Setting up Bunny for development

To make sure you have a working version of Bunny, follow the steps in the Quickstart.

If the bunny-daemon logs messages saying that it’s setting up a database connection, looking for and receiving jobs, and resolving queries, then it’s working. To double check, you can also use the bunny-cli to generate an output JSON.

Using the environment

Bunny uses poetry to manage its dependencies. To test any changes made to Bunny in development you can use either

poetry run bunny-[daemon/cli]

or start a poetry shell

poetry shell
bunny-[daemon/cli]

Architecture

The same core components are used by the daemon and command-line interface (CLI). CLI components are shown in blue, daemon components shown in green.

bunny-daemon

The daemon initialises by

• Connecting to the database
• Starting the logger
• Initialising the task API client
• Building the results modifiers
  • Reads the LOW_NUMBER_SUPPRESSION_THRESHOLD and ROUNDING_TARGET from settings
• Setting up the polling endpoint
  • Reads the COLLECTION_ID and TASK_API_TYPE from settings

Polling

The daemon sends GET requests through the task API client until the task API endpoint sends a job. It uses the JSON of the response body to execute a query. If the query executes successfully, it tries to POST the result through the task API client five times. After executing the query, it polls for jobs again.

The interval between polling requests is set through the POLLING_INTERVAL environment variable.

Bunny core

The components used by the daemon and CLI are held in the core library.

bunny-cli

The CLI:

• Connects to the database
• Parses command-line arguments
• Builds the results modifiers
  • Reads the LOW_NUMBER_SUPPRESSION_THRESHOLD and ROUNDING_TARGET from settings

It then executes the query, and writes obfuscated results to a JSON file.