, you should not be surprised to learn that you can create lists of objects. You can also create objects that contain lists (as attributes); you can create lists that contain lists; you can create objects that contain objects; and so on.

In this chapter we will look at some examples of these combinations, using Card objects as an example.

8.2. Card objects

If you are not familiar with common playing cards, now would be a good time to get a deck, or else this chapter might not make much sense. There are fifty-two cards in a deck, each of which belongs to one of four suits and one of thirteen ranks. The suits are Spades, Hearts, Diamonds, and Clubs (in descending order in bridge). The ranks are Ace, 2, 3, 4, 5, 6, 7, 8, 9, 10, Jack, Queen, and King. Depending on the game that you are playing, the rank of Ace may be higher than King or lower than 2.

If we want to define a new object to represent a playing card, it is obvious what the attributes should be: rank and suit. It is not as obvious what type the attributes should be. One possibility is to use strings containing words like "Spade" for suits and "Queen" for ranks. One problem with this implementation is that it would not be easy to compare cards to see which had a higher rank or suit.

An alternative is to use integers to encode the ranks and suits. By encode, we do not mean what some people think, which is to encrypt or translate into a secret code. What a computer scientist means by encode is to define a mapping between a sequence of numbers and the items I want to represent. For example:

Spades   -->  3
Hearts   -->  2
Diamonds -->  1
Clubs    -->  0

An obvious feature of this mapping is that the suits map to integers in order, so we can compare suits by comparing integers. The mapping for ranks is fairly obvious; each of the numerical ranks maps to the corresponding integer, and for face cards:

Jack   -->  11
Queen  -->  12
King   -->  13

The reason we are using mathematical notation for these mappings is that they are not part of the Python program. They are part of the program design, but they never appear explicitly in the code. The class definition for the Card type looks like this:

class Card:
    def __init__(self, suit=0, rank=0):
        self.suit = suit
        self.rank = rank

As usual, we provide an initialization method that takes an optional parameter for each attribute.

To create an object that represents the 3 of Clubs, use this command:

three_of_clubs = Card(0, 3)

The first argument, 0, represents the suit Clubs.

8.3. Class attributes and the __str__ method

In order to print Card objects in a way that people can easily read, we want to map the integer codes onto words. A natural way to do that is with lists of strings. We assign these lists to class attributes at the top of the class definition:

class Card:
    SUITS = ('Clubs', 'Diamonds', 'Hearts', 'Spades')
    RANKS = ('narf', 'Ace', '2', '3', '4', '5', '6', '7',
             '8', '9', '10', 'Jack', 'Queen', 'King']

    def __init__(self, suit=0, rank=0):
        self.suit = suit
        self.rank = rank

    def __str__(self):
        """
          >>> print(Card(2, 11))
          Queen of Hearts
        """
        return '{0} of {1}'.format(Card.RANKS[self.rank],
                                   Card.SUITS[self.suit])


if __name__ == '__main__':
    import doctest
    doctest.testmod()

Class attributes like Card.SUITS and Card.RANKS are defined outside of any method, and can be accessed from any of the methods in the class.

Inside __str__, we can use SUITS and RANKS to map the numerical values of suit and rank to strings. For example, the expression Card.SUITS[self.suit] means use the attribute suit from the object self as an index into the class attribute named SUITS, and select the appropriate string.

The reason for the "narf" in the first element in ranks is to act as a place keeper for the zero-eth element of the list, which will never be used. The only valid ranks are 1 to 13. This wasted item is not entirely necessary. We could have started at 0, as usual, but it is less confusing to encode 2 as 2, 3 as 3, and so on.

We have a doctest in the __str__ method to confirm that Card(2, 11) will display as “Queen of Hearts”.

8.4. Comparing cards

For primitive types, there are conditional operators ( <, >, ==, etc.) that compare values and determine when one is greater than, less than, or equal to another. For user-defined types, we can override the behavior of the built-in operators by providing a method named __cmp__. By convention, __cmp__ takes two parameters, self and other, and returns 1 if the first object is greater, -1 if the second object is greater, and 0 if they are equal to each other.

