treewide: use mkTarget (batch 2)#1362
Conversation
a698361 to
668e367
Compare
2c09de6 to
b7555bf
Compare
b7555bf to
ed85314
Compare
ed85314 to
f4ef0dc
Compare
trueNAHO
left a comment
There was a problem hiding this comment.
this should be rebased or merge committed, DO NOT SQUASH so any problematic commits can be reverted.
-- #1342
I did not think of that before merging the previous batch with a merge commit, but in this case, it might be better to apply these treewide mkTarget changes in one commit to prevent littering the commit history with noisy <SCOPE>: use mkTarget commits. Using smaller commits to "trivially" revert /modules/<SCOPE>/ changes is still trivial with git checkout <HASH_OF_THIS_SQUASH_MERGE>~ modules/<SCOPE>/. IMHO, it would be better to sqaush merge this PR.
Splitting this treewide patchset into smaller patchet would make more sense in much larger projects, like the Linux Kernel, where each patch would be reviewed by entirely different submodule maintainers. However, our project is far from being this big and the core maintainers essentially just review the entire PR as a whole. The /module/<SCOPE>/meta.nix module maintainers are also not really involved in these treewide changes, making this use case somewhat non-existent in our case.
f4ef0dc to
60b976e
Compare
Didn't realize that it's possible to revert edits to a specific file with git, in that case this should be squashed. |
|
Successfully created backport PR for |
|
this broke tofi. removed lib from {...}, still used in a couple calls. lib.toHexString, couple more |
|
@trueNAHO maybe revert tofi on branches as it doesnt build? (or throw a commit adding lib back, ig, dont see another problem) |
Fixes: 7ffb31d ("treewide: use mkTarget (batch 2) (nix-community#1362)")
@repparw, can you test #1370? That PR will be automatically merged after CI passes. Let me know if the issue is still not resolved after that. |
Things done
Notify maintainers