Skip to content

fix(betrieb): replace AppData quality delivery with fresh red-first commits - #1996

Merged
mergify[bot] merged 3 commits into
mainfrom
codex/kr10e2-quality-replacement
Oct 28, 2026
Merged

fix(betrieb): replace AppData quality delivery with fresh red-first commits#1996
mergify[bot] merged 3 commits into
mainfrom
codex/kr10e2-quality-replacement

Conversation

@Cuarroc

@Cuarroc Cuarroc commented Oct 28, 2026 •
edited
Loading

Copy link
Copy Markdown
Owner

Die beiden freigegebenen Planzeilen sind im bestehenden Ersatz-PR ergänzt.
Der von Root akzeptierte Quellstand bleibt in allen sechs Dateien unverändert.
Offen bleiben Roots abschließende Metadatenprüfung und die Integration über Mergify.

Report

Plan-ID: KR-10e-2-APPDATA-QUALITY

Replacement of closed #1975, under the Root replacement decision. Same logical owner srv-kr10e2-quality / internal coordination ticket 54c24620-dd9b-4078-ab1f-e11fe3dd9d57; new execution label srv-kr10e2-quality-replacement only. This is checked source and bounded metadata delivery; full product acceptance remains open. Root closed #1975 without merging at 2026-10-28T03:26:26Z and linked this replacement. Its branch, commits and worktree were left untouched.

Verified metadata handover (2026-10-28)

  • Source: 6590dba8e680b5f1489a43d7d614354900665f15; final metadata/remote head: aa69ed48b3da3ab88b25204918f5a2e8c14248d4; metadata tree: 205eb42bf1558ee3f85fb465f371866c17732c66.
  • Pinned source base and fresh main: 5bb04d671bf47e6efaf9e679637af6bfc2db73c4, CI40623276085 completed/success. After the own row commit, actual git merge-tree --write-tree HEAD origin/main exited 0 and exactly equaled the metadata tree: no conflict and no main merge performed. No automatic upstream delta was introduced in this phase. The older bc3-to-5bb inherited changes remain exactly as recorded above.
  • Only docs/PLAN.md changes since source6590 (+2/-1). P1/P2 each have nine cells. Removing P1 and reverting only P2 reproduces the frozen initial PLAN exactly (initial SHA-256 4425986336b574c5fde40b98d4b0893e8abf26b275f553404fe0c0f5f26c01ec). Parser inventory preserves every foreign ID/title/raw and normalized status/ordered dependency; 482 rows before, 483 after. P1 normalizes to in_review; P2 to done.
  • All six mode/blob pairs above and all 1,871 foreign paths except the authorized PLAN retain source6590 identity. No UI, test, style, DS, API, scan, probe, baseline, configuration, lockfile or dependency edit. No capture/runtime/review rerun for unchanged UI.
  • Prior jobs404/405 ended before this job started. Archived alive wrapper/native identities were reread; fresh process census found no known PID, group/session, worktree or target11/12 survivor. Neither preceding row attempt started a heavy phase. This is a finite termination observation, not universal child-reaping proof. No competing PLAN edit was started.
  • Actual own turn_context at 03:38:46.901Z: OpenAI/gpt-6.1-sol/high. Own native quota at 03:40:07.492Z: weekly61% used/39% remaining; current admission/rules and critical-existing-PR exception checked, no independent author launched. Startcheck0: 13.9GiB free, zero Cargo builds. Heavy phases used unchanged resource/ancestry guard and warm target10/jobs1; no CARGO_PROFILE override. No installation, agents, services or purchase.
Final metadata command / check Actual exit / observation
git diff --check 0
node scripts/plan-check.mjs --offline 0, no findings
Protected six-mode/blob, foreign manifest, nine cells, RAW, row inverse and parser inventory checks 0 each; 143 RAW total, no other row change
bash scripts/ci/secret-scan.sh immediately before normal commit 0
Normal git commit with honest documentation-only No-Test trailer 0; precommit lane0, 62s
git merge-tree --write-tree HEAD origin/main 0, exact metadata tree; no conflict/merge
bash scripts/ci/red-first.sh at final metadata head 0, all five original literal test specifications red/missing on base and green on head; only fresh branch trailers
bash scripts/ci/gates.sh lane prepush through ONE normal foreground push hook 0 at exact own head/worktree, 275s; 1,859 frontend and 2,331 Rust tests passed, seven existing Rust skips
git push -u origin codex/kr10e2-quality-replacement 0; one full foreground push, no imposed timeout, >=1200s allowance available, no light/bypass mode
git ls-remote origin refs/heads/codex/kr10e2-quality-replacement refs/heads/main 0; full remote head exactly aa69ed48b3da3ab88b25204918f5a2e8c14248d4, main5bb

