Как вызвать ошибку питон

How do I manually throw/raise an exception in Python?

Use the most specific Exception constructor that semantically fits your issue.

Be specific in your message, e.g.:

raise ValueError('A very specific bad thing happened.')

Don’t raise generic exceptions

Avoid raising a generic Exception. To catch it, you’ll have to catch all other more specific exceptions that subclass it.

Problem 1: Hiding bugs

raise Exception('I know Python!') # Don't! If you catch, likely to hide bugs.

For example:

def demo_bad_catch():
    try:
        raise ValueError('Represents a hidden bug, do not catch this')
        raise Exception('This is the exception you expect to handle')
    except Exception as error:
        print('Caught this error: ' + repr(error))

>>> demo_bad_catch()
Caught this error: ValueError('Represents a hidden bug, do not catch this',)

Problem 2: Won’t catch

And more specific catches won’t catch the general exception:

def demo_no_catch():
    try:
        raise Exception('general exceptions not caught by specific handling')
    except ValueError as e:
        print('we will not catch exception: Exception')
 

>>> demo_no_catch()
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "<stdin>", line 3, in demo_no_catch
Exception: general exceptions not caught by specific handling

Best Practices: raise statement

Instead, use the most specific Exception constructor that semantically fits your issue.

raise ValueError('A very specific bad thing happened')

which also handily allows an arbitrary number of arguments to be passed to the constructor:

raise ValueError('A very specific bad thing happened', 'foo', 'bar', 'baz') 

These arguments are accessed by the args attribute on the Exception object. For example:

try:
    some_code_that_may_raise_our_value_error()
except ValueError as err:
    print(err.args)

prints

('message', 'foo', 'bar', 'baz')    

In Python 2.5, an actual message attribute was added to BaseException in favor of encouraging users to subclass Exceptions and stop using args, but the introduction of message and the original deprecation of args has been retracted.

Best Practices: except clause

When inside an except clause, you might want to, for example, log that a specific type of error happened, and then re-raise. The best way to do this while preserving the stack trace is to use a bare raise statement. For example:

logger = logging.getLogger(__name__)

try:
    do_something_in_app_that_breaks_easily()
except AppError as error:
    logger.error(error)
    raise                 # just this!
    # raise AppError      # Don't do this, you'll lose the stack trace!

Don’t modify your errors… but if you insist.

You can preserve the stacktrace (and error value) with sys.exc_info(), but this is way more error prone and has compatibility problems between Python 2 and 3, prefer to use a bare raise to re-raise.

To explain — the sys.exc_info() returns the type, value, and traceback.

type, value, traceback = sys.exc_info()

This is the syntax in Python 2 — note this is not compatible with Python 3:

raise AppError, error, sys.exc_info()[2] # avoid this.
# Equivalently, as error *is* the second object:
raise sys.exc_info()[0], sys.exc_info()[1], sys.exc_info()[2]

If you want to, you can modify what happens with your new raise — e.g. setting new args for the instance:

def error():
    raise ValueError('oops!')

def catch_error_modify_message():
    try:
        error()
    except ValueError:
        error_type, error_instance, traceback = sys.exc_info()
        error_instance.args = (error_instance.args[0] + ' <modification>',)
        raise error_type, error_instance, traceback

And we have preserved the whole traceback while modifying the args. Note that this is not a best practice and it is invalid syntax in Python 3 (making keeping compatibility much harder to work around).

>>> catch_error_modify_message()
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "<stdin>", line 3, in catch_error_modify_message
  File "<stdin>", line 2, in error
ValueError: oops! <modification>

In Python 3:

raise error.with_traceback(sys.exc_info()[2])

Again: avoid manually manipulating tracebacks. It’s less efficient and more error prone. And if you’re using threading and sys.exc_info you may even get the wrong traceback (especially if you’re using exception handling for control flow — which I’d personally tend to avoid.)

Python 3, Exception chaining

In Python 3, you can chain Exceptions, which preserve tracebacks:

raise RuntimeError('specific message') from error

Be aware:

• this does allow changing the error type raised, and
• this is not compatible with Python 2.

Deprecated Methods:

These can easily hide and even get into production code. You want to raise an exception, and doing them will raise an exception, but not the one intended!

