Dataformer makes text processing simple through an interactive graphical user interface.

Dataformer currently uses poetry (install instructions) for Python packaging and dependency management and npm for managing the frontend. After installing these tools (and their dependencies), run

poetry install
cd frontend
npm install

in the repository root to setup the project. Poetry automatically sets up a virtual environment for Python, which can be activated by running

poetry shell

New Python dependencies can be added by running

poetry add <dependency>
poetry export -f requirements.txt --output requirements.txt

The second step exports the standard requirements.txt file, which can be used to setup the project in environments not supporting poetry.

New node dependencies can be added in a standard way by running

npm install <dependency>

You can either run the server and the frontend separately (probably most convenient for development) or together using a Docker container. In order to run the server, make sure that the virtual environment for the project is activated (i.e. run poetry shell if needed) and run

flask run

in the repository root. The server handles the API requests and is also set up to serve the latest built version of the React app (located in /frontend/build after running npm run build in /frontend). However, the build of the app will not refresh dynamically when making changes, so it might be more convenient to run

npm start

in /frontend in order to start the frontend development server.

If you want to run the application in a Docker container, you can run

docker build -t dataformer .
docker run -p 8000:8000 dataformer

in the repository root. This will build and start the application on localhost:8000.

You can format the Python code using Black by running

black .

You can format the React code using Prettier by running

npx prettier --write .

in the /frontend directory. If you are using Visual Studio Code, both Black and Prettier can be set up to automatically format the code on saving files. Alternately, the formatters can be configured as pre-commit Git hooks.