-
-
Notifications
You must be signed in to change notification settings - Fork 14.7k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[release-24.11] treewide: migrate fetchgit rev = "refs/tags/..." to tag #373256
base: release-24.11
Are you sure you want to change the base?
Conversation
1d06ad9
to
04a2e85
Compare
This is my manual diff when fixing merge conflicts https://gist.github.com/pbsds/6445e4c28eba8ee710d3c4a5b94bc6ba |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
CI is just codeowners noping out of requesting too many reviews.
I spot-chekced and it LGTM.
Can we stop backporting massive treewide changes? I am not looking forward to all the merge conflicts this will create on my stable fork that I then have to fix. Also as previously demonstrated this caused a bunch of regressions and I am pretty sure that others will happen on stable and for someone not super deep into nixpkgs they are not easily fixable. |
I don't think internal code organisation is part of the stable interface. I also don't think we should care about other people's forks to begin with; whether stable channel or not. Additionally the reason why this is even backported is because this would otherwise be a diff to unstable where it matters most (version info), making backports a pain for everyone because cherry-picks cause conflicts. If this wasn't the case, I wouldn't advocate for backporting this either.
All things considered, those were pretty minor, relied on questionable interfaces and were trivial to fix. Finding those is also part of the reason we have the unstable channel. All the ones found so far are also included here AFAICT.
Those are going to be extremely fringe at this point. I'd like to avoid those too for stable but the question is always: at what cost? I find the cost of not having automated backports for thousands of packages for the next 5 months too high to pay and would rather risk breaking extremely fringe usages of things that (IMHO) never should have been depended upon. You could of course convince me by volunteering to do the backports yourself and resolve conflicts manually but I doubt you want that. |
As I see, It prevent backporting bot working properly many times. |
This is a redo of #368177 on
release-24.11
, not because of "correctness" but to alleviate automatic backport conflicts.The script was adapted (see
data1c
) to only migrate packages that already usetag
on master:I went through the
nix-env
diff and fixed all eval failures in a separate commit.The remaining diff to
packages.json
after my fixups may be found here:https://gist.github.com/pbsds/465391bdf3edf34ad5cd73cbe4d92260#file-packages-diff
Also included are the
nix-diff
output and adiff -r
of the one changed storepath.I've looked into various fixups of the original treewide to see if they apply here:
release-24.11
v
prefix onrelease-24.11
(image)Things done
nix.conf
? (See Nix manual)sandbox = relaxed
sandbox = true
nix-shell -p nixpkgs-review --run "nixpkgs-review rev HEAD"
. Note: all changes have to be committed, also see nixpkgs-review usage./result/bin/
)Add a 👍 reaction to pull requests you find important.