Valid in Python 2, but not in Python 3 is the following:

raise ValueError, 'message' # Don't do this, it's deprecated!

Only valid in much older versions of Python (2.4 and lower), you may still see people raising strings:

raise 'message' # really really wrong. don't do this.

In all modern versions, this will actually raise a TypeError, because you’re not raising a BaseException type. If you’re not checking for the right exception and don’t have a reviewer that’s aware of the issue, it could get into production.

Example Usage

I raise Exceptions to warn consumers of my API if they’re using it incorrectly:

def api_func(foo):
    '''foo should be either 'baz' or 'bar'. returns something very useful.'''
    if foo not in _ALLOWED_ARGS:
        raise ValueError('{foo} wrong, use "baz" or "bar"'.format(foo=repr(foo)))

Create your own error types when apropos

«I want to make an error on purpose, so that it would go into the except»

You can create your own error types, if you want to indicate something specific is wrong with your application, just subclass the appropriate point in the exception hierarchy:

class MyAppLookupError(LookupError):
    '''raise this when there's a lookup error for my app'''

and usage:

if important_key not in resource_dict and not ok_to_be_missing:
    raise MyAppLookupError('resource is missing, and that is not ok.')

Данный урок посвящен исключениям и работе с ними. Основное внимание уделено понятию исключения в языках программирования, обработке исключений в Python, их генерации и созданию пользовательских исключений.

Исключения в языках программирования

Исключениями (exceptions) в языках программирования называют проблемы, возникающие в ходе выполнения программы, которые допускают возможность дальнейшей ее работы в рамках основного алгоритма. Типичным примером исключения является деление на ноль, невозможность считать данные из файла (устройства), отсутствие доступной памяти, доступ к закрытой области памяти и т.п. Для обработки таких ситуаций в языках программирования, как правило, предусматривается специальный механизм, который называется обработка исключений (exception handling).

Исключения разделяют на синхронные и асинхронные. Синхронные исключения могут возникнуть только в определенных местах программы. Например, если у вас есть код, который открывает файл и считывает из него данные, то исключение типа “ошибка чтения данных” может произойти только в указанном куске кода. Асинхронные исключения могут возникнуть в любой момент работы программы, они, как правило, связаны с какими-либо аппаратными проблемами, либо приходом данных. В качестве примера можно привести сигнал отключения питания.

В языках программирования чаще всего предусматривается специальный механизм обработки исключений. Обработка может быть с возвратом, когда после обработки исключения выполнение программы продолжается с того места, где оно возникло. И обработка без возврата, в этом случае, при возникновении исключения, осуществляется переход в специальный, заранее подготовленный, блок кода.

Различают структурную и неструктурную обработку исключений. Неструктурная обработка предполагает регистрацию функции обработчика для каждого исключения, соответственно данная функция будет вызвана при возникновении конкретного исключения. Для структурной обработки язык программирования должен поддерживать специальные синтаксические конструкции, которые позволяют выделить код, который необходимо контролировать и код, который нужно выполнить при возникновении исключительной ситуации.

В Python выделяют два различных вида ошибок: синтаксические ошибки и исключения.

Синтаксические ошибки в Python

Синтаксические ошибки возникают в случае если программа написана с нарушениями требований Python к синтаксису. Определяются они в процессе парсинга программы. Ниже представлен пример с ошибочным написанием функции print.

>>> for i in range(10):
    prin("hello!")

Traceback (most recent call last):
  File "<pyshell#2>", line 2, in <module>
    prin("hello!")
NameError: name 'prin' is not defined

Исключения в Python

Второй вид ошибок – это исключения. Они возникают в случае если синтаксически программа корректна, но в процессе выполнения возникает ошибка (деление на ноль и т.п.). Более подробно про понятие исключения написано выше, в разделе “исключения в языках программирования”.

Пример исключения ZeroDivisionError, которое возникает при делении на 0.

>>> a = 10
>>> b = 0
>>> c = a / b
Traceback (most recent call last):
  File "<pyshell#5>", line 1, in <module>
    c = a / b
ZeroDivisionError: division by zero

В Python исключения являются определенным типом данных, через который пользователь (программист) получает информацию об ошибке. Если в коде программы исключение не обрабатывается, то приложение останавливается и в консоли печатается подробное описание произошедшей ошибки с указанием места в программе, где она произошла и тип этой ошибки.