Root replacement decision

Root msg-0mvri0ujb-94d131d0, native 2026-10-28T02:32:01.943Z, authorized fresh equivalent-payload commits on current main after the accepted grouping split made retired Test-First trailers incompatible with the old branch. No public permalink to that native message was supplied; this identifier and the original PR preserve the disposition reference. This branch neither merges nor cherry-picks old history. Two fresh source commits and one documentation-only metadata commit follow the pinned base; no retired grouping trailer, fake test name, gate waiver or history rewrite.

Accepted source and behavior

  • Pinned new base: 5bb04d671bf47e6efaf9e679637af6bfc2db73c4.
  • Immutable accepted source candidate: 6590dba8e680b5f1489a43d7d614354900665f15 (tree b02d4487a9c6b06906b4fd5e5a2ef4ae3210d7a2).
  • Fresh assertion-RED commit: b7ad8d4c111df2142b43ee48ca6b7e99173f782f; minimal fix: 6590dba8e680b5f1489a43d7d614354900665f15.
  • Reference package: cd2cbfbd4464f03ab4385cef8cd0ffae29dd7e88, based on older main bc3a9ee6a26219a3a91c37ac21028d7a847b73e4.
  • Accepted source size: 139 additions + 1 deletion = 140 RAW, six paths including tests. Final metadata adds 2 lines and removes 1: 143 RAW total across seven paths, within the reserved 155 and hard 300. No exception.

Preserves partial => mindestens, complete => exact size, absent/null/malformed quality => explicit unknown completeness. Absent/null/invalid/negative/nonfinite sizes remain unknown; true zero remains zero. The original separate UNBOXED AppData description list, CPU-average wording, RAM wording, free-volume meaning, one snapshot read across rerender and the command whitelist remain. BetriebPanel.tsx also equals original reviewed da390b08 byte-for-byte. No boxed fact/grouping redesign, new scan, poll, IPC permission or concurrency behavior. Existing separate grouping assignment stays outside this delivery.

All six final mode/blob pairs equal cd2 exactly; the binary payload patch 5bb04d671bf47e6efaf9e679637af6bfc2db73c4..6590dba8e680b5f1489a43d7d614354900665f15 equals bc3a9ee6..cd2 restricted to these paths, byte-for-byte:

Path Git blob
src/lib/ipc.resource-quality.public.test.ts a1ae7bfbb2606ecc7dc86eb485783c51177d1ad5
src/lib/ipc.ts a0d6b273548a923d95e7a132eb1065b1f77cd562
src/screens/leitstand/betrieb/BetriebPanel.resources.public.test.tsx 48b83a224471e2860979584474a179c08d379222
src/screens/leitstand/betrieb/BetriebPanel.test.tsx d6262920b0f4583a417bb858a0c7ffb1fa850801
src/screens/leitstand/betrieb/BetriebPanel.tsx 99c53cc147255b49738cbb1962008068b8348052
src/screens/leitstand/betrieb/resources/texts.ts 04ffdae8654de21991b989a989706f2a59ea9fe9