Some types are completely ordered, which means that you can compare any two elements and tell which is bigger. For example, the integers and the floating-point numbers are completely ordered. Some sets are unordered, which means that there is no meaningful way to say that one element is bigger than another. For example, the fruits are unordered, which is why you cannot compare apples and oranges.

The set of playing cards is partially ordered, which means that sometimes you can compare cards and sometimes not. For example, you know that the 3 of Clubs is higher than the 2 of Clubs, and the 3 of Diamonds is higher than the 3 of Clubs. But which is better, the 3 of Clubs or the 2 of Diamonds? One has a higher rank, but the other has a higher suit.

In order to make cards comparable, you have to decide which is more important, rank or suit. To be honest, the choice is arbitrary. For the sake of choosing, we will say that suit is more important, because a new deck of cards comes sorted with all the Clubs together, followed by all the Diamonds, and so on.

With that decided, we can write __cmp__:

def __cmp__(self, other):
    # check the suits
    if self.suit > other.suit: return 1
    if self.suit < other.suit: return -1
    # suits are the same... check ranks
    if self.rank > other.rank: return 1
    if self.rank < other.rank: return -1
    # ranks are the same... it's a tie
    return 0

In this ordering, Aces appear lower than Deuces (2s).

8.5. Decks

Now that we have objects to represent Cards, the next logical step is to define a class to represent a Deck. Of course, a deck is made up of cards, so each Deck object will contain a list of cards as an attribute.

The following is a class definition for the Deck class. The initialization method creates the attribute cards and generates the standard set of fifty-two cards:

class Deck:
    def __init__(self):
        self.cards = []
        for suit in range(4):
            for rank in range(1, 14):
                self.cards.append(Card(suit, rank))

The easiest way to populate the deck is with a nested loop. The outer loop enumerates the suits from 0 to 3. The inner loop enumerates the ranks from 1 to 13. Since the outer loop iterates four times, and the inner loop iterates thirteen times, the total number of times the body is executed is fifty-two (thirteen times four). Each iteration creates a new instance of Card with the current suit and rank, and appends that card to the cards list.

The append method works on lists but not, of course, tuples.

8.6. Printing the deck

As usual, when we define a new type of object we want a method that prints the contents of an object. To print a Deck, we traverse the list and print each Card:

class Deck:
    ...
    def print_deck(self):
        for card in self.cards:
            print(card)

Here, and from now on, the ellipsis ( ...) indicates that we have omitted the other methods in the class.

As an alternative to print_deck, we could write a __str__ method for the Deck class. The advantage of __str__ is that it is more flexible. Rather than just printing the contents of the object, it generates a string representation that other parts of the program can manipulate before printing, or store for later use.

Here is a version of __str__ that returns a string representation of a Deck. To add a bit of pizzazz, it arranges the cards in a cascade where each card is indented one space more than the previous card:

class Deck:
    ...
    def __str__(self):
        s = ""
        for i in range(len(self.cards)):
            s += " " * i + str(self.cards[i]) + "\n"
        return s

This example demonstrates several features. First, instead of traversing self.cards and assigning each card to a variable, we are using i as a loop variable and an index into the list of cards.

Second, we are using the string multiplication operator to indent each card by one more space than the last. The expression " " * i yields a number of spaces equal to the current value of i.

Third, instead of using the print function to print the cards, we use the str function. Passing an object as an argument to str is equivalent to invoking the __str__ method on the object.

Finally, we are using the variable s as an accumulator. Initially, s is the empty string. Each time through the loop, a new string is generated and concatenated with the old value of s to get the new value. When the loop ends, s contains the complete string representation of the Deck, which looks like this:

>>> deck = Deck()
>>> print(deck)
Ace of Clubs
 2 of Clubs
  3 of Clubs
   4 of Clubs
     5 of Clubs
       6 of Clubs
        7 of Clubs
         8 of Clubs
          9 of Clubs
           10 of Clubs
            Jack of Clubs
             Queen of Clubs
              King of Clubs
               Ace of Diamonds

And so on. Even though the result appears on 52 lines, it is one long string that contains newlines.

8.7. Shuffling the deck

If a deck is perfectly shuffled, then any card is equally likely to appear anywhere in the deck, and any location in the deck is equally likely to contain any card.

