Useful additions to Django's default TestCase

Related tags

Testingtestingdjango
Overview

django-test-plus

Useful additions to Django's default TestCase from REVSYS

pypi build matrix demo

Rationale

Let's face it, writing tests isn't always fun. Part of the reason for that is all of the boilerplate you end up writing. django-test-plus is an attempt to cut down on some of that when writing Django tests. We guarantee it will increase the time before you get carpal tunnel by at least 3 weeks!

Support

Supports: Python 3.6, 3.7, 3.8, 3.9, and 3.10.

Supports Django Versions: 2.0, 2.1, 2.2, 3.0, 3.1. and 3.2.

Documentation

Full documentation is available at http://django-test-plus.readthedocs.org

Installation

$ pip install django-test-plus

Usage

To use django-test-plus, have your tests inherit from test_plus.test.TestCase rather than the normal django.test.TestCase::

from test_plus.test import TestCase

class MyViewTests(TestCase):
    ...

This is sufficient to get things rolling, but you are encouraged to create your own sub-classes for your projects. This will allow you to add your own project-specific helper methods.

For example, if you have a django project named 'myproject', you might create the following in myproject/test.py:

from test_plus.test import TestCase as PlusTestCase

class TestCase(PlusTestCase):
    pass

And then in your tests use:

from myproject.test import TestCase

class MyViewTests(TestCase):
    ...

This import, which is similar to the way you would import Django's TestCase, is also valid:

from test_plus import TestCase

pytest Usage

You can get a TestCase like object as a pytest fixture now by asking for tp. All of the methods below would then work in pytest functions. For example:

def test_url_reverse(tp):
    expected_url = '/api/'
    reversed_url = tp.reverse('api')
    assert expected_url == reversed_url

The tp_api fixture will provide a TestCase that uses django-rest-framework's APIClient():

def test_url_reverse(tp_api):
    response = tp_api.client.post("myapi", format="json")
    assert response.status_code == 200

Methods

reverse(url_name, *args, **kwargs)

When testing views you often find yourself needing to reverse the URL's name. With django-test-plus there is no need for the from django.core.urlresolvers import reverse boilerplate. Instead, use:

def test_something(self):
    url = self.reverse('my-url-name')
    slug_url = self.reverse('name-takes-a-slug', slug='my-slug')
    pk_url = self.reverse('name-takes-a-pk', pk=12)

As you can see our reverse also passes along any args or kwargs you need to pass in.

get(url_name, follow=True, *args, **kwargs)

Another thing you do often is HTTP get urls. Our get() method assumes you are passing in a named URL with any args or kwargs necessary to reverse the url_name. If needed, place kwargs for TestClient.get() in an 'extra' dictionary.:

def test_get_named_url(self):
    response = self.get('my-url-name')
    # Get XML data via AJAX request
    xml_response = self.get(
        'my-url-name',
        extra={'HTTP_X_REQUESTED_WITH': 'XMLHttpRequest'})

When using this get method two other things happen for you: we store the last response in self.last_response and the response's Context in self.context.

So instead of:

def test_default_django(self):
    response = self.client.get(reverse('my-url-name'))
    self.assertTrue('foo' in response.context)
    self.assertEqual(response.context['foo'], 12)

You can write:

def test_testplus_get(self):
    self.get('my-url-name')
    self.assertInContext('foo')
    self.assertEqual(self.context['foo'], 12)

It's also smart about already reversed URLs, so you can be lazy and do:

def test_testplus_get(self):
    url = self.reverse('my-url-name')
    self.get(url)
    self.response_200()

If you need to pass query string parameters to your url name, you can do so like this. Assuming the name 'search' maps to '/search/' then:

def test_testplus_get_query(self):
    self.get('search', data={'query': 'testing'})

Would GET /search/?query=testing.

post(url_name, data, follow=True, *args, **kwargs)

