Held to tests/vectors/slot-nil/reads.msgpack, written by the C# reference (SlotNilVectorTests): every accepted row
rewrites to the reference's bytes, nil in an int slot is refused. 94a26431c000c0 round-trips byte-identical.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
The cultcache-ts README, the DocumentPayload comment (comment only) and the cultmesh-browser README say plainly what the
opaque type guarantees and what it does not; documentPayloadBytes is the escape for bytes.
All commits build.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
The put is a union on `decoded`: true carries the document read through the registry's definition, false carries the
opaque DocumentPayload, so the owner read on the undecoded branch compiles without a cast (compile fixture in
payload-type.test.ts) and the three test casts are gone. A store holding a foreign parsing record, a matched record that
does not parse and a matched record that parses must return the last, which kills the survivors on the schema-matched
second find in resolveCultMeshStoreDocumentRecord.
All commits build.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
The put handler no longer plain-decodes such a payload, and does not refuse it either: five
existing server tests put through an empty registry, a catch-all server is a capability a general
library keeps, and the payload type already stops a plain read. CultMeshRudpDocumentPut.payload
documents the two cases; a handler that holds the definition reads it with decodeDocumentPayload.
Builds in every package; suites run at the end of the pass.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
DocumentPayload is an opaque object type over the same runtime Uint8Array, keyed by a string
property, and types CultCacheEnvelope.payload, CultNetRawDocumentRecord.payload and
CultMeshRealtimeFrame.payload. decodeDocumentPayload(definition, payload) and
encodeDocumentPayload(definition, value) are the only reader and writer; documentPayloadFromBytes
and documentPayloadBytes are the two named escapes at the edges where bytes enter and leave.
resolveDocumentFormatter leaves the package index. The cache, cultnet replication, the mesh read
paths, the provider command (now a CultCache document), and the browser's decodeCultNetPayload
(definition required, cultcache-ts/document entry) all go through the owner.
Compile-fail fixtures per payload-carrying package pin the type with expect-error lines.
check-typescript-tests-wired.mjs fails when a test file is run by no script; cultcache-ts now
runs dist-test/test/*.test.js, and the portability workflow runs the check and the cache suite.
Builds in every package (tsc, src and tests); the suites run once, at the end of the pass.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
The two payload-decode-owner text-scan tests go, as do the string-schemaId overloads and
branches of the cultMeshDocumentFrom* functions and statics, decodeCultMeshPayload,
decodeCultNetRawDocumentPayload, toUint8Array, replication's second decodeDocumentValue, and the
undefined branch of decodeDocumentPayload. decodeDocumentPayload now requires the definition and
also runs its schema (the one reader). A publication binding names a definition, never a bare
schema id; the RUDP put handler refuses a schema with no registered definition instead of
plain-decoding it. Builds in every package; tests not yet rerun.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
The guard test counted its package root from the compiled layout and failed under tsx and for a one-file src; it now walks up to
the package.json. The browser's non-messagepack refusal and the cache's codec-and-formatter refusal are pinned by message.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The RUDP put handler and the browser payload reader decoded a codec document with plain msgpack. cultcache-ts now
exports decodeDocumentPayload (definition or plain), the RUDP server reads a put through the registry's definition,
and decodeCultNetPayload takes the formatter. cultmesh-ts and cultmesh-browser no longer depend on @msgpack/msgpack:
frames and commands use cultnet-ts/wire, and a test per package scans src and the manifest. The envelope probe fixture
now runs the plain record first and last. Not yet built or tested on this commit; the verify job follows.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
cultmesh-ts reads document payloads through resolveDocumentFormatter at its five plain-msgpack sites; WireFloat carries a signalling flag and the float32 leaf refuses it; slots() refuses a key above 65535 at definition. The C# vector writer gains the 2^24+1 reads row and float32-nan.msgpack, whose committed files land in the next commit: this commit does not build its tests (union-parity numeric vectors, row count 160 and float32-nan.msgpack are absent).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
An odd kept value just past exact pins the tie rule (the Stryker survivor). The excess <= 0 guard was equivalent to the
general path for the only values that reach it, so it is deleted.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- cultnet-ts replication calls cultcache-ts's resolveDocumentFormatter (now exported) and loses its private
@msgpack/msgpack default, so a raw put and a snapshot of a codec document carry the codec's bytes.
- the hand-written numeric table is deleted; the tests read tests/vectors/numeric-leaves.
- an integer past 2^53 in a float32 slot rounds once (ties to even), not through a double.
- a union arm key is any Int32 (negative too, sorted numerically); a slot key is 0 to Int32 max.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The table is the reference's own answer for 120 one-slot reads and 46 writes, from a job-only probe of src/.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The codec path owns a small MessagePack reader and writer so it can tell an integer from a float and keep 64-bit
integers whole. Leaves int32, int64, float32, float64 replace number; every codec carries the reference's type name and
a slot record derives its catalog members from them. Not built or run at this commit.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A bound is released only when its violation exceeds both kktTol and the rounding
bound of its double gradient (Higham gamma_(n+1)); the cost-progress rule now judges
only passes that release nothing, and a release resets fMark. Fixes a cold start
stopping Converged with a violated bound (Soul W5, finding
aetheria-release:finding:cut-thrust-allocator-core.s1.solver-converged-early).
