Unittest integration

On PyDev 1.6.4, an improved unittest support was introduced. It allows using different test runners (the default PyDev test runner, nose or py.test) and allows seeing the results in a view (PyUnit view) with a red/green bar which also allows re-running tests.

Configuring test runners

The preferences page to configure the test runner is accessible from the PyUnit view at the dropdown menu > configure test runner preferences (or at window > preferences > PyDev > PyUnit)

Note: the flags to choose the tests in the test runner should not be specified (they are properly managed by PyDev in each test run)

Configuring the PyDev test runner

The options that the PyDev unittest accepts are:

--verbosity=number
Sets the verbosity level for the run

--jobs=number
The number of processes to be used to run the tests

--split_jobs=tests|module
if tests is passed (default), the tests will be split independently to each process
if module is passed, a given job will always receive all the tests from a module

An example of options that can be set in the preferences would be:

--verbosity=1 --jobs=2 --split_jobs=module

Configuring the Nose test runner

Note: the integration was tested with version 1.0

The options that the nose test runner accepts can be found at http://nose.readthedocs.io

An example of options that can be set in the preferences would be:

--verbosity=2 --processes=2

--verbosity=2 (set the verbosity level to 2)

--processes=2 (use 2 processes to run the tests)

Configuring the Py.test test runner

Note: the integration was tested with version 2.0

The options that the py.test test runner accepts can be found at http://pytest.org

An example of options that can be set in the preferences would be:

-maxfail=2 --tb=native

-maxfail=2 (stop at 2nd failure)

--tb=native (will show tracebacks in the default standard library formatting)

Note: currently when using the xdist plugin, the results won't be properly shown in the PyUnit view.

PyUnit view

The PyUnit view may be used to see the results of tests being run, their output, time, re-running tests, among others.

The most interesting features related to seeing the tests are:

• The results of tests are shown, along with a green/red bar depending whether all the tests succeeded or not (or if you're with Aptana Studio, following the color theme)
• The time a test took to run is shown (and may be used to sort the tree)
• The errors/output are shown by selecting the test run or just hovering over the items
• A filter to show only errors is available

And the most interesting actions are:

• A test session can be rerun
• A new test session can be created to rerun only the errors of the current run
• The test session can be stopped
• The results of a previously run test session can be seen again

Running and showing results in the PyUnit view

To show the results in the PyUnit view, a unittest run must be done. If running the tests when editing a module, Ctrl+F9 may be used to run the tests in that mode (and choosing which tests should be run), otherwise, right-click a folder or python file and choose Run as > Python Unittest