Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
18 changes: 18 additions & 0 deletions data/hypergraph/README.md
Original file line number Diff line number Diff line change
Expand Up @@ -32,6 +32,24 @@ workflow_dispatch, with read-only checkouts of public Math- and query-, and uplo
`math-hypergraph` artifact. Vertex keys are repo-prefixed (`main:`, `Math-:`, `query-:`). See `tools/hypergraph/README.md`.
The PNG is a build artifact (optional, `--png`), not committed.

## Two explicit source paths (2026-10-30)

Rows 67 and 258 now spell out their repository-root source paths. Their former
unit-local `reviews/REVIEW_A_V2.md` and
`companions/COLLISION_TRANSFER_THEOREM.md` spellings collided with configured
repository-root prefixes, producing two incorrect derived source vertices.
The [bounded repair record](../../docs/superpowers/plans/2026-10-29-source-path-repair.md)
pins the actual Math `9f8dbb4d` blobs and SHA-256 identities. Neither former
repository-root path exists at that cut; both intended unit-local sources do.

Only those two source-member identities change. All 344 other rows, every
non-source cell, existing implementation memberships, graph nodes, bounds and
scientific qualifications remain. Counts stay 346 units / 1437 vertices /
346 edges, source 220/126, implemented 298/48, placeholder 30 and bound 14.
This is a path correction, not new proof coverage, implementation applicability
or scientific acceptance. The builder, configuration, tests and gates are
unchanged. Historical snapshots below retain their original source cuts.

