This project consists of a webscraper to grab artists/albums/tracks/lyrics, as well as a frontend to access that data.

The database is pre-populated with over 1000 artists and 45000 tracks with lyrics, and the frontend allows you to search by:

• Artist name
• Album title
• Track title
• Word or phrase, searched against the lyrics

Requires Python 3.8+ (Python 3.10.1 used) Requires Docker v18 or higher (docker compose up -d)

• Creates Postgres and pgAdmin containers for app usage

Requires Itty3 module for Python (pip install itty3) Requires pg8000 module for Python (pip install pg8000)

python3 main.py

Open a browser to http://localhost:4000

python3 scraper.py

• Resolve accenting - websites simply drop the accents rather than accepting URI encoded characters
• Resolve TypeErroring in parse_tracks - this should be raising AttributeErrors if the object doesn't exist, so some conditional type conversion may need to take place to resolve this
• Implement unittesting
• Allow user-curated input
• Create handler for parsing only artists that have no albums and albums that have no tracks - use LEFT JOIN on IDs where album_title and track_title IS NULL
• Move from html.parser to LXML for ~25% speedup
• Backtracking refactor (page by page stepping)
• Consider edge-case handling by headless chrome instance (Playwright)