Be consistent on test tools and drivers
Bug #1490747 reported by
Dave McCowan
This bug affects 1 person
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
Barbican |
Fix Released
|
Medium
|
Fernando Diaz |
Bug Description
Most of the time Barbican uses testr to run unit tests and functional tests.
functionalte
tox -e py27
In one case Barbican uses nosetests
tox -e functional
It would be better to use one tool for all.
One reason is test attributes. Nose only knows about nose attributes. Testr only knows about testtools attributes.
To benefit from using attributes, we need to pick one flavor.
One good use of test attributes is to mark those tests that can not be run in parallel with a tag such as "no_parallel", then when running tests only run those tests not marked so with the --parallel parameter.
description: | updated |
Changed in barbican: | |
status: | New → In Progress |
assignee: | nobody → Fernando Diaz (diazjf) |
Changed in barbican: | |
importance: | Undecided → Medium |
Changed in barbican: | |
milestone: | none → liberty-rc1 |
Changed in barbican: | |
status: | Fix Committed → Fix Released |
Changed in barbican: | |
milestone: | liberty-rc1 → 1.0.0 |
To post a comment you must log in.
Fix available here: /review. openstack. org/#/c/ 218117/
https:/