## Determinant post-511 membership supplement (2026-10-28)

346 ledger rows /1437 vertices /346 ordered hyperedges. Existing row345
Expand Down
4 changes: 2 additions & 2 deletions data/hypergraph/math_coverage_ledger_20261023.md
Original file line number Diff line number Diff line change
Expand Up @@ -401,7 +401,7 @@ OBL OPEN; source accounting only, NOT A RESULT.
| 64 | main#527 | Math- | `frontiers/finite_l_d2_coefficient_20261018` | GB-8e | PROOF.md; REVIEW.md | frontiers/finite_l_d2_coefficient_20261018/coefficient.py; frontiers/finite_l_d2_coefficient_20261018/contract.py | — | N | orphan | 0 |
| 65 | main#527 | Math- | `frontiers/fixed_r_inverse_lifetime_20260930` | GB-8e | PROOF.md | frontiers/fixed_r_inverse_lifetime_20260930/inverse.py; frontiers/fixed_r_inverse_lifetime_20260930/test_inverse.py | — | N | orphan | 0 |
| 66 | main#527 | Math- | `frontiers/full_price_20260924` | GB-7/graph | PROOF.md | frontiers/full_price_20260924/full_price.py; frontiers/full_price_20260924/run_validation.py | math.p15-full-price | N | unbound | 0 |
| 67 | main#527 | Math- | `frontiers/full_spectrum_enclosure_method_20261018` | GB-8w | PROOF.md; history/ROOT_V1_ADVERSE_REVIEW.md; reviews/REVIEW_A_V2.md | MISSING | — | N | orphan | 0 |
| 67 | main#527 | Math- | `frontiers/full_spectrum_enclosure_method_20261018` | GB-8w | PROOF.md; history/ROOT_V1_ADVERSE_REVIEW.md; frontiers/full_spectrum_enclosure_method_20261018/reviews/REVIEW_A_V2.md | MISSING | — | N | orphan | 0 |
| 68 | main#527 | Math- | `frontiers/gaussian_a4_elder_sector_20261017` | GB-8m | PROOF.md; REVIEW.md; coauthor/DERIVATION.md | MISSING | — | N | orphan | 0 |
| 69 | main#527 | Math- | `frontiers/gaussian_short_bar_theory_20261017` | GB-8w | asymmetric_elder/PROOF.md; asymmetric_elder/REVIEW.md; asymmetric_elder/REVIEW_MANIFEST.json | MISSING | — | N | orphan | 0 |
| 70 | main#527 | Math- | `frontiers/intermediate_window_20260928` | GB-7/graph | PROOF.md | frontiers/intermediate_window_20260928/algebra.py; frontiers/intermediate_window_20260928/run_validation.py | math.d5-component.intermediate-window-proof | Y | bound | 0 |
Expand Down Expand Up @@ -592,7 +592,7 @@ OBL OPEN; source accounting only, NOT A RESULT.
| 255 | main#527 | Math- | `imports/gp_data_114_embedded_20260930` | GB-8t | MISSING | imports/gp_data_114_embedded_20260930/GP-DATA-114-v1.0_uniform_q_degree4_interval.py; imports/gp_data_114_embedded_20260930/inert_decoder.py | — | N | orphan | 0 |
| 256 | main#527 | Math- | `imports/hardening_ebedb780` | GB-8t | MISSING | imports/hardening_ebedb780/verify.py; .github/workflows/hardening-proof-custody.yml | — | N | orphan | 0 |
| 257 | main#527 | Math- | `imports/lifetime_parent_20260925` | GB-7/graph | ERRATUM_CONGRUENCE.md; ERRATUM_POINTER.md; MARKED_CYLINDER_CAP_PROOF.md; UNIFORM_MATRIX_CAP_AND_LIFETIME.md | imports/lifetime_parent_20260925/verify.py; .github/workflows/c6-fourier-cutoff.yml | math.d1-component.congruence-erratum, math.d1-component.marked-cylinder-cap, math.uniform-matrix-cap-lifetime | N | unbound | 0 |
| 258 | main#527 | Math- | `imports/transverse_contact_library_20260927` | GB-8t | companions/COLLISION_TRANSFER_THEOREM.md | imports/transverse_contact_library_20260927/companions/diagnostic_kernel.py; imports/transverse_contact_library_20260927/companions/exact_checks.py | — | N | orphan | 0 |
| 258 | main#527 | Math- | `imports/transverse_contact_library_20260927` | GB-8t | imports/transverse_contact_library_20260927/companions/COLLISION_TRANSFER_THEOREM.md | imports/transverse_contact_library_20260927/companions/diagnostic_kernel.py; imports/transverse_contact_library_20260927/companions/exact_checks.py | — | N | orphan | 0 |
| 259 | main#527 | Math- | `imports/upper2d_h5_ledgers_20260926` | GB-8t | REVIEW.md | imports/upper2d_h5_ledgers_20260926/verify.py; .github/workflows/h5-rim-contract.yml | — | N | orphan | 0 |
| 260 | main#527 | Math- | `imports/upper2d_stage_e_20260926` | GB-8t | REVIEW.md | imports/upper2d_stage_e_20260926/raw/B1_taxonomy/b1_falsifier.py; imports/upper2d_stage_e_20260926/raw/C1_alpha_intensity/c1_grid.py | — | N | orphan | 0 |
| 261 | main#527 | Math- | `coefficients/d3_l4_anisotropic_20261005` | GB-8h | PROOF.md | coefficients/d3_l4_anisotropic_20261005/certificate.py; coefficients/d3_l4_anisotropic_20261005/source_crosscheck.py | — | N | orphan | 0 |
Expand Down
2 changes: 1 addition & 1 deletion data/hypergraph/math_hypergraph.json

Large diffs are not rendered by default.

77 changes: 77 additions & 0 deletions docs/superpowers/plans/2026-10-29-source-path-repair.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,77 @@
# Two explicit source-path corrections

Unit: `CBZ0-SOURCE-PATH-REPAIR-20261029`. Actual source author:
OpenAI/Codex `/root/cbz_main_source_accounting`, source-inventory child.
Scientific effect **NONE**. This is artifact membership metadata only.

