Writing plugins

It is easy to implement local conftest plugins for your own project or pip-installable plugins that can be used throughout many projects, including third party projects. Please refer to Installing and Using plugins if you only want to use but not write plugins.

A plugin contains one or multiple hook functions. Writing hooks explains the basics and details of how you can write a hook function yourself. pytest implements all aspects of configuration, collection, running and reporting by calling well specified hooks of the following plugins:

• Pytest default plugin reference: loaded from pytest’s internal _pytest directory.
• external plugins: modules discovered through setuptools entry points
• conftest.py plugins: modules auto-discovered in test directories

In principle, each hook call is a 1:N Python function call where N is the number of registered implementation functions for a given specification. All specifications and implementations following the pytest_ prefix naming convention, making them easy to distinguish and find.

Plugin discovery order at tool startup

pytest loads plugin modules at tool startup in the following way:

• by loading all builtin plugins

• by loading all plugins registered through setuptools entry points.

• by pre-scanning the command line for the -p name option and loading the specified plugin before actual command line parsing.

• by loading all conftest.py files as inferred by the command line invocation:

  • if no test paths are specified use current dir as a test path
  • if exists, load conftest.py and test*/conftest.py relative to the directory part of the first test path.

  Note that pytest does not find conftest.py files in deeper nested sub directories at tool startup. It is usually a good idea to keep your conftest.py file in the top level test or project root directory.

• by recursively loading all plugins specified by the pytest_plugins variable in conftest.py files

conftest.py: local per-directory plugins

Local conftest.py plugins contain directory-specific hook implementations. Hook Session and test running activities will invoke all hooks defined in conftest.py files closer to the root of the filesystem. Example of implementing the pytest_runtest_setup hook so that is called for tests in the a sub directory but not for other directories:

a/conftest.py:
    def pytest_runtest_setup(item):
        # called for running each test in 'a' directory
        print ("setting up", item)

a/test_sub.py:
    def test_sub():
        pass

test_flat.py:
    def test_flat():
        pass

Here is how you might run it:

py.test test_flat.py   # will not show "setting up"
py.test a/test_sub.py  # will show "setting up"

Note

If you have conftest.py files which do not reside in a python package directory (i.e. one containing an __init__.py) then “import conftest” can be ambiguous because there might be other conftest.py files as well on your PYTHONPATH or sys.path. It is thus good practice for projects to either put conftest.py under a package scope or to never import anything from a conftest.py file.

Writing your own plugin

If you want to write a plugin, there are many real-life examples you can copy from:

• a custom collection example plugin: A basic example for specifying tests in Yaml files
• around 20 doc:builtin plugins which provide pytest’s own functionality
• many external plugins providing additional features

All of these plugins implement the documented well specified hooks to extend and add functionality.

Note

Make sure to check out the excellent cookiecutter-pytest-plugin project, which is a cookiecutter template for authoring plugins.

The template provides an excellent starting point with a working plugin, tests running with tox, comprehensive README and entry-pointy already pre-configured.

Also consider contributing your plugin to pytest-dev once it has some happy users other than yourself.

Making your plugin installable by others

If you want to make your plugin externally available, you may define a so-called entry point for your distribution so that pytest finds your plugin module. Entry points are a feature that is provided by setuptools. pytest looks up the pytest11 entrypoint to discover its plugins and you can thus make your plugin available by defining it in your setuptools-invocation:

# sample ./setup.py file
from setuptools import setup

setup(
    name="myproject",
    packages = ['myproject']

    # the following makes a plugin available to pytest
    entry_points = {
        'pytest11': [
            'name_of_plugin = myproject.pluginmodule',
        ]
    },
)

If a package is installed this way, pytest will load myproject.pluginmodule as a plugin which can define well specified hooks.

