Incorrect test duration recorded when test uses freezegun

[petedmarsh]

I came across this project yesterday and I tried it out on my Django project, I got some strange output:

77450953.5910s test_a (some.TestCase)
45914963.3745s test_b (other.TestCase)
...
1.6083s test_c (final.TestCase)

Now, my some of my tests are on the slow side but they don't take > 2 years to run!

I remember that some of my test cases use freezegun to test date/time sensitive code - I think that's likely the cause.

In my case other.TestCase.test_b uses freezegun:

class TestCase(TestCase):
    @freeze_time('2016-11-14 10:00')
    def test_b(self):
        #...

but some.TestCase.test_a does not - and when I run other.TestCase.test_b in isolation I get a reasonable recorded duration, so it seems that using freezegun in one test affects the recorded time of another.

[mjhea0]

Thanks for letting me know. I need to look at freeze gun and figure out what's going on. Give me a few days.

[petedmarsh]

I've had a look myself and I have a patch, will raise a PR later- fairly simple to fix fortunately!

…

On 15 Jun 2017 1:43 p.m., "Michael Herman" ***@***.***> wrote: Thanks for letting me know. I need to look at freeze gun and figure out what's going on. Give me a few days. — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#16 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ABIxepuUPaMThuBYLjPUq9XehFyxBHeZks5sESb_gaJpZM4N7Am_> .