Skip to content

Commit 045f193

Browse files
committed
docs(devlog): record r7 PASS and r8 NEAR-PASS, and close the last proof gap
r7 walked the whole sequence and found nothing. Its verdict could not be recorded because the goalplan's activeWorkPhaseId was null, and the review observer discards any sign-off whose round targets a work-phase that is not active (review-observer.ts:99-101). That is why r2 through r6 all had to be aborted as inconclusive after their findings were absorbed — the findings were real, the FSM record was not. Fixed. r8 then re-confirmed independently and found one thing worth having: step 3 asserted wire -> cancel -> tip but never VERIFIED_BASE -> wire, and the counts do not cover that gap because rev-list --count A..B counts commits reachable from B and not A even when A is not an ancestor of B. The reviewer demonstrated it against dev at 1645bb9, where --is-ancestor exits 1 while the three counts still sum. So a stack could have passed step 3 with its bottom not on the verified base — exactly the class of defect this plan kept failing on. The assertion is added with its reasoning inline so nobody deletes it as redundant later. Eight rounds, 26 findings, all absorbed. Four of them attacked the same question and the fourth failure was the signal the question was wrong, which is recorded in 009.
1 parent 86232fe commit 045f193

2 files changed

Lines changed: 82 additions & 3 deletions

File tree

Lines changed: 69 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,69 @@
1+
# 010a — Audit rounds r7 (PASS) and r8 (NEAR-PASS): the plan is executable
2+
3+
## Why two rounds recorded together
4+
5+
`r7` walked the whole sequence and returned **PASS** with no blocking findings. That
6+
verdict could not be recorded: the goalplan's `activeWorkPhaseId` was `null`, and the
7+
review observer discards a sign-off whose round targets a work-phase that is not
8+
active (`review-observer.ts:99-101`). Every earlier round had been silently
9+
discarded for the same reason, which is why `r2`-`r6` all had to be aborted as
10+
inconclusive after their findings were absorbed. Fixed by setting
11+
`activeWorkPhaseId = wp1-integration-roadmap`.
12+
13+
`r8` then re-confirmed independently and returned **NEAR-PASS** with exactly one
14+
finding.
15+
16+
## r7 (PASS) — what it verified
17+
18+
- The stack cut: both boundary subjects occur exactly once, the 39-commit range is
19+
linear with no merges, and `17 + 3 + 19 = 39`. Neither boundary commit can be
20+
dropped as empty — both boundary docs are absent from `dev`, and the overlapping
21+
EOF commit stays non-empty because of `emittedTerminal`.
22+
- All four `r6` findings are **structurally impossible** under the new procedure, not
23+
merely unlikely: commits are referenced, never copied or rearranged.
24+
- Retargeting is safe: a merge commit preserves PR1's commits, so after PR1 lands its
25+
tip remains the merge base of `dev` and PR2, leaving PR2's effective diff exactly
26+
`PR1_TIP..PR2_TIP`. The no-squash rule is what protects this.
27+
- PR1 without WP2b is a complete, correct change: PR1 makes a truncated turn
28+
reportable, WP2b later makes it report tokens. Nothing in PR1 imports the relocated
29+
helper, and PR1's rewritten test asserts terminal shape without requiring usage.
30+
- PR3's inventory is complete — 27 files, exactly the bridge/adapter work, the
31+
integration unit, and the late corrections `030` requires the PR body to disclose.
32+
33+
## r8 (NEAR-PASS) — the one finding, accepted
34+
35+
Step 3's proof was incomplete: it asserted `wire → cancel → tip` but never
36+
`VERIFIED_BASE → wire`. The counts do not cover that gap, because
37+
`git rev-list --count A..B` counts commits reachable from B and not A **even when A
38+
is not an ancestor of B**. The reviewer demonstrated it: against `dev` at
39+
`1645bb924`, `git merge-base --is-ancestor 1645bb924 dfb6fb884` exits 1 while the
40+
three counts still sum correctly.
41+
42+
So a stack could have passed step 3 while its bottom did not sit on the verified
43+
base — precisely the class of defect this plan has been failing on. Added:
44+
45+
git merge-base --is-ancestor "$VERIFIED_BASE" cursor-call-wire # exit 0
46+
47+
with the reasoning inline so a future reader does not delete it as redundant.
48+
49+
`r8` also independently re-confirmed both conflict resolutions, WP2b's resolver
50+
choice and re-export, and the governance position, and found no first-failing step
51+
once the assertion is added.
52+
53+
## Round tally
54+
55+
| Round | Verdict | Findings | Recorded |
56+
|-------|---------|----------|----------|
57+
| r1 | FAIL | 6 | aborted (observer gap) |
58+
| r2 | NEAR-PASS | 3 | aborted (observer gap) |
59+
| r3 | FAIL | 5 | aborted (observer gap) |
60+
| r4 | FAIL | 4 | aborted (observer gap) |
61+
| r5 | FAIL | 3 | aborted (observer gap) |
62+
| r6 | FAIL | 4 | aborted (observer gap) |
63+
| r7 | PASS | 0 | aborted (observer gap — cause found here) |
64+
| r8 | NEAR-PASS | 1 | recorded |
65+
66+
26 findings, every one verified against the tree and absorbed. Four of the eight
67+
rounds attacked the same question (how to split the stack) and the fourth failure
68+
was the signal that the question itself was wrong — recorded in `009`.
69+

