Experimental features

Introduction

This section provides an overview of experimental features in Schemathesis - features that are under development and available for testing, but not yet considered stable.

This section provides an overview of experimental features in Schemathesis. These are features that are under development, available for user testing but not yet stable. Experimental features offer a glimpse into upcoming functionalities and enhancements, providing users an opportunity to try out and contribute feedback to shape their final form.

Note

Experimental features can change or be removed in any minor version release.

Enabling Experimental Features

Schemathesis provides a few ways to enable experimental features: via the CLI, Python tests, and environment variables.

In CLI

To enable an experimental feature via the CLI, use the --experimental option.

For example, to enable experimental support for OpenAPI 3.1:

st run https://example.schemathesis.io/openapi.json --experimental=openapi-3.1

In Python Tests

To enable experimental features for your Schemathesis tests in Python, use the enable method on the desired experimental feature. It’s a good idea to put the code to enable the feature in a place that runs before your tests start. For those using pytest, putting this code in a conftest.py file at the root of your test directory would work well.

import schemathesis

# Globally enable OpenAPI 3.1 experimental feature
schemathesis.experimental.OPEN_API_3_1.enable()

Executing the above code will ensure that Schemathesis will utilize the enabled experimental features in all tests run afterwards.

Note

This action will apply globally, affecting all your tests. Use with caution.

Using Environment Variables

You can also enable experimental features through environment variables. This is particularly useful for CI/CD pipelines or when you don’t have direct control over CLI arguments or test code.

For example, to enable experimental support for OpenAPI 3.1:

export SCHEMATHESIS_EXPERIMENTAL_OPENAPI_3_1=true

This will enable the OpenAPI 3.1 experimental feature for any Schemathesis runs in the same environment.

Current Experimental Features

Open API 3.1

Provides partial support for OpenAPI 3.1. This includes compatible JSON Schema validation for API responses. Note that data generation is still compatible only with OpenAPI 3.0.

Enabling this feature also automatically enables UUID format support.

In CLI

st run https://example.schemathesis.io/openapi.json --experimental=openapi-3.1

In Python Tests

import schemathesis

# Globally enable OpenAPI 3.1 experimental feature
schemathesis.experimental.OPEN_API_3_1.enable()

Using Environment Variables

To enable OpenAPI 3.1 support via environment variables, you can set:

export SCHEMATHESIS_EXPERIMENTAL_OPENAPI_3_1=true

For more details, join the GitHub Discussion #1822.

Schema Analysis

Enables integration with Schemathesis.io to optimize test case generation for better bug detection. This feature infers appropriate data generation strategies, supports uncommon media types, adjusts schemas for faster data generation, and detects the web server to generate more targeted test data.

In CLI

st run https://example.schemathesis.io/openapi.json --experimental=schema-analysis

Using Environment Variables

To enable schema analysis support via environment variables, you can set:

export SCHEMATHESIS_EXPERIMENTAL_SCHEMA_ANALYSIS=true

For more details, join the GitHub Discussion #2056.

Stabilization of Experimental Features

Criteria for moving a feature from experimental to stable status include:

  • Full coverage of planned functionality

  • API design stability, assessed through user feedback and internal review

Providing Feedback

Feedback is crucial for the development and stabilization of experimental features. We encourage you to share your thoughts via GitHub Discussions

Note

When you use an experimental feature, a notice will appear in your test output, providing a link to the corresponding GitHub discussion where you can leave feedback.