Repository navigation
Conversation
|
Important Review skippedAuto reviews are disabled on base/target branches other than the default branch. Please check the settings in the CodeRabbit UI or the ⚙️ Run configurationConfiguration used: Central YAML (base), Organization UI (inherited) Review profile: CHILL Plan: Pro Plus Run ID: You can disable this status message by setting the Use the checkbox below for a quick retry:
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
Greptile SummaryCentralizes interactive rendering configuration without changing behavior.
Confidence Score: 5/5The PR appears safe to merge with no actionable defects identified. The sole renderer caller uses the new options type with values equivalent to the previous arguments, while the style extraction preserves the same rendering operations. Important Files Changed
Reviews (1): Last reviewed commit: "refactor(config): centralize interactive..." | Re-trigger Greptile |
Instruction counts
No instruction-count regression above 1%. Only instruction counts gate. Wall clock is shown for context — on identical hardware it moves 4-20% run to run. Measured by tak — instruction-counted CLI benchmarks, stored in this repository's git notes.
|
## Summary - return process statuses through `ExitCode` instead of calling the disallowed `std::process::exit` - rename enum variants so they satisfy Clippy without suppressions - normalize seccomp syscall constants across integer widths - use an async-aware mutex in the brew test that holds a guard across `await` - remove six targeted Clippy exclusions and fix the uncovered `mise-shim` warnings ## Validation - `mise run lint` - `cargo clippy --workspace --all-targets --all-features -- -D warnings` - `cargo test -p mise-shim -p vfox --all-features` (91 passed, 2 ignored) - `cargo test --all-features test_manager_repairs_linked_record_without_repouring` This is a prerequisite refactor in stack #11523; it targets #11522. *AI-assisted — Tool: Codex; model: unavailable/unavailable; version: unavailable.* <!-- CURSOR_SUMMARY --> --- > [!NOTE] > **Low Risk** > Behavioral change is mainly exit path ordering (destructors run before exit); exit codes and shell activation semantics are preserved. > > **Overview** > This PR clears several Clippy suppressions by adjusting how binaries exit, renaming a few enums, and tightening test synchronization. > > **Process exit:** `mise`, `mise-shim`, and `vfox` no longer call `std::process::exit`. They return `ExitCode` from `main` (with `exit::status` replacing `exit::terminate`), so teardown can run before the process ends. > > **Naming:** `ActivatePrelude` variants drop the `Env` suffix (`Set`, `Prepend`, `MovePrepend`), and the completion CLI shell enum uses `Powershell` instead of `PowerShell`, removing `enum_variant_names` allowances. > > **Other:** Seccomp rule keys use a small `syscall_number` helper instead of a conversion lint allow. Brew pour tests switch from `std::sync::Mutex` to `tokio::sync::Mutex` with `blocking_lock` / `.await` so the async test can hold the env lock across `await` without a clippy exception. > > <sup>Reviewed by [Cursor Bugbot](https://cursor.com/bugbot) for commit 479c2ea. Bugbot is set up for automated code reviews on this repo. Configure [here](https://www.cursor.com/dashboard/bugbot).</sup> <!-- /CURSOR_SUMMARY -->
Summary
RenderOptionsvalueRenderStylesclippy::too_many_argumentsexceptions without changing rendered outputValidation
mise run lintcargo check --all-featurescargo clippy -p mise-interactive-config --all-targets --all-features -- -D warningscargo test -p mise-interactive-config -- --skip schema::tests::test_setting_types --skip schema::tests::test_valid_settings(35 passed)The full crate suite also reports two pre-existing schema failures (
test_setting_typesandtest_valid_settings), reproduced unchanged on the parent branch in #11521.AI-assisted — Tool: Codex; model: unavailable/unavailable; version: unavailable.