[krassx]

Release 0.9.0

Version bump (Cargo.toml, Cargo.lock) and the changelog entry. Nothing else.

Added

• debug-files upload --type portable-pdb and convert --to bmf: Unity Mono Release builds symbolicate (Portable PDB / Mono: implement debug-files convert --to bmf (or a real portable-pdb upload) for Unity Release Mono #86, feat: upload Unity Mono portable pdbs as BMF symbols (#86) #99). New capability, so an integrator needs a 0.9.0 floor before using it.

Fixed

• bugsee-cli update on Windows sweeps leftover binaries from earlier updates (fix(update): sweep what an earlier Windows update left behind #103).
• A Rust panic no longer exits 101: 1 before any request, 50 after (fix(exit): a panic exits 1 before any request, 50 after, never 101 #98; it was already in Unreleased).

Changed

• self-replace 1.5.1 (chore(deps): bump self-replace from 1.5.0 to 1.5.1 in the cargo-minor-patch group #97); the Windows update path now has its own CI check on x64 and ARM64 (test(ci): exercise the Windows self-replace for real, as its own check #100, test(ci): say what the self-update test can and cannot observe #102, fix(update): sweep what an earlier Windows update left behind #103).

Checked here: cargo clippy --all-targets -- -D warnings, cargo test (all suites green, 692 unit tests), --version prints 0.9.0.

After merge: tag v0.9.0 on the merge commit. That runs release.yml, then npm-publish.yml and mirror-to-s3.yml chain off it. I have not tagged anything.

🤖 Generated with Claude Code

https://claude.ai/code/session_01WeiJxQWhjhQvvURAjiDkti

[claude]

Claude finished @krassx's task in 24s —— View job

---

Review of #104: chore(release): 0.9.0

No blocking issues found. I did not post inline comments because I found nothing that needs a code change.

I read the diff against origin/main and compared the changelog with the commits it summarizes. I did not run cargo test or clippy. The PR description says they were run, and I didn't re-run them.

What I checked

• Cargo.toml and Cargo.lock both say 0.9.0, and the lockfile change is only the bugsee-cli version line. cargo metadata --no-deps reports 0.9.0.
• Cargo.lock has self-replace at 1.5.1, which matches the "Changed" entry. Cargo.toml still requires "1", so no manifest change was needed.
• The commits between 0.8.2 and this release are Portable PDB / Mono: implement debug-files convert --to bmf (or a real portable-pdb upload) for Unity Release Mono #86/feat: upload Unity Mono portable pdbs as BMF symbols (#86) #99, fix(exit): a panic exits 1 before any request, 50 after, never 101 #98, chore(deps): bump self-replace from 1.5.0 to 1.5.1 in the cargo-minor-patch group #97, test(ci): exercise the Windows self-replace for real, as its own check #100, test(ci): say what the self-update test can and cannot observe #102 and fix(update): sweep what an earlier Windows update left behind #103. Each is covered in the changelog.
• --shipped-dir exists in src/cli/debug_files.rs, and ExitCode::InternalAfterSend = 50 exists in src/exit_code.rs, so the changelog text matches the code.
• The only remaining 0.8.2 references are the retired legacy npm package, which is meant to stay pinned there.
• The Unreleased heading is left empty and the new section is dated 2026-11-04, in the right order.

Nits (P3, optional)

• The "A Rust panic no longer exits 101" entry has no [#98] link, although it is in this release. Other entries link their PRs. It would be a one-line addition plus a footer definition.
• The new link definitions sit before [#28]/[#29], so the footer block is not in numeric order. It was already unsorted, so I'd leave it.
• The panic fix adds exit code 50 and changes the exit for a panic. The "Fixed" entry covers this and says older integrators treat 50 as a substantive failure. Integrators only need a 0.9.0 floor for the Unity/BMF feature, as the PR already says. If any integrator wants to branch on 50 explicitly, they will need the floor as well.

Overall risk: Low. The change is a version bump plus changelog.

Merge recommendation: Approve and merge. Tag v0.9.0 on the merge commit as described. release.yml generates the GitHub Release, then npm-publish.yml and mirror-to-s3.yml chain off it.

Most important to fix: nothing blocking. The #98 link is the only suggestion.

Positives

• The changelog says plainly that the Unity capability needs a new integrator floor, and that the worker side is only deployed to staging.
• It records the failure behavior (exit 11 uploads nothing, no partial results) and the Windows-only scope of the update sweep.
• The PR does not tag anything, which respects the rollout order in CLAUDE.md.

