Rebase the draft for "trying to only use next endpoint for the major data" onto master

[mk-pmb]

Thanks for inviting me! As discussed over in "trying to only use next endpoint for the major data", I'm trying to rebase your draft onto master and also apply the changes suggested in the review there.

I've pushed these branches that hopefully will help you see my progress so far, when you look at them in a railroad diagram (like with gitg):

• use-only-next-for-extra-info-base: The parent commit from which your original branch forked.
• use-only-next-for-extra-info-250319: My current progress in splitting your draft into small, understandable (for me) parts that I can then try to replicate based on master.
• use-only-next-for-extra-info-meh: Reverts the improvements, in order to become identical (empty diff) with your original draft. This helps me ensure I didn't forget anything, and also you can easily review exactly what I would like to omit or change.
  • The "[rejected]"/"[debated]" commits revert improvements that I have made based on the review. I marked those review comments with a rocket, so you can more easily see which still need your reply.
  • The bad swap commit most likely was a mistake and should be omitted.
  • I think we should also avoid the bad indent commit because it has a really ugly and confusing diff.

Next steps where I could use your help:

• Please review my splitting, and if you like it (i.e. makes sense, looks like I've understood what you meant to do), reset your draft branch to use-only-next-for-extra-info-250319.
• In the "[meh]" commit, the end after templated "watch" seems to be the end of a function. So, how about we avoid the huge "then" branch by instead inverting the condition, putting the short "else" branch first, repeat the templated "watch" and return early? That way, what would have become the "then" branch, can stay as is and doesn't need the ugly indent commit.
  • If it's trivial for you, please add an early return commit to your draft. If you don't have the capacity soon, I can instead try to learn enough crystal to try it myself.

[unixfox]

I'm really sorry I wasn't subscribed to any notification on this repo so I didn't get any notification about your issue.

I'll check this out hopefully in the coming days.

[unixfox]

I think it would be better for you to push into the branch use-only-next-for-extra-info directly because reviewing changes between the multiple branches is a bit of a nightmare. Hence why I did let you push to my fork.

[mk-pmb]

Hence why I did let you push to my fork.

Thanks, but I'm aiming for rather having a hard reset instead, once we agree on all details. For example the author attribution on the good commits is just a suggestion, I'll be happy to change it to what you'd like.

reviewing changes between the multiple branches is a bit of a nightmare.

I know. I had the same problem when I was trying to ensure I didn't miss any of your changes. That's why I made the effort of replicating your work exactly, bit-perfect. You can run
git diff use-only-next-for-extra-info use-only-next-for-extra-info-meh
to see that the diff is empty. The use…info-meh branch is the exact replica, just split into multiple commits that helped me decipher what you did. So the first point of agreement could be that you can reset your branch to use…info-meh.

However, the latest commits in there, with [rejected]/[bad]/[meh] tags, are commits that I assume you'd rather not keep (for the reasons above). When you omit them, you arrive at use…info-250319. So if you agree they should be omitted, you can reset your branch to use…info-250319.

[unixfox]

Thanks, but I'm aiming for rather having a hard reset instead, once we agree on all details.

Ok I didn't think you would get to that point. I thought you would continue from my existing commits.

Then please create a separate PR to the real invidious repo from your fork, it will be easier like this.

Thanks!

[mk-pmb]

Done: iv-org#5237
For license reasons it would be nice if you could state there that you endorse the authorship declaration in the commits. (Or say what I should change.)

[unixfox]

Thanks.

I'm not sure if I'm following you on the "licensing" stuff, all the code that get merged into the invidious project is under AGPL-3.0: https://github.com/iv-org/invidious/blob/master/LICENSE.

[mk-pmb]

IANAL but in my layman's understanding of German authorship protection law, I cannot grant AGPL license on the commits in your name unless I do it on your orders. Otherwise, my attempt to grant the license may be considered void in some countries.