Integration testing related queries
Hi everyone, I have few doubt's with respect to certain aspects of integration testing
1. Why is bottom up testing good for object oriented systems, real-time systems and systems with strict performance requirements?
2. How does call graph based integration testing overcome the decomposition(structural) problem and how is call graph different from a functional tree(of decomposition based methods)?
3. Most importantly, I have a doubt with path based integration testing. What does it mean when its said that path based integration also focuses on interaction among related modules instead of just the interfaces.
Thanks in advance.
To answer #1 ,
Bottom up is good when you want to have a high level of code coverage and also is easier to do because as you work you way up, you can reuse some of your fixtures when you start working on the upper layers.
Top down is good in that it works better with a TDD workflow where tests are written first. I think it let's your flush out your business requirements earlier and forces you to mock heavily, which indirectly ensures you follow the SOLID principles.
Originally Posted by RollsGhost
Refer this presentation you will get some idea.Integration testing(revisited)