At source6590, all 1,871 other base paths retain their exact modes/blobs; final metadata changes only the authorized PLAN path among them. There is intentionally no whole-tree identity claim against cd2. Automatic upstream changes from older main bc3 to the pinned base total 827 additions / 31 deletions across 16 paths; these are preserved, not replacement-package edits:

5	2	docs/PLAN.md
1	1	docs/plan/v2.0/plan.md
5	1	src/components/TerminalView.tsx
20	8	src/screens/gedaechtnis/LektionenFreigabe.tsx
1	0	src/screens/gedaechtnis/texts.ts
2	2	src/screens/leitstand/LeitstandRoute.test.tsx
40	9	src/screens/leitstand/LeitstandRoute.tsx
230	0	src/screens/leitstand/project-create.public.test.tsx
3	3	src/screens/terminal/TerminalRoute.test.tsx
27	2	src/screens/terminal/TerminalRoute.tsx
275	0	src/screens/terminal/TerminalSplit.browser.mjs
4	0	src/screens/terminal/split/texts.ts
2	1	src/screens/terminal/terminal.css
12	1	src/shell/mount/ShellMount.tsx
199	0	src/shell/mount/gedaechtnis.public.test.tsx
1	1	src/shell/mount/scope-entry.mount.test.tsx

Compared with cd2, the final tree differs on these same 16 foreign paths. PLAN additionally contains the two authorized P1/P2 updates described below. The six declared source blobs are the identity boundary; neither the upstream tree nor cd2's full PLAN was copied over the new base.

Original source evidence (bound to 6590dba)

Observed in the original source run on 2026-10-28; retained as source evidence, not a new runtime/design claim for metadata. Actual own native turn_context: OpenAI gpt-6.1-sol, effort high, checked before editing. Fresh publisher admission and supplied native usage/rules evidence were checked separately; no quota is inferred from a requested profile. Startcheck exited 0. The initial wrapper target was superseded for every heavy phase by the brief's warm target 10, jobs 1, with the unchanged shell-scoped resource/ancestry guard. No background command was launched; commands were awaited through completion.

Command / check Actual exit and result
bash scripts/sync.sh start 0; own worktree/branch clean before edits
npm run dev:start-check -- --min-free-gb 3.5 --max-cargo 2 0
gh run view 40623276085 --json status,conclusion,headSha 0; pinned main CI completed success
Public-tests-only run of the three targeted files on unchanged pinned main 1, five compiling behavioral failures, 24 existing tests passed; no missing export/import/type error or timeout
npm run test:unit -- src/screens/leitstand/betrieb/BetriebPanel.resources.public.test.tsx src/screens/leitstand/betrieb/BetriebPanel.test.tsx src/lib/ipc.resource-quality.public.test.ts after fix 0, 29/29 passed
bash scripts/ci/secret-scan.sh before each normal commit; both normal precommit hooks 0 each; typecheck also 0
git diff --check; six-blob/patch/foreign-mode/scope/RAW/fresh-history checks 0
bash scripts/ci/red-first.sh at exact source candidate 0, all five literal specifications checked on head, red/missing on merge-base; both commits carry exact named trailers
bash scripts/ci/gates.sh lane prepush, through ONE normal foreground push hook 0, full lane at exact candidate, 298 seconds; 1,859 frontend tests and 2,331 Rust tests passed; 7 pre-existing Rust skips
git push -u origin codex/kr10e2-quality-replacement 0, 2026-10-28T03:15:19Z–03:20:30Z; no light/bypass mode, no command timeout imposed
git ls-remote origin refs/heads/codex/kr10e2-quality-replacement refs/heads/main 0, remote branch exactly 6590dba8e680b5f1489a43d7d614354900665f15; main still 5bb04d671bf47e6efaf9e679637af6bfc2db73c4 at handover check
npm run design:gate -- --out <base-capture> and same for <head-capture> 1 / 1, 396 cases each, 20 stored failures each; one capture each
Unchanged strict paired comparator and archive revalidation 0, every case/metric, aggregate and stored-failure count nonincreasing, no exceptions

Test boundary: public IPC through the mocked Tauri transport, and rendered BetriebPanel through literal DOM text and list structure. The five literal public tests are retained from cd2; the existing rejected-read expectation changes from two unknown values to three because AppData adds the third value. No assertion is weakened, and no skip, retry or timeout is added. The retired grouping-only test remains private historical evidence outside this branch. No additional clean-main DiffView probe was run; the earlier single pass remains historical only.

