Create a Python virtual environment in your test that cleans up on teardown. The fixture has utility methods to install packages and list what's installed.
Install using your favourite package installer:
pip install pytest-virtualenv
# or
easy_install pytest-virtualenv
Enable the fixture explicitly in your tests or conftest.py (not required when using setuptools entry points):
pytest_plugins = ['pytest_virtualenv']
This fixture is configured using the following evironment variables
Setting | Description | Default |
---|---|---|
VIRTUALENV_FIXTURE_EXECUTABLE | Which virtualenv executable will be used to create new venvs | virtualenv |
Here's a noddy test case to demonstrate the basic fixture attributes.
For more information on path.py
see https://pathpy.readthedocs.io/
def test_virtualenv(virtualenv):
# the 'virtualenv' attribute is a `path.py` object for the root of the virtualenv
dirnames = virtualenv.virtualenv.dirs()
assert {'bin', 'include', 'lib'}.intersection(set(dirnames))
# the 'python' attribute is a `path.py` object for the python executable
assert virtualenv.python.endswith('/bin/python')
You can install packages by name and query what's installed.
def test_installing(virtualenv):
virtualenv.install_package('coverage', installer='pip')
# installed_packages() will return a list of `PackageEntry` objects.
assert 'coverage' in [i.name for i in virtualenv.installed_packages()]
Any packages set up in the test runner's python environment (ie, the same runtime that
py.test
is installed in) as source checkouts using python setup.py develop
will be
detected as such and can be installed by name using install_package
.
By default they are installed into the virtualenv using python setup.py develop
, there
is an option to build and install an egg as well:
def test_installing_source(virtualenv):
# Install a source checkout of my_package as an egg file
virtualenv.install_package('my_package', build_egg=True)
The test fixture has a run
method which allows you to run commands with the correct
paths set up as if you had activated the virtualenv first.
def test_run(virtualenv):
python_exe_path = virtualenv.python
runtime_exe = virtualenv.run("python -c 'import sys; print sys.executable'", capture=True)
assert runtime_exe == python_exe_path
The test fixture has a run_with_coverage
method which is like run
but runs the command
under coverage inside the virtualenv. This is useful for capturing test coverage on
tools that are being tested outside the normal test runner environment.
def test_coverage(virtualenv):
# You will have to install coverage first
virtualenv.install_package(coverage)
virtualenv.run_with_coverage(["my_entry_point", "--arg1", "--arg2"])