Open Stack

Hi folks,

I have a question for the refstack team.

Does the idempotent_id has to be unique for a single test? To give more
context: the ddt framework [1] will generate multiple similar tests with
different data set. In the current implementation it’s something like a for-
loop within the test itself. So refstack won’t recognize it.

DDT autogenerated tests will look like independent tests from the test runner
perspective. Introducing a mechanism to assign a unique idempotent_id would be
quite channeling (but not impossible). So main question: is refstack fine with
a idempotent_id that groups a set of tests?

Please have a look to the qa-spec for more detail [2] and sample
implementation [3].

Regards Marc
SAP SE

[1]: http://ddt.readthedocs.org/en/latest/
[2]: https://review.openstack.org/#/c/259934/
[3]: https://review.openstack.org/#/c/223953/