dor-py

Python code for the DOR project.

Setup

1. Clone the repo

git clone https://github.com/mlibrary/dor-py.git
cd dor-py

1. In the terminal, run the init.sh

./init.sh

This will:

• set up the initial environment variables file
• build the docker image
• install the python dependencies

./init.sh can be run more than once.

1. Edit .env with actual environment variables

2. In the app container, use poetry shell to enable the virtual environment. Otherwise use:

 docker compose run --rm app poetry run YOUR_COMMAND

Features and Tests

Behavior driven development (BDD) with pytest-bdd

Behave

Test driven development (TDD) with pytest

To run the features located in the 'features' directory and tests located in the tests directory, use the following command:

docker compose run --rm app poetry run pytest

PostgreSQL database

Some operations and tests rely on a PostgreSQL database that can be run using a separate Docker service. Use the following command to start up the database.

docker compose up db

REST API

The application provides a few REST API endpoints for reporting purposes. Use the following command to start up the server. API documentation for the routes will be available at http://0.0.0.0:8000/docs, or using a different port if you set the API_PORT environment variable to a different value.

docker compose up api

Generating Samples

To generate sample packages, on the command line run:

docker compose run --rm app poetry run dor samples generate --collid xyzzy --num-scans 5 --versions 1

This will generate a submission package in BagIt format for the collection xyzzy for one item made up of 5 page scans.

To see all options:

docker compose run --rm app poetry run dor samples generate --help

How does the versions option work? The first version of the item will contain all the scans. The next versions will only contain a random subset of updated scans.

Can you validate the METS2 documents? The schemas for METS2 and PREMIS3 are in the etc/xsd directory. With xmllint --schema you can only use one of these schemas, so validating a METS2 document will always complain about the PREMIS:object type.