text-extractor

The text-extractor service extracts plain text from files in different formats. The following formats are supported:

• PDF (*.pdf)
• Word (*.doc, *.docx)
• HTML (*.html, *.htm)

This service can be executed as a single Python script or within a Docker container.

How to build and run

To execute as a Python application:

python src/extractor/wsgi.py

or using gunicorn:

gunicorn --bind 0.0.0.0:5000 --chdir src/extractor wsgi:app

To build or rebuild the Docker image type (the HTTP_PROXY variable must be set if executed behind a proxy):

docker build --build-arg http_proxy=$HTTP_PROXY 
             --build-arg https_proxy=$HTTPS_PROXY
             --build-arg no_proxy=$NO_PROXY -t text-extractor .

To run the container as a service:

docker run --rm -p <port>:5000 --name text-extractor text-extractor

To run the container interactively:

docker run --interactive --entrypoint /bin/bash -t text-extractor

How to test

Check service status locally:

curl http://<host>:5000/v1/health # return service status 'UP'

How to test the extraction endpoint:

curl -F file=@<file>  http://<host>:<port>/api/v1.0/documents/extract