Factorize interface for test generators

[ChristianBirchler]

Hi, I am new to this framework. I like the code pipeline and it is easy to use. I felt free to share a few thoughts about the interface for generating the test cases :-).

As a new user of this code pipeline, I want to implement only the part of a test generator that describes the actual test case such that I don't need to focus on the test execution and reporting.

Suggestions for refactoring:

• Factorize the interface of the test generators into an abstract base class (e.g., start and __init__ methods) which have to be implemented by the test generators.
• Use dependency injection for road_points in the start method:

def start(road_points):
  # add points to `road_points`

• In case of multiple tests the data structure of road_points could look like road_points = [test1_list_of_points, test2_list_of_points]
• Extract the code for the test execution.
• Extract the code for reporting the test outcome.

These are just my thoughts. What do you think about this design option? Do you see some drawbacks in this?