Anyone know how nosetest's -m, -i and -e work?

问题

I am trying to get nosetests to identify my tests but it is not running any of my tests properly.

I have the following file structure

Project
    +----Foo/
          +----__init__.py
          +----bar.py
    +----test/
          +----__init__.py
          +----unit/
                +----__init__.py
                +----bar_test.py
          +----functional/
                +----__init__.py
                +----foo_test.py

Within bar_test.py

class BarTest(unittest.TestCase):
     def bar_1_test():
         ...

Within foo_test.py

class FooFTest.py
     def foo_1_test():
         ...

Using -m, -i, -e options of nosetests

What is the regex I need to run only unit tests (under unit/, tests in class BarTest)
What is the regex I need run only functional tests (under functional/, tests in class FooFTest)

I've tried various combinations and can't seem to get nosetests to do what I want consistently

回答1:

The easiest way to run only the tests under Project/test/unit is to use --where. For example:

nosetests --where=Project/test/unit

Then use --match (-m) or --exclude (-e) to refine the list, if needed.

If you still want to use the regex selectors, you can probably do it like this (not tested):

nosetests --match='^Foo[\b_\./-])[Tt]est'

Executing this script from the Project directory would run all tests that start with "Foo" and end in "[Tt]est".

As a general rule, you probably want to use either --match or --exclude, but not both. These parameters both specify the pattern of the function names to match. You can refine either one by using --ignore-files which, naturally, allows you to ignore whole files.