Иерархия исключений в Python

Существует довольно большое количество встроенных типов исключений в языке Python, все они составляют определенную иерархию, которая выглядит так, как показано ниже.

BaseException
+– SystemExit
+– KeyboardInterrupt
+– GeneratorExit
+– Exception
     +– StopIteration
     +– StopAsyncIteration
     +– ArithmeticError
     |    +– FloatingPointError
     |    +– OverflowError
     |    +– ZeroDivisionError
     +– AssertionError
     +– AttributeError
     +– BufferError
     +– EOFError
     +– ImportError
          +– ModuleNotFoundError
     +– LookupError
     |    +– IndexError
     |    +– KeyError
     +– MemoryError
     +– NameError
     |    +– UnboundLocalError
     +– OSError
     |    +– BlockingIOError
     |    +– ChildProcessError
     |    +– ConnectionError
     |    |    +– BrokenPipeError
     |    |    +– ConnectionAbortedError
     |    |    +– ConnectionRefusedError
     |    |    +– ConnectionResetError
     |    +– FileExistsError
     |    +– FileNotFoundError
     |    +– InterruptedError
     |    +– IsADirectoryError
     |    +– NotADirectoryError
     |    +– PermissionError
     |    +– ProcessLookupError
     |    +– TimeoutError
     +– ReferenceError
     +– RuntimeError
     |    +– NotImplementedError
     |    +– RecursionError
     +– SyntaxError
     |    +– IndentationError
     |         +– TabError
     +– SystemError
     +– TypeError
     +– ValueError
     |    +– UnicodeError
     |         +– UnicodeDecodeError
     |         +– UnicodeEncodeError
     |         +– UnicodeTranslateError
     +– Warning
          +– DeprecationWarning
          +– PendingDeprecationWarning
          +– RuntimeWarning
          +– SyntaxWarning
          +– UserWarning
          +– FutureWarning
          +– ImportWarning
          +– UnicodeWarning
          +– BytesWarning
          +– ResourceWarning

Как видно из приведенной выше схемы, все исключения являются подклассом исключения BaseException. Более подробно об иерархии исключений и их описании можете прочитать здесь.

Обработка исключений в Python

Обработка исключений нужна для того, чтобы приложение не завершалось аварийно каждый раз, когда возникает исключение. Для этого блок кода, в котором возможно появление исключительной ситуации необходимо поместить во внутрь синтаксической конструкции try…except.

print("start")
try:
   val = int(input("input number: "))
   tmp = 10 / val
   print(tmp)
except Exception as e:
   print("Error! " + str(e))
print("stop")

В приведенной выше программе возможных два вида исключений – это ValueError, возникающее в случае, если на запрос программы “введите число”, вы введете строку, и ZeroDivisionError – если вы введете в качестве числа 0.

Вывод программы при вводе нулевого числа будет таким.

start input number: 0 Error! stop

Если бы инструкций try…except не было, то при выбросе любого из исключений программа аварийно завершится.

print("start")
val = int(input(“input number: “))
tmp = 10 / val
print(tmp)
print("stop")

Если ввести 0 на запрос приведенной выше программы, произойдет ее остановка с распечаткой сообщения об исключении.

start

input number: 0

Traceback (most recent call last):

 File “F:/work/programming/python/devpractice/tmp.py”, line 3, in <module>

   tmp = 10 / val

ZeroDivisionError: division by zero

Обратите внимание, надпись stop уже не печатается в конце вывода программы.

Согласно документу по языку Python, описывающему ошибки и исключения, оператор try работает следующим образом:

• Вначале выполняется код, находящийся между операторами try и except.
• Если в ходе его выполнения исключения не произошло, то код в блоке except пропускается, а код в блоке try выполняется весь до конца.
• Если исключение происходит, то выполнение в рамках блока try прерывается и выполняется код в блоке except. При этом для оператора except можно указать, какие исключения можно обрабатывать в нем. При возникновении исключения, ищется именно тот блок except, который может обработать данное исключение.
• Если среди except блоков нет подходящего для обработки исключения, то оно передается наружу из блока try. В случае, если обработчик исключения так и не будет найден, то исключение будет необработанным (unhandled exception) и программа аварийно остановится.

