This project contains code and configuration to deploy a text extraction service for PDFs in a kubernetes cluster.
Note: nix is used to make building the Python applications and container images reproducible.
Steps to run the service:
nix develop(or use direnv)just workloadkubectl port-forward svc/api 8888:http
For example, you can use one of the commands below to test the service:
curl -F "file=@<path-to-file>" localhost:8888/queue-extraction-jobcurl -F "file=@$HOME/$(cd; nix run 'nixpkgs#fzf')" localhost:8888/queue-extraction-job