dirkf

This series has episode URLs like https://www.bbc.co.uk/iplayer/episode/p01sf9kq/shetland-series-2-1-raven-black-part-1. That should set `webpage_url_basename` to `shetland-series-2-1-raven-black-part-1` from which the `[\w-]+?-series-(?P\d+)-(?P\d+)(?:-[\w]+)*` can be parsed as before. As to why this isn't being extracted by...

Is there a particular reason why `traversal` has to be made public to the extractor, rather than just importing from `utils`? Asking for a friend ...

With apologies to @gormster, I've updated this PR with changes including the latest update from yt-dlp/yt-dlp/pull/10928. Since I only get geo-blocking I've also been able to improve that area over...

OP had to work hard to break this. There could be a warning not to do that, like "No title data available: check use of extractor-args". I can't think how...

Supposing that not only OP might have a flaky internet connection that needs to be unduly cosseted, isn't `--sleep-requests ...` designed for the situation? Otherwise, is it too harsh to...

Anyhow, the extractor should look at these HLS URLs (including the old one in case there are any pages still using it): ``` for source_url in traverse_obj(data, ( (('manifests', Ellipsis,...

Your connections to the YT server are repeatedly failing in an unusual (so much that this issue is the top and apparently the only valid GHit for `"[SYS] unknown error...

The underlying problem is that the Crypto/Cryptodome `AES.en/de/crypt()` APIs require a full block in CBC mode, where the caller has to provide the padding. The Python `aes_cbc_encrypt()` implementation in aes.py...

See also analysis at https://github.com/yt-dlp/yt-dlp/issues/3578#issuecomment-1134110546 and below.