I have an application written in C++. I use the boost unit test framework to do my unit testing. I want to add some functional/end-to-end testing. My app essentially outputs a file, so I want to run the application with a few different inputs (command line arguments) and check the results by parsing the output file.
My first thought is to use python to automate these tests. I went through a reasonably exhaustive list and here's my top 3 at first glance:
unittest
framework. Seems a valid option, but is it ok for functional/end-to-end testing?nose
. Seems good, but they clearly say it's not supported anymore, so I feel I should exclude it.py.test
. Seems fairly old, and less good than nose.
So after looking into this, I feel like unittest
is the best framework for my use case. However, I have a nagging feeling that I might regret my decision. I think this is mainly because the framework is called "unittest"! Is it really restricted to unit test? Will I regret my decision some time down the road?
Aucun commentaire:
Enregistrer un commentaire