Skip to content

evidence(§A.42, landing 3 of 3): land #1009's three SUBSTATE-1 Lean modules - #1100

Merged
amaybaum merged 2 commits into
mainfrom
evidence/a42-substate1-lean-20261104
Nov 4, 2026
Merged

amaybaum merged 2 commits into
mainfrom
evidence/a42-substate1-lean-20261104

Conversation

@amaybaum

@amaybaum amaybaum commented Nov 4, 2026 •

Copy link
Copy Markdown
Owner

Landing 3 of 3 in the owner-approved plan (amaybaum, Claude Code session chat, 2026-11-04: "Land the missing sources", then "Approve all three"). It was cut from main 9d022a65 and is now updated with main 5425dd02 (after #1097, #1098 and #1099). Its files are disjoint from theirs, but the evidence is not independent of #1099 (see below).

Head 27383a03: commit a025eb4e and one merge of main. The merge is an ordinary merge of main 5425dd02, for the up-to-date rule. It conflicts with nothing and changes none of this PR's files. The net diff against main is byte-identical to the reviewed 9d022a65..a025eb4e.

What it lands. Commit a025eb4e:

Why. The observer-relative-finiteness dependency audit on main (DEPENDENCY-AUDIT.md) refers to #1009's work; these modules are its Lean part.

Dependence on #1099, which is now merged (5425dd02). The audit's #1009 rows are fully on main once this PR lands too:

References inside the copied text (§A.42 item 8, adopted by #1097).

Scope.

Checks (local).

  • Census: the census tool passes. Its countercontrol, run on main's census plus the three modules, fails as expected with three UNCLASSIFIED modules.
  • Other repository checks: placement, duplicate, claims and voice pass.
  • Not run locally: the Lean build and axiom report. They run in CI (§A.40).
  • Closing keywords: none in the commit message or this description.

Gates.

  • Independent exact-head review, including the §A.42 item 8 trace. Review at a025eb4e: PASS, nothing blocking. The owner approved that head (amaybaum, session chat, 2026-11-04). For updated heads: "the previous exact-head approvals remain applicable only to those heads and their original gates. Updated heads require their own integration checks."

  • Independent delta review of the update at 27383a03: CONFIRMED WITH NITS, nothing blocking. It confirmed that:

    • the parents are a025eb4e and 5425dd02;
    • the net diff is byte-identical to the reviewed one;
    • the merge equals the automatic merge, and main's 259 changed paths do not overlap this PR's 5;
    • OIBridge.lean and the census are the reviewed blobs: three imports, each once; 122 families, with no duplicate names or keys;
    • the three module blobs are those at a025eb4e, 8af2d904 and d4a80fc8.

    Its nit N1, that the copied text cites the off-main build at a02e197d, is covered above. Its nit N2 is that the PR stays in draft until the required checks are green at this head.

  • Green CI at the exact head 27383a03, including the Mathlib bridge job's Build and Release gate steps.

  • Owner approval naming the exact head SHA.

  • Merge commit only.

🤖 Generated with Claude Code

https://claude.ai/code/session_013T6pfKuv5mi3MXtKjH2tZb

…odules

Byte-identical copies from 8af2d90 (unchanged at #1009's head d4a80fc):
OIBridge/ObservableFactorRecurrence.lean, ObservableMeasureReturn.lean and
GradientLinkFlatness.lean. OIBridge.lean imports them after
NativePadicReduction, as #1009's branch does after RootedRecurrenceBridge.
The census gains their two kernel-only families, verbatim from 8af2d90
(120 -> 122 families); every other entry is unchanged.

The observer-relative-finiteness dependency audit on main cites the first
two for its measured-return row and the third for its gauge-action control.
Their imports (RecurrenceHorizon, EquivalenceChain, Mathlib) are on main.
No proof is changed. If the build fails, the failure is reported and no
proof is repaired.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_013T6pfKuv5mi3MXtKjH2tZb
Brings in #1097 (AGENTS.md), #1098 (history merges) and #1099 (the
missing sources) so the branch is up to date for the required checks.
No file of this branch changes.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_013T6pfKuv5mi3MXtKjH2tZb
@amaybaum
amaybaum marked this pull request as ready for review November 4, 2026 16:51
@amaybaum
amaybaum merged commit 82f0e11 into main Nov 4, 2026
26 checks passed
@amaybaum
amaybaum deleted the evidence/a42-substate1-lean-20261104 branch November 4, 2026 16:51

amaybaum commented Nov 4, 2026

Copy link
Copy Markdown
Owner Author

Merged. Merge commit 82f0e119fd32122c00716060bbe4e93e97fb05f8, pinned to head 27383a036d76bf4fd72a7a9f35b281af405630b4.

Owner approval (amaybaum, Claude Code session chat, 2026-11-04): "Once all required CI checks on 27383a03 pass, I conditionally approve its normal merge, provided the exact head, reviewed file delta and mergeability remain unchanged."

Conditions checked immediately before the merge:

  • Head and base: the remote head was 27383a03 and the base was main 5425dd02.
  • File delta: git diff --binary origin/main 27383a03 was byte-identical to the reviewed 9d022a65..a025eb4e.
  • Mergeability: GitHub reported it mergeable and clean.
  • CI: run 41869638020 at 27383a03 concluded success, with 25 checks succeeded and 1 skipped by design. In it, the Mathlib bridge job's Build, Release gate and axiom-report steps all concluded success, read from the run's job records.

After the merge:

  • Ancestry: the parents are 5425dd02 and 27383a03, and the tree (6f7fa599) equals the head's tree.
  • Diff: git diff 5425dd02 82f0e119 touches exactly the census, OIBridge.lean and the three modules.
  • Census: tools/lean_manuscript_census.py passes on 82f0e119. It shows 122 families with no duplicate names, every module classified.
  • main CI on 82f0e119 is running. Its run will be cited, once its Mathlib steps are confirmed, by the correction PR's item C as the build that checks the landed modules. The a02e197d build stays as provenance for an earlier version that differs only in doc comments.

No theorem, definition, proof or import was changed. #1009 is untouched.


Generated by Claude Code

amaybaum pushed a commit that referenced this pull request Nov 4, 2026
Required changes R1-R5 and nits O1-O6, O8:
- continuous-substates: every written proof in the directory is listed as
  unreviewed on main; acceptance labels scoped to #1009 units, with review
  records elsewhere on main keeping their own status; later edits and the
  #1016 files noted. ANALYSIS.md: "shows" -> "argues".
- PE-01 REPLAY.json: witness_scope gains the limits of REVIEW.md:148-149,
  157-158 and 170-179.
- O-RET README: Q7 wording also frozen in [U, Q7] items and referenced in
  DESIGN-PULSE.md; REVIEW.md:151 superseded; the posted #914 wording; run
  not replayed.
- Census: RecursiveFiniteFactor and GradientLinkFlatness notes name their
  landing PRs (#1096, #1100), as for the two notes already corrected.
- PC-1 README row range; MR-1 README wording.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_013T6pfKuv5mi3MXtKjH2tZb
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.

2 participants