Program Arcade Games
With Python And Pygame

Chapter 18: Exceptions

Video: Exception Handling

When something goes wrong with your program, do you want to keep the user from seeing a red Python error message? Do you want to keep your program from hanging? If so, then you need exceptions.

Exceptions are used to handle abnormal conditions that can occur during the execution of code. Exceptions are often used with file and network operations. This allows code to gracefully handle running out of disk space, network errors, or permission errors.

18.1 Vocabulary

There are several terms and phrases used while working with exceptions. Here are the most common:

Most programming languages use the terms “throw” and “catch.” Unfortunately Python doesn't. Python uses “raise” and “exception.” We introduce the throw/catch vocabulary here because they are the most prevalent terms in the industry.

18.2 Exception Handling

The code for handling exceptions is simple. See the example below:

# Divide by zero
    x = 5 / 0
    print("Error dividing by zero")

On line two is the try statement. Every indented line below it is part of the “try block.” There may be no unindented code below the try block that doesn't start with an except statement. The try statement defines a section of code that the code will attempt to execute.

If there is any exception that occurs during the processing of the code the execution will immediately jump to the “catch block.” That block of code is indented under the except statement on line 4. This code is responsible for handling the error.

A program may use exceptions to catch errors that occur during a conversion from text to a number. For example:

# Invalid number conversion
    x = int("fred")
    print("Error converting fred to a number")

An exception will be thrown on line 3 because “fred” can not be converted to an integer. The code on line 5 will print out an error message.

Below is an expanded version on this example. It error-checks a user's input to make sure an integer is entered. If the user doesn't enter an integer, the program will keep asking for one. The code uses exception handling to capture a possible conversion error that can occur on line 5. If the user enters something other than an integer, an exception is thrown when the conversion to a number occurs on line 5. The code on line 6 that sets number_entered to True will not be run if there is an exception on line 5.

number_entered = False
while not number_entered:
    number_string = input("Enter an integer: ")
        n = int(number_string)
        number_entered = True
        print("Error, invalid integer")

Files are particularly prone to errors during operations with them. A disk could fill up, a user could delete a file while it is being written, it could be moved, or a USB drive could be pulled out mid-operation. These types of errors may also be easily captured by using exception handling.

# Error opening file
    my_file = open("myfile.txt")
    print("Error opening file")

Multiple types of errors may be captured and processed differently. It can be useful to provide a more exact error message to the user than a simple “an error has occurred.”

In the code below, different types of errors can occur from lines 5-8. By placing IOError after except on line 9, only errors regarding Input and Output (IO) will be handled by that code. Likewise line 11 only handles errors around converting values, and line 13 covers division by zero errors. The last exception handling occurs on line 15. Since line 15 does not include a particular type of error, it will handle any error not covered by the except blocks above. The “catch-all” except must always be last.

Line 1 imports the sys library which is used on line 16 to print the type of error that has occurred.

import sys

# Multiple errors
    my_file = open("myfile.txt")
    my_line = my_file.readline()
    my_int = int(s.strip())
    my_calculated_value = 101 / my_int
except IOError:
    print("I/O error")
except ValueError:
    print("Could not convert data to an integer.")
except ZeroDivisionError:
    print("Division by zero error")
    print("Unexpected error:", sys.exc_info()[0])

A list of built-in exceptions is available from this web address:

18.3 Example: Saving High Score

This shows how to save a high score between games. The score is stored in a file called high_score.txt.

Show how to use exceptions to save a high score for a game.

Sample Python/Pygame Programs
Simpson College Computer Science

def get_high_score():
    # Default high score
    high_score = 0

    # Try to read the high score from a file
        high_score_file = open("high_score.txt", "r")
        high_score = int(
        print("The high score is", high_score)
    except IOError:
        # Error reading file, no high score
        print("There is no high score yet.")
    except ValueError:
        # There's a file there, but we don't understand the number.
        print("I'm confused. Starting with no high score.")

    return high_score

def save_high_score(new_high_score):
        # Write the file to disk
        high_score_file = open("high_score.txt", "w")
    except IOError:
        # Hm, can't write it.
        print("Unable to save the high score.")

def main():
    """ Main program is here. """
    # Get the high score
    high_score = get_high_score()

    # Get the score from the current game
    current_score = 0
        # Ask the user for his/her score
        current_score = int(input("What is your score? "))
    except ValueError:
        # Error, can't turn what they typed into a number
        print("I don't understand what you typed.")

    # See if we have a new high score
    if current_score > high_score:
        # We do! Save to disk
        print("Yea! New high score!")
        print("Better luck next time.")

# Call the main function, start up the game
if __name__ == "__main__":

18.4 Exception Objects

More information about an error can be pulled from the exception object. This object can be retrieved while catching an error using the as keyword. For example:

    x = 5 / 0
except ZeroDivisionError as e:

The e variable points to more information about the exception that can be printed out. More can be done with exceptions objects, but unfortunately that is beyond the scope of this chapter. Check the Python documentation on-line for more information about the exception object.

18.5 Exception Generating

Exceptions may be generated with the raise command. For example:

# Generating exceptions
def get_input():
    user_input = input("Enter something: ")
    if len(user_input) == 0:
        raise IOError("User entered nothing")


Try taking the code above, and add exception handling for the IOError raised.

It is also possible to create custom exceptions, but that is also beyond the scope of this book. Curious readers may learn more by going to:

18.6 Proper Exception Use

Exceptions should not be used when if statements can just as easily handle the condition. Normal code should not raise exceptions when running the “happy path” scenario. Well-constructed try/catch code is easy to follow but code involving many exceptions and jumps in code to different handlers can be a nightmare to debug. (Once I was assigned the task of debugging code that read an XML document. It generated dozens of exceptions for each line of the file it read. It was incredibly slow and error-prone. That code should have never generated a single exception in the normal course of reading a file.)

18.7 Review

18.7.1 Multiple Choice Quiz

Click here for a multiple-choice quiz.

18.7.2 Short Answer Worksheet

Click here for the chapter worksheet.

You are not logged in. Log in here and track your progress.