dirkf

I've now had time to check the quoted excerpt, which I believe was the only part referencing the problem video. Confusingly, it seems to specify a playlist with no selected...

Simple: ```diff if not programme_id: programme_id = self._search_regex( - rf'"vpid"\s*:\s*"({self._ID_REGEX})"', webpage, 'vpid', fatal=False, default=None) + rf'"(?:vpid|tleoId)"\s*:\s*"({self._ID_REGEX})"', webpage, 'vpid', fatal=False, default=None) ``` More complex but more tightly targeted: 1. extract JSON...

You are reading it right. The simple fix doesn't work because we are only matching PIDs that are or were known programme PIDs (starting with p, b, l, j, r)...

Historically, see https://github.com/ytdl-org/youtube-dl/issues/30018 and linked issues/PR.

Apparently it gets output as `extractor_key`. Or `extractor`, which is the extractor name: ``` $ yt-dlp --print extractor_key --print extractor 'https://www.youtube.com/watch?v=XmG1QOWbb00') Youtube youtube $ ```

The error is a post mortem, reported when the HTTP downloader gets no data. 1. The fragment downloader tries to get fragment 5. 2. It invokes a specially configured HTTP...

One case seen before was with [format 22 for a newly uploaded video](https://github.com/ytdl-org/youtube-dl/issues/30990), when it was supposed that the format had not yet been fully generated. It's also possible that...

So if the hypothesis is correct, the logs are showing a failed HLS download because: * the manifest specifies a final 0-length or non-existent fragment, probably contrary to the HLS...

That certainly reads like the analysis of someone who could code the fix! As in the problem page of #15195, the `vpid` is available in hydration JSON but not being...

Even from UK it is possible to get 403 on some media links. The extractor skips 403 and 404 for HLS for this reason. In the S1E5 problem page, we...