[jdx]

https://entire.io/gh/jdx/mise/trails/445

Fixes the confusing output in discussion #13988: mise uninstall --dry-run java --all printed the same remove <path> row several times per version, plus a row for a version that isn't installed.

$ mise uninstall --dry-run java --all
java@temurin-11.0.30+7 remove ~/.local/share/mise/installs/java/temurin-11.0.30+7
java@temurin-11.0.30+7 remove ~/.local/share/mise/installs/java/temurin-11.0.30+7
java@temurin-27.0.0+35 remove ~/.local/share/mise/installs/java/temurin-27.0.0+35
...
java@27.0.0     remove ~/Library/Caches/mise/java/27.0.0

• A dry run now collects the install, download and cache paths and reports them in one message per version, instead of one progress-row update per path.
• mise uninstall --all skips a listed entry whose install directory doesn't exist (a configured request like 27 standing for temurin-27.0.0+35), so it no longer shows up as a second removal that only touches cache.

I reproduced the output on Linux without a TTY (clean there) and couldn't reproduce the repeated rows themselves, so the TTY redraw is read from the code rather than observed. The e2e test covers the one-row-per-version output.

AI-assisted — Tool: Claude Code; model: anthropic/claude-sonnet-5-5; version: claude-code_2-1-288_agent.

🤖 Generated with Claude Code

---

Note

Low Risk
Uninstall and dry-run display logic only; real uninstall behavior is unchanged aside from skipping non-existent install dirs under --all.

Overview
Fixes noisy and misleading mise uninstall --dry-run output, especially with --all.

Dry-run reporting now gathers install, download, and cache paths for each version and emits a single remove … line listing every path, instead of updating the progress row once per path (which duplicated rows on a TTY).

uninstall --all skips configured versions whose install directory is missing, so alias-style entries (e.g. a short 27 vs the real temurin-… install name) no longer appear as extra removals that only touch cache.

An e2e test asserts one dry-run line per version and that both install and cache paths are named.

^{Reviewed by Cursor Bugbot for commit f96f3d6. Bugbot is set up for automated code reviews on this repo. Configure here.}

Summary by CodeRabbit

• Bug Fixes
  • uninstall --all now skips versions that are not actually installed, preventing stale cache data from creating misleading removal entries.
  • Dry-run output now groups the paths that would be removed into a single progress message. It shows the installation and cache paths for each version once, making planned removals easier to review.

[coderabbitai]

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

• Configuration used: Repository YAML (base), Central YAML (inherited), Organization UI (inherited)
• Review profile: CHILL
• Plan: Advanced
• Run ID: d9eb8cd0-e787-4550-91cf-b88611a33337

📥 Commits

Reviewing files that changed from the base of the PR and between 5fb85b9 and f96f3d6.

📒 Files selected for processing (1)

• e2e/cli/test_uninstall

Included review availability: This review used your included allowance. Your plan provides up to 10 included reviews per hour; 9 remain after this review.

---

📝 Walkthrough

Walkthrough

The uninstall flow skips missing installation paths in --all mode. Dry-run removal collects existing paths and reports them together. An end-to-end test checks the output for two installed versions.

Changes

Uninstall dry-run handling

Layer / File(s)	Summary
Filter missing candidates src/cli/uninstall.rs	In --all mode, candidates whose installation paths do not exist are skipped and logged at debug level.
Aggregate dry-run paths and verify output src/backend/mod.rs, e2e/cli/test_uninstall	Dry-run removal collects existing paths, including dangling links, and reports them in one message. The test checks that each of two installed dummy versions appears once and that its install and cache paths appear in the output.

Priority: ⬇️ Low

Estimated code review effort: 2 (Simple) | ~10 minutes

Change: Bug fix

Merge Risk: ⚪ Minimal · up to f96f3

Dry-run output groups the relevant paths for each removal candidate, and --all skips candidates without an installation directory. No actionable merge-blocking risk remains.

🚥 Pre-merge checks | ✅ 4 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Docstring Coverage	⚠️ Warning	Docstring coverage is 0.00% which is insufficient. The required threshold is 80.00%. Docstring coverage is scoped to functions touched by this diff. Analyzed 2 functions across 2 files. (1 skipped: 1 …	Write docstrings for the functions missing them to satisfy the coverage threshold.

✅ Passed checks (4 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title clearly describes the main change: report each version once during an uninstall dry run.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.

Full details: Docstring Coverage

Explanation

Docstring coverage is 0.00% which is insufficient. The required threshold is 80.00%. Docstring coverage is scoped to functions touched by this diff. Analyzed 2 functions across 2 files. (1 skipped: 1 unsupported.)

• Fix all pre-merge checks with AI

• Autopilot · Keep fixing CodeRabbit findings and required CI, and resolving merge conflicts

---

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.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}

[greptile-apps]

Confidence Score: 5/5

[Medium risk] Changes how the uninstall command reports what it would remove.

The PR appears safe to merge; no new blocking issue was found.

What we checked:

• Extra download path breaks test: Successful installs clean up downloads. The test runs in an isolated environment without enabling download retention.

Summary

Dry-run uninstall now reports removal paths in one message per version. --all also skips entries with no install path.

• The latest changes strengthen the test for both versions and their install and cache paths.
• The earlier unnumbered test-coverage thread was resolved by jdx after adding cache directories and path assertions. Download-path coverage remains outside this test; no duplicate finding is posted.
• No new actionable issues were found.

_{Reviews (2) · Last reviewed commit: "test(uninstall): assert the dry run name..."}

[github-actions]

Instruction counts

benchmark	trend	instructions	Δ	wall (min)	Δ
env	▆▄▁▄▃▃▃▂▃█▆█	27,155,323 → 27,173,660	+0.07%	18.50 → 19.48ms	+5.30%
hook-env	▃▁▃▅▅▄▃▄▅█▇▆	29,153,349 → 29,150,352	-0.01%	19.41 → 19.98ms	+2.96%
ls	▃▁▂▂▂▂▁▁▂███	30,864,295 → 30,857,761	-0.02%	20.18 → 21.20ms	+5.02%
registry	█▄▁▃▃▄▄▄▄▄▄▁	39,952,703 → 39,868,049	-0.21%	17.07 → 17.70ms	+3.71%
startup	█▅▁▂▂▂▁▂▁▂▃▃	10,829,168 → 10,829,420	+0.00%	11.50 → 12.27ms	+6.68%

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.}

_{f96f3d660cbd vs 171902a41a63 · measured on the runner, not pushed to the history.}