Total number of run test cases is calculated incorrectly

[Roman Podoliaka]

Affected versions: 0.0.16 and 0.0.17

Consider the following test suite:

import unittest

class Test(unittest.TestCase):
    def test_case1(self):
        self.assertTrue(2 > 1)

    def test_case2(self):
        self.assertTrue(2 > 1)

.testr.conf:

[DEFAULT]
test_command=python -m subunit.run discover . $LISTOPT $IDOPTION
test_id_option=--load-list $IDFILE
test_list_option=--list

Running of "testr run" shows:
(venv)rpodolyaka@rpodolyaka-pc:~/sandbox/tesrt$ testr run
running=python -m subunit.run discover .
Ran 4 (+2) tests in 0.000s (-0.000s)
PASSED (id=143)

So the number of run test cases is doubled to the actual number of test cases.

--parallel and --concurrency X have no effect on this.

[Robert Collins (lifeless)]

Thanks, will get onto this soon.

[David Kranz (david-kranz)]

Robert thinks this may also be the cause of duplicate log messages at the same time for each tempest test as seen in

http://logs.openstack.org/91/43791/3/check/gate-tempest-devstack-vm-full/07c8c7f/console.html

[Robert Collins (lifeless)]

Thanks, will get onto this soon.

[Jonathan Lange (jml)]

https://github.com/testing-cabal/testtools/issues/54 was reported. Might be root cause?

[Robert Collins (lifeless)]

It is the root cause.

[Robert Collins (lifeless)]

https://github.com/testing-cabal/testtools/pull/61