diff --git a/research/quadratic-hyperbolic-radical-width/PROOF.md b/research/quadratic-hyperbolic-radical-width/PROOF.md new file mode 100644 index 0000000000..af2bdaaf6a --- /dev/null +++ b/research/quadratic-hyperbolic-radical-width/PROOF.md @@ -0,0 +1,123 @@ +# Exact phase-preserving width for H plus one radical + +Let V have ordered binary basis a,b,z. Its alternating Gram form has +B(a,b)=1 and B(a,z)=B(b,z)=0. Write the prescribed quadratic refinement as + + q(xa+yb+tz) = xy + alpha*x + beta*y + gamma*t in F2. + +The three parameters alpha,beta,gamma range independently over F2. A faithful +phase-preserving realization is an injective linear map f:V -> F2^m with + + f(v) dot f(w) = B(v,w), + wt(f(v)) = 2*q(v) modulo 4. + +Every image therefore has even weight. On the even-weight space E_m define +Q(u)=wt(u)/2 modulo 2. The elementary weight identity gives + + Q(u+v) = Q(u)+Q(v)+u dot v. + +Consequently checking the basis Gram matrix and the three basis phases checks +the prescribed refinement on every vector. The exact minimum m is + +| gamma | alpha,beta | Minimum m | +|---|---|---:| +| 0 | 1,1 | 4 | +| 0 | 0,0 or 1,0 or 0,1 | 6 | +| 1 | any | 5 | + +This is a finite-dimensional exact theorem, supported below by explicit lower +bounds and attaining maps. It answers only the first working example of Open +next question 1 in Nick Petalas's PR6456. That predecessor's general Gram-only +minimum is cited context; this proof does not require replaying it. + +## Lower bounds + +For m>0, E_m has dimension m-1. An injective three-dimensional image therefore +needs m>=4. At m=0 there is no injective image either. + +At m=4 an injective image must equal E_4. The radical of the dot form on E_4 +is the line generated by 1111: the perpendicular complement of the even-weight +hyperplane is the all-ones line, which itself has even weight. Injectivity and +Gram preservation send the nonzero radical vector z to 1111. Its Q is zero, +so gamma=1 cannot occur at m=4. This proves m>=5 when gamma=1. + +Suppose gamma=0 and (alpha,beta)!=(1,1). At least one of a,b, call it v, has +q(v)=0. The vectors v and z are distinct nonzero vectors and B(v,z)=0. Their +images must therefore be distinct nonzero even vectors of Q=0 that are +orthogonal. In five coordinates, every such vector has weight four. Two +distinct weight-four vectors in F2^5 have intersection of size three, so their +dot product is one. This is impossible. Padding by zero coordinates preserves +all conditions, so the impossibility at width five excludes every smaller +width too. Thus these three phase choices need m>=6. + +These are direct exclusions of widths below the displayed minima. + +## Attaining maps + +Integers below encode binary coordinate vectors with the low bit first. +The following rows are the images of a,b,z, in that order. + +| alpha,beta,gamma | m | Images | +|---|---:|---| +| 0,0,0 | 6 | 15,23,57 | +| 1,0,0 | 6 | 24,23,57 | +| 0,1,0 | 6 | 15,24,57 | +| 1,1,0 | 4 | 3,5,15 | +| 0,0,1 | 5 | 15,23,3 | +| 1,0,1 | 5 | 12,23,3 | +| 0,1,1 | 5 | 15,20,3 | +| 1,1,1 | 5 | 12,20,3 | + +For the first row, the weights are 4,4,4, pairwise intersections have sizes +3,2,2, and the three vectors are independent. Replacing the first or second +image by their XOR, 24, gives the next two rows. Since q(a+b)=1 in this seed, +these replacements give the indicated phases while preserving the Gram form +and injectivity. + +At width four the seed 3,5,15 has weights 2,2,4 and intersection parities +1,0,0. The three vectors are independent, giving the fourth row. + +At width five the seed 15,23,3 has weights 4,4,2 and intersection parities +1,0,0. Its third vector is independent of its first two. Adding that third +vector to the first or second vector preserves every Gram entry and toggles +exactly that vector's Q, because its pairing with the third vector is zero. +This gives all four gamma=1 rows. The polarization identity extends every +basis check to all of V. The lower bounds prove each minimum is exact. + +## Finite corroboration and interpretation + +The standalone checker uses two enumerations. The first visits every ordered +triple of even vectors at each width 0 through 6, including zero and dependent +triples before rejecting them. There are 37,450 candidate triples. The second +visits every multiset of m column vectors in F2^3 for the same widths, 3,003 +multisets in total. For a multiset with counts c_j, it adds m!