Assert for exact facts match on fact related tests #36
Conversation
Add the list of facts that RHO scan's for and update the fact related tests to match for the exact match of the expected facts. Closes #32
| @@ -45,12 +53,15 @@ def test_fact_list_filter(): | |||
| :steps: Run ``rho fact list --filter <filter>`` | |||
| :expectedresults: Only the facts that match the filter are printed. | |||
| """ | |||
| rho_fact_list = pexpect.spawn('rho fact list --filter connection') | |||
| facts = random.choice((RHO_CONNECTION_FACTS, RHO_JBOSS_FACTS)) | |||
| fact_filter = facts[0].split('.')[0] | |||
There was a problem hiding this comment.
So it looks like you are randomly chooses which to test.
What motivated that choice? Will it be easy to tell which once failed if we get a failure?
I would think instead to parametrize the test on this list and test both, that way if we get failures easy to figure out which one failed?
Maybe it would be clear if it failed, I'm not 100% certain what failure out put would be for this test.
There was a problem hiding this comment.
The initial thought for this test was to ensure the fact list can be filtered, so it does really matter which fact subset we are looking for. I decided to use random to make sure we can test different values and try to reduce the chance to work only for an specific subset.
Having a parametrized test is a good idea. But I would recommend something slightly different, since it accepts a regex I would parametrize by a string as a field or a real regex. This improvement deserve an issue by its own I think.
If you agree, let's open an issue to test using a string or a regex and we can leave this as it is and improve later. The goal here is to match for the exact expected facts.
There was a problem hiding this comment.
Just created #37 to track this
Add the list of facts that RHO scan's for and update the fact related
tests to match for the exact match of the expected facts.
Closes #32