Skip to content

Release 0.29.2 - #2072

Merged
JacobStephens2 merged 1 commit into
mainfrom
release-0.29.2
Nov 12, 2026
Merged

JacobStephens2 merged 1 commit into
mainfrom
release-0.29.2

Conversation

@JacobStephens2

Copy link
Copy Markdown
Owner

Summary

The agent summary was unavailable, so this is GitHub's generated notes.

What's Changed

Full Changelog: v0.29.1...v0.29.2

Version diff

diff --git a/Cargo.lock b/Cargo.lock
index eae25a9e..93338a5b 100644
--- a/Cargo.lock
+++ b/Cargo.lock
@@ -2432,3 +2432,3 @@ dependencies = [
 name = "thirdshift"
-version = "0.29.1"
+version = "0.29.2"
 dependencies = [
diff --git a/Cargo.toml b/Cargo.toml
index 7fc675fb..d524428f 100644
--- a/Cargo.toml
+++ b/Cargo.toml
@@ -2,3 +2,3 @@
 name = "thirdshift"
-version = "0.29.1"
+version = "0.29.2"
 edition = "2024"

@JacobStephens2 JacobStephens2 added the in-progress A Claim: thirdshift has taken this issue label Nov 12, 2026
@JacobStephens2

Copy link
Copy Markdown
Owner Author

Review run. No fixes to push. Spec axis used the pull request body alone; no closing issue was linked.

Review of PR #2072 (Release 0.29.2)

Neither reviewer found anything, so no files changed and nothing was committed. Local validation did not pass: the full Rust suite failed on 2 tests in tests/process.rs. I reran those tests on the base commit and one of the two base runs failed the same way, so the flakiness predates this PR rather than coming from it.

Fixed point: refs/remotes/origin/5aa8a7657ca04be24c4da31e9d9a24bfea534125 does not exist in this worktree. The commit it names, 5aa8a765, does exist and is origin/main, so I used that. The diff is one commit, e83e7fe8 Release 0.29.2, which changes Cargo.toml and Cargo.lock (0.29.1 → 0.29.2).

Spec source: no closing issue is linked, so the pull request body alone served as the spec.

Standards

No findings (no hard violations and no smells).

  • scripts/release.sh defines a release PR as one that "bumps the version and nothing else". This diff changes only the [package] version line in Cargo.toml and the thirdshift entry in Cargo.lock, and the two agree.
  • Running bump_version's own awk programs on the base files reproduces HEAD byte for byte (toml-ok and lock-ok).
  • rg -n "0\.29\.[12]" finds the version only at Cargo.toml:3 and Cargo.lock:2433, so no other file needed bumping.
  • The rules in CODING_STANDARDS.md and AGENTS.md (generated prompts, generated manual, site pages) don't apply to a version-only change.

Files read: AGENTS.md, CODING_STANDARDS.md, docs/testing.md, scripts/release.sh, Cargo.toml, Cargo.lock.

Spec

No findings.

Files read: .thirdshift-review-input.md, Cargo.toml, Cargo.lock, scripts/release.sh, AGENTS.md.

Fixed

Nothing. There were no findings, so no files changed and there is no commit.

Unaddressed findings

  • Standards: none.
  • Spec: none.

Review coverage

  • Changed files: Cargo.toml and Cargo.lock, both read by both reviewers.
  • Generated-output exclusions: none.
  • Unread changed files: none.
  • Reports: written to .thirdshift-review-Jrwriu/standards.md, spec.md and summary.json ({"standards":0,"spec":0}).

Validation

I checked that the PR was OPEN and not a draft before each validation step. The extra-commands list was empty, so there were no extra commands to run.

  1. scripts/affected-tests main: exit 1, a bad selection.
    • Local main is stale (17c206f, behind origin/main), so the script picked 51 paths instead of the 2 this PR changes.
    • npm test --prefix site then failed because the marked package isn't installed: ERR_MODULE_NOT_FOUND: Cannot find package 'marked'.
    • The script stops at the first failure, so nothing else ran. I don't count this as a run against the branch.
  2. scripts/affected-tests origin/main: exit 1.
    • It selected the full Rust suite (as it does for any Cargo.* change), which cargo test ran (nextest isn't installed).
    • Everything passed except 2 tests in --test process, which failed with 238 passed and 2 failed. The failed tests were findings::fixing_dispatches_each_reproduced_finding_and_keeps_unreproduced_findings_private and findings::merged_fix_is_rechecked_after_a_later_fix_fails.
    • Both failed the same way: "the Security fix publishing session did not stage its proposed public issue text", so the command exited 1 instead of 0.
  3. cargo test --test process findings:: rerun on the branch:
    • First rerun: exit 101, 2 failed (a different second test, closing_rechecks_each_merged_base_and_leaves_a_failing_finding_for_the_day_shift).
    • Second rerun: exit 101, 1 failed (merged_fix_is_rechecked_after_a_later_fix_fails).
  4. The same command on base 5aa8a765, in a throwaway worktree I've since removed:
    • Run 1: exit 0, 18 passed.
    • Run 2: exit 101, fixing_dispatches... failed.

These tests don't read the package version, and they fail intermittently on the base commit too, so the failures are a flake that predates this PR. I didn't file an issue for it because it isn't a review finding.

GitHub CI's test, site-test and console-test jobs were still pending when I checked.

Security

No unaddressed introduced findings.

CI passed or absent at e83e7fe

@JacobStephens2 JacobStephens2 removed the in-progress A Claim: thirdshift has taken this issue label Nov 12, 2026
@JacobStephens2
JacobStephens2 merged commit 89dda44 into main Nov 12, 2026
16 checks passed
@JacobStephens2
JacobStephens2 deleted the release-0.29.2 branch November 12, 2026 00:48
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant