A catalogue service with CRUD operations.

Needs Python >= 3.8.

Install requirements and package:

pip install -r requirements.txt
python setup.py install

The application is configured using environment variables:

• SECRET_KEY (optional): a randomly-generated secret used for encryption
• DATABASE_URL: the connection URL for PostGIS (>=3.0) backend (postgresql://host:port/database)
• DATABASE_USERNAME: the username for database connection
• DATABASE_PASSWORD_FILE: the file containing the password for database connection
• SQLALCHEMY_TRACK_MODIFICATIONS: A True/False flag
• SERVER_NAME (optional): the server name (host:port) under which the service is accessed.
• SWAGGER_UI_DOC_EXPANSION: one of none, list or full
• RESTX_VALIDATE: a True/False flag
• RESTX_MASK_SWAGGER: a True/False flag
• ERROR_404_HELP: a True/False flag
• FLASK_DEBUG: a True/False flag (should be False in a production environment!)

For convenience, these variables can be kept in an enviroment-like file, say config-development.py or config-testing.py. Look at the example at config.py.example.

To initialize the database schema (this only needs SQLALCHEMY_DATABASE_URI variable to be set):

./generate-db-schema.py    

Run a development server (environment variables must be set in current shell):

./wsgi.py

If environment variables are kept in a seperate file, say config-development.py, run by pointing to that file:

env FILE_CONFIG=config-development.py ./wsgi.py

To run with Docker we must prepare a docker-compose recipe.

Copy .env.example into .env. Edit as needed.

The additional environment variables here (i.e. not described in section 2):

• FLASK_ENV: One of production or development
• VERSION: The semantic version of the application (used to tag the Docker image)
• DATABASE_INITIALIZE_SCHEMA: If True, will initialize database schema before starting the WSGI server

Copy docker-compose.yml.example into docker-compose.yml. Edit as needed. You will at least need to configure the network (inside docker-compose.yml) to attach to.

For example, you can create a private network named opertusmundi_network:

docker network create --attachable opertusmundi_network

Build the image:

docker-compose build

Initialize database schema (if not initialized before):

docker-compose run --rm catalogueapi python -c 'import catalogueapi; catalogueapi.generate_db_schema()'

Run:

docker-compose up -d

The API documentation is using the OpenAPI standard. You can brownse the documentation here.

Install the additionaly required packages for testing ('nose' etc.):

pip install -r requirements-testing.txt

Prepare a environment file, say config-testing.py, analogous to the one used for running the application (start by copying the example config.py.example).

Generate database schema for testing database (if not already):

env FILE_CONFIG=config-testing.py ./generate-db-schema.py

Run nose tests:

env FILE_CONFIG=config-testing.py nosetests