Do I have to manually figure out which videos were deleted?

[github-account1111]

I like the idea of separating recents from non-recents, but I do not understand what is supposed to happen after the fact.

Say, it's been a few months, I've accumulated multiple videos, a few were deleted - let's be optimistic and say most weren't. Do I then have to go through all those videos and determine which ones can and can't be deleted and then transfer whichever ones remain into their respective folders?

If the answer is yes, that's quite cumbersome and error-prone.

If not, is there some functionality inside the script that takes care of this?

[TheFrenchGhosty]

Yes, and yes it's not ideal.

[93-infinity]

I'm also interested in this, but I don't fully understand what you're asking.

Are you saying this is your situation? (Let's assume only one video per year, named exactly as thus, for simplicity's sake)

Archivist	Recent
----	2021
2020	2020
----	2019
2018	2018

For example, 2021 is still too recent to be archived. 2020 and 2018 have not been deleted so the "recent" can assumed to be inferior to the "archivist" version. 2019 has obviously been deleted within a month since it only exists in "recent" and "archivist" did not catch it. Obviously side-by-side it's easy to see 2019 has been deleted, but, are you are saying:

1. it's hard to determine if 2019 should be moved,
2. you are afraid you might accidentally move 2018 recent->archive, replacing a high quality version with an inferior version?

[github-account1111]

Yes, nicely put. It's both. Both can happen, and both would be undesirable. It's easy to determine side-by-side what should happen in your example, but, like you said, it's a simplified version of what is likely going to happen.

Having a check in place every time the script runs to move what can be moved (2019), keep what is still too recent (2021) and delete what has been archived (2020 and 2018) would be nice. The latter is also important because I don't want duplicates (or low res - high res pairs) sitting on my drive.

[jacroe]

Using something like --dateafter and --datebefore might could help with something like this. I've got something in testing that should download <30 days versions and then the 30>= day versions while deleting the previous ones.

And by "testing" I mean that it doesn't work at all, but at least I don't know why it doesn't work.

[TheFrenchGhosty]

@jacroe It should be possible using 2 different archive.log files... but it seems over complicated...

[jacroe]

@TheFrenchGhosty It's the only automated way I can think of that will get some version of a copy of everything as it's posted.

So definitely complicated. And definitely slower since it's running through everything twice. But I'm not sure if I care? It's a robot; tedious things are exactly what they're made for 🤖

[TheFrenchGhosty]

A script has been written and will be released soon.

[jacroe]

@TheFrenchGhosty Did the script get released? I couldn't immediately find it if it did, and I just wanted to make sure I didn't miss it. No worries at all if it didn't

[TheFrenchGhosty]

@jacroe It hasn't been released no... and I don't remember where I put it... oops.

(I almost certainly still have it somewhere in my archive)

[TheFrenchGhosty]

@jacroe Found it in my archive, here it is:

youtubeavailable.sh.zip

Note: it's an highly experimental version that worked when I tested it back then, I haven't used it since I made it back in august 2023 so it might be highly imperfect.

If I remember correct, you have to run it in the folder where the "channel name" folders are. The list of unavailable videos will be outputted to unavailable_videos.txt