/product(c_j!) +labelled matrices. Column span checks injectivity; integer column incidences +check parity, Gram entries and phases. Both methods must give identical +embedding counts at every width for every phase choice. Neither census uses +the predicted minimum to prune its search. + +The checker also verifies every vector phase and all ordered vector pairings +for each displayed witness, and rejects phase loss, collapsed radical, odd +weight, wrong Gram data, invalid dimensions and malformed input controls. +Finite corroboration does not constitute formal proof-kernel acceptance. + +Compared with the cited Gram-only width four, the extra requirement costs zero, +one or two coordinates. Thus a phase-preserving native label reduction cannot +infer width four from Gram data alone, even on this smallest degenerate example. +The coordinate cost is exact. Native physical realization, source/read +incidence, dirty restoration, nested frames, precision, owner allocation and +the complete recursive bill remain separate obligations. No kappa improvement +or complete multiplication theorem follows from this example. + +## Credit + +This contribution was produced by Nick Petalas. Apache-2.0. Nick Petalas's +PR6456 posed this phase-preserving first example after its Gram-only theorem. +Maximus Putnam / Maximooch's PR831 supplies the earlier alternating trace and +phase-interface motivation, retaining Joel Pulikkan / GamingPuzzled and the +endpoint-normal-form lineage. THU New Media Research Center / thu-nmrc's +PR5464 is related degenerate-embedding background retained by PR6456. Those +results are cited, not audited here. Polarization, binary even-weight spaces +and elementary code arguments are classical. The new repository contribution +is the exact eight-case phase-sensitive minimum and its standalone certificate; +no literature-priority claim is made. diff --git a/research/quadratic-hyperbolic-radical-width/README.md b/research/quadratic-hyperbolic-radical-width/README.md new file mode 100644 index 0000000000..d2d9516ae2 --- /dev/null +++ b/research/quadratic-hyperbolic-radical-width/README.md @@ -0,0 +1,116 @@ +# Exact phase-preserving width for a hyperbolic pair and one radical + +For the three-dimensional binary form `H direct sum 0`, prescribe + + q(xa+yb+tz) = xy + alpha*x + beta*y + gamma*t. + +An injective linear realization must preserve its Gram form and satisfy +`wt(f(v)) = 2*q(v) modulo 4` on every vector. Its exact minimum coordinate +width is: + +| Radical phase gamma | Hyperbolic basis phases alpha,beta | Minimum | +|---|---|---:| +| 0 | 1,1 | 4 | +| 0 | 0,0 or 1,0 or 0,1 | 6 | +| 1 | any | 5 | + +Claim type: scoped written proof with exhaustive finite corroboration and +explicit minimum embeddings. The prescribed phases can therefore cost one or +two coordinates beyond the cited Gram-only minimum four. This settles the +first working example of Open next question 1 in +[PR6456](https://github.com/CrocSwap/integer-mult-bounds/pull/6456), rather than +auditing that package. [PROOF.md](PROOF.md) proves the lower bounds and lists +attaining maps for all eight phase choices. + +There is no new kappa, physical supplier, selected-frontier change or complete +multiplication theorem. This is a phase-sensitive label interface result. +Physical realization and a complete signed