PATo API

Source code

https://github.com/DiamondLightSource/pato-backend

Particle Analysis and Tomography Data API.

Configuration

The API supports a configuration file, that follows the example set in config.json, but most importantly, two environment variables need to be set:

• SQL_DATABASE_URL: The URL for the database
• CONFIG_PATH: Path for the configuration file

Additionally, the following environment variables can be set to enable message queue support:

• MQ_USER: Message queue user
• MQ_PASS: Message queue password

As for the JSON configuration file, details are as follows:

• auth

  • endpoint: URL for the chosen auth method
  • read_all_perms: Permission groups that should be granted full read permissions
  • type: Authentication type. Can be micro or dummy
  • beamline_mapping: A map of permission groups and the beamlines that permission group should be allowed to viewing

• mq

  • host: Message queue host
  • port: Mesasge queue port
  • queue: Queue name
  • vhost: Message queue virtual host
  • consumer_queue: Queue to consume messages from (for notifications)

• ispyb

  • pool: Connection pool size
  • overflow: Connection pool overflow max size

• facility

  • contact_email: Contact email to be used across the application
  • smtp_server: SMTP server host
  • smtp_port: SMTP port to be used for emailing reports
  • active_session_cutoff: Time, in weeks, to be used as a threshold for determining if a session is active or not, following the end of the first processing pipeline. For example, a session would be considered inactive if there were no new actions for the past 5 weeks, by default.
  • users_only_on_industrial: Hide industrial session details from staff, only display it to users/staff directly listed as part of that visit

Deployment

Running development server on your machine:

1. Install the package with pip install . or pip install -e .
2. Set the SQL_DATABASE_URL environment variable according to your database's location
3. Run uvicorn with uvicorn pato.main:app --reload --port 8000 --host 0.0.0.0

Optional dependencies:

You might want to test out RabbitMQ integration. In order to do that, you need to point to a valid MQ instance. You can run one locally with podman run -d -e RABBITMQ_DEFAULT_USER=guest -e RABBITMQ_DEFAULT_PASS=guest -p 5672:5672 rabbitmq:3

Testing

• Build the mock ISPyB database (in database, or from the built Docker image)

• Run with podman run -p 3306:3306 --detach --name diamond-ispyb localhost/diamond-ispyb

  • You may change the port or where the container itself runs, just remember to update .test.env

• Run tox -e pytest