Exact-source design proof

Original base 86401bd81ec62e5e684903f4d29db18730dfbd86 versus exact source 6590dba8e680b5f1489a43d7d614354900665f15 in verified Git archives. Same accepted probe from 07d8706b, Git blob 7dabe132b1c3697862345abcc2e8ada9926b26d1, SHA-256 9d408cad19ce6248e881fc1a7df6b4840593c09e978f4cb8afe7130b5bcc1d18. Only the older-base archive receives that declared probe overlay; head already contains it. Every other archive blob was verified before and after. Chromium 153.0.8010.12, identical 396 cases: 0/2/12 items, 1440×900 and 1920×1080, light/dark. Comparator SHA-256 c8df62853a8deb17f538a569e094125a11118bb19d0a1824ae562a46daefc9d0 unchanged.

Metric Original base New exact source
slack 222 210
nesting 2532 2316
tabs 132 132
eyebrows 376 376
grids 20 20
stored failures 20 20

No case increases. Every Betrieb metric matches the original base, including nesting 8→8 for two items and 28→28 for twelve items in both sizes/themes. Reductions elsewhere are inherited upstream: twelve classic-settings nesting reductions and twelve memory-route slack reductions; none offsets an increase. Both absolute gates remain red, while strict nonincrease is green.

All four actual final-source two-item Betrieb screenshots were visually inspected. Full App-Daten-Verzeichnis and 0,4 GiB (Vollständigkeit unbekannt) are readable in the separate unboxed list at both widths/themes. The existing free-volume value remains ellipsized at 1440; no broad free-disk readability claim. The older nb3 comparison and screenshots are historical cd2 evidence, not substituted for these new-head captures.

Architecture drift: no new occurrence and no allowlist change. No native, design-system, stylesheet, harness, baseline, dependency, lockfile, threshold or configuration edit.

Current metadata and review disposition

P1/P2 are delivered on this same PR. P1 names accepted source 6590dba8e680b5f1489a43d7d614354900665f15 and #1996, with #1975 solely closed replacement history. It retains honest unknown/partial/complete sizes, true zero, one read and the original unboxed presentation. The separate grouping assignment 348 remains blocked until this delivery is adopted. No new source claim is added.

P2 is byte-for-byte the Root-approved KR-10e-1-RESOURCE-READ row from old cd2: bounded #1927 merged bc95e556bc3f5ef5fa6fb2d9412183dc152a4ef3, 2026-10-26T16:56:38Z, head 3a6d73475a375e59a6030bf054b2927f697e46ed; Root acceptance. Rechecked live. Its goal/title and ordered dependencies are preserved. No other historical row or whole parent is closed.

Source accepted by Root at 6590dba after Kimi Code k3.1-pro (Moonshot) APPROVE and Root's source check. A GLM attempt was unavailable and does not count. Final metadata acceptance remains PENDING with Root. Review: ausstehend (Stufe C). R1975-a1/D1 and a2 grouping remain SPLIT OUT; R1975-A1..A3 remain unchanged follow-ups. No author review, Ready, labels, queue, closure or manual merge action.

Goal boundary: current-thread get_goal returned null at entry. The supported API has no goal-id/get-by-id/cross-session resume parameter. Prior active/unfulfilled goal 01a17ffc-1891-7b80-8771-cca6cf301f8d remains reconciled to the internal ticket and was not completed or replaced. A bounded current-execution metadata goal covers the checked handover only; parent/product/native/UAT/Glass acceptance remains open.