## Source and cause

Main base `11e28815163e15223e8827fe6ff47a1faaf36ebd`, tree
`0b531a91bb5daa494ffd02ec1b7481c45a595992`. Its hypergraph inputs and
outputs retain their exact historical `55961773` bytes.
The configured root-directory resolver treats `reviews/` and `companions/`
as repository-root paths. Two ledger cells instead use those prefixes for
directories local to their unit, producing incorrect derived members.
Leave that deterministic resolver and all its tests unchanged; make these
two source spellings explicit in the existing ledger.

At Math `9f8dbb4d18b9f3ebd9ffac6b0caa42df9f343966`:

| Row | Correct repository-root path | Git blob | SHA-256 |
|---|---|---|---|
| 67 | `frontiers/full_spectrum_enclosure_method_20261018/reviews/REVIEW_A_V2.md` | `d3d957783fc89d2bdfb63e99b6c82e5412b6da62` | `8f8870fac9937a87a9cc2748a2fd7ea76f50958d713649004bcec294b921e360` |
| 258 | `imports/transverse_contact_library_20260927/companions/COLLISION_TRANSFER_THEOREM.md` | `c8901cd18513ee498e6c5ec23e2b9f346d930f53` | `4f09c1305266264f9030c317dc81912d132590d6882bebf7396adc07f6ac852b` |

The former derived paths `reviews/REVIEW_A_V2.md` and
`companions/COLLISION_TRANSFER_THEOREM.md` do not exist in that Math tree.
These are path corrections, not newly discovered proofs or implementations.

## Execution and evidence

1. Preserve the original inputs/outputs outside the checkout. Read the exact
current Math blobs and reproduce both incorrect memberships. Run the
existing hypergraph suite: observed baseline **19 passed**.
2. Replace only the two source spellings in ledger rows 67 and 258. Before
regeneration, run the existing `build --check`; require an actual stale
output failure. Rebuild once using the unchanged pinned builder.
3. Run the same 19 tests, `build --check` and `validate`. Compare complete
before/after rows and edges: 344 other rows unchanged; every non-membership
cell unchanged; only the two incorrect typed source vertices/roles replaced.
Preserve 346 units, 1437 vertices, 346 edges, source 220/126,
implementation 298/48, placeholder 30 and bound 14.
4. Obtain a fresh whole-branch engineering review; publish this disjoint
source branch after refreshing Main and the complete open-PR path fence.
Root performs any later current-composition integration and hosted checks.

Verification uses the existing pinned `tools/hypergraph/requirements.txt` in
an external virtual environment. No builder, test, gate, bound, graph node,
scientific status, canonical baseline surface or unrelated raw row changes.
No mathematical program, Lean/kernel replay, source applicability review or
full federation gate is executed by this metadata repair.

The bounded local evidence lives at
`/workspace/scratch/cbz-20261029/main-source-evidence/`; this disposable path
is an execution locator, not a required runtime dependency. The authoritative
source identities and reproducible existing commands are recorded here.

Ruling: use the owner's supplied inline two-row plan, isolate the source branch,
and keep canonical foundation/tooling ownership with Root. The owner directive
supersedes historical review-family and human-approval ceremony. Preserve
actual tests, failures, source cuts and scientific qualifications.

## Observed scoped validation, 30 October 2026 UTC

The two actual native source blobs match the table above (11,240 and 8,914
bytes). Both erroneous root paths are absent. Existing hypergraph controls
passed 19/19 before the edit and 19/19 after it. After only the two ledger
cells changed, `build --check` returned 1 with `STALE`. The one subsequent
`build` returned 0; final `build --check` returned 0 and MATCH; `validate`
returned 0 and `[]`.

Complete structured comparison confirms only two source vertex IDs and their
row-role keys changed; all vertex metadata and other edge data remain exact.
All 344 other raw rows and every other cell in rows 67/258 remain exact.
`query_results.json` was regenerated by the existing builder but its bytes
are unchanged, so it creates no commit delta. All documented counts remain
unchanged. The scoped diff has no builder/test/gate changes. Full repository,
hosted current-composition and federation qualification are separate.
Loading