Not clear how to setup YT-DLP

[maelp]

I see that there is a yt-dlp integration, but:

• first it's not clear from the interface (only by searching the doc)
• and it's not clear how to use it? (for now when I go on a Youtube (or another supported site) link and bookmark in Hoarder, I'm not asked to "archive with yt-dlp" or whatever
• perhaps there should be a specific "max asset size" for video content (since it will be much higher than "normal bookmark content")
• when it tries to "hoard" a Youtube page, it seems the youtube website has some anti-bot protection, so I only see "please login" or "accept cookie" page, no content

Perhaps there should be a way to connect to the Puppeteer instance remotely so we can login to a few websites with our credentials?

[MohamedBassem]

You enable it with CRAWLER_VIDEO_DOWNLOAD=true in the env file. If the yt-dlp finds a video there, it'll download it. You can configure max video size with CRAWLER_VIDEO_DOWNLOAD_MAX_SIZE.

[maelp]

weird I added those, but it's still not clear where the video is downloaded or how to access it? perhaps it's because of issues with cookies on Youtube website, where it asks to validate cookies, but because I can't share my Chrome cookies with the Puppeteer instance it is "locked" ?

[maxysoft]

For me the video download is not working. This is an example of pages with youtube videos. Maybe yt-dlp cannot find the video? Should be useful an extended log report if the download fails.

2025-04-30T07:45:04.995Z error: [VideoCrawler][125] Failed to download a file from "https://arstechnica.com/science/2025/04/fireflys-rocket-suffers-one-of-the-strangest-launch-failures-weve-ever-seen/" to "/tmp/video_downloads/ec922aa6-fff8-4d0c-b042-07b47e7c0e12"
2025-04-30T07:45:04.995Z info: [VideoCrawler][125] Video Download Completed successfully

2025-04-30T08:12:38.301Z info: [VideoCrawler][382] Attempting to download a file from "https://tanelpoder.com/psnapper/" to "/tmp/video_downloads/b73f8c68-9668-4a3c-b0d2-8e1383aae761" using the following arguments: "https://tanelpoder.com/psnapper/,-f,best[filesize<100M],-o,/tmp/video_downloads/b73f8c68-9668-4a3c-b0d2-8e1383aae761,--no-playlist"
2025-04-30T08:12:44.005Z error: [VideoCrawler][382] Failed to download a file from "https://tanelpoder.com/psnapper/" to "/tmp/video_downloads/b73f8c68-9668-4a3c-b0d2-8e1383aae761"
2025-04-30T08:12:44.006Z info: [VideoCrawler][382] Video Download Completed successfully

[hvmzx]

Pretty unclear from the doc how this is supposed to work, once downloaded, how are the videos accessible/viewable ? I checked the logs and they are stored in "/tmp/video_downloads/" but after that, no idea how to access them