Use newer multithreaded youtube downloader from scraperlib

[satyamtg]

We now have support for downloadeing youtube videos on a fixed number of threads in scraperlib. Hence, we shall use that instead of the current approach of downloading in batches

[rgaudin]

We definitely should but we shall keep an alternative method for S3 download/upload (IO bound) and ffmpeg (cpu bound)

[stale[bot]]

This issue has been automatically marked as stale because it has not had recent activity. It will be now be reviewed manually. Thank you for your contributions.

[kelson42]

@benoit74 Still a valid issue? Or maybe already fixed in 2.3.0?

[benoit74]

If I get it correctly, this is still not implemented:

• goal of this issue is to use https://github.com/openzim/python-scraperlib/blob/7d498319baadba715316c15cf9857ff2f6974a00/src/zimscraperlib/download.py#L18C7-L18C24
• currently downloads are still not done with it: https://github.com/openzim/youtube/blob/main/scraper/src/youtube2zim/scraper.py#L739-L740

[benoit74]

Let's focus on new UI for 3.0

[benoit74]

Unfortunately, youtube downloader from scraperlib does not permit concurrent downloads from s3 cache, so it is way more complex to implement than needed.

Closing this issue for now, we have to way for scraperlib to properly support downloads from s3 as well