K-pop Data API and Dataset Generator
A TypeScript library that scrapes K-pop idol and group information from online sources to create comprehensive JSON datasets. The package provides:
- Data scraping with caching
- Advanced fuzzy search with multi-language support
- TypeScript type definitions
- Built-in dataset access
bun add @castdrian/kdapi# Run scraper in debug mode (5 samples per category)
kdapi scrape --debug
# Run full scraper with caching
kdapi scrape --cache
# Force refresh all profiles
kdapi scrape --force
# Configure batch size and delays
kdapi scrape --batch-size 10 --delay 3000import { search, getItemById } from '@castdrian/kdapi';
const results = search('stayc', { // Supports Korean, Japanese, Chinese characters
type: 'all', // 'idol' | 'group' | 'all'
limit: 10, // Max number of results
threshold: 0.4 // Match quality threshold (0-1)
});
// The search is performed across multiple fields:
// - For idols: stage name, full name, native name, Korean, Japanese, and Chinese names
// - For groups: group names, fandom name, company names, and member names
// Each result includes:
// - item: The matched idol or group
// - type: 'idol' or 'group'
// - Additional metadata about match quality
// Get specific idol/group by ID
const item = getItemById('some-uuid');- Bun
- TypeScript
# Clone the repository
git clone https://github.com/castdrian/kdapi.git
# Install dependencies
bun install
# Run scraper in debug mode
bun run cli.ts scrape --debugHTML responses are cached in:
/cache/idols/- Idol profile pages/cache/groups/- Group profile pages
Data is saved to:
/data/idols.json- Idol data/data/groups.json- Group data/data/metadata.json- Dataset statistics
MIT