Star systems info extractor for Dyson Sphere Program

This tool will automagically convert a table with information about a celestial body to a plain text table in a Google Document.

Common prerequisites

• AWS account and credentials for use of AWS Textract
• Google Developer app with Google Sheets API enabled and OAuth configured
  • Redirect URL of the app should point to the GET /access endpoint of the configured server
• Dedicated Google Spreadsheet for writing

Docker prerequisites

• Docker (optionally with docker-compose) installed and working

Docker setup

• Download the application code and build the container with docker build -t dsp-textract .
• Create a persistent local directory to be mounted to the container.
  • For docker-compose, this directory must be named config.local/, this name is recommended
• Optionally create a persistent log directory which will be mapped to log directory in the container
  • For docker-compose, this directory is named log
• Copy config/config.local.neon.template from the source code to config.local.neon in the local config directory
• Point sheet_client_config and sheet_token_config parameters in config.local.neon to files in config.local/ directory
• Set up remaining parameters in config.local.neon file
• Start the container
  • Either with docker-compose
    • The application will be exposed on port 8080
  • Or with docker run where you point your local config directory to config.local in the container.
    • The server in the container runs on port 8080, point wherever you like

Manual prerequisites

• A server capable of serving PHP pages, PHP 8, composer

Manual setup

• Install composer dependencies
• Create log, temp/cache directories
• Set up the server to point to the public directory of the app
• Copy config/config.local.neon.template to config/config.local.neon and fill with information
  • sheet_client_config is a path to Google app config json.
    • Example value can be %root_dir%/config/credentials.json
  • Other options should be self-explanatory
  • %root_dir% can be used to point to the root directory of the app

Using the app

• Set up Google Docs access token by visiting the GET /access endpoint on the configured server
• Send your screenshots to the POST /document endpoint of the configured server
  • The endpoint expects a single POST field of name dataurl with Base64-encoded data-url of the image

Screenshots

• It is necessary to provide just a cropped screenshot of the details table, otherwise the result is uncertain
• Very useful tool for this can be CloudShot along with my Generic API Upload Plugin which send exactly the data the app expects

Results

• Each solar system will have a separate sheet in the document
• Each star and planet of the system will take two columns with two empty columns on the right for notes
  • The order of bodies on the sheet is Star/Giant/Black hole and then orbiting bodies I-V
• On subsequent screenshots of the same object table, the data of the object will be overwritten

Example Google Spreadsheet from my current playthrough.

Debugging

• The app is configured for production mode
• To switch the app to a development mode, create a .dev file in the local config directory
• Warning: development mode may leak source code and credentials. Only local use is encouraged. Use lvh.me domain which points to 127.0.0.1 for oauth redirect url and/or the whole app.