# Changelog of Advanced Youtube Scraper (`stefanie-rink/youtube-scraper`) Actor

- **URL**: https://apify.com/stefanie-rink/youtube-scraper/changelog.md
- **Full Actor documentation**: https://apify.com/stefanie-rink/youtube-scraper.md

### 2026-09-17

- Added support for current YouTube channel video cards and alternate page formats.
- Set the public-page cookie preference to reject optional cookies.
- Corrected consent detection so incidental links do not block valid video data.
- Restored subscriber counts and legacy comment extraction.
- Added video IDs, durations, and thumbnail URLs to results.
- Added a run summary. Failed inputs now fail the run and preserve completed results.
- Updated the Apify SDK to 4.0.2 and added automatic payment limit handling.
- Prepared tier prices of $0.50, $0.45, and $0.40 per 1,000 videos for the rental migration.

### 2026-04-18

- Fixed the default no-input run so Apify health checks complete successfully.
- Replaced the old brittle scraping flow with parsing based on current `ytInitialData`, `ytInitialPlayerResponse`, and `ytcfg` payloads.
- Fixed search and channel pagination to use YouTube continuation endpoints.
- Normalized channel root URLs to `/videos` automatically.
- Added support for bare channel handles and scheme-less channel URLs such as `@Google` and `youtube.com/@Google`.
- Removed unrelated result collection caused by loose video ID extraction from raw HTML.
- Search and channel runs now emit list-page items first and only use watch pages for optional enrichment.
- Added lightweight per-unique-channel subscriber hydration with a bounded cache for search results.
- Added explicit blocked/interstitial watch-page detection and fallback logging.
- Added proxy configuration through environment variables with redacted logging.
- Restored best-effort top-level comment scraping through watch-page continuation requests while keeping fail-soft behavior.
- Made related videos fail soft instead of failing the full run.
- Updated the local/runtime dependency pin to `apify == 1.7.2`.
- Updated the README and `.actor` metadata to match the current actor behavior.

### 2023-05-18

- Added channel video scraping through `/videos` URLs.
- Added the old `clean` input flag.
- Fixed missing `related_videos` in output when no related videos were found.