‎devlog/_plan/260818_cursor_call_integration/030_phase3.md‎

Lines changed: 13 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -84,8 +84,10 @@ anything.
8484
git branch cursor-call-wire <PR1_TIP>
8585
git branch cursor-call-cancel <PR2_TIP>
8686

87-
3. Prove the stack mechanically — all four must pass:
87+
3. Prove the stack mechanically — all three ancestry assertions plus the count
88+
identity must pass:
8889

90+
git merge-base --is-ancestor "$VERIFIED_BASE" cursor-call-wire # exit 0
8991
git merge-base --is-ancestor cursor-call-wire cursor-call-cancel # exit 0
9092
git merge-base --is-ancestor cursor-call-cancel cursor-call # exit 0
9193
git rev-list --count "$VERIFIED_BASE"..cursor-call-wire
@@ -94,6 +96,14 @@ anything.
9496
# the counts must sum to:
9597
git rev-list --count "$VERIFIED_BASE"..cursor-call
9698

99+
The FIRST assertion is not redundant (audit `r8`): `rev-list --count A..B` counts
100+
commits reachable from B and not A even when A is not an ancestor of B, so the
101+
three counts can sum correctly while the bottom of the stack does not actually sit
102+
on the verified base. Demonstrated: against `dev` at `1645bb924`,
103+
`git merge-base --is-ancestor 1645bb924 dfb6fb884` exits 1 while the counts still
104+
add up. Only the ancestry chain `VERIFIED_BASE → wire → cancel → tip`, together
105+
with the counts, establishes the partition.
106+
97107
4. Push the two new branches and open the PRs bottom-up.
98108

99109
## Policy constraints (`AGENTS.md`)
@@ -127,11 +137,11 @@ gh pr view <n> --json body
127137
```
128138

129139
PR1 base `dev`; PR2 base `cursor-call-wire`; PR3 base `cursor-call-cancel`; step 3's
130-
four checks recorded; all three template sections non-thin in each.
140+
ancestry chain and count identity recorded; all three template sections non-thin in
141+
each.
131142

132143
## Fallback
133144

134145
If step 3 fails — which would mean the rebase did not preserve order as expected —
135146
open ONE PR from `cursor-call` to `dev` and say why in the body. Do not invent a
136147
sixth splitting scheme.
137-

0 commit comments

Comments
 (0)