Requiring/Loading plugins in a test module or conftest file

You can require plugins in a test module or a conftest file like this:

pytest_plugins = "name1", "name2",

When the test module or conftest plugin is loaded the specified plugins will be loaded as well. You can also use dotted path like this:

pytest_plugins = "myapp.testsupport.myplugin"

which will import the specified module as a pytest plugin.

Accessing another plugin by name

If a plugin wants to collaborate with code from another plugin it can obtain a reference through the plugin manager like this:

plugin = config.pluginmanager.getplugin("name_of_plugin")

If you want to look at the names of existing plugins, use the --traceconfig option.

Testing plugins

pytest comes with some facilities that you can enable for testing your plugin. Given that you have an installed plugin you can enable the testdir fixture via specifying a command line option to include the pytester plugin (-p pytester) or by putting pytest_plugins = "pytester" into your test or conftest.py file. You then will have a testdir fixture which you can use like this:

# content of test_myplugin.py

pytest_plugins = "pytester"  # to get testdir fixture

def test_myplugin(testdir):
    testdir.makepyfile("""
        def test_example():
            pass
    """)
    result = testdir.runpytest("--verbose")
    result.fnmatch_lines("""
        test_example*
    """)

Note that by default testdir.runpytest() will perform a pytest in-process. You can pass the command line option --runpytest=subprocess to have it happen in a subprocess.

Also see the RunResult for more methods of the result object that you get from a call to runpytest.

Writing hook functions

hook function validation and execution

pytest calls hook functions from registered plugins for any given hook specification. Let’s look at a typical hook function for the pytest_collection_modifyitems(session, config, items) hook which pytest calls after collection of all test items is completed.

When we implement a pytest_collection_modifyitems function in our plugin pytest will during registration verify that you use argument names which match the specification and bail out if not.

Let’s look at a possible implementation:

def pytest_collection_modifyitems(config, items):
    # called after collection is completed
    # you can modify the ``items`` list

Here, pytest will pass in config (the pytest config object) and items (the list of collected test items) but will not pass in the session argument because we didn’t list it in the function signature. This dynamic “pruning” of arguments allows pytest to be “future-compatible”: we can introduce new hook named parameters without breaking the signatures of existing hook implementations. It is one of the reasons for the general long-lived compatibility of pytest plugins.

Note that hook functions other than pytest_runtest_* are not allowed to raise exceptions. Doing so will break the pytest run.

firstresult: stop at first non-None result

Most calls to pytest hooks result in a list of results which contains all non-None results of the called hook functions.

Some hook specifications use the firstresult=True option so that the hook call only executes until the first of N registered functions returns a non-None result which is then taken as result of the overall hook call. The remaining hook functions will not be called in this case.

hookwrapper: executing around other hooks

New in version 2.7: (experimental)

pytest plugins can implement hook wrappers which wrap the execution of other hook implementations. A hook wrapper is a generator function which yields exactly once. When pytest invokes hooks it first executes hook wrappers and passes the same arguments as to the regular hooks.

At the yield point of the hook wrapper pytest will execute the next hook implementations and return their result to the yield point in the form of a CallOutcome instance which encapsulates a result or exception info. The yield point itself will thus typically not raise exceptions (unless there are bugs).

Here is an example definition of a hook wrapper:

import pytest

@pytest.hookimpl(hookwrapper=True)
def pytest_pyfunc_call(pyfuncitem):
    # do whatever you want before the next hook executes

    outcome = yield
    # outcome.excinfo may be None or a (cls, val, tb) tuple

    res = outcome.get_result()  # will raise if outcome was exception
    # postprocess result

Note that hook wrappers don’t return results themselves, they merely perform tracing or other side effects around the actual hook implementations. If the result of the underlying hook is a mutable object, they may modify that result but it’s probably better to avoid it.

