A private microservice to support The Microsetta Initiative
The private microservice depends on Postgres 9.5. For OSX users, we recommend installing Postgres.app. For Linux users, please consult documentation for the Linux distribution in use.
Create a new conda environment containing flask and other necessary packages:
conda create -n microsetta-private-api flask psycopg2 natsort pycryptodome
Once the conda environment is created, activate it:
conda activate microsetta-private-api
Ensure that the conda-forge channel has been added to the conda install and run:
conda install -c conda-forge python-dateutil pycountry coveralls pytest-cov
Install connexion version 2.0 (which supports the OpenAPI Specification 3.0) as well as the Swagger UI:
pip install connexion[swagger-ui]
Install the JSON Web Tokens library with cryptography support:
pip install pyjwt[crypto]
Install Redis and Celery with Redis support for out-of-band compute:
conda install redis
pip install celery[redis]Then install the microsetta-private-api in editable mode:
pip install -e .
In the activated conda environment, initialize a test database:
python microsetta_private_api/LEGACY/build_db.py
Then we'll initiate a Celery worker (Example shown is with embedded Celery Beat scheduler):
celery -A microsetta_private_api.celery_worker.celery worker -B --loglevel=info
*** OR (Allowing for multiple workers)***
celery -A microsetta_private_api.celery_worker.celery worker --loglevel=info (Start 1 worker)
celery -A microsetta_private_api.celery_worker.celery beat --loglevel=info (Start the Celery Beat scheduler)
Next, start the microservice using flask's built-in server by running, e.g.,
python ./microsetta_private_api/server.py
which will start the server on http://localhost:8082 . Note that this usage is suitable for development ONLY--real use of the service would require a production-level server.
The Swagger UI should also be available at http://localhost:8082/api/ui .