Refactor Serial Tests pt1

[k-macmillan]

User Story - Business Need

Some tests require serial execution due to the called functions operating on more than what the test has control over. These are usually valuable integration tests with the database; we do not want to remove them.

Other tests marked serial could be refactored to not be dependent on serial execution. Here are some of the reasons a test may have been marked serial:

• The reusable fixture uses hardcoded references
• Multiple rows being found when one was expected
• Test does not make good use of freezetime

Update all non-dao tests that must remain serial with a comment: # requires serial execution - <method_name>, where <method_name> is the method within the test that requires serial execution. These should only be due to the query, and where freezing time does not work.

• Ticket is understood, and QA has been contacted (if the ticket has a QA label).

User Story(ies)

As a VA Notify engineer
I want to limit serial tests
So that we can keep as many tests parallel as possible

Additional Info and Resources

Engineering Checklist

• Update the serial tests in all folders other than tests/app/dao/
• Any that cannot be updated should be labelled # requires serial execution - <method_name> with the method that cannot be ran in parallel

Acceptance Criteria

• All serial tests that remain, other than those in tests/app/dao, have been labelled # requires serial execution - <method_name> or have @pytset.mark.serial removed