Plex/nose Example

About

This should show how to get Plex and nose working together.

Requirements

You are expected to have Plex Media Server installed, of course.

Setup

This very repository is a functioning Gamespot.com channel. Please feel free to clone and start hacking.

Ubuntu

git clone git://github.com/mikew/plex-nose-example.git
cd plex-nose-example/
./run-ubuntu.sh Contents/Tests/test_....py # run specific tests
./run-ubuntu.sh Contents/Tests/ # run all tests

OS X

git clone git://github.com/mikew/plex-nose-example.git
cd plex-nose-example/
./run-osx.sh Contents/Tests/test_....py # run specific tests
./run-osx.sh Contents/Tests/ # run all tests

Windows

Note: Tests seem to run much faster when the server is left running.

git clone git://github.com/mikew/plex-nose-example.git
cd plex-nose-example\
.\run-windows.bat Contents\Tests\test_....py # run specific tests
.\run-windows.bat Contents\Tests # run all tests

Other Operating Systems

To run the tests, you need to grab the environment variables that Plex channels are run with. Install plex-nose-example.bundle as you would any other Plex channel, visit /video/plex-nose-example/env and note the contents of the message attribute. You are free to shut down the server after this.

In addition to this, you must find the path to the media server's Framework.bundle/.../Versions/2/Python and assign it to ENV[PLEX_FRAMEWORK_PATH]

env PLEX_FRAMEWORK_PATH="..." ... python Contents/Tests/nose_runner.py /full/path/to/Contents/Tests/test_....py

Note the /full/path/to/... requirement.

Integrating with existing channels

No modification to existing code is necessary, but you will need some files from this repository:

Contents/Libraries/Shared/nose/
Contents/Libraries/Shared/plex_nose.py
Contents/Libraries/Shared/spec.py
Contents/Tests/nose_runner.py
run-osx.sh
run-ubuntu.sh

Writing Tests

Nose is included and used to run the tests, so you can write super-quick tests:

import plex_nose

@plex_nose.sandbox
def test():
    ok_('some_data' not in Dict)

Just be sure to use the plex_nose.sandbox decorator.

Or, to make things even easier you can just extend plex_nose.TestCase and stop using the decorator:

import plex_nose

class TestMyChannel(plex_nose.TestCase):
    def test():
        ok_('some_data' not in Dict)

Note that the function definition is def test(): and not def test(self): as you would expect. This is due to the fact that different parts of channel code are run in different sandboxes, there is no concept of self.

Either way, the tests are run in the exact* environment as if they were launched by the media server, but in a fraction of the time.

* Well, not exact. The Dict is cleaned after each run, and the log file is ./test.log during the test runs.

Helpers

As mentioned earlier, nose is included, so eq_ and ok_ are made available to tests. In addition, there is also eqL_, eqF_ and eqcb_. These tests against localization keys and callbacks.

class TestCaseHelpers(plex_unit.TestCase):
    def test_eqL():
        subject = L('some-key')
        eqL_(subject, 'some-key')
        # vs. eq_(subject._key, 'some-key')

    def test_eqF():
        subject = F('formatted-string', 'param')
        eqF_(subject, 'formatted-string')
        # vs. eq_(subject._key.string1._key, 'formatted-string')

    def test_eqcb():
        subject = DirectoryObject(key = Callback(Videos, section = 'all'))
        eqcb_(subject.key, Videos, section = 'all')
        # vs. eq_(subject.key, Callback(Videos, section = 'all'))

Code written in the tests won't have access to file or open, which is necessary to mock data in tests some times. Use plex_nose.publish_local_file in these situations.

class TestPublishLocalFile(plex_nose.TestCase):
    @classmethod
    def setUpClass(cls):
        plex_nose.publish_local_file('Contents/Tests/all_videos.html',
            name = 'all_videos')

    def test():
        import mock
        mocked = HTML.ElementFromString(all_videos)
        @mock.patch.object(HTML, 'ElementFromURL', return_value = mocked)
        def test_inner(*a):
            subject = HTML.ElementFromURL(...)
            # TODO: write actual tests here

Mock is located in Contents/Libraries/Shared, but not required to run the tests.

Guardfile

When using the Guardfile files saved to Contents/Tests will be run automatically. Files saved to Contents/Code will also have the corresponding test run, should one exist.

To get up and running with the Guardfile, all you need is ruby and bundler installed.

Ubuntu

sudo apt-get -y install ruby1.9.1 ruby1.9.1-dev
sudo gem install bundler rake
bundle install --path .bundle/gems/
bundle exec guard --clear -n f

OS X

sudo gem install bundler
bundle install --path .bundle/gems/
bundle exec guard --clear -n f