Quepid Elasticsearch Proxy

Elasticsearch proxy for Quepid.

A lot of Quepid users are surprised by the fact that Quepid works with Elasticsearch directly from the browser. Elasticsearch are rarely available publicly or can be tunneled to the labeler's computer.

Quepid Elasticsearch Proxy can be deployed next to your Quepid and can proxy requests to Elasticsearch. The proxy is protected with basic auth.

Run with Docker

The image of the proxy is publicly available on https://quay.io/amboss-mededu/quepid_es_proxy. Currently the image is not publicly available. Please build it locally.

To run the proxy docker execute

docker run \
-e "PROXY_USERNAME=username_is_here" \
-e "PROXY_PASSWORD=password_is_here" \
-e "ES_HOST=example-elasticsearch-domain.eu-west-1.es.amazonaws.com" \
-e "ES_PORT=443" \
-e "ES_USE_SSL=true" \
-e "WEB_CONCURRENCY=2" \
-p 5000:5000 \
quepid_es_proxy

The proxy is now available with basic auth now on http://username_is_here:password_is_here@localhost:5000/. Use this address in Quepid instead of Elasticsearch.

If you run Elasticsearch locally in Docker, use Docker DNS in ES_HOST. You might also need to specify a Docker network in which Elasticsearch is running.

docker run \
-e "PROXY_USERNAME=username_is_here" \
-e "PROXY_PASSWORD=password_is_here" \
-e "ES_HOST=elasticsearch_docker_name" \
-e "ES_PORT=9200" \
-e "ES_USE_SSL=false" \
-e "WEB_CONCURRENCY=2" \
-p 5000:5000 \
--network="elasticsearch-docker-network" \
quepid_es_proxy

Run locally with Python virtual environment.

Proxy uses Python 3.11.

First prepare a virtual environment make prepare-env. The proxy will be available with the default credentials on http://lab_user:jhHB73bYBKk6G^@localhost:5000/.

To run the proxy locally execute make run-service. Modify PROXY_USERNAME and PROXY_PASSWORD environment variables in the run-server make target if you need.

Development

make doit

To sort imports, format the code in the consistent way and execute flake8 and mypy checks run make doit.

Formatting

The code base is formatted with https://github.com/psf/black.

Imports

Imports are sorted with https://github.com/PyCQA/isort.

Pre-commit hooks.

1. Install pre-commit. On Mac use brew install pre-commit, on Linux pip install pre-commit.
2. Install hooks pre-commit install.