To shuffle the deck, we will use the randrange function from the random module. With two integer arguments, a and b, randrange chooses a random integer in the range a <= x < b. Since the upper bound is strictly less than b, we can use the length of a list as the second parameter, and we are guaranteed to get a legal index. For example, this expression chooses the index of a random card in a deck:

random.randrange(0, len(self.cards))

An easy way to shuffle the deck is by traversing the cards and swapping each card with a randomly chosen one. It is possible that the card will be swapped with itself, but that is fine. In fact, if we precluded that possibility, the order of the cards would be less than entirely random:

class Deck:
    ...
    def shuffle(self):
        import random
        num_cards = len(self.cards)
        for i in range(num_cards):
            j = random.randrange(i, num_cards)
            self.cards[i], self.cards[j] = self.cards[j], self.cards[i]

Rather than assume that there are fifty-two cards in the deck, we get the actual length of the list and store it in num_cards.

For each card in the deck, we choose a random card from among the cards that haven’t been shuffled yet. Then we swap the current card ( i) with the selected card ( j). To swap the cards we use a tuple assignment:

self.cards[i], self.cards[j] = self.cards[j], self.cards[i]

8.8. Removing and dealing cards

Another method that would be useful for the Deck class is remove, which takes a card as a parameter, removes it, and returns True if the card was in the deck and False otherwise:

class Deck:
    ...
    def remove(self, card):
        if card in self.cards:
            self.cards.remove(card)
            return True
        else:
            return False

The in operator returns True if the first operand is in the second, which must be a list or a tuple. If the first operand is an object, Python uses the object’s __cmp__ method to determine equality with items in the list. Since the __cmp__ in the Card class checks for deep equality, the remove method checks for deep equality.

To deal cards, we want to remove and return the top card. The list method pop provides a convenient way to do that:

class Deck:
    ...
    def pop(self):
        return self.cards.pop()

Actually, pop removes the last card in the list, so we are in effect dealing from the bottom of the deck.

One more operation that we are likely to want is the boolean function is_empty, which returns true if the deck contains no cards:

class Deck:
    ...
    def is_empty(self):
        return (len(self.cards) == 0)

8.9. Inheritance

The language feature most often associated with object-oriented programming is inheritance. Inheritance is the ability to define a new class that is a modified version of an existing class.

The primary advantage of this feature is that you can add new methods to a class without modifying the existing class. It is called inheritance because the new class inherits all of the methods of the existing class. Extending this metaphor, the existing class is sometimes called the parent class. The new class may be called the child class or sometimes subclass.

Inheritance is a powerful feature. Some programs that would be complicated without inheritance can be written concisely and simply with it. Also, inheritance can facilitate code reuse, since you can customize the behavior of parent classes without having to modify them. In some cases, the inheritance structure reflects the natural structure of the problem, which makes the program easier to understand.

On the other hand, inheritance can make programs difficult to read. When a method is invoked, it is sometimes not clear where to find its definition. The relevant code may be scattered among several modules. Also, many of the things that can be done using inheritance can be done as elegantly (or more so) without it. If the natural structure of the problem does not lend itself to inheritance, this style of programming can do more harm than good.

In this chapter we will demonstrate the use of inheritance as part of a program that plays the card game Old Maid. One of our goals is to write code that could be reused to implement other card games.

8.10. A hand of cards

For almost any card game, we need to represent a hand of cards. A hand is similar to a deck, of course. Both are made up of a set of cards, and both require operations like adding and removing cards. Also, we might like the ability to shuffle both decks and hands.

A hand is also different from a deck. Depending on the game being played, we might want to perform some operations on hands that don’t make sense for a deck. For example, in poker we might classify a hand (straight, flush, etc.) or compare it with another hand. In bridge, we might want to compute a score for a hand in order to make a bid.

This situation suggests the use of inheritance. If Hand is a subclass of Deck, it will have all the methods of Deck, and new methods can be added.

In the class definition, the name of the parent class appears in parentheses:

class Hand(Deck):
    pass

This statement indicates that the new Hand class inherits from the existing Deck class.

