Study material generated by AI from this openly licensed book (CC BY-NC 3.0). Not affiliated with or endorsed by the publisher.
Think Python, 2nd Edition, by Allen B. Downey, Green Tea Press, licensed under CC BY-NC 3.0. Read the original free
Chapter 17 · pp. 183–192
Object-oriented programming (OOP) is a style of writing programs in which data and the operations on that data are grouped together. In this chapter, functions that work on a programmer-defined type are turned into methods that belong to that type's class. The change is mostly in syntax, but it makes the program's structure easier to see. (p. 183)
Python is an object-oriented language, so it has features that support OOP. A program in this style has three defining traits (p. 183):
Here is an example. A Time class models the time of day as people record it. Functions that work with times, such as printing a time, model what people do with times. In the same way, Point and Rectangle classes model the mathematical ideas of a point and a rectangle. (p. 183)
Notice that earlier programs used classes like these but were not truly object-oriented. They kept the types and the functions that use them as separate pieces and did not state how the two are related. Turning those functions into methods makes the relationship explicit. (p. 183)
You do not strictly need these features. Most of them are another way to write things you could already do with ordinary functions. They are still worth using because the new syntax is often shorter and shows the program's structure more honestly. (p. 183)
For example, suppose a class definition is followed by a list of separate functions. Nothing in the code says that they belong together. You have to read each function and notice that every one takes a Time object as an argument. That observation is the reason methods exist. (pp. 183–184)
A method is a function that is associated with a particular class. You have already met methods on strings, lists, dictionaries and tuples. Now you can define your own for types you create. (p. 184)
Methods mean the same thing as functions, but they differ in two ways of writing them (p. 184):
Converting a function to a method is a mechanical process: you follow a fixed series of steps. If you can convert in both directions, you can choose whichever form suits the job. (p. 184)
As an ordinary function, printing a time looks like this (p. 184):
class Time:
"""Represents the time of day."""
def print_time(time):
print('%.2d:%.2d:%.2d' % (time.hour, time.minute, time.second))
The format %.2d prints an integer using at least two digits, adding a leading zero if needed. To call the function, you pass a Time object as the argument:
>>> start = Time()
>>> start.hour = 9
>>> start.minute = 45
>>> start.second = 0
>>> print_time(start)
09:45:00
To make print_time a method, move its definition inside the class. The only change is the indentation, which now places the function under class Time: (p. 184):
class Time:
def print_time(time):
print('%.2d:%.2d:%.2d' % (time.hour, time.minute, time.second))
Once print_time is inside the class, you can call it in two ways. (p. 184)
Function syntax (less common). Use dot notation with the class name, and pass the object as an ordinary argument:
>>> Time.print_time(start)
09:45:00
Here Time is the class name, print_time is the method name, and start is the argument. (p. 184)
Method syntax (shorter and more common). Put the object before the dot:
>>> start.print_time()
09:45:00
Both calls give the same result. In the second form, the object written before the dot is handed to the method as its first argument. This is why the call needs no parentheses content, even though the method definition has a parameter. (p. 184)
Time object named start and a method print_time defined in the class, write both ways to call it. Which one is more common?print_time into a method of Time?