Configure Mocks (Part 1)

00:00 All right, we’re going to switch gears into talking about how you can configure your Mock objects. I’m going to start an IPython session. And let’s do from unittest.mock, import the Mock class.

00:15 What we’ve done so far is something like mock = Mock() and then mock.side_effect is maybe a KeyError exception. And we’ve kind of configured our Mock attributes line by line, like .call_count = 0 and mock.return_value = True.

00:40 And then we can use our Mock object and see how it raises a KeyError because we’ve set the .side_effect. Rather than setting these attributes of the Mock object line by line, we can set them when we first initiate the Mock object by using the .__init__() method that comes with this Mock class.

01:02 I’ll clear the screen and show you what I mean. Instead of doing it line by line, we can just say mock = Mock() and then within the parentheses, we can define our return_value, our side_effect, or whatever we want to do. Let’s say the side_effect is an AttributeError.

01:24 And then when we press Enter, we have these attributes set already on our Mock object. So when we call mock(), we see that it’s an AttributeError now.

01:38 That’s one way you can configure your mocks—during the initialization of the Mock object. Another way is by using a function called .configure_mock(),

01:53 and this is useful for when you want to change an attribute of a Mock object that already exists. So for example, when we call mock(), it raises an AttributeError—but what if we didn’t want that anymore, or what if we wanted to change that?

02:12 What we could do is say mock.configure_mock(), and then here we can set the side_effect again. Maybe we don’t want to have a side effect this time.

02:26 So we configured Mock to have no side effect, and this time when we call mock() the exception isn’t raised and it returns True because we originally set the .return_value

02:39 to be True, right? So you can configure your Mock through the .__init__() method or use this .configure_mock() in the same way, setting the attributes within the parentheses.

02:52 But this is a little more dynamic—you can do it with existing Mock objects. And just to verify that one more time, let’s change the .return_value.

03:00 We can say mock.configure_mock(return_value=6).

03:10 When we call mock() again, it returns 6.