How tcstenungsund/schedule works
Project type and purpose
• This is a static, full-stack-in-the-browser scheduling application published via GitHub Pages.
• It uses TypeScript-based web components (built with Rollup), plain HTML/CSS, and markdown files (schedules and assignments) as its data source.
Main components of the system
a. Build pipeline
• TypeScript sources (components/src)
• Rollup configuration (components/rollup.config.mjs)
• Output bundle (components/dist/index.js + .d.ts types)
b. Hosting / Delivery
• GitHub Pages (serving the static site)
c. Client runtime (Browser)
• HTML shell pages load CSS and the JS bundle
• JS bundle instantiates web components:
• <schedule-preview> (loads schedule markdown)
• <assignment> (loads individual assignment markdown)
• <navbar> (navigation UI)
• md-converter (converts markdown to HTML at runtime)
• APIs (schedule-api.ts, assignment-api.ts) that fetch raw .md via HTTP
d. Data layer
• Markdown files under md/ (templates, assignments, schedules)
• CSS under css/
• Static assets under resources/
Relationships & interactions
• Build: components/src (.ts) → Rollup → components/dist (index.js)
• Deployment: dist/ + HTML/CSS/md/resources → GitHub Pages
• At runtime: Browser requests an HTML page → CSS + dist/index.js + markdown assets → JS APIs fetch .md files over HTTP → md-converter parses content → web components render schedules/assignments
Architectural patterns & design principles
• Component-based UI (web components in TypeScript)
• Static-site-as-application: no custom backend; markdown is the “API”
• Separation of concerns:
• Build/configuration (Rollup + tsconfig.json)
• Presentation (HTML + CSS)
• Data/model (markdown + lightweight JS APIs)