NICHT ABGEDECKT

  • P1/P2 metadata is delivered and source6590 is Root-accepted; final Root Tier C metadata acceptance and integration remain open. fix(betrieb): retain AppData measurement quality #1975 is closed without merge. This worker performs no Ready, label, queue, closure or integration action.
  • Both absolute design gates remain red with 20 stored failures. Proven: strict paired nonincrease and inspected AppData wording only.
  • Linux does not compile/run #[cfg(windows)] tests (run_in_pty, npm_shim, ConPTY, DPAPI) or Windows clippy arms. The Windows lane runs in the merge queue.
  • Prepush excludes linux-lane Browser-Smoke, frontend build and workflow gates. Existing Rust skips remain unchanged. Offline plan-check excludes GitHub/age/running-job checks.
  • Updater endpoint, OmniRoute and mirror state are not checked by these gates. lage-jetzt mirror is unavailable; live GitHub/main and dispatch artifacts were used.
  • No native directory scan, CPU/disk/performance measurement, installed/effective, real keyboard/AT or UAT acceptance. Command elapsed times measure checks, not native product performance. Full KR-10e, whole Glass and parent acceptance remain OPEN.

Root review record — replacement source 5bb04d6...6590dba (2026-10-28)

  • fix(betrieb): retain AppData measurement quality #1975 closed by Root at 2026-10-28T03:26:26Z with a link to this PR; branch codex/kr10e2-quality kept (remote head 64ee625).
  • Reviewer: Kimi Code k3.1-pro (Moonshot), non-OpenAI; author family OpenAI. Full replacement diff (140 RAW): APPROVE, findings none. A GLM attempt failed (local Ollama not reachable) and does not count.
  • Root source check at 6590dba: BetriebPanel.tsx maps diskAppPartial true → "mindestens", false → exact, null → "Vollständigkeit unbekannt"; ipc.ts accepts only a boolean, otherwise null; the AppData row is an unboxed KeyValue, so no grouping was added (a1/a2 split out); BetriebPanel.test.tsx:334 2→3 counts the new honest unknown fact and does not weaken anything.
  • [SUPERSEDED 2026-10-28T04:08Z] Still open: own row and review-ok
  • Source accepted by Root at 6590dba.

Root acceptance (INTEGRIEREN) — head aa69ed4 (2026-10-28)

Root record boundary: the source record above is preserved verbatim; its then-pending serial row has now been delivered by this metadata commit. Root's final metadata check and review-ok/Mergify remain pending.

@Cuarroc
test(resources): specify AppData quality and original unboxed display
b7ad8d4 
Fresh assertion-red evidence on the pinned replacement base; preserve the five literal public regression names.

Test-First: src/lib/ipc.resource-quality.public.test.ts::retains only boolean app data quality and leaves absent or malformed quality unknown
Test-First: src/screens/leitstand/betrieb/BetriebPanel.resources.public.test.tsx::labels partial app data as a lower bound and never assumes missing quality is complete
Test-First: src/screens/leitstand/betrieb/BetriebPanel.resources.public.test.tsx::keeps absent and invalid app data sizes unknown instead of inventing zero
Test-First: src/screens/leitstand/betrieb/BetriebPanel.resources.public.test.tsx::keeps a rejected resource read unknown while preserving queue and RAM error states
Test-First: src/screens/leitstand/betrieb/BetriebPanel.resources.public.test.tsx::keeps AppData in its original unboxed description list
@Cuarroc
fix(resources): retain AppData measurement quality without grouping
6590dba 
Restore exactly the accepted six-file source payload as fresh content. Keep null and invalid sizes unknown, preserve true zero and one snapshot read, and leave AppData unboxed.

