-
-
Notifications
You must be signed in to change notification settings - Fork 14.8k
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
lib.packagesFromDirectoryRecursive: mark directories with recurseIntoAttrs
#369421
base: master
Are you sure you want to change the base?
lib.packagesFromDirectoryRecursive: mark directories with recurseIntoAttrs
#369421
Conversation
Found by `deadnix` 🖤
in preparation for adding more tests for `lib.packagesFromDirectoryRecursive`
…en `newScope` is provided Co-authored-by: Rebecca Turner <[email protected]>
Fixes a bug preventing `recurseIntoDirectory` from changing the `directory` argument. Moreover, `processDir` now cannot capture arguments from the `packagesFromDirectoryRecursive` call, entirely preventing this class of bug from reoccurring should new parameters be added etc.
…oAttrs` This makes the `recurseIntoAttrs` behaviour consistent regardless of whether `newScope` is provided, and is more coherent with nixpkgs convention.
Issue exposed when `packagesFromDirectoryRecursive` set `recurseForDerivations`, as QA automation could then find the `libspl` derivation.
This comment was marked as outdated.
This comment was marked as outdated.
Unfortunately, |
I've been trying my hardest to nuke |
d59d323
to
f23618e
Compare
Entirely fair, and I won't stand in the way of removing it. I'm just going for the lowest-touch change that won't break things when setting Going back to the issue at hand, in principle we could tar the |
Yeah, I think we should totally have recursion enabled! The review comment I left was asking what issues you had with the default |
This comment was marked as outdated.
This comment was marked as outdated.
Oops, sorry I completely missed that one 😅 |
Doesn't this break everyone's usage of |
By “this,” do you mean applying
The scope-related machinery is optional, and not used by default; there might be a misunderstanding rather anything “breaking.” |
It's very likely that I misunderstood the diff. I thought the optional |
Ah, no, not at all. Some arguments default to |
I see! Thank you for explaining; since this doesn't alter |
Thanks for clarifying your concerns. ❤️ This draft PR does alter the behaviour, but only in that attrsets corresponding to directories are tagged with |
Many FreeBSD-specific drvs with undefined platform fail to build on Linux.
Many NetBSD-specific drvs with undefined platform fail to build on Linux.
The snapshot URL returned error 400 “Bad request,” likely due to CGit expiring snapshots.
f23618e
to
a77d3bf
Compare
|
@rhelmot I'd like a new review, but you can wait until the blocker is merged if that's easier. (At least GitHub won't show the parent PR in the diff) |
Split off #359984, at this requires fixes in the BSD package sets.
Things done
nix-shell -p nixpkgs-review --run "nixpkgs-review rev HEAD"
. Note: all changes have to be committed, also see nixpkgs-review usageAdd a 👍 reaction to pull requests you find important.