Pseudonymization tool

Web-based tool for data de-identification.

Requirements

Windows

Install WSL
Install Windows Terminal (optional, needs Windows 10.0.18362.0 or higher)
Install Python 3.7 (to Windows)
Install IDE (optional)
- VSCode
- PyCharm with WSL interpret (works only in PyCharm Professional)
Install Docker on WSL 2

Dependencies

make
docker and docker-compose
Python 3.7

Pseudonymization tool uses Flask framework and runs in Python's venv. The venv is usually auto created by make, but you can also create it manually using make venv.

Configuration

The app needs PostgreSQL and Redis to work correctly, a connection details has to be provided in env variables:

# Required params
DB_USER=postgres
DB_PASSWORD=postgres
APP_SECRET_KEY=USE_YOUR_SECRET_KEY
# Optional params, not recommended with docker-compose (or make)
DB_HOST=db
DB_NAME=psan_db
CELERY_REDIS=redis://localhost:6379

Runtime

The application expects the .env file in the project root that contains the configuration.

PSAN tool is available on http://localhost:5000/
SQL adminer is available on http://localhost:5050/ (only in debug)

Debug runtime

You can start the application in debug mode (with PostreSQL and Redis) in Docker containers using make docker-debug. This configuration enables debug mode in flask and sets runtime code updates using bind mount.

Tests

The repository has some unit tests in folder tests. These tests could be executed using make docker-test or without docker (in venv) using run_tests.sh in the project root.

Production

You can start the application (with PostreSQL and Redis) in Docker containers using docker-compose up in the project root. You can also start the app without using docker (in venv) using run.sh.

ELITR/pseudonymizer