[derive] Allow clippy::empty_enums on generated field tokens - #3600
[derive] Allow clippy::empty_enums on generated field tokens#3600githubrobbi wants to merge 1 commit into
clippy::empty_enums on generated field tokens#3600Conversation
The derives emit one uninhabited `enum ẕ<field> {}` per field as a
type-level field token (`derive_has_field_struct_union`). Since
`never_type` stabilized in rust-lang/rust#155499, `clippy::empty_enums`
(guarded on that feature) fires on every empty enum, and because derive
output carries call-site spans the diagnostic lands on the user's
`#[derive(FromBytes)]` line — a downstream crate denying `clippy::nursery`
cannot compile on nightly >= 2026-08-25 without an allow around every
derive site.
Add the lint to the `#[allow(...)]` list of the generated `const _` block,
next to the existing `clippy::missing_inline_in_public_items`, so the
exemption is scoped to macro output only.
Fixes google#3414
|
Thanks for your pull request! It looks like this may be your first contribution to a Google open source project. Before we can look at your pull request, you'll need to sign a Contributor License Agreement (CLA). View this failed invocation of the CLA check for more information. For the most up to date status, view the checks section at the bottom of the pull request. |
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## main #3600 +/- ##
=======================================
Coverage 91.85% 91.85%
=======================================
Files 20 20
Lines 6093 6093
=======================================
Hits 5597 5597
Misses 496 496 ☔ View full report in Codecov by Harness. 🚀 New features to boost your workflow:
|
![chatgpt-codex-connector[bot]](https://avatars.githubusercontent.com/in/1144995?s=60&v=4){kind=small}
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: f56ef68da8
ℹ️ About Codex in GitHub
Codex has been enabled to automatically review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
When you sign up for Codex through ChatGPT, Codex can also answer questions or update the PR, like "@codex address that feedback".
| // (rust-lang/rust#155499), `clippy::empty_enums` fires on every | ||
| // empty enum, and because derive output carries call-site spans | ||
| // the lint lands on the user's `#[derive(...)]` line. See #3414. | ||
| clippy::empty_enums, |
There was a problem hiding this comment.
When a downstream crate runs Clippy 1.92 or earlier with -D warnings, this generated attribute is itself a hard error because those releases register the singular clippy::empty_enum and report clippy::empty_enums as an unknown lint. Since const_block wraps every derive expansion, this makes otherwise valid Zerocopy derives fail across older supported toolchains; include unknown_lints in the generated allow list so the unavailable plural spelling is ignored there.
AGENTS.md reference: zerocopy/AGENTS.md:L13-L17
Useful? React with 👍 / 👎.
| // (rust-lang/rust#155499), `clippy::empty_enums` fires on every | ||
| // empty enum, and because derive output carries call-site spans | ||
| // the lint lands on the user's `#[derive(...)]` line. See #3414. | ||
| clippy::empty_enums, |
There was a problem hiding this comment.
Adding this token changes every expansion produced by const_block, but all 33 checked-in files under zerocopy-derive/src/output_tests/expected still omit it. The output-test harness performs exact pretty-printed token comparison, so cargo test -p zerocopy-derive --lib output_tests::test_known_layout_struct already fails on this added attribute, and the other snapshot-based expansion tests will fail similarly; regenerate and commit the expected expansion files.
Useful? React with 👍 / 👎.
2 participants
Fixes #3414.
What breaks
Since rust-lang/rust#155499 stabilized
never_type(nightly-2026-08-25 onward),clippy::empty_enums— which is guarded oncx.tcx.features().never_type()— fires on every uninhabited enum,#![feature(never_type)]or not. The derives emit oneenum ẕ<field> {}per struct/union field as a type-level field token (derive_has_field_struct_unioninderive/try_from_bytes.rs), and derive output carries call-site spans, so the diagnostic lands on the user's#[derive(FromBytes)]line:A downstream crate that denies
clippy::nursery(orpedanticon older clippy, where the lint lived) cannot pass clippy on current nightlies without an#[allow]around every derive site. Clippy's external-macro suppression does not help because the span is the call site.Fix
Add
clippy::empty_enumsto the#[allow(...)]list of the generatedconst _: () = { … }block inutil.rs, next to the existingclippy::missing_inline_in_public_items. The exemption is scoped to macro output only; user code keeps the lint.Verification
Downstream workspace with
clippy::nursery = deny(14FromBytesderive sites):cargo +nightly-2026-08-29 clippy --workspace --all-targets --all-features -- -D warningsgoes from 14 errors to clean with this change applied via[patch.crates-io].