Handling errors: try and except

You have written a calculator. The user types “twelve” instead of a number, or divides by zero — and the program stops with a red error message. Nobody likes a program like that! A good program is ready for mistakes: it recognises the problem, explains it clearly to the user and carries on.

What are exceptions?

There are two kinds of errors. Syntax errors (SyntaxError) are mistakes in how the code is written, and the program does not even start. When the code is written correctly but something goes wrong while it runs, Python raises an exception. If the exception is not caught, the program stops and Python shows a “traceback” — an error report:

Traceback (most recent call last):
  File "main.py", line 2, in <module>
    print(10 / count)
ZeroDivisionError: division by zero

try and except

We put code that might fail inside a try block. If an exception happens, Python skips the rest of the try block and jumps to the matching except block. After that, the program continues normally.

text = 'abc'
try:
    number = int(text)
    print('Number:', number)
except ValueError:
    print('That is not a number!')
print('The program continues.')

That is not a number!
The program continues.

One try can have several except blocks — each catches its own type of exception. Writing except ... as e stores the exception itself in the variable e; if you print it, you see Python's explanation.

def safe_divide(a, b):
    try:
        return a / b
    except ZeroDivisionError:
        print('Cannot divide by zero!')
    except TypeError as e:
        print('Wrong type:', e)
    return None

print(safe_divide(10, 2))
print(safe_divide(5, 0))
print(safe_divide(5, 'x'))

5.0
Cannot divide by zero!
None
Wrong type: unsupported operand type(s) for /: 'int' and 'str'
None

else and finally

• The else block runs only if the try block finished without an error — put the work with the successful result there.
• The finally block always runs — whether there was an error or not. It is usually used to wrap things up: closing a file, writing a message.

for value in ['42', 'seven']:
    try:
        n = int(value)
    except ValueError:
        print(f'{value}: not a number')
    else:
        print(f'{value}: doubled is {n * 2}')
    finally:
        print('checked', value)

42: doubled is 84
checked 42
seven: not a number
checked seven

raise: raising your own exception

Sometimes everything is fine for Python but not for your program: for example, an age cannot be negative. In such cases you can raise an exception yourself with raise. The code below imitates a user's answers and accepts the first valid age:

answers = ['abc', '-3', '14']
for answer in answers:
    try:
        age = int(answer)
        if age <= 0:
            raise ValueError('age must be positive')
    except ValueError as e:
        print(f'Bad input {answer}: {e}')
    else:
        print('Accepted age:', age)
        break

Bad input abc: invalid literal for int() with base 10: 'abc'
Bad input -3: age must be positive
Accepted age: 14

data = ['12', 'seven', '30', '4.5', '8']
total = 0
# for each item: try to add int(item) to total,
# on ValueError print Skipped: and the item
# print Total: and total

Skipped: seven
Skipped: 4.5
Total: 50

# define sell(stock, count) here

stock = 100
for count in [30, -5, 90, 50]:
    try:
        stock = sell(stock, count)
        print('Sold', count, '- left:', stock)
    except ValueError as e:
        print('Error:', e)

Sold 30 - left: 70
Error: count must be positive
Error: not enough in stock
Sold 50 - left: 20