[coderabbitai]

Warning

Review limit reached

You've used all free OSS reviews for now. Wait for the free limit to reset to keep reviewing this public repository.

Next included review available in 39 minutes.

Check out review usage here.

View limit details

Limit details: You’ve used the included review currently available.

Learn how review limits work.

Review configuration:

Run configuration

• Source mode: Auto
• Repository: Unavailable
• Branch: Unavailable
• File: No configuration file from this repository applied
• Commit: Unavailable
• Layers: Organization UI
• Review profile: CHILL
• Plan: Advanced
• Run ID: af872422-135f-4942-aabd-624b1e7f5c60

Commits

Reviewing files that changed from the base of the PR and between 65d3213 and db5a3cd.

Files selected for processing: 1

• CHANGELOG.md

No actionable comments were generated in the recent review.

Summary

Summary by CodeRabbit

• New Features
  • Added Unity Mono Release symbolication support using portable PDB files.
• Bug Fixes
  • Improved cleanup of leftover updater files on Windows.
  • Panics now return an exit code based on whether a request may have been sent.
• Changed
  • Updated the CLI to version 0.9.0.
  • Expanded Windows CI coverage to x64 and ARM64.

The changelog adds 0.9.0 release notes for Unity Mono Release symbolication, Windows updater cleanup, panic exit-code mapping, a dependency update, and Windows CI coverage. Cargo.toml updates the package version from 0.8.2 to 0.9.0.

Possibly related PRs

• bugsee/bugsee-cli#99: Adds portable PDB upload and conversion commands for Unity Mono symbolication, which the release notes describe.
• bugsee/bugsee-cli#98: Adds panic exit handling with the exit-code mapping recorded in the release notes.
• bugsee/bugsee-cli#103: Adds Windows cleanup of leftover self-replace files, which the release notes describe.

Change: Other

Merge Risk: Minimal · previous review 65d32

The release notes and package version consistently describe the intended 0.9.0 release.

Pre-merge checks: 5

Check name	Status	Explanation
Title check		The title "chore(release): 0.9.0" identifies the main change, a release version bump to 0.9.0. The Cargo.toml version change from 0.8.2 to 0.9.0 and the CHANGELOG.md entry for 0.9.0 both match it. The…
Description check		The description states that the change is a version bump in Cargo.toml and Cargo.lock plus a changelog entry. It then lists the release's Added, Fixed, and Changed items, which match the changelog sum…
Docstring Coverage		No eligible functions found in the changed files. Skipping docstring coverage check. Docstring coverage is scoped to functions touched by this diff with cyclomatic complexity of at least 2. Documentat…
Linked Issues check		Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check		Check skipped because no linked issues were found for this pull request.

Full details: Title check

Explanation

The title "chore(release): 0.9.0" identifies the main change, a release version bump to 0.9.0. The Cargo.toml version change from 0.8.2 to 0.9.0 and the CHANGELOG.md entry for 0.9.0 both match it. The title does not list the release contents, but the author's description covers those details, and the title does not need to.

Full details: Description check

Explanation

The description states that the change is a version bump in Cargo.toml and Cargo.lock plus a changelog entry. It then lists the release's Added, Fixed, and Changed items, which match the changelog summary. It also gives verification steps and post-merge tagging notes. It is directly related to the changeset.

Full details: Docstring Coverage

Explanation

No eligible functions found in the changed files. Skipping docstring coverage check. Docstring coverage is scoped to functions touched by this diff with cyclomatic complexity of at least 2. Documentation-only edits and deletions within surviving functions are included; unchanged diff context and fully deleted functions are excluded. Analyzed 0 functions across 0 files. (2 skipped: 2 unsupported.)

Finishing Touches

Generate unit tests (beta)

• Commit to this branch
• Create a new PR

• Autofix · 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

Recent review info

Run configuration

• Source mode: Auto
• Repository: Unavailable
• Branch: Unavailable
• File: No configuration file from this repository applied
• Commit: Unavailable
• Layers: Organization UI
• Review profile: CHILL
• Plan: Advanced
• Run ID: e1e09f17-61a2-48b9-94fb-f155171d190d

Commits

Reviewing files that changed from the base of the PR and between bc0ec81 and 65d3213.