Our post() method takes a named URL, an optional dictionary of data you wish to post and any args or kwargs necessary to reverse the url_name. If needed, place kwargs for TestClient.post() in an 'extra' dictionary.:

def test_post_named_url(self):
    response = self.post('my-url-name', data={'coolness-factor': 11.0},
                         extra={'HTTP_X_REQUESTED_WITH': 'XMLHttpRequest'})

NOTE Along with the frequently used get and post, we support all of the HTTP verbs such as put, patch, head, trace, options, and delete in the same fashion.

get_context(key)

Often you need to get things out of the template context:

def test_context_data(self):
    self.get('my-view-with-some-context')
    slug = self.get_context('slug')

assertInContext(key)

You can ensure a specific key exists in the last response's context by using:

def test_in_context(self):
    self.get('my-view-with-some-context')
    self.assertInContext('some-key')

assertContext(key, value)

We can get context values and ensure they exist, but we can also test equality while we're at it. This asserts that key == value:

def test_in_context(self):
    self.get('my-view-with-some-context')
    self.assertContext('some-key', 'expected value')

assert_http_###_<status_name>(response, msg=None) - status code checking

Another test you often need to do is check that a response has a certain HTTP status code. With Django's default TestCase you would write:

from django.core.urlresolvers import reverse

def test_status(self):
    response = self.client.get(reverse('my-url-name'))
    self.assertEqual(response.status_code, 200)

With django-test-plus you can shorten that to be:

def test_better_status(self):
    response = self.get('my-url-name')
    self.assert_http_200_ok(response)

Django-test-plus provides a majority of the status codes assertions for you. The status assertions can be found in their own mixin and should be searchable if you're using an IDE like pycharm. It should be noted that in previous versions, django-test-plus had assertion methods in the pattern of response_###(), which are still available but have since been deprecated. See below for a list of those methods.

Each of the assertion methods takes an optional Django test client response and a string msg argument that, if specified, is used as the error message when a failure occurs. The methods, assert_http_301_moved_permanently and assert_http_302_found also take an optional url argument that if passed, will check to make sure the response.url matches.

If it's available, the assert_http_###_<status_name> methods will use the last response. So you can do:

def test_status(self):
    self.get('my-url-name')
    self.assert_http_200_ok()

Which is a bit shorter.

The response_###() methods that are deprecated, but still available for use, include:

  • response_200()
  • response_201()
  • response_204()
  • response_301()
  • response_302()
  • response_400()
  • response_401()
  • response_403()
  • response_404()
  • response_405()
  • response_409()
  • response_410()

All of which take an optional Django test client response and a str msg argument that, if specified, is used as the error message when a failure occurs. Just like the assert_http_###_<status_name>() methods, these methods will use the last response if it's available.

get_check_200(url_name, *args, **kwargs)

GETing and checking views return status 200 is a common test. This method makes it more convenient::

def test_even_better_status(self):
    response = self.get_check_200('my-url-name')

make_user(username='testuser', password='password', perms=None)

When testing out views you often need to create various users to ensure all of your logic is safe and sound. To make this process easier, this method will create a user for you:

def test_user_stuff(self)
    user1 = self.make_user('u1')
    user2 = self.make_user('u2')

If creating a User in your project is more complicated, say for example you removed the username field from the default Django Auth model, you can provide a Factory Boy factory to create it or override this method on your own sub-class.

To use a Factory Boy factory, create your class like this::

from test_plus.test import TestCase
from .factories import UserFactory


class MySpecialTest(TestCase):
    user_factory = UserFactory

    def test_special_creation(self):
        user1 = self.make_user('u1')

NOTE: Users created by this method will have their password set to the string 'password' by default, in order to ease testing. If you need a specific password, override the password parameter.

You can also pass in user permissions by passing in a string of '<app_name>.<perm name>' or '<app_name>.*'. For example:

user2 = self.make_user(perms=['myapp.create_widget', 'otherapp.*'])

print_form_errors(response_or_form=None)

When debugging a failing test for a view with a form, this method helps you quickly look at any form errors.

