Skip to content

fix: 0.13 backport, two Windows P0 gateway fixes (#37927, #38025) - #39397

Merged
bolichen97 merged 2 commits into
release/0.13.0from
fix/backport-0.13-p0-windows
Nov 19, 2026
Merged

bolichen97 merged 2 commits into
release/0.13.0from
fix/backport-0.13-p0-windows

Conversation

@bolichen97

Copy link
Copy Markdown
Contributor

Problem / Motivation

Goal: 0.13 Windows users get two P0 fixes that main merged after the 0.13 fork.

Neither fix is on release/0.13.0, and neither was in #39387.

Why it matters

Both bugs leave a Windows install with no working gateway. 0.13 is the next stable line.

Not a goal

Any other fix. The open P0 #38811 has no merged fix on main yet.

What changed (motivation → approach → change)

Two git cherry-pick -x commits on release/0.13.0 (2835f547d0), in main's merge order. Both applied without a conflict. Each keeps its original author and main's message, with the cherry picked from line.

Backwards compatibility

Compatible: both picks are code already on main.

Removes nothing: neither pick hides, deletes or tightens anything.

Tests

On this branch:

  • test/test_feishu_gateway.py: 20 passed.
  • node --test on gateway-stop.test.js, windows-port.test.js and windows-recovery-port-probe.test.js: 105 passed, 1 skipped, 0 failed.
  • flake8 7.1.0 and mypy on the changed Python files: exit 0.

Proof: not provable -- other: a backport of fixes whose proof ran on their own main PRs

Manual verification

N/A: each fix was verified on its own main PR. This PR only carries them.

Related Issues

Part of #38021
Part of #37902

Sibling sweep

Siblings: none -- a backport; each fix's sweep is recorded on its own main PR

Pattern harvest

Not generalizable: a backport; the patterns are recorded on each main PR

Checklist

  • At most two commits (one is the norm), with a Conventional Commits title
  • Existing tests pass and new tests added for new functionality
  • Self-review completed; code follows project style guidelines
  • Documentation updated (if applicable)
  • No secrets, credentials, or internal references in the diff

jeeshofone and others added 2 commits November 19, 2026 04:45
…not trip the stall watchdog (#37927)

Fixes #37902

(cherry picked from commit b5d70c5)
@bolichen97
bolichen97 requested review from a team and KiroCrewCaptain as code owners November 19, 2026 04:54
@bolichen97
bolichen97 requested review from dodgeblaster and removed request for a team November 19, 2026 04:54
@github-actions github-actions Bot added the readiness: checking Automated validation is still running label Nov 19, 2026
@github-actions

Copy link
Copy Markdown
Contributor

UX Review (Opus 5.5) — ✅ PASS

UX-level review of b5698d5eba14fda6f31a3acf1054f1530039764e — updated in place on each push. A BLOCK verdict blocks PR readiness; PASS does not.

UX-Verdict: PASS

Nothing to check. The diff adds or changes no user-visible control or string: the Electron changes only touch log lines and how the free port is detected, and the Feishu change only moves an import off the event loop.

[UX-REVIEWED] b5698d5

Description read

[DESCRIPTION-READ] ac0a17ff4a84522bc62e8c46345547a4e0f50050fed3982d966f3a90f0f4400d

That sha256 names everything this verdict read: the PR description, plus the 3 evidence file(s) the workflow collected from it and handed this review. It covers that evidence because the media strip replaces every attachment URL with the same placeholder, so an edit swapping one attachment for another would leave the description's captured text identical while what this review judged changed. Recompute it with .github/scripts/pr-description-capture.sh over the same inputs; if it differs, the description or that evidence changed after this verdict was formed and any finding drawn from either is unproven.

@github-actions

Copy link
Copy Markdown
Contributor

Design Review (Opus 5.5) — ✅ PASS

Design-level review of b5698d5eba14fda6f31a3acf1054f1530039764e — updated in place on each push. A BLOCK verdict blocks PR readiness; PASS does not.

Design-Verdict: PASS

Nothing blocks: these are two clean cherry-picks of fixes already on main, and their Windows branches are covered by node tests that inject the platform.

Notes

  • The windowsHide: true change in windows-port.js and the Feishu live.snapshot() seed-and-replay of the allow-list both come with these picks, but the description doesn't mention either. Name both under "What changed" so 0.13 reviewers know an allow-list authorization change is shipping.