Tests: the W5 problem (cold == re-solve at 1e-12 and 1e-9) and the seed-141 noise
case. Builds; verified on Yggdrasil in the following report.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Stale logs are deleted before launch; the results file must be newer than the launch. All commits build (no source change).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
packages/cultmath/scripts/unity-smoke/ holds the good compute shader, the
deliberately broken control, the EditMode test (FindKernel, dispatch, readback),
and run-unity-smoke.ps1, which builds a scratch project outside the repo with a
file: path to the package, takes the Unity path as a parameter, requires all
three tests to pass, and deletes the project. The package README's Unity section
names it as the pre-tag step. Not -nographics: GetComputeShaderMessageCount
reports 0 for a broken shader until FindKernel or Dispatch compiles it. Ran once
on Starfire with Unity 6000.3.24f1: 3 passed, control rejected. Builds.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Asserts per output (value, and each snoise_grad component) that it differs
from 0.3.0 on the tie set, and pins the 0.4.0 values as goldens beside the
0.3.0 ones, so reverting either math.cs:614 or :720 alone fails the test. The
comment says inputs within a few float32 ulps of the set can tie too. Not yet
run when committed; the next commit-free verification run follows.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
CHANGELOG and package README: the noise enclosures rest on SNOISE_LIPSCHITZ,
SNOISE2_LIPSCHITZ and SNOISE_HESSIAN, all empirical with a margin; the README
bullet no longer says "with a proof" or "every point". Breaking: inputs within a
few float32 ulps of the tie set can also change (probe: none at 1e-5 or more,
a handful at 1e-6 near magnitude 30). Text only; builds.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
CHANGELOG and README no longer claim iv_snoise_ball and iv_fbm_ball enclose
every input unconditionally: they rest on SNOISE_LIPSCHITZ, an empirical bound
with a margin, and the tests sample. Docs only; builds.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The Breaking entry now names the whole tie set (inputs whose y-x and z-x are
whole numbers, every integer point included, change by up to 0.776), with
examples and a consumer warning; design.md and the NoiseGradTests comment drop
the narrower origin-and-(t,t,t) wording. Docs and test comment only; builds.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Inputs whose coordinate differences y-x and z-x are whole numbers move from the
0.3.0 value; inputs off that set keep its bits. The goldens were computed by
the cultmath-unity-v0.3.0 tag's tracked CultMath.dll. Builds.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Copied from artifacts/cultmath-unity-build of the release script's first run
(netstandard2.1, non-incremental, Source Link off, ContinuousIntegrationBuild).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Bumps CultMath.csproj, the Unity package.json, the org.gamecult.cultlib
dependency and the semver-policy paragraph to 0.4.0. CHANGELOG states the snoise
tie rule under Breaking and the additions since 0.3.0. The tracked DLL and pdb
still read 0.3.0 at this commit, so the release script's parity check fails here.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The release flips at 1.05x and 0.95x of the computed relative gradient at four
weights either side of 1e-6; all-free conditioned solves change between the
default and 1e-12; the omitted tolerance releases at 1.1e-6 and not at 0.9e-6.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Default and explicit-default are bit-identical; a light term under a heavy row
and an allocator-shaped problem are start-dependent at the default and solved
tight at 1e-12; invalid tolerances leave x untouched; the noise-floor fixtures
stop on progress at 1e-12 and 1e-30 (at most 6 iterations); the tolerance path
allocates nothing; the original ten-parameter signature still exists.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
New overload with kktRelativeTolerance before maxIterations; the existing
signature forwards with the new public DefaultKktRelativeTolerance (1e-6), which
replaces the private constant. A tolerance that is not finite and positive is
InvalidInput with x untouched. The default path is bit-identical.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Records status, iteration count and every float bit of x for the A54/B5 fixture,
three seeded random problems and a Duel allocator problem, taken at f7ab6212.
The kktRelativeTolerance cut must keep these literals unchanged.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Re-measured at c01ff025 with IntervalSkipHalvesEvaluations: (c) affine 3.57x and 3.46x, c-band 3.95x
and 3.54x, efficiencies and range-ceiling ratios as printed; the (b) rows and the DeepWellCullMap
figures (4.23, 3.60, 2.02) are unchanged. The sentence that justified deleting the affine detail
ball is gone with the deletion.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Deleting the affine gate's detail ball (51a4034c) gave up 6 of 5933 provable culls inside the fog
(first: (b) inside the fog [affine]); the gate never gives up a provable cull, and a 1.4% probe
saving does not buy breaking that. The detail ball is back in the affine gate, and the gate test
now includes the fog interior, the only field where the interval gate's detail ball decides, so
saturating either gate's detail ball fails the test.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