Example usage:

class MyFormTest(TestCase):

    self.post('my-url-name', data={})
    self.print_form_errors()

    # or

    resp = self.post('my-url-name', data={})
    self.print_form_errors(resp)

    # or

    form = MyForm(data={})
    self.print_form_errors(form)

Authentication Helpers

assertLoginRequired(url_name, *args, **kwargs)

This method helps you test that a given named URL requires authorization:

def test_auth(self):
    self.assertLoginRequired('my-restricted-url')
    self.assertLoginRequired('my-restricted-object', pk=12)
    self.assertLoginRequired('my-restricted-object', slug='something')

login context

Along with ensuring a view requires login and creating users, the next thing you end up doing is logging in as various users to test your restriction logic:

def test_restrictions(self):
    user1 = self.make_user('u1')
    user2 = self.make_user('u2')

    self.assertLoginRequired('my-protected-view')

    with self.login(username=user1.username, password='password'):
        response = self.get('my-protected-view')
        # Test user1 sees what they should be seeing

    with self.login(username=user2.username, password='password'):
        response = self.get('my-protected-view')
        # Test user2 see what they should be seeing

Since we're likely creating our users using make_user() from above, the login context assumes the password is 'password' unless specified otherwise. Therefore you you can do:

def test_restrictions(self):
    user1 = self.make_user('u1')

    with self.login(username=user1.username):
        response = self.get('my-protected-view')

We can also derive the username if we're using make_user() so we can shorten that up even further like this:

def test_restrictions(self):
    user1 = self.make_user('u1')

    with self.login(user1):
        response = self.get('my-protected-view')

Ensuring low query counts

assertNumQueriesLessThan(number) - context

Django provides assertNumQueries which is great when your code generates a specific number of queries. However, if this number varies due to the nature of your data, with this method you can still test to ensure the code doesn't start producing a ton more queries than you expect:

def test_something_out(self):

    with self.assertNumQueriesLessThan(7):
        self.get('some-view-with-6-queries')

assertGoodView(url_name, *args, **kwargs)

This method does a few things for you. It:

  • Retrieves the name URL
  • Ensures the view does not generate more than 50 queries
  • Ensures the response has status code 200
  • Returns the response

Often a wide, sweeping test like this is better than no test at all. You can use it like this:

def test_better_than_nothing(self):
    response = self.assertGoodView('my-url-name')

Testing DRF views

To take advantage of the convenience of DRF's test client, you can create a subclass of TestCase and set the client_class property:

from test_plus import TestCase
from rest_framework.test import APIClient


class APITestCase(TestCase):
    client_class = APIClient

For convenience, test_plus ships with APITestCase, which does just that:

from test_plus import APITestCase


class MyAPITestCase(APITestCase):

    def test_post(self):
        data = {'testing': {'prop': 'value'}}
        self.post('view-json', data=data, extra={'format': 'json'})
        self.assert_http_200_ok()

Note that using APITestCase requires Django >= 1.8 and having installed django-rest-framework.

Testing class-based "generic" views

The TestCase methods get() and post() work for both function-based and class-based views. However, in doing so they invoke Django's URL resolution, middleware, template processing, and decorator systems. For integration testing this is desirable, as you want to ensure your URLs resolve properly, view permissions are enforced, etc. For unit testing this is costly because all these Django request/response systems are invoked in addition to your method, and they typically do not affect the end result.

Class-based views (derived from Django's generic.models.View class) contain methods and mixins which makes granular unit testing (more) feasible. Quite often your usage of a generic view class comprises an override of an existing method. Invoking the entire view and the Django request/response stack is a waste of time when you really want to call the overridden method directly and test the result.

CBVTestCase to the rescue!

As with TestCase above, have your tests inherit from test_plus.test.CBVTestCase rather than TestCase like so:

from test_plus.test import CBVTestCase

class MyViewTests(CBVTestCase):

Methods

get_instance(cls, initkwargs=None, request=None, *args, **kwargs)

This core method simplifies the instantiation of your class, giving you a way to invoke class methods directly.

