Technical details

Pdf article

Backend

Requirements

  • Maintenance costs (including server and hosting fees) should be zero.
  • The quote database itself be independent of any particular UI system.
  • It should be easy to contribute to the quote database - but such contributions should be verifiable.
  • Data format should be easily parsable in a variety of machine languages.

Current solution

  • Quotes are stored as simple markdown files with TOML metadata, so as to capture annotations such as meter, topic, poetic ornamentation, rasas, bhAva-s, sources, secondary sources, etc..
  • Within each markdown file, the text and various commentaries or presentations are stored under html details tags (which contain relevant content in markdown format).
  • For location of databases, see contribution page.

Setting up dev environment

Code locus: TW

git clone --recurse-submodules https://github.com/subhAShita/subhaashita_py.git
pip install -e .

UI

  • Report issues other than the below on github.

Setting up dev environment

git clone --recurse-submodules https://github.com/subhAShita/subhAShita.github.io.git
cd subhAShita.github.io.git
hugo server --renderToDisk

With this, you should be able to browse the site and navigate to saMskRtam/padyam/random on the browser, which corresponds to saMskRtam/padyam/random.md.
JS Code mainly resides in saMskRtam/padyam/quote-helper.js.

If you’re editing the theme javascript, you should do this in a separate terminal window -

cd subhAShita.github.io.git/themes/sanskrit-documentation-theme-hugo/webpack_src
npm install
npm run watch

TODOs

  • UI
    • प्रतिमाला - अनुष्टुभ्-अपाकरणे विकल्पः। अक्षरालाभेऽक्षरान्तरग्रहणम्।
    • Access keys/ shortcuts
    • TODOs in quote-helper.js.
    • Better randomization.
  • backend
    • Dump subhAShita pages for indexing.
  • AI to annotate with topics and translations (backend).
  • AI translations - dharmamitra, bing etc..