engpass.radar — Swiss Drug Shortage Tracker

A modern dashboard that tracks current drug shortages in Switzerland. Data is scraped daily from drugshortage.ch and displayed with KPI cards, full-text search, status/company filters, sortable table, and a detail drawer.

---

Features

• Daily scrape — cheerio-based parser pulls the full ASP.NET table from drugshortage.ch
• KPI cards — active shortage count, top manufacturer, ATC group coverage, average days since report
• Search — real-time full-text search across drug name, company, and ATC code
• Filters — filter by status level (1–5) and manufacturer
• Sortable table — sort by any column; URL-driven so links are shareable
• Detail drawer — click any row to see GTIN, pharmacode, delivery date, mutation history, and source link
• Cron endpoint — POST /api/scrape secured with a Bearer token, wired to Vercel Cron

---

Tech Stack

Layer	Technology
Framework	Next.js 14 (App Router, Server Components)
Language	TypeScript 5
Styling	Tailwind CSS + shadcn/ui
Scraping	cheerio (HTML), SheetJS (XLSX), fast-xml-parser (XML)
ORM	Prisma 7 with PrismaPg Driver Adapter
Database	Supabase PostgreSQL
Testing	Jest + ts-jest
Deployment	Vercel

---

Getting Started

Prerequisites

• Node.js 18+
• npm 9+

Installation

git clone https://github.com/brainbytes-dev/engpassradar.git
cd drugshortage
npm install

Environment Variables

Create a .env.local file in the project root (see .env.example):

DATABASE_URL=postgresql://...          # Pooled connection (Supavisor port 6543)
DIRECT_URL=postgresql://...            # Direct connection (port 5432) for migrations
CRON_SECRET=your-secret-here

CRON_SECRET protects the POST /api/scrape endpoint. Use a long random string in production.

Run Locally

# Fetch current shortage data
npm run scrape

# Start development server
npm run dev

Open http://localhost:3000.

---

Scripts

Command	Description
npm run dev	Start Next.js development server
npm run build	Production build
npm run start	Start production server
npm run lint	Run ESLint
npm test	Run Jest test suite
npm run scrape	Fetch and store current shortages

---

Project Structure

src/
├── app/
│   ├── api/
│   │   ├── scrape/route.ts           # POST — cron trigger (Bearer auth)
│   │   └── shortages/route.ts        # GET  — search, filter, paginate
│   ├── firma/[slug]/page.tsx         # Firma-Profil (ISR)
│   ├── medikament/[slug]/page.tsx    # Medikament-Detail (ISR)
│   ├── methodik/page.tsx             # Methodik & Datenquellen
│   ├── layout.tsx
│   └── page.tsx                      # Dashboard (Server Component)
├── components/
│   ├── filter-bar.tsx                # Status + company dropdowns
│   ├── kpi-cards.tsx                 # 4-card KPI grid
│   ├── search-bar.tsx                # Full-text search input
│   ├── shortage-drawer.tsx           # Detail slide-out
│   ├── shortages-table.tsx           # Sortable, paginated table
│   └── status-badge.tsx              # Color-coded status badge
├── lib/
│   ├── db.ts                         # Prisma data layer
│   ├── scraper.ts                    # cheerio HTML parser + off-market parsers
│   ├── slug.ts                       # URL slug helper
│   └── types.ts                      # TypeScript interfaces
└── scripts/
    └── scrape.ts                     # CLI scrape script
prisma/
└── schema.prisma                     # DB schema (7 models, all indexes)
tests/
├── api/shortages.test.ts
└── lib/
    ├── db.test.ts
    └── scraper.test.ts

---

Database Schema

Data is persisted in Supabase PostgreSQL via Prisma ORM (prisma/schema.prisma).

Table	Content
shortages	Active & historical shortages (isActive flag), with slug for URL routing
overview_stats	Aggregated stats per scrape run
scrape_runs	Scrape execution log
alternatives_cache	Cached alternatives from drugshortage.ch (GTIN key)
bwl_shortages	Federal supply disruptions from bwl.admin.ch (GTIN join)
oddb_products	128k CH medications: ingredient, Swissmedic-Nr, composition
off_market_drugs	Off-market drugs: AUSSER_HANDEL (permanently withdrawn) and VERTRIEBSEINSTELLUNG (sales discontinued)

All indexes are defined in prisma/schema.prisma — never in loose SQL files.

Data Layer

src/lib/db.ts is the single data access layer:

• upsertShortages(incoming) — batch upsert with slug generation, soft-deletes missing entries (isActive: false)
• queryShortages(query) — filters, sorts, and paginates active records (status 1–5)
• getKPIStats() — aggregates KPI metrics
• getFirmaList() — unique sorted manufacturer list for the filter dropdown
• upsertOffMarketDrugs(entries) — upserts off_market_drugs rows by (gtin, category)
• queryOffMarketDrugs(query) — paginated query with category filter
• getOffMarketStats() — counts by category

---

Cron Job (Vercel)

vercel.json configures a daily scrape at 03:00 UTC:

{
  "crons": [
    {
      "path": "/api/scrape",
      "schedule": "0 3 * * *"
    }
  ]
}

Set CRON_SECRET in your Vercel environment variables to match your .env.local value.

---

Deployment

1. Push to GitHub
2. Import the repo at vercel.com/new
3. Add CRON_SECRET as an environment variable
4. Deploy — Vercel runs the daily cron automatically

---

Contributing

See CONTRIBUTING.md.

---

License

MIT — © 2026 BrainBytes