Repository containing code to extract recipes from printed recipe cards. Repository contains the following:

• kook/ - code to extract recipes from printed recipes cards

  • reads scans directly from Apps/arbakker-kook directory
  • requires kook/.env file with ACCESS_TOKEN set

  Run with: python3 kook/main.py

• webapp/ - webapp code

  Run with:

  npm install # Install dependencies
  npm run serve # Compiles and hot-reloads for development
  npm run build # Compiles and minifies for production
  npm run lint # Lints and fixes files

• scripts/ - Bash utility scripts; for copying over recipes and scans

  • clear-processed-recipes.sh - remove processed recipe index and recipe json files from webapp, so next run of kook/main.py reprocesses all recipes

The kook/ code base requires installation of:

tesseract 4.1.1
 leptonica-1.79.0
...

• add fuzzy search title/subtitle to frontpage
• add home button from recipe page
• sort steps on order title
• replace comma step title with nr.
• order recipes on title alpahbetically
• add copy to clipboard on ingredients list
• require password before accessing page
• read scans directly from dropbox
• process only unprocessed recipes/scans
• generate index specifically for search
• run python processing from Docker image
  • run python processing with GH Actions
    • Redeploy webapp after running gh action processing
      • GH Action processing after files is uploaded to dropbox folder
• add check to verify if recipes are converted correctly, skip failing recipes
• add keywords/tags/to recipes
• remove allcaps phrases like VREE from steps