B074JG8W8Z EBOK by Powerful Python 2nd Edition

Author:Powerful Python, 2nd Edition
Language: eng
Format: azw3, epub
Publisher: Powerful Python Press
Published: 0101-01-01T00:00:00+00:00

Exceptions Are Objects

An exception is an object: an instance of an exception class. KeyError, IndexError, TypeError and ValueError are all built-in classes, which inherit from a base class called Exception. Writing code like except KeyError: means "if the exception just raised is of type KeyError, run this block of code."

So far, we haven’t dealt with those exception objects directly. And often, you don’t need to. But sometimes you want more information about what happened, and capturing the exception object can help. Here’s the structure:

try: do_something() except ExceptionClass as exception_object: handle_exception(exception_object)

where ExceptionClass is some exception class, like KeyError, etc. In the except block, exception_object will be an instance of that class. You can choose any name for that variable; no one actually calls it exception_object, preferring shorter names like ex, exc, or err. The methods and contents of that object will depend on the kind of exception, but almost all will have an attribute called args. That will be a tuple of what was passed to the exception’s constructor. The args of a KeyError, for example, will have one element - the missing key:

# Atomic numbers of noble gasses. nobles = {'He': 2, 'Ne': 10, 'Ar': 18, 'Kr': 36, 'Xe': 54} def show_element_info(elements): for element in elements: print('Atomic number of {} is {}'.format( element, nobles[element])) try: show_element_info(['Ne', 'Ar', 'Br']) except KeyError as err: missing_element = err.args[0] print('Missing data for element: ' + missing_element)

Running this code gives you the following output:

Atomic number of Ne is 10 Atomic number of Ar is 18 Missing data for element: Br

The interesting bit is in the except block. Writing except KeyError as err stores the exception object in the err variable. That lets us look up the offending key, by peeking in err.args. Notice we could not get the offending key any other way, unless we want to modify show_element_info (which we may not want to do, or perhaps can’t do, as described before).

Let’s walk through a more sophisticated example. In the os module, the makedirs function will create a directory:

# Creates the directory "riddles", relative # to the current directory. import os os.makedirs("riddles")

By default, if the directory already exists, makedirs will raise FileExistsError:[3] Imagine you are writing a web application, and need to create an upload directory for each new user. That directory should not exist; if it does, that’s an error and needs to be logged. Our upload-directory-creating function might look like this:

# First version.... import os import logging UPLOAD_ROOT = "/var/www/uploads/" def create_upload_dir(username): userdir = os.path.join(UPLOAD_ROOT, username) try: os.makedirs(userdir) except FileExistsError: logging.error( "Upload dir for new user already exists")

It’s great we are detecting and logging the error, but the error message isn’t informative enough to be helpful. We at least need to know the offending username, but it’s even better to know the directory’s full path (so you don’t have to dig in the code to remind yourself what UPLOAD_ROOT was set to).

Fortunately, FileExistsError objects have an attribute called filename. This is a string, and the path to the already-existing directory. We can

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.