Skip to content

EMG-001 P2-a v9: carried test folds #5678, #5784, lab guard #5879 - #6031

Closed
noahgift wants to merge 4 commits into
mainfrom
emg/5875-p2-a-v9
Closed

noahgift wants to merge 4 commits into
mainfrom
emg/5875-p2-a-v9

Conversation

@noahgift

Copy link
Copy Markdown
Member

Linear rebuild of #6023 (v8, DIRTY after main moved) on main f56911b (emg/5875-p2-a-v9 @4a441e3005); only scripts/guards_nightly_manifest.txt conflicted, resolved as the union of both rows. Carries the #6008 rebuild:: 4 commits, 1 PMAT-Ticket each (PMAT-5678 x2, PMAT-5784, PMAT-5879), no merge. v6 failed workspace-test on a real defect: the #5724 carry brought contracts/train-classify-train-loop-v1.yaml without its four _5686 tests, so strict_test_binding saw 1 dangling reference; the new PMAT-5678 commit restores the tests unchanged from #5724 (4 pass; strict_test_binding passes locally). The other v6 red, aprender-train distributed_integration test_communication_computation_overlap, is a timing test that failed twice and passed on try 3. Supersedes #6008 (v6), #5999 (v5), #5985 (v4, workspace-test red: the ONT witness was stale after the fold; the witness regen is now in the PMAT-5784 commit; ont5_consistency_gate 5/5 ok locally), #5942 and v3. Nothing dropped: where a carried test overlapped #5935, the diff is kept (tests in a nested mod, contracts union-resolved, profile cmd renumbered to 827). Derived files (README, roadmap, contracts.nt, witness) regenerated: no diff.

Touches a LAB guard (PMAT-5879, nightly only) and CI explicit-test fragments. Seat: w5404 approved v4 (handoff/w5404-5942v4-seat-20261022.md). Not armed here.

keep-open: #5678 tracked by its epic, evidence only
keep-open: #5784 tracked by its epic, evidence only
keep-open: #5879 tracked by its epic, evidence only
keep-open: #5875 epic
ont-delta: none

🤖 Generated with Claude Code

t and others added 4 commits October 23, 2026 01:34
…kept as nested mod (Refs #5678)

Rebuild of the PMAT-5678 fold from #5942 on top of main after #5935 landed.
model_card.rs tests sit in mod carried_5724; classification-finetune-v1 keeps
main's FALSIFY-CLASS-010 and the carry's apply-preconditions row as CLASS-011;
profile cmd renumbered to 827 (814 is taken).

PMAT-Ticket: PMAT-5678
PMAT-Batch: BATCH-aprender-5875-p2
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
(cherry picked from commit f815efe)
Carried from #5840; witness, contracts.nt and README count regenerated on current main.

PMAT-Ticket: PMAT-5784
PMAT-Batch: BATCH-aprender-5875-p2
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
(cherry picked from commit 3e2ecee)
#5875 (Refs #5879)

Carried from #5880. The scan step keeps running after a red step, opens one LAB ticket (owner aprender-flow, label lab-cov-scan) on a red night, and reads open issues through a GraphQL page loop.

PMAT-Ticket: PMAT-5879
PMAT-Batch: BATCH-aprender-5875-p2
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
(cherry picked from commit f03cf93)
…-v1 cites (Refs #5678)

The #5724 carry brought contracts/train-classify-train-loop-v1.yaml onto main
without its four `_5686` tests (classify_trainer_tests.rs), so the contract's
`cargo test -p aprender-train --lib _5686` ran nothing and
strict_test_binding failed with 1 dangling reference. The tests are taken
unchanged from #5724 (batch/test-fold-branch-only-0345 @002a4bd15c).

PMAT-Ticket: PMAT-5678
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
(cherry picked from commit 60a24cb)

@noahgift noahgift left a comment

Copy link
Copy Markdown
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Seat 1 (aprender-w5404, non-author): APPROVE @4a441e3005538b449fec9f1b75148b1af9c52b8e — test-only/LAB carry (4 commits, 4 tickets: #5678, #5784, #5879 and the #5686 ClassifyTrainer tests; 45 files, no non-test src change except tiny step_profiler.rs/certification_processing.rs/panels.rs/model_card.rs/loadtest.rs test additions). Measured on this head: every new test target passes — cgp quant_profile_output 6, serve qwen35_load_failclosed_5678 4, train execute_plan_guards_5674 4, orchestrate repl_slash 10, contracts-cli certify_analyze_config_5622 6, profile run_config_errors_5713 5, train --lib _5686 4, sparse bsr_new_5632 6, explain tui 32, present-widgets model_card 88, test-lib loadtest 52, cbtop panels 99; check_coverage_single_report.sh --self-test PASS (3 rows); its guards-nightly.yml step is nightly-only with a one-ticket owner (C324 LAB). EMG-7 GREEN (4 commits, one PMAT-Ticket each, 0 merges). Not measured: clippy, the pre-commit complexity hook, strict_test_binding / readme_contract on the edited contracts (classification-finetune-v1, qwen35-shapes-v1, distributed-training-v1 + 4 new ones), and README count sync. Note this is NOT the 75-line #6023 I seated; it is a larger fold. Needs a second non-author seat.

@noahgift
noahgift marked this pull request as ready for review October 23, 2026 00:05
@github-actions

Copy link
Copy Markdown

§13.11 rung 1 — quorum shadow verdict

S13-SHADOW pr=6031 head=4a441e3005538b449fec9f1b75148b1af9c52b8e verdict=REFUSE class=Q1 arm_rc=1

Shadow mode: this records a verdict and merges nothing. A refusal
to arm is not a block (§13 adds zero rows to §7) — the pull request is
exactly as green as it was.

@noahgift

Copy link
Copy Markdown
Member Author

Snapshot: head 4a441e3. Superseded by #6076 (cop order, aprender-57). Branch kept; closing.

@noahgift noahgift closed this Oct 23, 2026
auto-merge was automatically disabled October 23, 2026 04:14

Pull request was closed

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