Bielemetrics Kinexon API Wrapper

Introduction

Welcome to the Kinexon REST API Authentication and Usage Example for Python.

Due to the lack of available documentation on the Kinexon API (i'm not mad, just disappointed), I am providing this open source script to assist you.

This package provides a wrapper for the Kinexon API, including functions for authentication and data retrieval. It simplifies interacting with the Kinexon REST API, making it easier to authenticate and retrieve data.

Overview

Currently a WIP. If you find this repo useful, consider leaving a star, or not. Your choice. I am not affiliated with Kinexon.

Used in the soon-to-be-opensource™ framework BieLeMetrics, which is a Python project for automated multimodal data processing applied in the paper Expected Goals Prediction in Professional Handball using Synchronized Event and Positional Data.

This example demonstrates a comprehensive approach for interacting with the Kinexon REST API. It includes a two-step authentication process, session management with HTTP cookies, and API key handling for secure data access from the Kinexon Cloud.

Key Features

Two-Step Authentication: First, authenticate using HTTP Basic Authentication, followed by a session-based login.
Session Management: Utilizes HTTP cookies to maintain session state across multiple API requests.
API Key Handling: Demonstrates secure usage of API keys for accessing protected endpoints.
Retrieval Examples: Shows how to fetch and download positional information and further data.

Installation

Clone the repository:

git clone https://github.com/mad4ms/kinexon-api-access-python.git

You can install the package using pip:

cd kinexon-api-access-python
pip install .

Usage

Preparation of Credentials & Environment Variables

Set the following environment variables with your credentials and API information:

export USERNAME_KINEXON_SESSION='your_session_username' # Username for the popup on the website
export PASSWORD_KINEXON_SESSION='your_session_password' # Password for the popup on the website
export ENDPOINT_KINEXON_SESSION='https://your_session_endpoint' # Endpoint for the session login

export USERNAME_KINEXON_MAIN='your_main_username' # Username for the main login site
export PASSWORD_KINEXON_MAIN='your_main_password' # Password for the main login site
export ENDPOINT_KINEXON_MAIN='https://your_main_endpoint' # Endpoint for the main login

export API_KEY_KINEXON='your_api_key' # API key created in profile settings
export ENDPOINT_KINEXON_API='https://your_api_endpoint' # Base URL for the Kinexon API

Explanation:

USERNAME_KINEXON_SESSION and PASSWORD_KINEXON_SESSION: Credentials for initial authentication using HTTP Basic Authentication. ENDPOINT_KINEXON_SESSION is usually https://hbl-cloud.kinexon.com/
USERNAME_KINEXON_MAIN and PASSWORD_KINEXON_MAIN: Credentials for the main site authentication using a POST request payload. ENDPOINT_KINEXON_MAIN is usually https://hbl-cloud.kinexon.com/checklogin/
KINEXON_API_KEY: API key created in the user profile under the Teams tab. Ensure to confirm your password again to save your profile and store the API_KEY. ENDPOINT_KINEXON_API is usually https://hbl-cloud.kinexon.com/public/v1/

Hack for faster auth:

Create a .env file in the repo with credentials (will be ignored by .gitignore):

USERNAME_KINEXON_SESSION='your_session_username'
PASSWORD_KINEXON_SESSION='your_session_password'
...

python-dotenv package will detect .env file and use it. No need for cumbersome environment entries.

Authentication

First, load your credentials and authenticate with the Kinexon API.

import requests
from bielemetrics_kinexon_api_wrapper import login

# Load credentials from environment variables
credentials = load_credentials()

# Create a session by logging in
session_kinexon = login(credentials)

You should see a successful login message.

Fetching Data

You can fetch different types of data using the provided functions.

Fetch Team IDs (fetch_data.py)

from bielemetrics_kinexon_api_wrapper import get_available_metrics_and_events

metrics_and_events = get_available_metrics_and_events(session)
...
print(metrics_and_events["metrics"])

Troubleshooting

Failed to Login: Double-check your credentials. Ensure they match those provided by the Kinexon Cloud and are correctly set to the environment.
Connection Errors: Ensure you have an active internet connection.
API Changes: Be aware that the Kinexon API is young and prone to changes.

Contributing

Feel free to contribute to this project. You're welcome to fork, submit PRs, or open issues.

License

This project is licensed under the MIT License.