The Hand constructor initializes the attributes for the hand, which are name and cards. The string name identifies this hand, probably by the name of the player that holds it. The name is an optional parameter with the empty string as a default value. cards is the list of cards in the hand, initialized to the empty list:

class Hand(Deck):
    def __init__(self, name=""):
       self.cards = []
       self.name = name

For just about any card game, it is necessary to add and remove cards from the deck. Removing cards is already taken care of, since Hand inherits remove from Deck. But we have to write add:

class Hand(Deck):
    ...
    def add(self,card):
        self.cards.append(card)

Again, the ellipsis indicates that we have omitted other methods. The list append method adds the new card to the end of the list of cards.

8.11. Dealing cards

Now that we have a Hand class, we want to deal cards from the Deck into hands. It is not immediately obvious whether this method should go in the Hand class or in the Deck class, but since it operates on a single deck and (possibly) several hands, it is more natural to put it in Deck.

deal should be fairly general, since different games will have different requirements. We may want to deal out the entire deck at once or add one card to each hand.

deal takes two parameters, a list (or tuple) of hands and the total number of cards to deal. If there are not enough cards in the deck, the method deals out all of the cards and stops:

class Deck :
    ...
    def deal(self, hands, num_cards=999):
        num_hands = len(hands)
        for i in range(num_cards):
            if self.is_empty(): break   # break if out of cards
            card = self.pop()           # take the top card
            hand = hands[i % num_hands] # whose turn is next?
            hand.add(card)              # add the card to the hand

The second parameter, num_cards, is optional; the default is a large number, which effectively means that all of the cards in the deck will get dealt.

The loop variable i goes from 0 to nCards-1. Each time through the loop, a card is removed from the deck using the list method pop, which removes and returns the last item in the list.

The modulus operator ( %) allows us to deal cards in a round robin (one card at a time to each hand). When i is equal to the number of hands in the list, the expression i % nHands wraps around to the beginning of the list (index 0).

8.12. Printing a Hand

To print the contents of a hand, we can take advantage of the printDeck and __str__ methods inherited from Deck. For example:

>>> deck = Deck()
>>> deck.shuffle()
>>> hand = Hand("frank")
>>> deck.deal([hand], 5)
>>> print(hand)
Hand frank contains
2 of Spades
 3 of Spades
  4 of Spades
   Ace of Hearts
    9 of Clubs

It’s not a great hand, but it has the makings of a straight flush.

Although it is convenient to inherit the existing methods, there is additional information in a Hand object we might want to include when we print one. To do that, we can provide a __str__ method in the Hand class that overrides the one in the Deck class:

class Hand(Deck)
    ...
    def __str__(self):
        s = "Hand " + self.name
        if self.is_empty():
            s = s + " is empty\n"
        else:
            s = s + " contains\n"
        return s + Deck.__str__(self)

Initially, s is a string that identifies the hand. If the hand is empty, the program appends the words is empty and returns s.

Otherwise, the program appends the word contains and the string representation of the Deck, computed by invoking the __str__ method in the Deck class on self.

It may seem odd to send self, which refers to the current Hand, to a Deck method, until you remember that a Hand is a kind of Deck. Hand objects can do everything Deck objects can, so it is legal to send a Hand to a Deck method.

In general, it is always legal to use an instance of a subclass in place of an instance of a parent class.

8.13. The CardGame class

The CardGame class takes care of some basic chores common to all games, such as creating the deck and shuffling it:

class CardGame:
    def __init__(self):
        self.deck = Deck()
        self.deck.shuffle()

This is the first case we have seen where the initialization method performs a significant computation, beyond initializing attributes.

To implement specific games, we can inherit from CardGame and add features for the new game. As an example, we’ll write a simulation of Old Maid.

The object of Old Maid is to get rid of cards in your hand. You do this by matching cards by rank and color. For example, the 4 of Clubs matches the 4 of Spades since both suits are black. The Jack of Hearts matches the Jack of Diamonds since both are red.

To begin the game, the Queen of Clubs is removed from the deck so that the Queen of Spades has no match. The fifty-one remaining cards are dealt to the players in a round robin. After the deal, all players match and discard as many cards as possible.

When no more matches can be made, play begins. In turn, each player picks a card (without looking) from the closest neighbor to the left who still has cards. If the chosen card matches a card in the player’s hand, the pair is removed. Otherwise, the card is added to the player’s hand. Eventually all possible matches are made, leaving only the Queen of Spades in the loser’s hand.