Для указания набора исключений, который должен обрабатывать данный блок except их необходимо перечислить в скобках (круглых) через запятую после оператора except.

Если бы мы в нашей программе хотели обрабатывать только ValueError и ZeroDivisionError, то программа выглядела бы так.

print("start")
try:
   val = int(input("input number: "))
   tmp = 10 / val
   print(tmp)
except(ValueError, ZeroDivisionError):
   print("Error!")
print("stop")

Или так, если хотим обрабатывать ValueError, ZeroDivisionError по отдельность, и, при этом, сохранить работоспособность при возникновении исключений отличных от вышеперечисленных.

print("start")
try:
   val = int(input("input number: "))
   tmp = 10 / val
   print(tmp)
except ValueError:
   print("ValueError!")
except ZeroDivisionError:
   print("ZeroDivisionError!")
except:
   print("Error!")
print("stop")

Существует возможность передать подробную информацию о произошедшем исключении в код внутри блока except.

rint("start")
try:
   val = int(input("input number: "))
   tmp = 10 / val
   print(tmp)
except ValueError as ve:
   print("ValueError! {0}".format(ve))
except ZeroDivisionError as zde:
   print("ZeroDivisionError! {0}".format(zde))
except Exception as ex:
   print("Error! {0}".format(ex))
print("stop")

Использование finally в обработке исключений

Для выполнения определенного программного кода при выходе из блока try/except, используйте оператор finally.

try:
   val = int(input("input number: "))
   tmp = 10 / val
   print(tmp)
except:
   print("Exception")
finally:
  print("Finally code")

Не зависимо от того, возникнет или нет во время выполнения кода в блоке try исключение, код в блоке finally все равно будет выполнен.

Если необходимо выполнить какой-то программный код, в случае если в процессе выполнения блока try не возникло исключений, то можно использовать оператор else.

try:
   f = open("tmp.txt", "r")
   for line in f:
       print(line)
   f.close()
except Exception as e:
   print(e)
else:
   print("File was readed")

Генерация исключений в Python

Для принудительной генерации исключения используется инструкция raise.

Самый простой пример работы с raise может выглядеть так.

try:
   raise Exception("Some exception")
except Exception as e:
   print("Exception exception " + str(e))

Таким образом, можно “вручную” вызывать исключения при необходимости.

Пользовательские исключения (User-defined Exceptions) в Python

В Python можно создавать собственные исключения. Такая практика позволяет увеличить гибкость процесса обработки ошибок в рамках той предметной области, для которой написана ваша программа.

Для реализации собственного типа исключения необходимо создать класс, являющийся наследником от одного из классов исключений.

class NegValException(Exception):
   pass

try:
   val = int(input("input positive number: "))
   if val < 0:
       raise NegValException("Neg val: " + str(val))
   print(val + 10)
except NegValException as e:
  print(e)

P.S.

Если вам интересна тема анализа данных, то мы рекомендуем ознакомиться с библиотекой Pandas. На нашем сайте вы можете найти вводные уроки по этой теме. Все уроки по библиотеке Pandas собраны в книге “Pandas. Работа с данными”.

<<< Python. Урок 10. Функции в Python   Python. Урок 12. Ввод-вывод данных. Работа с файлами>>>

Python exception handling is the process of identifying and responding to errors in a program. In other words, it is a way to deal with errors that might occur in your program. In this article, you will learn how to handle errors in Python by using the Python try and except keywords. It will also teach you how to create custom exceptions, which can be used to define your own specific error messages.

What is an exception?

An exception is a condition that arises during the execution of a program. It is a signal that something unexpected happened. Python represents exceptions by an object of a specific type.

In Python, all built-in, non-system-exiting exceptions are derived from the Exception class. Exceptions have their own, descriptive names. For example, if you try to divide a number by zero, you will get a ZeroDivisionError exception, which is also a subclass of the Exception class.

For a complete hierarchy of all exceptions, you can view the Python manual if you’re interested. Here’s a small excerpt from this hierarchy, to illustrate:

BaseException
 +-- SystemExit
 +-- KeyboardInterrupt
 +-- Exception
      +-- ArithmeticError
      |    +-- FloatingPointError
      |    +-- OverflowError
      |    +-- ZeroDivisionError
      .....

