Kartka is a tool for people who hate paper. It provides a workflow for scanning, indexing, and then uploading letters you receive for later retrieval.

Kartka uses (your own) Google Drive for storage, and Sonic for efficient and simple indexing.

Kartka

positional arguments:
  {check,ingest,scan,search,hydrate}
                        The mode to use
    check               check connections
    ingest              ingest a letter
    scan                start a scan 
    search              search for letters
    hydrate             hydrate sonic from drive

optional arguments:
  -h, --help            show this help message and exit
  --config CONFIG       Kartka configuration file location

./kartka.py ingest scan_page_1.png scan_page_2.png

> ./kartka.py search mortgage

2020-12-22 1512  -> https://drive.google.com/file/d/<id>/view?usp=sharing
2020-11-18 1048  -> https://drive.google.com/file/d/<id>/view?usp=sharing
2020-11-12 1142  -> https://drive.google.com/file/d/<id>/view?usp=sharing

There are 3 basic commands:

ingest: Ingest one or more files from the command line, index them, and then upload them.

scan: Begin a 'scan' workflow where you can scan your pages into a configured directory, and then ingest once finished.

search: Search for letters that contain all of the provided keywords. Supports search suggestions via autocomplete.

To use kartka, you will need the following:

• A functioning Sonic instance: https://github.com/valeriansaliou/sonic

• Tesseract OCR: https://github.com/madmaze/pytesseract#installation

• Install the python requirements.txt, via a venv or otherwise: https://pip.pypa.io/en/stable/user_guide/#requirements-files

• A drive-enabled credentials.json: I use https://developers.google.com/drive/api/v3/quickstart/python because I'm lazy.

Then copy kartka.cfg.example to kartka.cfg and configure it with your preferred settings. I personally like to run my Sonic instance on a Raspberry Pi (for convenience), but it can be ran locally as well.

Move the credentials.json to your data_dir (configured in the cfg above).

Now you're ready to go! On first run kartka will run the auth-flow on your new drive-app. After this, it will re-use the stored token.