In our computer simulation of the game, the computer plays all hands. Unfortunately, some nuances of the real game are lost. In a real game, the player with the Old Maid goes to some effort to get their neighbor to pick that card, by displaying it a little more prominently, or perhaps failing to display it more prominently, or even failing to fail to display that card more prominently. The computer simply picks a neighbor’s card at random.

8.14. OldMaidHand class

A hand for playing Old Maid requires some abilities beyond the general abilities of a Hand. We will define a new class, OldMaidHand, that inherits from Hand and provides an additional method called remove_matches:

class OldMaidHand(Hand):
    def remove_matches(self):
        count = 0
        original_cards = self.cards[:]
        for card in original_cards:
            match = Card(3 - card.suit, card.rank)
            if match in self.cards:
                self.cards.remove(card)
                self.cards.remove(match)
                print("Hand {0}: {1} matches {2}".format(self.name, card, match)
                count = count + 1
        return count

We start by making a copy of the list of cards, so that we can traverse the copy while removing cards from the original. Since self.cards is modified in the loop, we don’t want to use it to control the traversal. Python can get quite confused if it is traversing a list that is changing!

For each card in the hand, we figure out what the matching card is and go looking for it. The match card has the same rank and the other suit of the same color. The expression 3 - card.suit turns a Club (suit 0) into a Spade (suit 3) and a Diamond (suit 1) into a Heart (suit 2). You should satisfy yourself that the opposite operations also work. If the match card is also in the hand, both cards are removed.

The following example demonstrates how to use remove_matches:

>>> game = CardGame()
>>> hand = OldMaidHand("frank")
>>> game.deck.deal([hand], 13)
>>> print(hand)
Hand frank contains
Ace of Spades
 2 of Diamonds
  7 of Spades
   8 of Clubs
    6 of Hearts
     8 of Spades
      7 of Clubs
       Queen of Clubs
        7 of Diamonds
         5 of Clubs
          Jack of Diamonds
           10 of Diamonds
            10 of Hearts
>>> hand.remove_matches()
Hand frank: 7 of Spades matches 7 of Clubs
Hand frank: 8 of Spades matches 8 of Clubs
Hand frank: 10 of Diamonds matches 10 of Hearts
>>> print(hand)
Hand frank contains
Ace of Spades
 2 of Diamonds
  6 of Hearts
   Queen of Clubs
    7 of Diamonds
     5 of Clubs
      Jack of Diamonds

Notice that there is no __init__ method for the OldMaidHand class. We inherit it from Hand.

8.15. OldMaidGame class

Now we can turn our attention to the game itself. OldMaidGame is a subclass of CardGame with a new method called play that takes a list of players as a parameter.

Since __init__ is inherited from CardGame, a new OldMaidGame object contains a new shuffled deck:

class OldMaidGame(CardGame):
    def play(self, names):
        # remove Queen of Clubs
        self.deck.remove(Card(0,12))

        # make a hand for each player
        self.hands = []
        for name in names:
            self.hands.append(OldMaidHand(name))

        # deal the cards
        self.deck.deal(self.hands)
        print("---------- Cards have been dealt")
        self.printHands()

        # remove initial matches
        matches = self.removeAllMatches()
        print("---------- Matches discarded, play begins")
        self.printHands()

        # play until all 50 cards are matched
        turn = 0
        numHands = len(self.hands)
        while matches < 25:
            matches = matches + self.playOneTurn(turn)
            turn = (turn + 1) % numHands

        print("---------- Game is Over")
        self.printHands()

The writing of printHands() is left as an exercise.

Some of the steps of the game have been separated into methods. remove_all_matches traverses the list of hands and invokes remove_matches on each:

class OldMaidGame(CardGame):
    ...
    def remove_all_matches(self):
        count = 0
        for hand in self.hands:
            count = count + hand.remove_matches()
        return count

count is an accumulator that adds up the number of matches in each hand and returns the total.

When the total number of matches reaches twenty-five, fifty cards have been removed from the hands, which means that only one card is left and the game is over.

