Migrating to PRAW 7.X#

Exception Handling#

Class APIException has also been renamed to RedditAPIException. Importing APIException will still work, but is deprecated, but will be removed in PRAW 8.0.

PRAW 7 introduced a fundamental change in how exceptions are received from Reddit’s API. Reddit can return multiple exceptions for one API action, and as such, the exception RedditAPIException serves as a container for each of the true exception objects. These objects are instances of RedditErrorItem, and they contain the information of one “error” from Reddit’s API. They have the three data attributes that APIException used to contain.

Most code regarding exceptions can be quickly fixed to work under the new system. All of the exceptions are stored in the items attribute of the exception as entries in a list. In the example code below, observe how attributes are accessed.

try:
    reddit.subreddit("test").submit("Test Title", url="invalidurl")
except APIException as exception:
    print(exception.error_type)

This can generally be changed to:

try:
    reddit.subreddit("test").submit("Test Title", url="invalidurl")
except RedditAPIException as exception:
    print(exception.items[0].error_type)

However, this should not be done, as this will only work for one error. The probability of Reddit’s API returning multiple exceptions, especially on submit actions, should be addressed. Rather, iterate over the exception, and do the action on each item in the iterator.

try:
    reddit.subreddit("test").submit("Test Title", url="invalidurl")
except RedditAPIException as exception:
    for subexception in exception.items:
        print(subexception.error_type)

Alternatively, the exceptions are provided to the exception constructor, so printing the exception directly will also allow you to see all the exceptions.

try:
    reddit.subreddit("test").submit("Test Title", url="invalidurl")
except RedditAPIException as exception:
    print(exception)