Hook function ordering / call example

For any given hook specification there may be more than one implementation and we thus generally view hook execution as a 1:N function call where N is the number of registered functions. There are ways to influence if a hook implementation comes before or after others, i.e. the position in the N-sized list of functions:

# Plugin 1
@pytest.hookimpl(tryfirst=True)
def pytest_collection_modifyitems(items):
    # will execute as early as possible

# Plugin 2
@pytest.hookimpl(trylast=True)
def pytest_collection_modifyitems(items):
    # will execute as late as possible

# Plugin 3
@pytest.hookimpl(hookwrapper=True)
def pytest_collection_modifyitems(items):
    # will execute even before the tryfirst one above!
    outcome = yield
    # will execute after all non-hookwrappers executed

Here is the order of execution:

1. Plugin3’s pytest_collection_modifyitems called until the yield point because it is a hook wrapper.
2. Plugin1’s pytest_collection_modifyitems is called because it is marked with tryfirst=True.
3. Plugin2’s pytest_collection_modifyitems is called because it is marked with trylast=True (but even without this mark it would come after Plugin1).
4. Plugin3’s pytest_collection_modifyitems then executing the code after the yield point. The yield receives a CallOutcome instance which encapsulates the result from calling the non-wrappers. Wrappers shall not modify the result.

It’s possible to use tryfirst and trylast also in conjunction with hookwrapper=True in which case it will influence the ordering of hookwrappers among each other.

Declaring new hooks

Plugins and conftest.py files may declare new hooks that can then be implemented by other plugins in order to alter behaviour or interact with the new plugin:

pytest_addhooks(pluginmanager)

called at plugin registration time to allow adding new hooks via a call to pluginmanager.add_hookspecs(module_or_class, prefix).

Hooks are usually declared as do-nothing functions that contain only documentation describing when the hook will be called and what return values are expected.

For an example, see newhooks.py from xdist: pytest distributed testing plugin.

Optionally using hooks from 3rd party plugins

Using new hooks from plugins as explained above might be a little tricky because of the standard validation mechanism: if you depend on a plugin that is not installed, validation will fail and the error message will not make much sense to your users.

One approach is to defer the hook implementation to a new plugin instead of declaring the hook functions directly in your plugin module, for example:

# contents of myplugin.py

class DeferPlugin(object):
    """Simple plugin to defer pytest-xdist hook functions."""

    def pytest_testnodedown(self, node, error):
        """standard xdist hook function.
        """

def pytest_configure(config):
    if config.pluginmanager.hasplugin('xdist'):
        config.pluginmanager.register(DeferPlugin())

This has the added benefit of allowing you to conditionally install hooks depending on which plugins are installed.

pytest hook reference

Initialization, command line and configuration hooks

pytest_load_initial_conftests(args, early_config, parser)

implements the loading of initial conftest files ahead of command line option parsing.

pytest_cmdline_preparse(config, args)

(deprecated) modify command line arguments before option parsing.

pytest_cmdline_parse(pluginmanager, args)

return initialized config object, parsing the specified args.

pytest_namespace()

return dict of name->object to be made globally available in the pytest namespace. This hook is called at plugin registration time.

pytest_addoption(parser)

register argparse-style options and ini-style config values.

Warning

This function must be implemented in a plugin and is called once at the beginning of a test run.

Implementing this hook from conftest.py files is strongly discouraged because conftest.py files are lazily loaded and may give strange unknown option errors depending on the directory py.test is invoked from.

Parameters:parser – To add command line options, call parser.addoption(...). To add ini-file values call parser.addini(...).

Options can later be accessed through the config object, respectively:

• config.getoption(name) to retrieve the value of a command line option.
• config.getini(name) to retrieve a value read from an ini-style file.

The config object is passed around on many internal objects via the .config attribute or can be retrieved as the pytestconfig fixture or accessed via (deprecated) pytest.config.

pytest_cmdline_main(config)

called for performing the main command line action. The default implementation will invoke the configure hooks and runtest_mainloop.

pytest_configure(config)

called after command line options have been parsed and all plugins and initial conftest files been loaded. This hook is called for every plugin.

pytest_unconfigure(config)

called before test process is exited.

