Skip to content

Merge dev → main (automated) - #2122

Merged
SorraTheOrc merged 80 commits into
mainfrom
release/dev-to-main-20261013224224
Oct 13, 2026
Merged

SorraTheOrc merged 80 commits into
mainfrom
release/dev-to-main-20261013224224

Conversation

@SorraTheOrc

Copy link
Copy Markdown
Member

Automated release created by ship skill.\n\nIncludes CHANGELOG.md with work-item summaries from this release.

pi added 30 commits October 9, 2026 20:28
…nsitions (tests + implementation; delivers WL-0MT2Q9FUZ002MZU2, WL-0MSJM4EIV001A0V9)
…T004H8XJ)

Analyse git commit history to quantify test code rewrite patterns
and evaluate whether test-driven development is delivering value.

Primary findings:
- Rewrite ratio: 21.1% (low, positive)
- Test-first dominance: 86.0%
- Overall: TDD practice is reasonably effective

Includes 10 distinct analysis dimensions:
1. Primary metric (rewrite ratio)
2. Temporal trends by quarter
3. Test area breakdown (9 categories)
4. Per-file stability (top 20 volatile)
5. Rewrite intensity distribution
6. Test-first signal
7. Source volatility correlation
8. Efficiency metrics
9. Commit patterns
10. Domain-specific churn
The list.ts --stage in_review sort comparator called compareGroupableItems()
which handles both bucket and within-bucket sorting, but then discarded
the within-bucket result when items shared the same bucket (bucketCmp === 0),
falling through to sortByPriorityAndDate() which uses a different priority
ordering (descending by createdAt) instead of the within-bucket spec
(high→medium→low by updatedAt).

Replace with compareInReviewItems() which correctly sorts by bucket first,
then by priority (high→medium→low), then by updatedAt (older first), and
finally by id as a deterministic tie-break — matching the specification
in WL-0MSLPM5ZB003TADT and satisfying AC1–AC3.

WL-0MUWDW0N6003KF6G: In Review selection list poorly sorted
…L-0MUY9GDPG0062NWF)

Raise DEFAULT_BACKOFF_CONFIG.maxDelayMs from 60 000 to 600 000 (10 minutes)
in retry-logic.ts, and align per-category maxDelayMs in error-patterns.ts
for serverError and timeout to match.

Updated test expectations across retry-loop.test.ts, retry-after.test.ts,
retry-after-header.test.ts, and settings.test.ts to reflect the new 10-minute
ceiling. Updated comments and doc strings accordingly.
….sh (WL-0MUYTCF4D000WVKR)

Add process group management and output redirection to the test so that
the sleep infinity grandchild spawned by run-in-pane.sh --exec is properly
terminated when the test completes. This fixes hangs in piped invocations
(e.g. bash test_run_in_pane.sh 2>&1 | tail) where the orphaned sleep
infinity kept the pipe open indefinitely.

Changes:
- set -m: enable job control so background processes get their own PGID
- cleanup: kill -TERM - to terminate the entire process group
- redirect stdout/stderr of the HERDR_PANE_ID test case to /dev/null
  to break pipe inheritance that prevents EOF on piped invocations
pi added 28 commits October 12, 2026 18:13
@SorraTheOrc
SorraTheOrc merged commit 1ab082f into main Oct 13, 2026
@SorraTheOrc
SorraTheOrc deleted the release/dev-to-main-20261013224224 branch October 13, 2026 22:46
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