[whywhywhywhy]

Wouldn't risk trying to extract from IG too much, I used to yt-dlp from it a lot and use scripts to extract the images because I like to archive references, nothing on a massive scale we're talking <20 times a month and I got a warning that I could lose my username if I "use automated scraping tools".

[tecleandor]

Oh! Were you using your user cookie? I use yt-dlp a couple times a month, but I think I'm always unauthenticated (although I guess they could match my IP address in their logs)

[whywhywhywhy]

I think the urls might be specific to your session in some cases. I’ve noticed some old IG posts I’ve put in bookmarking services say they’re expired.

So passing to yt-dlp would be traceable back to me in that case.

Definitely wasn’t cookies that’s what surprised me.

[tecleandor]

Ah OK, I was thinking about youtube, I re-read it and saw noticed you were talking IG. Yeah, I've noticed some IG URLs are suspicious: different from other users or opening related info that it shouldn't be there.

[prmoustache]

Why would you use an account to do that?

[whywhywhywhy]

I didn’t that’s what surprised me. I think some urls are unique to your session.

Not that it should matter I don’t think I’m the one at fault here.

[heap_perms]

AFAIK you need a username to be able to extract anything from instagram at all.

[Marsymars]

You don't, I don't have an IG account but my wife occasionally sends me links to reels that I use yt-dlp to download.