In case your knowledge about objects, classes, and inheritance is a bit rusty, you may want to read my article on objects and classes and my article on inheritance first.

When something unexpected occurs, we can raise an exception at the point of the error. When an exception is raised, Python stops the current flow of execution and starts looking for an exception handler that can handle it. So what is an exception handler? Here’s where the try and except statements come into play.

As demonstrated in the illustration, we can create a code block by starting with a try statement. This means: try to run this code, but an exception might occur.

After our try block, one or more except blocks must follow. This is where the magic happens. These except blocks can catch an exception, as we usually call this. In fact, many other programming languages use a statement called catch instead of except. Each except block can handle a specific type of exception.

Remember: classes are hierarchical. For that reason, exceptions are too. Hence, except blocks must go from most specific, like a ZeroDivisionError, to less specific, like an ArithmeticError.

To demonstrate this, imagine what happens when we would start with an except block that catches Exception. This first block would catch basically everything, because most exceptions inherit from this one, rendering the other except blocks useless.

Now let’s first go back to raising an exception. When an exception is raised, the exception handler that’s able to handle the exception can be nearby, but it can also be in a different module. What’s important to realize, is that Python won’t just scan your code at random for an exception handler. Instead, the handler should be somewhere in the call stack.

Please forget about else and finally for now. I’ll explain them in detail further down in this article. We first need to discuss call stacks, to truly understand how an exception finds its way to an exception handler.

Call stack

A call stack is an ordered list of functions that are currently being executed. For example, you might call function A, which calls function B, which calls function C. We now have a call stack consisting of A, B, and C. When C raises an exception, Python will look for an exception handler in this call stack, going backward from end to start. It can be in function C (closest to the exception), in function B (somewhat farther), in function A, or even at the top level of the program where we called function A.

If Python finds a suitable except block, it executes the code in that block. If it doesn’t find one, Python handles the exception by itself. This means it will print the exception and exit the program since it has no clue what to do with it.

I hope you’re still with me! If not, no worries. The examples on this page will hopefully make all this more clear. You might want to revisit this section once you’ve finished the entire article.

Catching exceptions with try except

Let’s finally write some actual code! To handle an exception, we need to catch it. As we just learned, we can catch an exception by using the try and except keywords. When an exception occurs while we are inside the try block, the code in the except block is executed.

A simple example

Let’s try a simple example first. As you hopefully know, we can’t divide by the number zero. If we do so anyway, Python will throw and exception called ZeroDivisionError, which is a subclass of ArithmeticError:

try:
    print(2/0)
except ZeroDivisionError:
    print("You can't divide by zero!")

If you call a Python function inside the try block, and an exception occurs in that function, the flow of code execution stops at the point of the exception and the code in the except block is executed. Try doing this again without try and except. You’ll see that Python prints the exception for us. You can do so in the following code crumb:

Also, note that Python prints the error to stderr if you don’t handle the exception yourself. In the crumb above, this is visible because the output appears in an ‘Error’ tab instead of an ‘Output’ tab.

Catching IOError

Let’s try another, more common example. After all, who divides a number by zero, right?

Exceptions are likely to occur when interacting with the outside world, e.g., when working with files or networks. For example, if you try to open a file with Python, but that file doesn’t exist, you will get an IOError exception. If you don’t have access to a file due to permissions, you will again get an IOError exception. Let’s see how to handle these exceptions.

Assignment

Please do the following:

1. Run the code below, and notice the file name (it doesn’t exist). See what happens.
2. Alter the file name to myfile.txt file and rerun the code. What happens now?

Alternatively, here’s the code to copy/paste:

try:
    # Open file in read-only mode
    with open("not_here.txt", 'r') as f:
        f.write("Hello World!")
except IOError as e:
    print("An error occurred:", e)

Answers

The file is not found in the first case. You should get this output:

An error occurred: [Errno 2] No such file or directory: 'not_here.txt'

You’ll still get an error after creating the file in the second case. This time, we’re trying to write to a file that is opened in read-only mode. For more information on these modes, read the article on opening, reading, and writing files with Python. The error should look like this:

An error occurred: not writable

Although this is an error, it’s not written to the stderr output of the operating system. That’s because we handled the exception ourselves. If you removed the try.. except from the code completely and then try to write to the file in read-only mode, Python will catch the error, force the program to terminate, and show this message:

Traceback (most recent call last):
  File "tryexcept.py", line 3, in <module>
    f.write("Hello World!")
io.UnsupportedOperation: not writable

The finally and else blocks

Remember the other two blocks that I asked you to forget for a while? Let’s look at those now, starting with the finally block.