bill remain missing. + +## Evidence and reproduction + +From the repository root, with CPython 3.12 and its standard library: + +```sh +/Users/nick/git/integer-mult-bounds/.venv/bin/python3.12 -I -W error -B research/quadratic-hyperbolic-radical-width/verify.py +/Users/nick/git/integer-mult-bounds/.venv/bin/python3.12 -I -O -W error -B research/quadratic-hyperbolic-radical-width/verify.py +/Users/nick/git/integer-mult-bounds/.venv/bin/python3.12 -B scripts/research_package.py --package research/quadratic-hyperbolic-radical-width +/Users/nick/git/integer-mult-bounds/.venv/bin/python3.12 -B scripts/check_research_commit.py --package research/quadratic-hyperbolic-radical-width --check-source-history +``` + +Other CPython 3.12 executables can replace the local interpreter path. The +standalone verifier is read-only and Git-free. It emits a generated JSON +receipt bound to all six package-file hashes and the marker +`PASS_QUADRATIC_HYPERBOLIC_RADICAL_WIDTH`. + +The finite domain is every width from zero through six and all eight refinements. +The literal row enumeration visits 37,450 even-vector triples, including zero +and dependent triples. A second enumeration visits all 3,003 column multisets +and weights them by their exact multinomial multiplicities. Their complete +phase-by-width embedding counts agree. There are 3,744 accepted ordered +embeddings across these widths. The first positive width for each phase equals +the proved minimum. Neither search prunes on the claimed answer. + +All eight displayed witnesses pass every vector phase and every ordered vector +pairing. Twelve controls reject lost phases, collapsed radicals, wrong Gram +data, odd images and malformed inputs. Explicit checks stay active under +optimized Python. Receipts are computed, never handwritten. + +The selected baseline binding command passed as source/metadata orientation. +No predecessor mathematical verifier, selected cold replay, native/caller +program, Lean or Docker was run. Independent human and formal proof-kernel +review are unclaimed. Hosted CI is recorded separately in the contribution PR. + +## Scope and source credit + +This contribution was produced by Nick Petalas. Apache-2.0. +Credit Nick Petalas's PR6456 for the Gram-only width problem and the explicit +phase-preserving first example; Maximus Putnam / Maximooch's PR831 for the +alternating trace and phase-interface motivation, with the retained Joel +Pulikkan / GamingPuzzled and endpoint-normal-form lineage; and THU New Media +Research Center / thu-nmrc's PR5464 for the related degenerate embedding +background retained there. These are cited premises and context. No predecessor +executable is imported. Binary polarization and even-weight code arguments +are classical; no literature-priority claim is made. + +[SOURCES.json](SOURCES.json) pins the exact predecessor README/proof, catalogue +and selected-record provenance at canonical base +`e54523517d0f03e58352e647c403a5af3e525a57`. The historical card is +`complex-rank-shape-decoder`, priority revision 8. [TASK.json](TASK.json) +records the change from complete decoder accounting to this geometric +subproblem. The selected record remains #7000 with its five assumptions. +The [advance claim](https://github.com/CrocSwap/integer-mult-bounds/discussions/1166#discussioncomment-19571816) +has the same mathematical scope. + +Live claims, account PRs, open/merged searches and canonical main README +searches were checked. The overlap helper reported incomplete sparse local +coverage and no usable cached Discussion snapshot; live GraphQL reads and +direct main searches supplemented it. Arbitrary historical diffs, unmaterialized +nested notes and unpublished work remain outside that coverage. + +## Verification bounds + +Every verifier function is shorter than sixty physical lines. There is no +recursion, subprocess, network, dynamic code