This changelog highlights releases that materially changed Genius Scraper for users. Dates use UTC; routine maintenance is intentionally omitted.
Fixes
- Search now returns the number of songs you ask for (up to 500 per query) instead of stopping at the first 3–6 matches.
- Lyrics are filled again for every song that has them; they were often empty after Genius started blocking direct song page loads.
- Song URLs resolve reliably, and a run with no valid input fails at once with a clear message and no charge.
Features
- New Search type option: find songs by a line of their lyrics.
- Each record says which search query found it (
searchQuery).
- The run's status message says how many songs were delivered against how many were asked for, and why it stopped.
- Server API: look up one song or a quick search over HTTP without starting a run.
Reliability
- Bulk search runs now process multiple queries through bounded concurrent queues, reducing timeouts while preserving query and relevance order.
- Large lyric and artist runs flush results in memory-bounded batches to prevent out-of-memory failures.
Improvements
- Search results now prioritize the closest canonical song match, so exact-title queries return the expected original before translations, covers, and alternate versions while preserving all matches.
- Added complete dataset and output field descriptions for clearer API, Console, and AI-tool discovery.
Changes
- Improved dataset table column labels in the Apify Console.
- Added ready-to-run public task examples for common song scraping workflows.
Reliability
- Restored residential proxy routing and browser-like request headers to improve Genius extraction reliability.
Features
- Initial release for extracting Genius song lyrics, metadata, release dates, media links, writer credits, and pageview statistics.