Conversation
There was a problem hiding this comment.
Code Review
This pull request updates mise to version 2026.4.4 and adds several new packages to the aqua-registry, including oyo, jiq, dealve-tui, jsongrep, lazyjira, and vize. Feedback was provided to correct registry configurations where override blocks were unintentionally overwriting global architecture mappings or providing redundant definitions. A consistency improvement for environment naming was also suggested.
| overrides: | ||
| - goos: linux | ||
| goarch: amd64 | ||
| replacements: | ||
| linux: unknown-linux-musl | ||
| - goos: linux | ||
| goarch: arm64 | ||
| replacements: | ||
| linux: unknown-linux-gnu |
There was a problem hiding this comment.
In aqua, providing a replacements map within an override block completely replaces the top-level replacements map for that specific environment. Since your top-level map contains architecture mappings (amd64: x86_64, arm64: aarch64), these will be lost in the Linux overrides, leading to incorrect asset names (e.g., using amd64 instead of x86_64). You should include the architecture replacements in these override blocks to ensure the correct binaries are fetched.
overrides:
- goos: linux
goarch: amd64
replacements:
amd64: x86_64
linux: unknown-linux-musl
- goos: linux
goarch: arm64
replacements:
arm64: aarch64
linux: unknown-linux-gnu| overrides: | ||
| - goos: windows | ||
| replacements: {} |
There was a problem hiding this comment.
The override for Windows with replacements: {} is redundant and potentially breaking. In aqua, an empty replacements map in an override replaces the top-level map, which in this case would prevent the amd64 to x86_64 mapping from occurring for Windows users. Since the top-level replacements already handle the architecture mapping correctly and do not interfere with Windows (as there is no windows key in the top-level map), this block should be removed.
| supported_envs: | ||
| - darwin | ||
| - windows | ||
| - amd64 |
There was a problem hiding this comment.
For consistency with the v0.1.1 block (line 25), amd64 should be specified as linux/amd64. While amd64 is technically valid in aqua to represent the architecture across any OS, being explicit improves maintainability and clarity, especially since darwin and windows are already listed separately.
- linux/amd64
Greptile SummaryThis is a routine release PR for mise version 2026.4.4, aggregating a substantial batch of features and bug fixes. Version bumps are consistent across The main substantive changes are seven aqua registry entries (six new tools, one updated). Key items to note:
Confidence Score: 5/5This release PR is safe to merge; all version bumps are consistent and the aqua registry entries follow established patterns. Only P2 style findings remain — the jiq Windows files/src discrepancy may be intentional if the upstream zip is flat, and the aqua-registry version discrepancy appears to be a deliberate independent versioning choice. Neither blocks the release. crates/aqua-registry/aqua-registry/pkgs/bellicose100xp/jiq/registry.yaml (Windows files src field) Important Files Changed
Flowchart%%{init: {'theme': 'neutral'}}%%
flowchart TD
A[Release 2026.4.4] --> B[Version Bumps]
A --> C[Auto-generated Artifacts]
A --> D[Aqua Registry Changes]
B --> B1["Cargo.toml → 2026.4.4"]
B --> B2["default.nix → 2026.4.4"]
B --> B3["snapcraft.yaml → 2026.4.4"]
B --> B4["packaging/rpm/mise.spec → 2026.4.4"]
C --> C1[Cargo.lock]
C --> C2["completions/*"]
C --> C3[mise.lock]
D --> D1["ahkohd/oyo (new, darwin-only)"]
D --> D2["bellicose100xp/jiq (new, ⚠ Windows files/src?)"]