generation or artifact write. +Widths are fixed at zero through six; row points number at most 32, column +tuples contain at most six entries, spans contain at most eight vectors, and +the receipt reads exactly six files of at most 128 KiB each. All loops have +finite bounds from those domains. Functions have direct calls and scoped local +state. Explicit argument and invariant checks remain enabled with `-O`. + +Power-of-ten rule 3 has a managed-runtime deviation: CPython allocates integers, +sets, lists and JSON objects during verification. The fixed domains above bound +live resources; fixed allocation is not claimed. Warning-strict normal and +optimized execution and source AST inspection are the available diagnostics. +No repository-wide static-analysis configuration change is included. + +## Open next questions + +1. Determine the exact phase-preserving width for two hyperbolic blocks and one + radical direction. Separate the radical phase and the quadratic type of the + nonsingular part first. Higher-dimensional embedding bounds are missing. +2. Instantiate one minimum embedding in an actual source/read interface with + specified phases and chronological frames. Native delivery, dirty restoration, + owner allocation, precision and a complete signed bill are still blockers. + +Structured first steps and blockers are in TASK.json. No follow-on scope is +reserved. Arbitrary binary forms and larger radicals are outside this result. diff --git a/research/quadratic-hyperbolic-radical-width/SOURCES.json b/research/quadratic-hyperbolic-radical-width/SOURCES.json new file mode 100644 index 0000000000..6c815382ea --- /dev/null +++ b/research/quadratic-hyperbolic-radical-width/SOURCES.json @@ -0,0 +1,30 @@ +{ + "schema_version": 1, + "base_commit": "e54523517d0f03e58352e647c403a5af3e525a57", + "files": [ + { + "path": "certificates/selected-result.json", + "usage": "provenance", + "git_blob": "75d77b328ea02ded535e7f23aee5b1e55d0bb1a8", + "sha256": "154c3d30455d778fd3343b6b4851f0fa4665f0f7625b0adb3675e2bafbbf6bf1" + }, + { + "path": "docs/research/agent-task-catalog.json", + "usage": "provenance", + "git_blob": "3404fbdd5e7529a0674eef26114b5071e4586cec", + "sha256": "ead6cdc9d0af6d58e8a49744d8ec8c7c15e13d2ba93c276c434bcf9b51b93a81" + }, + { + "path": "research/faithful-alternating-width/PROOF.md", + "usage": "input", + "git_blob": "e07287c5e124c056ad8b2ed60badaeb47217a2c8", + "sha256": "3c8755deee7f9aad54226962dcb119ae1b62bd6e1308dd6a080ed98a432f77eb" + }, + { + "path": "research/faithful-alternating-width/README.md", + "usage": "input", + "git_blob": "84d8e856ef08a6b49dcab9d1234b1b8fe941b8ea", + "sha256": "0330190b71c3f61c9fd8258acc1825ed415b59614e22ab6db75cbe2215cea6b0" + } + ] +} diff --git a/research/quadratic-hyperbolic-radical-width/TASK.json b/research/quadratic-hyperbolic-radical-width/TASK.json new file mode 100644 index 0000000000..5fa57c446c --- /dev/null +++ b/research/quadratic-hyperbolic-radical-width/TASK.json @@ -0,0 +1,87 @@ +{ + "schema_version": 1, + "task_id": "complex-rank-shape-decoder", + "priority_revision": 8, + "base_commit": "e54523517d0f03e58352e647c403a5af3e525a57", + "original_card": { + "id": "complex-rank-shape-decoder", + "question": "Can a legal decoder or coframe change improve the full signed moment, including changed helper targets?", + "starting_points": [ + "docs/research/current-priorities.md", + "docs/research/research-family-status.md", + "certificates/selected-result.json", + "docs/research/pr-intake-evidence/pr5599-completion/PROOF-REVIEW.md", + "docs/research/recursive-invoice-preflight.md" + ], + "methods": [ + "Construct and independently check one source-bound instance of the stated first experiment.", + "Seek an exact counterexample or target-specific complete-budget obstruction to that same proposed change." + ], + "acceptance_gate": "A source-bound arbitrary-input identity and