crystal_analysis_webservice

This is a webservice built using FastAPI. It exposes the great Julia topology analysis code developed by Lionel Zoubritzky.

We will add more tools, like zeo++ in due course.

The service also uses file-based caching, using the md5 hash of fileContent as key.

Usage

To be usable on Heroku or Dokku, which use PORTenvironmental variables, you need to either create this environmental variable or put it into an .env file. You can start by renaming .env.sample to .env.

Development

For local development, the run_docker.sh script uses 8091.

./build_docker # builds the docker image
./run_docker # starts the service

Deployment

For production, you may want to use Docker compose

docker-compose up --build -d

To test the image, you can go to localhost:$PORT/docs.

You could for example, use the request body from one of the files in the folder .example_requests.

The analysis takes some time (1--2 min) and should result in an object of the following form

{
  "rcsrName": "string",
  "apiVersion": "string",
  "rcsrLink": "string"
}

Settings

For the image to work, you need to specify a few settings. You can do this, for example, via an .env file:

• CONCURRENCY_LIMIT limits the number of concurrent connections before we raise a 503 error.
• WORKERS number of workers, we use on CPU per worker
• PORT will be the port number at which you can reach the app

Docs

You find docs on http://127.0.0.1:$PORT/docs.