Skip to content

ops(flake-review): add flake-review workflow#142

Open
ojsef39 wants to merge 2 commits intoInrixia:masterfrom
ojsef39:master
Open

ops(flake-review): add flake-review workflow#142
ojsef39 wants to merge 2 commits intoInrixia:masterfrom
ojsef39:master

Conversation

@ojsef39
Copy link
Contributor

@ojsef39 ojsef39 commented Feb 17, 2026

No description provided.

Copilot AI review requested due to automatic review settings February 17, 2026 20:27
Copy link

Copilot AI left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Pull request overview

This PR adds a new GitHub Actions workflow that integrates flake-review for automated Nix flake validation on pull requests. The workflow uses a reusable workflow from an external repository to review Nix flake changes across both macOS (aarch64-darwin) and Linux (x86_64-linux) platforms, which aligns with this project's Nix flake-based architecture.

Changes:

  • Added flake-review workflow triggered on pull requests to master and dev branches
  • Configured appropriate permissions for reading repository contents and commenting on pull requests
  • Set up matrix builds for both Darwin and Linux systems to ensure cross-platform compatibility

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

@Inrixia
Copy link
Owner

Inrixia commented Feb 17, 2026

Please review failing builds

@xaiyadev
Copy link
Contributor

3 Minutes for a check step sounds really ruff tbf...

Also the conflict will happen until #141 is not fixed

@ojsef39
Copy link
Contributor Author

ojsef39 commented Feb 17, 2026

3 Minutes for a check step sounds really ruff tbf...

Also the conflict will happen until #141 is not fixed

yeah, nix install takes forever on macOS, unfortunately and a bigger runner costs money :(
also in that run the package wasnt cached since i had something changed, maybe ill change the workflow to just use the script directly instead of nix packaged that should be wayyy faster

@Inrixia
Copy link
Owner

Inrixia commented Feb 17, 2026

@xaiyadev thx thought so. I'll consider this blocked by that then just to verify

@Inrixia
Copy link
Owner

Inrixia commented Feb 17, 2026

@ojsef39 yea ideally runtime should be ~30s like Linux.

Caching and using preinstalled things where possible would be preferable.

@ojsef39
Copy link
Contributor Author

ojsef39 commented Feb 17, 2026

@ojsef39 yea ideally runtime should be ~30s like Linux.

Caching and using preinstalled things where possible would be preferable.

I'll try my best, not sure how much i can do tho since most of the time is used by installing nix itself, ill give an update tomorrow :)

edit just looked again and yeah majority of time is in the review step itself, maybe i can get rid of some dependencies or something to make that faster

edit2 the pipeline failed because i had wrong cache entries in cachix, seems like the native-linux-builder features works against me with this so i started using a vm but forgot about the cache

@Inrixia
Copy link
Owner

Inrixia commented Feb 21, 2026

FYI I don't get notifications on comment edits. Idm it but might be nicer to just add new comments

@ojsef39
Copy link
Contributor Author

ojsef39 commented Feb 27, 2026

Sorry I haven't responded until now, I've tried some things, but the results were always pretty much the same, but i still have one or two ideas i want to try this weekend :)

@Inrixia
Copy link
Owner

Inrixia commented Feb 27, 2026

No problem and no rush :)

@Inrixia
Copy link
Owner

Inrixia commented Mar 10, 2026

@frostplexx see failing runs, may be a red herring though

@ojsef39
Copy link
Contributor Author

ojsef39 commented Mar 10, 2026

@frostplexx see failing runs, may be a red herring though

This is just a fluke probably, since the branch here is outdated. I will take another look into it today evening since the other PR made it easier to get this working nicely :)

@frostplexx
Copy link
Contributor

The Hash for the DMG thats failing in nix darwin review should be the correct one:

❯ nix-hash --type sha256 --to-sri $(nix-prefetch-url https://download.tidal.com/desktop/TIDAL.arm64.dmg --type sha256)
path is '/nix/store/gkciyxjizjpmhriyv6lkysncw3ca8x47-TIDAL.arm64.dmg'
sha256-77Z5i4m2nQtfw7CAZqWNybYzpUZG+mEqltlK/llKQJ0=

No idea where the workflow is getting sha256-w5tQscUkhxpWOToAP4oIJJstCNFIdosebTyDI1zFIAE= from.

As for the failing has its the exact same one that was used previous for linux (sha256-Oj34rQbKbsHnqPdVv+ti8z+gZTT+VOsDxg/MQ22sLRQ=) so no idea why it thinks its wrong.

Maybe @ojsef39 has more insight?

@ojsef39
Copy link
Contributor Author

ojsef39 commented Mar 10, 2026

Maybe @ojsef39 has more insight?

it probably also broke because its using the main branch here, while i have some changes in other branches n branch here, while i have some changes in other branches. Ill see later :)

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

5 participants