Skip to content

Make authenticated requests for data to the Hakai API with Python

License

Notifications You must be signed in to change notification settings

HakaiInstitute/hakai-api-client-py

Repository files navigation

Hakai Api Python Client

This project exports a single Python class that can be used to make HTTP requests to the Hakai API resource server. The exported Client class extends the functionality of the Python requests library to supply Hakai OAuth2 credentials with url requests.

PyPI tests License: MIT

Table of Contents

Installation

Quickstart

Methods

API endpoints

Advanced usage

Contributing

Installation

Python 3.8 or higher is required. Install with pip:

pip install hakai-api

Quickstart

from hakai_api import Client

# Get the api request client
client = Client()  # Follow stdout prompts to get an API token

# Make a data request for chlorophyll data
url = '%s/%s' % (client.api_root, 'eims/views/output/chlorophyll?limit=50')
response = client.get(url)

print(url)  # https://hecate.hakai.org/api/eims/views/output/chlorophyll...
print(response.json())
# [{'action': '', 'event_pk': 7064, 'rn': '1', 'date': '2012-05-17', 'work_area': 'CALVERT'...

Methods

This library exports a single client name Client. Instantiating this class produces a requests.Session client from the Python requests library. The Hakai API Python Client inherits directly from requests.Session thus all methods available on that parent class are available. For details see the requests documentation.

The hakai_api Client class also contains a property api_root which is useful for constructing urls to access data from the API. The above Quickstart example demonstrates using this property to construct a url to access project names.

API endpoints

For details about the API, including available endpoints where data can be requested from, see the Hakai API documentation.

Advanced usage

You can specify which API to access when instantiating the Client. By default, the API uses https://hecate.hakai.org/api as the API root. It may be useful to use this library to access a locally running API instance or to access the Goose API for testing purposes. If you are always going to be accessing data from a locally running API instance, you are better off using the requests.py library directly since Authorization is not required for local requests.

from hakai_api import Client

# Get a client for a locally running API instance
client = Client("http://localhost:8666")
print(client.api_root)  # http://localhost:8666

You can also pass in the credentials string retrieved from the hakai API login page while initiating the Client class.

from hakai_api import Client

# Pass a credentials token as the Client Class is initiated
client = Client(credentials="CREDENTIAL_TOKEN")

Finally, you can set credentials for the client class using the HAKAI_API_CREDENTIALS environment variable. This is useful for e.g. setting credentials in a docker container. The value of the environment variable should be the credentials token retrieved from the Hakai API login page.

Contributing

See CONTRIBUTING

License

See LICENSE