Posted on by Miro Boro

Exceptions in Python

1. In Python, there is a distinction between two kinds of errors:

  • syntax errors (parsing errors), which occur when the parser comes across a statement that is incorrect. For example:
    Trying to execute the following line:

    will cause a SyntaxError, and result in the following (or similar) message being displayed in the console:

    Pay attention to the arrow – it indicates the place where the Python parser has run into trouble. In our case, it’s the missing double quote. Did you notice it?

  • exceptions, which occur even when a statement/expression is syntactically correct; these are the errors that are detected during execution when your code results in an error which is not uncoditionally fatal. For example:
    Trying to execute the following line:


    will cause a ZeroDivisionError exception, and result in the following (or similar) message being displayed in the console:

    Pay attention to the last line of the error message – it actually tells you what happened. There are many different types of exceptions, such as ZeroDivisionError, NameError, TypeError, and many more; and this part of the message informs you of what type of exception has been raised. The preceding lines show you the context in which the exception has occured.

 

2. You can “catch” and handle exceptions in Python by using the try-except block. So, if you have a suspicion that any particular snippet may raise an exception, you can write the code that will gracefully handle it, and will not interrupt the program. Look at the example:


The code above asks the user for input until they enter a valid integer number. If the user enters a value that cannot be converted to an int, the program will print Warning: the value entered is not a valid number. Try again..., and ask the user to enter a number again. What happens in such a case?

  1. The program enters the while loop.
  2. The try block/clause is executed. The user enters a wrong value, for example: hello!.
  3. An exception occurs, and the rest of the try clause is skipped. The program jumps to the except block, executes it, and then continues running after the try-except block.

If the user enters a correct value and no exception occurs, the subsequent instructions in the try block are executed.

 

3. You can handle multiple exceptions in your code block. Look at the following examples:

You can use multiple except blocks within one try statement, and specify particular exception names. If one of the except branches is executed, the other branches will be skipped. Remember: you can specify a particular built-in exception only once. Also, don’t forget that the default (or generic) exception, that is the one with no name specified, should be placed at the bottom of the branch (use the more specific exceptions first, and the more general last).

You can also specify and handle multiple built-in exceptions within a single except clause:


 

4. Some of the most useful Python built-in exceptions are: ZeroDivisionError, ValueError, TypeError, AttributeError, and SyntaxError. One more exception that, in our opinion, deserves your attention is the KeyboardInterrupt exception, which is raised when the user hits the interrupt key (CTRL-C or Delete). Run the code above and hit the key combination to see what happens.

To learn more about the Python built-in exceptions, consult the official Python documentation.

 

5. Last but not least, you should remember about testing and debugging your code. Use such debugging techniques as print debugging; if possible – ask someone to read your code and help you to find bugs in it or to improve it; try to isolate the fragment of code that is problematic and susceptible to errors: test your functions by applying predictable argument values, and try to handle the situations when someone enters wrong values; comment out the parts of the code that obscure the issue. Finally, take breaks and come back to your code after some time with a fresh pair of eyes.

 

Some useful exceptions

Let’s discuss in more detail some useful (or rather, the most common) exceptions you may experience.

ZeroDivisionError

This appears when you try to force Python to perform any operation which provokes division in which the divider is zero, or is indistinguishable from zero. Note that there is more than one Python operator which may cause this exception to raise. Can you guess them all?

Yes, they are: /, //, and %.

ValueError

Expect this exception when you’re dealing with values which may be inappropriately used in some context. In general, this exception is raised when a function (like int() or float()) receives an argument of a proper type, but its value is unacceptable.

TypeError

This exception shows up when you try to apply a data whose type cannot be accepted in the current context. Look at the example:


You’re not allowed to use a float value as a list index (the same rule applies to tuples, too). TypeError is an adequate name to describe the problem, and an adequate exception to raise.

 

AttributeError

This exception arrives – among other occasions – when you try to activate a method which doesn’t exist in an item you’re dealing with. For example:


The third line of our example attempts to make use of a method which isn’t contained in the lists. This is the place where AttributeError is raised.

SyntaxError

This exception is raised when the control reaches a line of code which violates Python’s grammar. It may sound strange, but some errors of this kind cannot be identified without first running the code. This kind of behavior is typical of interpreted languages – the interpreter always works in a hurry and has no time to scan the whole source code. It is content with checking the code which is currently being run. An example of such a category of issues will be presented very soon.

It’s a bad idea to handle this exception in your programs. You should produce code that is free of syntax errors, instead of masking the faults you’ve caused.

 

Exercise 1

What is the output of the following program if the user enters 0?


 

The program will output: Very bad input....

 

Ex. 2

Assuming the user enters 2 as input, what is the output of the following code snippet?

