AdMotiv

LauzHack 5.0 hackathon (2020)

Stack

• The backend uses different python libraries, the most relevant ones being:
  • fastapi
  • flair
  • google-api
  • todoist
  • bs4
• The chrome extension is a derivation of the work done by add-art

Install

Backend

The Backend directory contains all the logic related to the Python backend.

First you will need to define your personal tokens for external APIs:

• Todoist: enter your token in todoist_token.txt.
• Google Calendar: enter your token object in credentials.json and start the app: you will be prompted to follow a link and sign in with your Google account. Once that is done, your OAuth token will be saved in the token.pickle for further reuse.

Then you will need to (manually) download the machine learning model that we are using. Download tars-base.pt and place it under resources.

To launch the app you have two options (choose one):

• Docker: simply execute docker-compose up and the container should prepare itself and bootstrap.
• or local environment: create a Python virtual environment (use 3.7, issues were encountered with newer versions), install the requirements with pip install -r requirements.txt, enter the app directory and start the server with uvicorn main:app --reload --host 0.0.0.0 --ssl-keyfile=../secrets/key.pem --ssl-certfile=../secrets/cert.pem

Browser extension

The ChromePlugin is a self contained extension for the chrome/chromium browsers.

Head to chrome://extensions, activate the "Developer mode" and click on "Load unpacked". Select the ChromePlugin directory and your done.