The finally block in try-except

The finally block is executed regardless of whether an exception occurs or not. Finally blocks are useful, for example, when you want to close a file or a network connection regardless of what happens. After all, you want to clean up resources to prevent memory leaks.

Here’s an example of this at work, in which we open a file without using the with statement, forcing us to close it ourselves:

try:
    # Open file in write-mode
    f = open("myfile.txt", 'w')
    f.write("Hello World!")
except IOError as e:
    print("An error occurred:", e)
finally:
    print("Closing the file now")
    f.close()

You can also try this interactive example:

You should see the ‘Closing the file now’ message printed on your screen. Now change the writing mode to ‘r’ instead of ‘w’. You’ll get an error since the file does not exist. Despite this exception, we try to close the file anyway thanks to the finally block. This in turn goes wrong too: a NameError exception is thrown because the file was never opened, and hence f does not exist. You can fix this with a nested try.. except NameError. Try it for yourself.

The else block in try-except

In addition to the except and finally blocks, you can add an else block. The else block executes only when no exception occurs. So it differs from the finally block, since finally executes even if an exception occurs.

When should you use the else block? And why shouldn’t you just add extra code to the try block? Good questions!

According to the Python manual, using the else clause is better than adding additional code to the try clause. But why? The reasoning is that it avoids accidentally catching an exception that wasn’t raised by the code being protected by the try and except statements in the first place. I admit I don’t use else blocks very often myself. Also, I find them somewhat confusing, especially for people coming from other languages.

Common Python exceptions

Some exceptions are so common that you’ll inevitably encounter them. Here are a few of the most common ones:

Exception name	When you’ll encounter it	Example situation that raises the exception
SyntaxError	Raised when there is an error in Python syntax. If not caught, this exception will cause the Python interpreter to exit.	pritn('test')
KeyError	Raised when a key is not found in a dictionary.	d = { 'a': 1} d['b']
IndexError	Raised when an index is out of range.	lst = [1, 2, 3] lst[10]
KeyboardInterrupt	Raised when the user hits the interrupt key (Ctrl+C)	Pressing control+c

If you like, you can try to evoke these exceptions intentionally. I promise you, you will encounter these countless times in your Python programming career. Understanding what they mean and when they occur will greatly help you debug your code.

Exception best practices

Now that we know the mechanics of handling exceptions, there are a couple of best practices I’d like to share with you.

Don’t use blank except blocks

I’ve written about this in the blog post ‘How not to handle exceptions in Python‘. Don’t use a blank block when you want to catch a broad range of exceptions. By this, I mean something like:

try:
    ...
except:
    print("An error occurred:")

You might encounter this in code samples on the web. If you do, make a habit of improving the exception handling. Why should you, and how can you improve code like the example above?

All exceptions, including system exceptions, inherit from a class called BaseException. If an except clause mentions a particular class, that clause also handles any exception classes derived from that class. An empty except is equivalent to except BaseException, hence it will catch all possible exceptions.

So although the syntax is allowed, I don’t recommend it. E.g., you’ll also catch KeyboardInterrupt and SystemExit exceptions, which prevent your program from exiting. Instead, use a try block with a list of explicit exceptions that you know you can handle. Or, if you really need to, catch the Exception base class to handle almost all the regular exceptions, but not the system ones.

If you’re feeling adventurous, you can try to catch all exceptions and see what happens:

from time import sleep

while True:
    try:
        print("Try and stop me")
        sleep(1)
    except:
        print("Don't stop me now, I'm having such a good time!")

