Skip to content

Runnable examples (sync and async)

Copy-paste friendly scripts under examples/ — each feature ships both YouTube and AsyncYouTube variants.

The repository includes short scripts you can run against a live YouTube connection. Feature examples expose both APIs in one file:

  • Sync — run_sync() with YouTube (default)
  • Async — run_async() with AsyncYouTube (--async flag)

Install

pip install ytscrape
pip install "ytscrape[async]"   # for --async and the concurrency example

uv add ytscrape
uv add "ytscrape[async]"

Clone the repo (or open it after install) and run from the project root:

python examples/01_search_videos.py
python examples/01_search_videos.py --async
python examples/10_async_concurrency.py

Catalogue

Script Topic
01_search_videos.py Search with SearchFilter.VIDEOS
02_search_channels_playlists.py Channels and playlists
03_video_details.py video() metadata
04_pagination.py Transparent vs manual pages
05_language_region.py language / region / Locale
06_error_handling.py ParseError, RequestError, YtScraperError
07_video_comments.py Comments, replies, CommentSort.NEWEST
08_channel_details.py channel() metadata
09_transcript.py Caption tracks and transcripts
10_async_concurrency.py asyncio.gather + max_concurrency (async only)

Sync ↔ async mapping

Sync Async
with YouTube() as yt: async with AsyncYouTube() as yt:
for item in yt.search(...) async for item in await yt.search(...)
yt.video(...) / yt.channel(...) await yt.video(...) / await yt.channel(...)
for c in yt.comments(...) async for c in await yt.comments(...)
yt.transcript(...) await yt.transcript(...)

Models (Video, Comment, VideoDetails, Transcript, …) are identical — only I/O and iteration differ. See the Async API guide for concurrency limits, retries, and fan-out patterns.

Warning

Examples call live YouTube endpoints. Prefer small max_results, reuse one client, and avoid aggressive parallel crawls.

Next steps