Files ignored due to path filters: 1

• Cargo.lock is excluded by !**/*.lock

Files selected for processing: 2

• CHANGELOG.md
• Cargo.toml

Included review availability: This review used your included allowance. Your plan provides up to 1 included review per hour; 0 remain after this review.

[cursor]

Stale comment

Release PR: version bump 0.8.2 → 0.9.0 and fold of [Unreleased] into dated 0.9.0 notes. Reviewed against every commit since v0.8.2, the shipping code for those changes, dist plan, and this PR's CI logs — not the diff alone.

Findings

P3 Low — 0.9.0 footnote definitions sit under 0.7.7, so the GitHub Release will not hyperlink them

Location: CHANGELOG.md (defs after the 0.7.7 section; they belong immediately after the 0.9.0 bullets)

Problem: cargo-dist copies only the matching version section into the GitHub Release body. This PR's plan job already shows announcement_changelog ending at the Changed paragraph, with no [#86]: / [#99]: / [#103]: definitions. v0.8.2's published notes did include [#96]: because that definition sat under the 0.8.2 section (same for 0.8.1/[#78], 0.8.0/[#69]).

Impact: The public v0.9.0 announcement is the GitHub Release. [#86], [#99], [#103], etc. will render as unresolved markdown references instead of links.

Scenario: Tag v0.9.0 → release.yml announce uses announcement_github_body from plan → Release notes contain [#99] with no definition.

Fix: Move those six definitions to sit after the 0.9.0 Changed bullet (before ## [0.8.2]). Optionally cite #98 on the panic bullet, which currently has no PR link.

CI (not caused by this diff)

test (windows-latest) failed: cli::debug_files::sourcemap_upload_tests::uploads_overlap_unless_concurrency_is_one expected the first PUT after 4 POSTs, observed ["POST", "PUT", ...]. The 150ms POST delay is a timing flake on a loaded Windows runner. This PR does not touch that test. ubuntu-latest and windows-11-arm passed; self-update jobs on both Windows hosts passed.

Checks that held

• Commits since v0.8.2 are all in the notes: #99/#86 (portable-pdb → BMF), #98 (panic 1/50), #103 (Windows leftover sweep), #97 (self-replace 1.5.1), #100/#102 (Windows self-update CI).
• Claims match the shipping code (convert_all stages then publishes; mismatched pdb is exit 11 and uploads nothing; leftover sweep is 32 lowercase letters; lockfile is self-replace 1.5.1).
• 0.9.0 is the right bump: new opt-in command; exit 50 is additive; 101 → 1/50 is the #98 fix. Auto-update stays on cli/v0.x/.
• Cargo.toml / Cargo.lock / cargo metadata are 0.9.0. dist plan: announcement_title 0.9.0 - 2026-11-04, tag v0.9.0. Empty [Unreleased] kept, same shape as 0.8.2.

Overall

1. Overall risk: Low
2. Merge recommendation: Approve the bump once test (windows-latest) is green (re-run that job). Move the 0.9.0 footnote definitions under that section before tagging so the GitHub Release links resolve.
3. Most important to fix: Green CI, then the footnote placement. Do not bump integrator floors until the worker side of portable-pdb is in production — the notes already say it is on staging, and --type portable-pdb is opt-in.
4. What looks good: Tight release PR; notes match git and code; semver and cargo-dist announcement extraction are correct; Windows leftover sweep and panic-exit fold are accurate.

  

_{Sent by Cursor Automation: Bugsee code review}

[claude]

Claude finished @krassx's task in 20s —— View job

---

Re-review of #104 after the latest push: no new findings

I re-read git diff origin/main...HEAD at db5a3cd. I did not run cargo test or clippy, and I did not modify any files.

Earlier findings now fixed

• Footnote placement (cursor, P3): the 0.9.0 link definitions now sit directly under the 0.9.0 ### Changed bullets, before ## [0.8.2] (CHANGELOG.md:47-53). cargo-dist copies only the matching version section into the GitHub Release body, so [#86], [#99], [#103] and the rest will resolve as links in the v0.9.0 notes.
• Missing #98 link (P3 nit from both earlier reviews): the panic bullet now reads "no longer exits 101 ([fix(exit): a panic exits 1 before any request, 50 after, never 101 #98])", and [#98] is defined at line 49.
• I grepped for each of the seven definitions (Portable PDB / Mono: implement debug-files convert --to bmf (or a real portable-pdb upload) for Unity Release Mono #86, chore(deps): bump self-replace from 1.5.0 to 1.5.1 in the cargo-minor-patch group #97, fix(exit): a panic exits 1 before any request, 50 after, never 101 #98, feat: upload Unity Mono portable pdbs as BMF symbols (#86) #99, test(ci): exercise the Windows self-replace for real, as its own check #100, test(ci): say what the self-update test can and cannot observe #102, fix(update): sweep what an earlier Windows update left behind #103). Each appears exactly once, so there are no duplicates and no unresolved references.

