This script scans from the ADF (Automatic Document Feeder), preprocesses it and uploads it to Docspell.

• A running Docspell instance.

• Any Linux distro (tested on Ubuntu 20.04)

• Install system requirements (apt command on Ubuntu:)

  sudo apt install sane python3 python3-pip libmagickwand-dev img2pdf sane-utils

• See if scanimage is able to find your scanner

  scanimage -L

  • if scanner is not found -> check sane-find-scanner
    • Maybe also as sudo to see if it's an privileges problem
    • If yes try sudo adduser <username> lp and logout/login and restart scanner

• Clone the repository

  git clone https://github.com/eresturo/scanadf2docspell

• Install requirements

  cd scanadf2docpsell
  pip3 install -r requirements.txt

• Generate a API-Key, as described here

• create a config file custom.conf

  api_key: 'YOUR_API_KEY'
  docspell_url: 'http://YOUR_DOCSPELL_URL'

Unfortunately, scanimage arguments are partially dependent on the scanner. Therefore, a customized configuration per scanner may be necessary. For this purpose, you can specify with which command a flatbed, with which a single-page adf scan and with which a duplex adf scan can be triggered.

Note: This list is far from complete. Which scanner do you use and which configuration is needed? Please tell us in an issue or pull request.

• insert a document in your scanner and run

  ./scan.py

• See help for further options.

  ./scan.py --help

Suggestions, feature requests, ideas to improve the script? Feel free to open an issue or send a pull request :)