Test-First: src/lib/ipc.resource-quality.public.test.ts::retains only boolean app data quality and leaves absent or malformed quality unknown
Test-First: src/screens/leitstand/betrieb/BetriebPanel.resources.public.test.tsx::labels partial app data as a lower bound and never assumes missing quality is complete
Test-First: src/screens/leitstand/betrieb/BetriebPanel.resources.public.test.tsx::keeps absent and invalid app data sizes unknown instead of inventing zero
Test-First: src/screens/leitstand/betrieb/BetriebPanel.resources.public.test.tsx::keeps a rejected resource read unknown while preserving queue and RAM error states
Test-First: src/screens/leitstand/betrieb/BetriebPanel.resources.public.test.tsx::keeps AppData in its original unboxed description list
@Cuarroc Cuarroc mentioned this pull request Oct 28, 2026
Closed
@Cuarroc
docs(plan): record accepted AppData source and resource-read delivery
aa69ed4 
Add the bounded KR-10e-2 delivery row for replacement PR #1996 and
restore the exact Root-accepted KR-10e-1 resource-read metadata.
Preserve all source blobs and other plan rows.

No-Test: Documentation-only P1/P2 metadata; accepted source 6590dba remains byte-identical.
@Cuarroc
Cuarroc marked this pull request as ready for review October 28, 2026 04:06
@Cuarroc Cuarroc added the review-ok Pipeline: review disposition complete, no open high finding label Oct 28, 2026
@chatgpt-codex-connector

chatgpt-codex-connector Bot commented Oct 28, 2026 •
edited
Loading

Copy link
Copy Markdown

Codex Review Summary

This comment shows the latest Codex review activity on this pull request.

Review Status Commit Review trigger
📝 Code Review ✅ Completed 2026-10-28T04:07:32.066417Z aa69ed4 Draft marked ready
ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

  • Open a pull request for review
  • Mark a draft as ready
  • Comment "@codex review" or "@codex security review".

Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings.

@mergify mergify Bot mentioned this pull request Oct 28, 2026
Closed
30 of 46 tasks
@mergify mergify Bot added the queued label Oct 28, 2026
@Cuarroc Cuarroc mentioned this pull request Oct 28, 2026
Merged
@mergify
mergify Bot merged commit 9c40bcc into main Oct 28, 2026
25 checks passed
@mergify
mergify Bot deleted the codex/kr10e2-quality-replacement branch October 28, 2026 04:28
@mergify

mergify Bot commented Oct 28, 2026

Copy link
Copy Markdown
Contributor

✅ Merged into main as 9c40bcc at 04:28 UTC

13 min in the queue, 12 min of it CI.

So far

Merge conditions: 9 of 9 met

This pull request

Condition Required by
✅ github-review-approved GitHub repository ruleset rule main-pr-required-checks
✅ label != do-not-merge —

The batch, tested on draft #1997

Condition Required by
✅ check-success = gates (linux) —
✅ check-success = gates (windows) —
✅ check-success = red-first —
✅ any of: check-success = @github-actions/gates (linux), check-neutral = @github-actions/gates (linux), check-skipped = @github-actions/gates (linux) GitHub repository ruleset rule main-pr-required-checks
✅ any of: check-success = @github-actions/gates (windows), check-neutral = @github-actions/gates (windows), check-skipped = @github-actions/gates (windows) GitHub repository ruleset rule main-pr-required-checks
✅ any of: check-success = @github-actions/red-first, check-neutral = @github-actions/red-first, check-skipped = @github-actions/red-first GitHub repository ruleset rule main-pr-required-checks
✅ any of: check-neutral = @mergify/Mergify Merge Protections, check-skipped = @mergify/Mergify Merge Protections, check-success = @mergify/Mergify Merge Protections GitHub repository ruleset rule main-pr-required-checks

Times in UTC · dashboard · how the merge queue works

@mergify mergify Bot removed the queued label Oct 28, 2026
@Cuarroc Cuarroc mentioned this pull request Oct 28, 2026
Merged
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

No reviews

Assignees

No one assigned

Labels

review-ok Pipeline: review disposition complete, no open high finding

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

1 participant

@Cuarroc