legal producer/consumer boundary with the complete paid and productive target invoice, including restoration, ownership, native alignment and extra charges. Price under the actual current model; historical ceiling.py assumptions must be established before use. A favorable optimistic relaxation keeps a route open but is not a completed supplier. A scoped local result remains useful; expansion requires the family-level budget gate.", + "first_step": { + "command": [ + "python3", + "-B", + "scripts/research_selected_baseline.py" + ], + "expected": "PASS_RESEARCH_BASELINE_BINDINGS_ONLY", + "inspect": "Read revision 8, the selected review and live exact-source handoffs. Then follow this card's bounded first_experiment.", + "scope": "Source-bound orientation only, not a supplier replay or theorem proof." + }, + "first_experiment": "Price one complete changed itinerary with both paid and productive histograms. Check #5693/#5698 for overlap and their exact hypotheses before another coframe search. Use the signed invoice preflight where its boundary contract agrees; an inconclusive concave comparison is not a power crossing. Stop if an optimistic complete budget fails at the chosen target or native delivery remains unspecified.", + "bottleneck_link": "Judge the actual current supplier contract and complete signed target/paid bill; frozen-profile ceilings require an applicability proof.", + "allowed_variations": [ + "One exact source head, helper/cut region or quantified proposition and changed hypothesis; check live handoffs first. Different seeds are not different scopes." + ], + "scope_change_triggers": [ + "Changing source head, helper/cut domain, operation alphabet or caller boundary without recording the new scope.", + "Treating an archived proposal or local numerical pass as complete supplier admission." + ], + "overlap_terms": [ + "coframe", + "signed moment", + "5693", + "5698" + ], + "existing_checks": [ + { + "scope": "Selected PR5599: complete helper-subspace construction and independent caller replay; five framework assumptions remain.", + "evidence": "docs/research/pr-intake-evidence/pr5599-completion/PROOF-REVIEW.md", + "remaining": "The next experiment must identify uncovered source-specific work. Successor archive intake does not transfer this admission.", + "status": "reviewed", + "source_commit": "09dd0965711ca7eeb169057a420abab1c9052137", + "selected_record_sha256": "2a0220e866417774606e0f991e41b9c1d26b1f43b97857238cf478dbc7d330b5" + } + ] + }, + "scope_status": "changed", + "actual_scope": "Exact minimum faithful phase-preserving binary width for H direct sum 0 in dimension three, for all eight quadratic refinements.", + "scope_change_reason": "This is the first working example explicitly left open by PR6456. It adds quadratic phases to its Gram-only contract; it is a geometric interface result rather than a complete decoder signed-moment improvement.", + "bottleneck_link": "Native phase-compatible labels must preserve weight modulo four as well as Gram pairings. The exact extra width is zero, one or two coordinates on this first degenerate example.", + "budget_minutes": 60, + "stop_rule": "Ship this first exact eight-case result; do not extend to larger forms.", + "outcome": "completed", + "evidence": "PROOF.md and standalone dual census through width six, explicit minimum embeddings and rejection controls.", + "author": "Nick Petalas", + "ai_assistance": "Independent human and formal proof-kernel review are unclaimed.", + "open_questions": [ + { + "question": "What is the exact phase-preserving width for two hyperbolic blocks plus one radical direction?", + "why_it_matters": "The first example shows that Gram-only minima can miss two physical label coordinates.", + "first_step": "Separate radical phase zero and one, then classify the two quadratic types on the nonsingular part.", + "blocked_by": "This package