You’ll probably need to close your terminal to stop this program. Now change the except block to catch Exception. You will still catch almost all exceptions, but the program will exit on system exceptions like KeyboardInterrupt and SystemExit:

from time import sleep

while True:
    try:
        print("Try and stop me")
        sleep(1)
    except Exception:
        print("Something went wrong")

It’s better to ask for forgiveness

In Python, you’ll often see a pattern where people simply try if something works, and if it doesn’t, catch the exception. In other words, it’s better to ask for forgiveness than permission. This is in contrast to other languages, where you preferably ask for permission. E.g., in Java, exceptions can slow down your program and you “ask for permission” by doing checks on an object instead of simply trying.

To make this more concrete: in Python, we often just try to access the key in a dictionary. If the key doesn’t exist, we’ll get an exception and handle it. Suppose we just converted some externally provided JSON to a dictionary, and now start to use it:

import json

user_json = '{"name": "John", "age": 39}'
user = json.loads(user_json)

try:
    print(user['name'])
    print(user['age'])
    print(user['address'])
    ...
except KeyError as e:
    print("There are missing fields in the user object: ", e)
    # Properly handle the error
    ...

This will print the error:

There are missing fields in the user object: 'address'

We could have added three checks (if 'name' in user, if 'age' in user, etc.) to make sure that all the fields are there. But this is not a good practice. It potentially introduces a lot of code just to check if keys exist. Instead, we ask for forgiveness in our except block once, which is much cleaner and more readable. And if you worry about performance: exceptions don’t take up that many CPU cycles in Python. Lots of comparisons are in fact slower than catching a single exception (if it occurs at all!).

Create custom exceptions

All built-in, non-system-exiting exceptions are derived from the Exception class as we learned before. All user-defined exceptions should also be derived from this class. So if we want to create our own exceptions, we need to create a subclass of the Exception class.

For example, if you want to create an exception that indicates that a user was not found, you can create a UserNotFoundError exception. This would, in its most basic form, look like this:

class UserNotFoundError(Exception):
    pass

This inherits all the properties and methods of Exception, but we give it a new name to distinguish it from the Exception class. This way, we’ll be able to specifically catch it with an except block.

The name of this exception clearly tells us the type of problem that was encountered, so as an added bonus, it functions as a form of code documentation as well. Just like well-named variables and functions, a well-named exception can be a big difference when reading back your code.

We’ll use this class in the example that follows.

Raising (or throwing) exceptions

We know some built-in exceptions and how to create custom exceptions. We also know how to catch exceptions with try and except. What’s left, is what’s called raising or throwing an exception. You can raise an exception yourself with the raise keyword.

In the example below, we use your previously defined UserNotFoundError. We call a function, fetch_user, that fetches some user data from an imaginary database. If the user is not found, this database returns None. We decided that we don’t want to return None, which would force the caller to check for None every time. Instead, we use our custom UserNotFoundError.

class UserNotFoundError(Exception):
    pass

def fetch_user(user_id):
    # Here you would fetch from some kind of db, e.g.:
    # user = db.get_user(user_id)

    # To make this example runnable, let's set it to None
    user = None

    if user == None:
        raise UserNotFoundError(f'User {user_id} not in database')
    else:
        return user

users = [123, 456, 789]
for user_id in users:
    try:
        fetch_user(user_id)
    except UserNotFoundError as e:
        print("There was an error: ", e)

Assignment

Here’s a little assignment. We could have used a regular Exception object instead. That way, we don’t need to define a custom one. Why is this a bad idea?

Answer

You can in fact raise a regular exception, e.g. with raise Exception('User not found'). But if you do, you need to catch all exceptions of type Exception. And as we know, there are a lot of those. Chances are you inadvertently catch some other exception that you’re not able to handle. For example, the database client might throw a DatabaseAuthenticationError which is also a subclass of Exception.

How to print a Python exception

You can print exceptions directly as long as you catch them properly. You may have seen examples of this above already. To be clear, here’s an example of how to catch and print an exception:

try:
    ...
except Exception as e:
    print("There was an error: ", e)

If you’d like to print the call stack, just like Python does when you don’t catch the exception yourself, you can import the traceback module:

import traceback

try:
    ...
except Exception:
    traceback.print_exc()

Keep learning

Here are some more resources to deepen your knowledge:

• My blog article ‘How not to handle exceptions in Python‘
• Introduction to Python functions
• Objects and classes, and Python inheritance
• The official documentation on exceptions.
• The official documentation on errors.