Python for Programmers: Raising Exceptions

• When something goes wrong, Python indicates the error by raising ("throwing") an exception. Execution stops, and Python prints a message telling us what happened.

• For example, invalid syntax raises an exception. (You can type error when a code example will cause an error.)

• >

  sum = 2 +*/- 3

  Result:

  SyntaxError: invalid syntax (<string>, line 1)

• We can raise our own exceptions with a raise statement. raise SomeExceptionType(some_message) raises the provided exception. Most exceptions take one argument: a message describing what happened.

• In the next example, we raise ValueError("Incorrect password"). ValueError is one of Python's many built-in exception types.

• >

  provided_password = "hunter3"
  correct_password = "hunter2"
  
  if provided_password != correct_password:
    raise ValueError("Incorrect password")
  "Access granted"

  Result:

  ValueError: Incorrect password

• Exceptions are values, just like 12 or "abc". Simply creating a ValueError doesn't raise it, and it won't stop execution.

• >

  provided_password = "hunter3"
  correct_password = "hunter2"
  
  if provided_password != correct_password:
    # Create an exception, but don't raise it. An easy mistake to make!
    ValueError("Incorrect password")
  "Access granted"

  Result:

  'Access granted'

• Python has a rich set of built-in exception types. Here are some examples:

• TypeError means that we provided data of the wrong type. For example, adding a number to a string raises a TypeError exception.

• >

  "a" + 1

  Result:

  TypeError: can only concatenate str (not "int") to str

• ValueError means that an argument had the correct type, but an incorrect value.

• >

  my_number = 5
  if my_number % 2 != 0:
    raise ValueError("expected an even number")
  my_number / 2

  Result:

  ValueError: expected an even number

• ZeroDivisionError means that we tried to divide by zero.

• >

  5 / 0

  Result:

  ZeroDivisionError: division by zero

• There are many more built-in exception types. We won't go through all of them here, but several will show up throughout this course.

• Python also has a generic exception type, Exception.

• >

  raise Exception("some generic error")

  Result:

  Exception: some generic error

• However, it's better to be specific when possible. ValueError makes sense in many situations, since most errors are caused by an incorrect value of some kind.

• Later in the course, we'll define our own exception types. But we need to see many more Python features before we get there.

• Finally, why does Python talk about "raising" exceptions, when most programming languages talk about "throwing" them? There's no deep reason behind it. It's just a difference in terminology, like Python having "lists" instead of "arrays".