Need help with pytest-watch?
Click the “chat” button below for chat support from the developer who created it, or find similar developers for support.

About the developer

562 Stars 48 Forks MIT License 183 Commits 40 Opened issues


Local continuous test runner with pytest and watchdog.

Services available


Need anything else?

Contributors list

pytest-watch -- Continuous pytest runner

Current version on PyPI Say Thanks!

pytest-watch a zero-config CLI tool that runs pytest, and re-runs it when a file in your project changes. It beeps on failures and can run arbitrary commands on each passing and failing test run.


Whether or not you use the test-driven development method, running tests continuously is far more productive than waiting until you're finished programming to test your code. Additionally, manually running

each time you want to see if any tests were broken has more wait-time and cognitive overhead than merely listening for a notification. This could be a crucial difference when debugging a complex problem or on a tight deadline.


$ pip install pytest-watch


$ cd myproject
$ ptw
 * Watching /path/to/myproject

Note: It can also be run using its full name


Now develop normally and check the terminal every now and then to see if any tests are broken. Alternatively, pytest-watch can notify you when tests pass or fail:

  • OSX

$ ptw --onpass "say passed" --onfail "say failed"
  $ ptw --onpass "growlnotify -m \"All tests passed!\"" \
        --onfail "growlnotify -m \"Tests failed\""

using GrowlNotify.

  • Windows
  > ptw --onfail flash

using Console Flash

You can also run a command before the tests run, e.g. seeding your test database:

$ ptw --beforerun

Or after they finish, e.g. deleting a sqlite file. Note that this script receives the exit code of

as an argument.
$ ptw --afterrun

You can also use a custom runner script for full

$ ptw --runner "python"

Here's an minimal runner script that runs

and prints its exit code:

import sys import pytest

print('pytest exited with code:', pytest.main(sys.argv[1:]))

Need to exclude directories from being observed or collected for tests?

$ ptw --ignore ./deep-directory --ignore ./integration_tests

See the full list of options:

$ ptw --help
Usage: ptw [options] [--ignore ...] [...] [-- ...]

Options: --ignore

Ignore directory from being watched and during collection (multi-allowed). --ext Comma-separated list of file extensions that can trigger a new test run when changed (default: .py). Use --ext=* to allow any file (including .pyc). --config Load configuration from file instead of trying to locate one of the implicit configuration files. -c --clear Clear the screen before each run. -n --nobeep Do not beep on failure. -w --wait Waits for all tests to complete before re-running. Otherwise, tests are interrupted on filesystem events. --beforerun Run arbitrary command before tests are run. --afterrun Run arbitrary command on completion or interruption. The exit code of "pytest" is passed as an argument. --onpass Run arbitrary command on pass. --onfail Run arbitrary command on failure. --onexit Run arbitrary command when exiting pytest-watch. --runner Run a custom command instead of "pytest". --pdb Start the interactive Python debugger on errors. This also enables --wait to prevent pdb interruption. --spool Re-run after a delay (in milliseconds), allowing for more file system events to queue up (default: 200 ms). -p --poll Use polling instead of OS events (useful in VMs). -v --verbose Increase verbosity of the output. -q --quiet Decrease verbosity of the output (precedence over -v). -V --version Print version and exit. -h --help Print help and exit.


CLI options can be added to a

section in your pytest.ini file to persist them in your project. For example:
# pytest.ini

[pytest] addopts = --maxfail=2

[pytest-watch] ignore = ./integration-tests nobeep = True


  • xdist offers the
    ) option (and distributed testing options). This instead re-runs only those tests which have failed until you make them pass. This can be a speed advantage when trying to get all tests passing, but leaves out the discovery of new failures until then. It also drops the colors outputted by pytest, whereas pytest-watch doesn't.
  • Nosey is the original codebase this was forked from. Nosey runs nose instead of pytest.


  1. Check the open issues or open a new issue to start a discussion around your feature idea or the bug you found
  2. Fork the repository, make your changes, and add yourself to
  3. Send a pull request

If you want to edit the README, be sure to make your changes to
and run the following to regenerate the
$ pandoc -t rst -o README.rst

If your PR has been waiting a while, feel free to ping me on Twitter.

Use this software often? Say Thanks! :smiley:

We use cookies. If you continue to browse the site, you agree to the use of cookies. For more information on our use of cookies please see our Privacy Policy.