The variable turn keeps track of which player’s turn it is. It starts at 0 and increases by one each time; when it reaches numHands, the modulus operator wraps it back around to 0.

The method playOneTurn takes a parameter that indicates whose turn it is. The return value is the number of matches made during this turn:

class OldMaidGame(CardGame):
    ...
    def play_one_turn(self, i):
        if self.hands[i].is_empty():
            return 0
        neighbor = self.find_neighbor(i)
        pickedCard = self.hands[neighbor].popCard()
        self.hands[i].add(pickedCard)
        print("Hand", self.hands[i].name, "picked", pickedCard)
        count = self.hands[i].remove_matches()
        self.hands[i].shuffle()
        return count

If a player’s hand is empty, that player is out of the game, so he or she does nothing and returns 0.

Otherwise, a turn consists of finding the first player on the left that has cards, taking one card from the neighbor, and checking for matches. Before returning, the cards in the hand are shuffled so that the next player’s choice is random.

The method find_neighbor starts with the player to the immediate left and continues around the circle until it finds a player that still has cards:

class OldMaidGame(CardGame):
    ...
    def find_neighbor(self, i):
        numHands = len(self.hands)
        for next in range(1,numHands):
            neighbor = (i + next) % numHands
            if not self.hands[neighbor].is_empty():
                return neighbor

If find_neighbor ever went all the way around the circle without finding cards, it would return None and cause an error elsewhere in the program. Fortunately, we can prove that that will never happen (as long as the end of the game is detected correctly).

We have omitted the print_hands method. You can write that one yourself.

The following output is from a truncated form of the game where only the top fifteen cards (tens and higher) were dealt to three players. With this small deck, play stops after seven matches instead of twenty-five.

>>> import cards
>>> game = cards.OldMaidGame()
>>> game.play(["Allen","Jeff","Chris"])
---------- Cards have been dealt
Hand Allen contains
King of Hearts
 Jack of Clubs
  Queen of Spades
   King of Spades
    10 of Diamonds

Hand Jeff contains
Queen of Hearts
 Jack of Spades
  Jack of Hearts
   King of Diamonds
    Queen of Diamonds

Hand Chris contains
Jack of Diamonds
 King of Clubs
  10 of Spades
   10 of Hearts
    10 of Clubs

Hand Jeff: Queen of Hearts matches Queen of Diamonds
Hand Chris: 10 of Spades matches 10 of Clubs
---------- Matches discarded, play begins
Hand Allen contains
King of Hearts
 Jack of Clubs
  Queen of Spades
   King of Spades
    10 of Diamonds

Hand Jeff contains
Jack of Spades
 Jack of Hearts
  King of Diamonds

Hand Chris contains
Jack of Diamonds
 King of Clubs
  10 of Hearts

Hand Allen picked King of Diamonds
Hand Allen: King of Hearts matches King of Diamonds
Hand Jeff picked 10 of Hearts
Hand Chris picked Jack of Clubs
Hand Allen picked Jack of Hearts
Hand Jeff picked Jack of Diamonds
Hand Chris picked Queen of Spades
Hand Allen picked Jack of Diamonds
Hand Allen: Jack of Hearts matches Jack of Diamonds
Hand Jeff picked King of Clubs
Hand Chris picked King of Spades
Hand Allen picked 10 of Hearts
Hand Allen: 10 of Diamonds matches 10 of Hearts
Hand Jeff picked Queen of Spades
Hand Chris picked Jack of Spades
Hand Chris: Jack of Clubs matches Jack of Spades
Hand Jeff picked King of Spades
Hand Jeff: King of Clubs matches King of Spades
---------- Game is Over
Hand Allen is empty

Hand Jeff contains
Queen of Spades

Hand Chris is empty

So Jeff loses.

8.16. Glossary

encode
To represent one set of values using another set of values by constructing a mapping between them.
class attribute
A variable that is defined inside a class definition but outside any method. Class attributes are accessible from any method in the class and are shared by all instances of the class.
accumulator
A variable used in a loop to accumulate a series of values, such as by concatenating them onto a string or adding them to a running sum.
inheritance
The ability to define a new class that is a modified version of a previously defined class.
parent class
The class from which a child class inherits.
child class
A new class created by inheriting from an existing class; also called a subclass.