pook PyPI Coverage Status Documentation Status Stability Code Climate Python Versions

Versatile, expressive and hackable utility library for HTTP traffic mocking and expectations made easy in Python. Heavily inspired by gock.

To get started, read the documentation, how it works, FAQ or examples.


  • Simple, expressive and fluent API.

  • Provides both Pythonic and chainable DSL API styles.

  • Full-featured HTTP response definitions and expectations.

  • Matches any HTTP protocol primitive (URL, method, query params, headers, body…).

  • Full regular expressions capable mock expectations matching.

  • Supports most popular HTTP clients via interceptor adapters.

  • Configurable volatile, persistent or TTL limited mocks.

  • Works with unittest and pytest.

  • First-class JSON & XML support matching and responses.

  • Supports JSON Schema body matching.

  • Works in both runtime and testing environments.

  • Can be used as decorator and/or via context managers.

  • Supports real networking mode with optional traffic filtering.

  • Map/filter mocks easily for generic or custom mock expectations.

  • Custom user-defined mock matcher functions.

  • Simulated raised error exceptions.

  • Network delay simulation (only available for aiohttp).

  • Pluggable and hackable API.

  • Customizable HTTP traffic mock interceptor engine.

  • Supports third-party mocking engines, such as mocket.

  • Fits good for painless test doubles.

  • Does not support WebSocket traffic mocking.

  • Works with +3.8 (including PyPy).

  • Dependency-less: just 3 small dependencies for JSONSchema, XML tree comparison, and URL parsing.

Supported HTTP clients

pook can work with multiple mock engines, however it provides a built-in one by default, which currently supports traffic mocking in the following HTTP clients:

More HTTP clients can be supported progressively.

Note: only recent HTTP client package versions were tested.


Using pip package manager (requires pip 1.8+):

pip install --upgrade pook

Or install the latest sources from Github:

pip install -e git+git://

Getting started

See ReadTheDocs documentation:

Documentation Status


See annotated API reference documention.


See examples documentation for full featured code and use case examples.

Basic mocking:

import pook
import requests

def test_my_api():
    mock = pook.get('', reply=404, response_json={'error': 'not found'})

    resp = requests.get('')
    assert resp.status_code == 404
    assert resp.json() == {"error": "not found"}
    assert mock.calls == 1

Using the chainable API DSL:

import pook
import requests

def test_my_api():
    mock = (pook.get('')
              .json({'error': 'not found'}))

    resp = requests.get('')
    assert resp.json() == {"error": "not found"}
    assert mock.calls == 1

Using the decorator:

import pook
import requests

@pook.get('', reply=204)
@pook.get('', reply=200)
def fetch(url):
    return requests.get(url)

res = fetch('')
print('#1 status:', res.status_code)

res = fetch('')
print('#2 status:', res.status_code)

Simple unittest integration:

import pook
import unittest
import requests

class TestUnitTestEngine(unittest.TestCase):

    def test_request(self):
        res = requests.get('')
        self.assertEqual(res.status_code, 204)

    def test_request_with_context_manager(self):
        with pook.use():
            pook.get('', reply=204)
            res = requests.get('')
            self.assertEqual(res.status_code, 204)

Using the context manager for isolated HTTP traffic interception blocks:

import pook
import requests

# Enable HTTP traffic interceptor
with pook.use():
    pook.get('', reply=204)

    res = requests.get('')
    print('#1 status:', res.status_code)

# Interception-free HTTP traffic
res = requests.get('')
print('#2 status:', res.status_code)

Example using mocket Python library as underlying mock engine:

import pook
import requests
from mocket.plugins.pook_mock_engine import MocketEngine

# Use mocket library as underlying mock engine

# Explicitly enable pook HTTP mocking (optional)

# Target server URL to mock out
url = ''

# Define your mock
mock = pook.get(url,
                reply=404, times=2,
                headers={'content-type': 'application/json'},
                response_json={'error': 'foo'})

# Run first HTTP request
assert mock.calls == 1

# Run second HTTP request
res = requests.get(url)
assert mock.calls == 2

# Assert response data
assert res.status_code == 404
assert res.json() == {'error': 'foo'}

# Explicitly disable pook (optional)

Example using Hy language (Lisp dialect for Python):

(import [pook])
(import [requests])

(defn request [url &optional [status 404]]
  (doto (.mock pook url) (.reply status))
  (let [res (.get requests url)]
    (. res status_code)))

(defn run []
  (with [(.use pook)]
    (print "Status:" (request "" :status 204))))

;; Run test program
(defmain [&args] (run))


Clone the repository:

git clone

Use [hatch]( to configure the environment by running the test suite:

hatch run test

Install the pre-commit hook:

hatch run lint:install

Lint the code:

hatch run lint:run

Run tests on all supported Python versions and implementations (this requires your host operating system to have each implementation available):

hatch run test:test

To run tests only for a specific version, affix the version designation to the environment name (the left side of the :):

hatch run test.pypy3.10:test

Generate documentation:

hatch run docs:build


MIT - Tomas Aparicio

Indices and tables