Returns an instance of cls, initialized with initkwargs. Sets request, args, and kwargs attributes on the class instance. args and kwargs are the same values you would pass to reverse().

Sample usage:

from django.views import generic
from test_plus.test import CBVTestCase

class MyClass(generic.DetailView)

    def get_context_data(self, **kwargs):
        kwargs['answer'] = 42
        return kwargs

class MyTests(CBVTestCase):

    def test_context_data(self):
        my_view = self.get_instance(MyClass, {'object': some_object})
        context = my_view.get_context_data()
        self.assertEqual(context['answer'], 42)

get(cls, initkwargs=None, *args, **kwargs)

Invokes cls.get() and returns the response, rendering template if possible. Builds on the CBVTestCase.get_instance() foundation.

All test_plus.test.TestCase methods are valid, so the following works:

response = self.get(MyClass)
self.assertContext('my_key', expected_value)

All test_plus TestCase side-effects are honored and all test_plus TestCase assertion methods work with CBVTestCase.get().

NOTE: This method bypasses Django's middleware, and therefore context variables created by middleware are not available. If this affects your template/context testing, you should use TestCase instead of CBVTestCase.

post(cls, data=None, initkwargs=None, *args, **kwargs)

Invokes cls.post() and returns the response, rendering template if possible. Builds on the CBVTestCase.get_instance() foundation.

Example:

response = self.post(MyClass, data={'search_term': 'revsys'})
self.response_200(response)
self.assertContext('company_name', 'RevSys')

All test_plus TestCase side-effects are honored and all test_plus TestCase assertion methods work with CBVTestCase.post().

NOTE: This method bypasses Django's middleware, and therefore context variables created by middleware are not available. If this affects your template/context testing you should use TestCase instead of CBVTestCase.

get_check_200(cls, initkwargs=None, *args, **kwargs)

Works just like TestCase.get_check_200(). Caller must provide a view class instead of a URL name or path parameter.

All test_plus TestCase side-effects are honored and all test_plus TestCase assertion methods work with CBVTestCase.post().

assertGoodView(cls, initkwargs=None, *args, **kwargs)

Works just like TestCase.assertGoodView(). Caller must provide a view class instead of a URL name or path parameter.

All test_plus TestCase side-effects are honored and all test_plus TestCase assertion methods work with CBVTestCase.post().

Development

To work on django-test-plus itself, clone this repository and run the following commands:

$ pip install -r requirements.txt
$ pip install -e .

NOTE: You will also need to ensure that the test_project directory, located at the root of this repo, is in your virtualenv's path.

Keep in touch!

If you have a question about this project, please open a GitHub issue. If you love us and want to keep track of our goings-on, here's where you can find us online:

Owner
REVSYS
Django, Python, ReactJS, PostgreSQL, Kubernetes and Ops. We offer technical support and consultation for complex open source systems.
REVSYS
FaceBot is a script to automatically create a facebook account using the selenium and chromedriver modules.

FaceBot is a script to automatically create a facebook account using the selenium and chromedriver modules. That way, we don't need to input full name, email and password and date of birth. All will

Fadjrir Herlambang 2 Jun 17, 2022
Whatsapp messages bulk sender using Python Selenium.

Whatsapp Sender Whatsapp Sender automates sending of messages via Whatsapp Web. The tool allows you to send whatsapp messages in bulk. This program re

Yap Yee Qiang 3 Jan 23, 2022
0hh1 solver for the web (selenium) and also for mobile (adb)

0hh1 - Solver Aims to solve the '0hh1 puzzle' for all the sizes (4x4, 6x6, 8x8, 10x10 12x12). for both the web version (using selenium) and on android

Adwaith Rajesh 1 Nov 05, 2021
FakeDataGen is a Full Valid Fake Data Generator.