D --> D3["kurama/dealve-tui (new)"]
D --> D4["micahkepe/jsongrep (new)"]
D --> D5["textfuel/lazyjira (new)"]
D --> D6["ubugeeei/vize (new)"]
D --> D7["sigstore/cosign (updated)"]
Reviews (37): Last reviewed commit: "chore: release 2026.4.4" | Re-trigger Greptile |
Hyperfine Performance
|
| Command | Mean [ms] | Min [ms] | Max [ms] | Relative |
|---|---|---|---|---|
mise-2026.4.3 x -- echo |
22.8 ± 0.5 | 21.7 | 26.1 | 1.00 |
mise x -- echo |
23.3 ± 0.6 | 22.3 | 27.8 | 1.02 ± 0.03 |
mise env
| Command | Mean [ms] | Min [ms] | Max [ms] | Relative |
|---|---|---|---|---|
mise-2026.4.3 env |
22.0 ± 0.5 | 21.3 | 25.8 | 1.00 |
mise env |
22.8 ± 0.7 | 21.8 | 32.3 | 1.03 ± 0.04 |
mise hook-env
| Command | Mean [ms] | Min [ms] | Max [ms] | Relative |
|---|---|---|---|---|
mise-2026.4.3 hook-env |
22.8 ± 0.5 | 21.9 | 26.3 | 1.00 |
mise hook-env |
23.4 ± 0.5 | 22.2 | 24.9 | 1.03 ± 0.03 |
mise ls
| Command | Mean [ms] | Min [ms] | Max [ms] | Relative |
|---|---|---|---|---|
mise-2026.4.3 ls |
20.5 ± 0.5 | 19.6 | 25.6 | 1.00 |
mise ls |
21.0 ± 0.4 | 20.1 | 22.6 | 1.03 ± 0.03 |
xtasks/test/perf
| Command | mise-2026.4.3 | mise | Variance |
|---|---|---|---|
| install (cached) | 150ms | 151ms | +0% |
| ls (cached) | 79ms | 79ms | +0% |
| bin-paths (cached) | 83ms | 83ms | +0% |
| task-ls (cached) | 798ms | 797ms | +0% |
## Summary - Adds a verification step to the `release` job that explicitly checks all upstream jobs (`build-tarball-linux`, `build-tarball-macos`, `build-tarball-windows`, `e2e-linux`, `rpm`, `deb`) returned `success` - Prevents releases when matrix variants (e.g. armv7) fail — currently the `!cancelled() && !failure()` condition treats skipped downstream jobs as passing, allowing the release to proceed ## Context PR #8881 (release 2026.4.4) is currently mergeable despite `build-tarball-linux-armv7` and `build-tarball-linux-armv7-musl` failing. The armv7 failures cause `e2e-linux`, `rpm`, and `deb` to be skipped, but `!failure()` doesn't consider skipped jobs as failures. ## Test plan - [ ] Verify the release workflow still succeeds when all targets build correctly - [ ] Verify the release job fails early if any build target fails (e.g. armv7) 🤖 Generated with [Claude Code](https://claude.com/claude-code) <!-- CURSOR_SUMMARY --> --- > [!NOTE] > **Medium Risk** > Medium risk because it changes release pipeline gating and could block publishing if any upstream job is skipped or misreported, but it does not touch product/runtime code. > > **Overview** > **Prevents publishing releases when prerequisites are incomplete.** The `release` workflow now runs an explicit verification step that checks `needs.*.result` for `build-tarball-*`, `e2e-linux`, `rpm`, and `deb`, and fails the job with a clear error message if any did not return `success`. > > This closes a gap where `!cancelled() && !failure()` could allow the release to proceed when upstream jobs were *skipped* (e.g., due to matrix variant failures) rather than failed. > > <sup>Reviewed by [Cursor Bugbot](https://cursor.com/bugbot) for commit ba5b580. Bugbot is set up for automated code reviews on this repo. Configure [here](https://www.cursor.com/dashboard/bugbot).</sup> <!-- /CURSOR_SUMMARY --> Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
fe40254 to
2ba226c
Compare
ccbccb9 to
6313db3
Compare
87aca7f to
f35402a
Compare
🚀 Features
dependsfield for user-specified tool dependencies by @cprecioso in #8776🐛 Bug Fixes
mise lock <tool>runs by @altendky in #8599🚜 Refactor
📚 Documentation
🧪 Testing
📦️ Dependency Updates
📦 Registry
Chore
New Contributors
📦 Aqua Registry Updates
New Packages (6)
ahkohd/oyobellicose100xp/jiqkurama/dealve-tuimicahkepe/jsongreptextfuel/lazyjiraubugeeei/vizeUpdated Packages (1)
sigstore/cosign