 

The code will raise a TypeError exception

  • The correct answer is TypeError. A TypeError exception is raised whenever an operation is performed on an incorrect data type. In our example, the input function assigns the value 2, entered by the user, into the variable x, as a string: x = '2'.
  • When evaluating the next print statement, the Python interpreter will raise a TypeError exception, since the division operator does not support string-type operands. In this case, the interpreter raises the TypeError exception before the ZeroDivisionError exception:print('2' / 0) -> TypeError .

 

Ex. 3

What is the output of the following code snippet?

An error occurred

A SyntaxError exception

Division by zero is not allowed

The value entered is incorrect

 

Correct answer

A SyntaxError exception

  • The code snippet raises a SyntaxError exception because the default except block (the one without an exception name) must not be specified first but always last.

 

Ex. 4

What is the output of the following code snippet?

Correct answer

A ValueError execption

  • The syntax of the range() function is range(start, stop, step).
  • range() returns a sequence of numbers containing all values from start to stop (excluded), where step specifies the increment.
  • If the step value is 0, Python will raise a ValueError exception.

 

Ex. 5

Which of the following code snippets raises a ValueError exception?

(Select two answers)


Correct selections:

  • The Python interpreter generates a ValueError exception when a function receives an argument of the correct type but the wrong value, or, when you try to perform an operation on a value that does not exist.
  • In line 4 and 6, we find the function int(), which converts a number or a string to an integer. Line 6 raises a TypeError because the argument to the int() function (a list) is incorrect. Line 4 raises a ValueError exception because the argument is of the correct type (a string), but the "Python" value cannot be converted to an integer.
  • The code in line 2 and 9 uses the index() method, which takes an element of a list as an argument and returns the index of its first occurrence. In line 2, the code raises a ValueError exception because the value 5, passed as an argument to the index() method, does not exist in the l list. Finally, line 9 raises an AttributeError exception because dictionaries do not support the index() method.

 

Ex. 6

What is the output of the following code snippet?

Ppython

Python

A TypeError exception

pPython

 

Correct answer

A TypeError exception

The interpreter generates a TypeError exception because strings in Python are immutable objects and, therefore, cannot be modified by item assignment.

 

Ex. 7

Which of the following code snippets causes a SyntaxError exception?

 

  • No response raises SyntaxError exception.
  • In Python, if you want to place multiple statements on a single physical line, you must separate them with a semicolon (;), which indicates the end of a statement.
  • However, it is good practice to write a single instruction in a single physical line as far as possible.

 

Ex. 8

What is the output of the following code snippet?


The code will raise the ValueError exception

The code will raise the AttributeError exception

The code will raise the TypeError exception

The code will raise the SyntaxError exception

Correct answer

The code will raise the AttributeError exception

  • When we try to use a method on an object that doesn’t support it, Python raises an AttributeError exception. In our case, tuples are immutable objects and do not support the append method.
  • Additional noteYou can easily append new elements to a tuple by using the + operator. If you try to run the following code, you will see that the print statement will display the tuple tpl with the new 4 element added at the end.

    However, even in this case, tuples are always immutable. In fact, line 2 of the code does not append a new element to the old tuple, but creates a new tuple that is assigned to the same variable tpl. In other words, while at the beginning tpl points to the tuple (1, 2, 3), at the end it points to the new tuple (1, 2, 3, 4).

 

Ex. 9

What is the output of the following code snippet?

pass

Nothing

A SyntaxError exception

None

 

Correct answer

A SyntaxError exception

  • Python doesn’t support empty code blocks. To write empty code blocks you can use the statement pass.
  • pass is a special statement that does nothing.  So, you can define an empty function as follows:

 

Ex. 10

Which of the following statements correctly handle multiple exceptions in a single except block?

except TypeError, ZeroDivisionError:

except (TypeError, ZeroDivisionError):

except TypeError, ZeroDivisionError

except (TypeError, ZeroDivisionError)

Correct answer

except (TypeError, ZeroDivisionError):

  • To properly handle multiple exceptions in a single except block, we can group them into a parenthesized tuple, followed by a colon:except (except1, except2, ..., exceptionN):.

 

Ex. 11

What is the output of the following code snippet?

A TypeError exception

None

7

3

Correct answer

A TypeError exception

  • In Python, a function without a return statement (or with a return statement without a value or expression) returns None.
  • None is a Python keyword and is used to represent the absence of a value. It can be assigned to a variable and can be used in comparison operations, but cannot be used as an operand in arithmetic expressions.
  • For this reason, the print statement of the code will raise a TypeError exception:

    step by step

    Step 1 –  print(x + fun(2))

    Step 2 – print(3 + None)

    Output – TypeError