FakeDataGen is a Full Valid Fake Data Generator. This tool helps you to create fake accounts (in Spanish format) with fully valid data. Within this in

Joel GM 64 Dec 12, 2022
Pytest support for asyncio.

pytest-asyncio: pytest support for asyncio pytest-asyncio is an Apache2 licensed library, written in Python, for testing asyncio code with pytest. asy

pytest-dev 1.1k Jan 02, 2023
AllPairs is an open source test combinations generator written in Python

AllPairs is an open source test combinations generator written in Python

Robson Agapito Correa 5 Mar 05, 2022
a plugin for py.test that changes the default look and feel of py.test (e.g. progressbar, show tests that fail instantly)

pytest-sugar pytest-sugar is a plugin for pytest that shows failures and errors instantly and shows a progress bar. Requirements You will need the fol

Teemu 963 Dec 28, 2022
Main purpose of this project is to provide the service to automate the API testing process

PPTester project Main purpose of this project is to provide the service to automate the API testing process. In order to deploy this service use you s

4 Dec 16, 2021
Network automation lab using nornir, scrapli, and containerlab with Arista EOS

nornir-scrapli-eos-lab Network automation lab using nornir, scrapli, and containerlab with Arista EOS. Objectives Deploy base configs to 4xArista devi

Vireak Ouk 13 Jul 07, 2022
A web scraping using Selenium Webdriver

Savee - Images Downloader Project using Selenium Webdriver to download images from someone's profile on https:www.savee.it website. Usage The project

Caio Eduardo Lobo 1 Dec 17, 2021
This file will contain a series of Python functions that use the Selenium library to search for elements in a web page while logging everything into a file

element_search with Selenium (Now With docstrings 😎 ) Just to mention, I'm a beginner to all this, so it it's very possible to make some mistakes The

2 Aug 12, 2021
Fail tests that take too long to run

GitHub | PyPI | Issues pytest-fail-slow is a pytest plugin for making tests fail that take too long to run. It adds a --fail-slow DURATION command-lin

John T. Wodder II 4 Nov 27, 2022
pytest plugin that let you automate actions and assertions with test metrics reporting executing plain YAML files

pytest-play pytest-play is a codeless, generic, pluggable and extensible automation tool, not necessarily test automation only, based on the fantastic

pytest-dev 67 Dec 01, 2022
Python dilinin Selenium kütüphanesini kullanarak; Amazon, LinkedIn ve ÇiçekSepeti üzerinde test işlemleri yaptığımız bir case study reposudur.

Python dilinin Selenium kütüphanesini kullanarak; Amazon, LinkedIn ve ÇiçekSepeti üzerinde test işlemleri yaptığımız bir case study reposudur. LinkedI

Furkan Gulsen 8 Nov 01, 2022
MongoDB panel for the Flask Debug Toolbar

Flask Debug Toolbar MongoDB Panel Info: An extension panel for Rob Hudson's Django Debug Toolbar that adds MongoDB debugging information Author: Harry

Cenk Altı 4 Dec 11, 2019
UX Analytics & A/B Testing

UX Analytics & A/B Testing

Marvin EDORH 1 Sep 07, 2021
This project demonstrates selenium's ability to extract files from a website.

This project demonstrates selenium's ability to extract files from a website. I've added the challenge of connecting over TOR. This package also includes a personal archive site built in NodeJS and A

2 Jan 16, 2022
Bayesian A/B testing

bayesian_testing is a small package for a quick evaluation of A/B (or A/B/C/...) tests using Bayesian approach.

Matus Baniar 35 Dec 15, 2022
GitHub action for AppSweep Mobile Application Security Testing

GitHub action for AppSweep can be used to continuously integrate app scanning using AppSweep into your Android app build process

Guardsquare 14 Oct 06, 2022
Using openpyxl in Python, performed following task

Python-Automation-with-openpyxl Using openpyxl in Python, performed following tasks on an Excel Sheet containing Product Suppliers along with their pr

1 Apr 06, 2022