Still true

• The diff is only CHANGELOG.md, Cargo.toml and Cargo.lock. Both version fields say 0.9.0, and the lockfile change is just the bugsee-cli version line.
• [Unreleased] is left empty above the dated 0.9.0 section.
• The older [0.x.y]: compare links stop at 0.7.7. Neither 0.8.x nor 0.9.0 has one, so 0.9.0 follows the existing convention and I'd leave it.

CI

• All five runs on the new head were queued or in progress when I checked, so there is no result for this commit yet.
• The earlier test (windows-latest) failure was sourcemap_upload_tests::uploads_overlap_unless_concurrency_is_one. It is a timing-sensitive test this PR doesn't touch. If it fails again, re-run that job. Wait for green before tagging.

Summary

1. Overall risk: Low.
2. Merge recommendation: Approve once CI is green. Re-run the Windows job if the same overlap test flakes.
3. Most important to fix: nothing outstanding in the diff.
4. Positives:
   • The release PR is tight and carries no code changes.
   • The notes match the commits since 0.8.2 and the shipping code.
   • It states the integrator floor for the Unity/BMF capability and that the worker side is only on staging.
   • Nothing is tagged ahead of the rollout order in CLAUDE.md.

[cursor]

Re-review of #104 at db5a3cd (prior 65d3213). Release PR: version bump 0.8.2 → 0.9.0 and fold of [Unreleased] into dated 0.9.0 notes. Rechecked against every commit since v0.8.2, the shipping code for those changes, this head's dist plan, and current CI — not the diff alone.

Findings

No P0–P3 remaining.

The previous P3 (0.9.0 footnote definitions sitting under 0.7.7) is fully addressed. The seven defs now sit immediately after the 0.9.0 Changed bullets (CHANGELOG.md:47-53), the panic bullet cites [#98], and this head's plan job announcement_changelog includes [#86]: through [#103]:. Tagging v0.9.0 will therefore hyperlink those references in the GitHub Release.

Checks that held

• Commits since v0.8.2 are all in the notes: #99/#86 (portable-pdb → BMF), #98 (panic 1/50), #103 (Windows leftover sweep), #97 (self-replace 1.5.1), #100/#102 (Windows self-update CI).
• Claims still match the shipping code (convert_all stages then publishes; mismatched pdb is exit 11 and uploads nothing; leftover sweep is this binary's stem plus a 32-letter lowercase id; lockfile is self-replace 1.5.1; catch_unwind maps panic to 1/50).
• 0.9.0 is the right bump: new opt-in command; exit 50 is additive; 101 → 1/50 is the #98 fix. Auto-update stays on cli/v0.x/.
• Cargo.toml / Cargo.lock / cargo metadata are 0.9.0. dist plan: announcement_title 0.9.0 - 2026-11-04, tag v0.9.0. Empty [Unreleased] kept.

CI

On this head (41856153783): test (windows-latest) passed, as did ubuntu-latest, windows-11-arm, both Windows self-update jobs, clippy, rustfmt, and the Windows/Linux e2e legs. macos was still queued when this review ran; that is runner capacity, not this diff. The earlier sourcemap-concurrency flake on 65d3213 did not reproduce.

Overall

1. Overall risk: Low
2. Merge recommendation: Approve. Tag v0.9.0 on the merge commit as described in the PR; release.yml then npm-publish.yml / mirror-to-s3.yml chain off it.
3. Most important to fix: nothing outstanding in the diff. Do not bump integrator floors until the worker side of portable-pdb is in production — the notes already say it is on staging, and --type portable-pdb is opt-in.
4. What looks good: Tight release PR; the footnote follow-up matches how cargo-dist extracts the GitHub Release body; notes match git and code; semver is correct; Windows leftover sweep and panic-exit fold remain accurate.

  

_{Sent by Cursor Automation: Bugsee code review}