[DESIGN-REVIEWED] b5698d5

Description read

[DESCRIPTION-READ] 245cb658209173b02bf90dedd7f060171f4456804f34247de9f259f8daa40291

That sha256 names everything this verdict read: the PR description, plus the 1 evidence file(s) the workflow collected from it and handed this review. It covers that evidence because the media strip replaces every attachment URL with the same placeholder, so an edit swapping one attachment for another would leave the description's captured text identical while what this review judged changed. Recompute it with .github/scripts/pr-description-capture.sh over the same inputs; if it differs, the description or that evidence changed after this verdict was formed and any finding drawn from either is unproven.

@github-actions

Copy link
Copy Markdown
Contributor

First Principles Review (Opus 5.5) — ✅ PASS

Premise-level review of b5698d5eba14fda6f31a3acf1054f1530039764e — why this exists and whether the shipped surface is the smallest honest version. Updated in place on each push. A BLOCK verdict blocks PR readiness; PASS does not.

First-Principles-Verdict: PASS

This is a clean backport of two P0 Windows fixes that are already on main. Every item fixes the mechanism, not a symptom. The description leaves out two items, and neither carries a premise risk.

Not checked: whether main's #37927 and #38025 carry the Feishu allow-list re-seed and the windowsHide change (shallow clone).

Notes

  • Item 2 — undeclared: the description says only "import lark-oapi off the event loop". The diff also changes which allow-list the Feishu transport authorizes at startup (seed_open_ids = list(snap.feishu.allowed_open_ids) plus live.watch().replay(dispatcher._config_sub)). It is needed because the new await opens a window where an edit can be dropped, and it reuses the existing ConfigWatcher.replay (config/live.py:514). Nothing should be removed; this is only a gap in the description.

What this change ships

Inventory (5 items) — 3 justified

Intent: give 0.13 Windows users a gateway that starts with Feishu enabled and when netstat times out on a free port. FIX.

  1. With Feishu enabled, the gateway loads the lark-oapi SDK off the event loop, so the stall watchdog no longer restarts it in a loop — justified
  2. The Feishu allow-list in force at startup comes from the live config, and an edit made during the SDK load is replayed — undeclared
  3. On Windows, when netstat cannot finish, the launcher binds the port itself and starts the gateway if the bind succeeds — justified
  4. Netstat/taskkill probes no longer open a console window each time — undeclared
  5. The Windows install guide describes the new bind fallback — justified

[FIRST-PRINCIPLES-REVIEWED] b5698d5

Description read

[DESCRIPTION-READ] be5e176404dfdff7913e71110a15beee3da8dabe3e5b2e3514738083e455b593

That sha256 names the PR description this verdict read. Recompute it with .github/scripts/pr-description-capture.sh; if it differs, the description changed after this verdict was formed and any finding drawn from the description is unproven.

@github-actions

Copy link
Copy Markdown
Contributor

Opus 5.5 Review — ✅ no blocking findings

Reviewed b5698d5eba14fda6f31a3acf1054f1530039764e — this comment is updated in place on each push.

Review details

No findings.

[OPUS-REVIEWED] b5698d5

Verdict parsed from the review's SHA-scoped output markers for commit b5698d5eba14fda6f31a3acf1054f1530039764e.

False positive or not applicable? A repository writer can comment:
/ai-review override fable b5698d5eba14fda6f31a3acf1054f1530039764e false-positive: <reason, 20 to 500 characters>

@bolichen97
bolichen97 merged commit 34fda1f into release/0.13.0 Nov 19, 2026
35 of 36 checks passed
@bolichen97
bolichen97 deleted the fix/backport-0.13-p0-windows branch November 19, 2026 04:57
@github-actions github-actions Bot removed the readiness: checking Automated validation is still running label Nov 19, 2026
@github-actions

Copy link
Copy Markdown
Contributor

GPT 6.1 Review — ✅ no blocking findings

GPT 6.1 completed its review of b5698d5eba14fda6f31a3acf1054f1530039764e and found no blocking issues.

This comment is updated in place on each push.

Review details

No findings.
[GPT-REVIEWED] b5698d5

False positive or not applicable? A repository writer can comment:
/ai-review override gpt b5698d5eba14fda6f31a3acf1054f1530039764e false-positive: <reason, 20 to 500 characters>

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.

3 participants