Rest-Framework-Smoke

Smoke tests for API built with Django Rest Framework.

Installation

pip install rest-framework-smoke

Usage

Full example located at testproject.testapp.tests

from rest_framework.test import APITestCase

from rest_framework_smoke.tests.mixins import ReadOnlyViewSetTestsMixin
from rest_framework_smoke.tests.schemas import get_object_schema
from testproject.testapp import models

TASK_SCHEMA = {
    "id": {"type": ["number"]},
    "name": {"type": ["string"]},
}

PROJECT_SCHEMA = {
    "id": {"type": ["number"]},
    "name": {"type": ["string"]},
    "task_set": {
        "type": ["array"],
        "minItems": 1,
        "items": get_object_schema(TASK_SCHEMA)
    }
}


class ProjectViewSetTestCase(ReadOnlyViewSetTestsMixin, APITestCase):
    object_name = 'project'
    basename = 'projects'
    schema = details_schema = PROJECT_SCHEMA
    pagination_schema = None

    @classmethod
    def setUpTestData(cls):
        cls.project = models.Project.objects.create(name='project')
        cls.task = models.Task.objects.create(name='task', project=cls.project)

Happy API testing :)

About schema checks

Rest-Framework-Smoke uses jsonschema to validate API response format. When we check format, we should pay attention to:

• no unexpected properties are found (is so, they are not validated by schema)
• there are no missing properties (missing properties are not validated)
• arrays are not empty (because there is nothing to check in empty arrays)
• all values are not null (because null values mostly are null by default, and other type variants will never appear in schema validation code)

So there are two helpers in rest_framework_smoke.tests.schemas to enforce these constraints (and they are used internally for format tests):

• get_object_schema
• get_array_schema