proves only the three-dimensional case; higher-dimensional quadratic embedding bounds are absent." + }, + { + "question": "Can an actual native source/read interface use one of these phase-preserving minimum embeddings?", + "why_it_matters": "A coordinate minimum can affect the supplier only through a legally realized and priced interface.", + "first_step": "Specify one exact three-dimensional label space, required quadratic phases and chronological input/output frames.", + "blocked_by": "Native delivery, dirty restoration, owner allocation, precision and a complete signed bill remain missing." + } + ], + "coordination": { + "claim": "https://github.com/CrocSwap/integer-mult-bounds/discussions/1166#discussioncomment-19571816", + "source_pr": 6456, + "source_author": "petalas", + "main_predecessor_author": "petalas" + } +} diff --git a/research/quadratic-hyperbolic-radical-width/VERIFY.json b/research/quadratic-hyperbolic-radical-width/VERIFY.json new file mode 100644 index 0000000000..2410ab7d18 --- /dev/null +++ b/research/quadratic-hyperbolic-radical-width/VERIFY.json @@ -0,0 +1,37 @@ +{ + "schema_version": 1, + "status": "ready", + "python": "3.12", + "requirements": [], + "inputs": [ + "research/faithful-alternating-width/PROOF.md", + "research/faithful-alternating-width/README.md" + ], + "commands": [ + { + "argv": [ + "{python}", + "-I", + "-W", + "error", + "-B", + "{package}/verify.py" + ], + "timeout_seconds": 60, + "checks": "Complete even-row and column-multiset censuses at widths zero through six; all eight minima and full witness phases/pairings; twelve rejecting controls; generated file-hash-bound receipt." + }, + { + "argv": [ + "{python}", + "-I", + "-O", + "-W", + "error", + "-B", + "{package}/verify.py" + ], + "timeout_seconds": 60, + "checks": "Repeat the same complete mathematical checks with optimized Python; explicit contracts stay enabled." + } + ] +} diff --git a/research/quadratic-hyperbolic-radical-width/verify.py b/research/quadratic-hyperbolic-radical-width/verify.py new file mode 100644 index 0000000000..d125792ac6 --- /dev/null +++ b/research/quadratic-hyperbolic-radical-width/verify.py @@ -0,0 +1,183 @@ +#!/usr/bin/env python3 +"""Exact phase-preserving widths for H plus one radical. Apache-2.0.""" + +import hashlib +import itertools +import json +import math +from pathlib import Path + + +def require(condition, message): + if not condition: + raise ValueError(message) + + +def dot(left, right): + return (left & right).bit_count() % 2 + + +def span(vectors): + values = {0} + for vector in vectors: + values |= {value ^ vector for value in tuple(values)} + return values + + +def combine(rows, value): + result = 0 + for index in range(3): + if value >> index & 1: + result ^= rows[index] + return result + + +def phase(phases, value): + return ((value & 1) * (value >> 1 & 1) + + sum(phases[index] * (value >> index & 1) + for index in range(3))) % 2 + + +def validate_embedding(phases, width, rows): + require(type(width) is int and 0 <= width <= 6, "width outside 0..6") + require(len(phases) == 3 and all(type(x) is int and x in (0, 1) + for x in phases), "invalid phases") + require(len(rows) == 3 and all(type(x) is int and 0 <= x < 1 << width + for x in rows), "invalid image rows") + require(all(row.bit_count() % 2 == 0 for row in rows), "odd image") + images = tuple(combine(rows, value) for value in range(8)) + require(len(set(images)) == 8, "noninjective image") + for value, image in enumerate(images): + require(image.bit_count() // 2 % 2 == phase(phases, value), + "quadratic phase mismatch") + for other, other_image in enumerate(images): + target = ((value & 1) * (other >> 1 & 1) + + (other & 1) * (value >> 1 & 1)) % 2 + require(dot(image, other_image) == target, "Gram mismatch") + + +def row_census(): + """Enumerate literal even-weight row triples, including zero rows.""" + counts = [] + candidates = 0 + for width in range(7): + table = [0] * 8 + points = tuple(x for x in range(1 << width) if x.bit_count() % 2 == 0) + for first, second, radical in itertools.product(points, repeat=3): + candidates += 1 + if not dot(first, second) or dot(first, radical) or dot(second, radical): + continue + if len(span((first, second, radical))) != 8: + continue + index = (first.bit_count() // 2 % 2 + + 2 * (second.bit_count() // 2 % 2) + + 4 * (radical.bit_count() // 2 % 2)) + table[index] += 1 + counts.append(table) + require(candidates == 37450, "incomplete row census") + return counts, candidates + + +def column_census(): + """Enumerate column multisets and count their labelled permutations.""" + counts = [] + candidates = 0 + for width in range(7): + table = [0] * 8 + for columns in itertools.combinations_with_replacement(range(8), width): + candidates += 1 + weights = tuple(sum(column >> bit & 1 for column in columns) + for bit in range(3)) + if any(weight % 2 for weight in weights) or len(span(columns)) != 8: + continue + pairings = tuple(sum((column >> a & 1) * (column >> b & 1) + for column in columns) % 2 + for a, b in ((0, 1), (0, 2), (1, 2))) + if pairings != (1, 0, 0): + continue + index = sum((weight // 2 % 2) << bit + for bit, weight in enumerate(weights)) + multiplicity = math.factorial(width) + for value in range(8): + multiplicity //= math.factorial(columns.count(value)) + table[index] += multiplicity + counts.append(table) + require(candidates == 3003, "incomplete column census") + return counts, candidates + + +def minimum_and_witness(phases): + first, second, radical = phases + if radical: + return 5, (15 ^ (3 * first), 23 ^ (3 * second), 3) + if first == second == 1: + return 4, (3, 5, 15) + if first: + return 6, (24, 23, 57) + if second: + return 6, (15, 24, 57) + return 6, (15, 23, 57) + + +def rejection_controls(): + fixtures = ( + ((0, 0, 0), 4, (3, 5, 15)), + ((1, 1, 1), 4, (3, 5, 15)), + ((1, 1, 0), 4, (3, 5, 0)), + ((1, 1, 0), 4, (3, 5, 6)), + ((1, 1, 0), 6, (3, 12, 39)), + ((1, 1, 0), 4, (3, 5, 7)), + ((0, 0, 0), 5, (15, 23, 57)), + ((0, 0, 0), 7, (15, 23, 57)), + ((0, 0, 2), 6, (15, 23, 57)), + ((0, 0), 6, (15, 23, 57)), + ((0, 0, 0), 6, (15, 23)), + ((0, 0, 0), 6, (15, 23, -1)), + ) + for phases, width, rows in fixtures: + try: + validate_embedding(phases, width, rows) + except ValueError: + continue + raise ValueError("accepted a rejecting fixture") + return len(fixtures) + + +def file_hashes(): + root = Path(__file__).resolve().parent + result = {} + for name in ("README.md", "PROOF.md", "TASK.json", "SOURCES.json", + "VERIFY.json", "verify.py"): + with (root / name).open("rb") as source: + data = source.read(131073) + require(len(data) <= 131072, "receipt file exceeds 128 KiB") + result[name] = hashlib.sha256(data).hexdigest() + return result + + +def main(): + rows, row_candidates = row_census() + columns, column_candidates = column_census() + require(rows == columns, "independent census disagreement") + results = [] + for phases in itertools.product(range(2), repeat=3): + index = sum(value << bit for bit, value in enumerate(phases)) + minimum, witness = minimum_and_witness(phases) + observed = next(width for width in range(7) if rows[width][index]) + require(observed == minimum, "incorrect minimum") + validate_embedding(phases, minimum, witness) + results.append({"phases": phases, "minimum": minimum, + "witness": witness, + "embeddings_by_width": [table[index] for table in rows]}) + receipt = {"result": "PASS_QUADRATIC_HYPERBOLIC_RADICAL_WIDTH", + "row_candidates": row_candidates, + "column_multisets": column_candidates, + "ordered_embeddings": sum(map(sum, rows)), + "minimum_cases": len(results), "results": results, + "rejection_controls": rejection_controls(), + "package_sha256": file_hashes()} + print(json.dumps(receipt, indent=2, sort_keys=True)) + + +if __name__ == "__main__": + main()