Generic “runtest” hooks

All runtest related hooks receive a pytest.Item object.

pytest_runtest_protocol(item, nextitem)

implements the runtest_setup/call/teardown protocol for the given test item, including capturing exceptions and calling reporting hooks.

Parameters:

• item – test item for which the runtest protocol is performed.
• nextitem – the scheduled-to-be-next test item (or None if this is the end my friend). This argument is passed on to pytest_runtest_teardown().

Return boolean:

True if no further hook implementations should be invoked.

pytest_runtest_setup(item)

called before pytest_runtest_call(item).

pytest_runtest_call(item)

called to execute the test item.

pytest_runtest_teardown(item, nextitem)

called after pytest_runtest_call.

Parameters:nextitem – the scheduled-to-be-next test item (None if no further test item is scheduled). This argument can be used to perform exact teardowns, i.e. calling just enough finalizers so that nextitem only needs to call setup-functions.

pytest_runtest_makereport(item, call)

return a _pytest.runner.TestReport object for the given pytest.Item and _pytest.runner.CallInfo.

For deeper understanding you may look at the default implementation of these hooks in _pytest.runner and maybe also in _pytest.pdb which interacts with _pytest.capture and its input/output capturing in order to immediately drop into interactive debugging when a test failure occurs.

The _pytest.terminal reported specifically uses the reporting hook to print information about a test run.

Collection hooks

pytest calls the following hooks for collecting files and directories:

pytest_ignore_collect(path, config)

return True to prevent considering this path for collection. This hook is consulted for all files and directories prior to calling more specific hooks.

pytest_collect_directory(path, parent)

called before traversing a directory for collection files.

pytest_collect_file(path, parent)

return collection Node or None for the given path. Any new node needs to have the specified parent as a parent.

For influencing the collection of objects in Python modules you can use the following hook:

pytest_pycollect_makeitem(collector, name, obj)

return custom item/collector for a python object in a module, or None.

pytest_generate_tests(metafunc)

generate (multiple) parametrized calls to a test function.

After collection is complete, you can modify the order of items, delete or otherwise amend the test items:

pytest_collection_modifyitems(session, config, items)

called after collection has been performed, may filter or re-order the items in-place.

Reporting hooks

Session related reporting hooks:

pytest_collectstart(collector)

collector starts collecting.

pytest_itemcollected(item)

we just collected a test item.

pytest_collectreport(report)

collector finished collecting.

pytest_deselected(items)

called for test items deselected by keyword.

pytest_report_header(config, startdir)

return a string to be displayed as header info for terminal reporting.

pytest_report_teststatus(report)

return result-category, shortletter and verbose word for reporting.

pytest_terminal_summary(terminalreporter)

add additional section in terminal summary reporting.

And here is the central hook for reporting about test execution:

pytest_runtest_logreport(report)

process a test setup/call/teardown report relating to the respective phase of executing a test.

You can also use this hook to customize assertion representation for some types:

pytest_assertrepr_compare(config, op, left, right)

return explanation for comparisons in failing assert expressions.

Return None for no custom explanation, otherwise return a list of strings. The strings will be joined by newlines but any newlines in a string will be escaped. Note that all but the first line will be indented sligthly, the intention is for the first line to be a summary.

Debugging/Interaction hooks

There are few hooks which can be used for special reporting or interaction with exceptions:

pytest_internalerror(excrepr, excinfo)

called for internal errors.

pytest_keyboard_interrupt(excinfo)

called for keyboard interrupt.

pytest_exception_interact(node, call, report)

called when an exception was raised which can potentially be interactively handled.

This hook is only called if an exception was raised that is not an internal exception like skip.Exception.

pytest_enter_pdb()

called upon pdb.set_trace(), can be used by plugins to take special action just before the python debugger enters in interactive mode.

Reference of objects involved in hooks