Elamur Groupants⌕
☰

world-stepgame-lexical

The groupant beside world/src/stepgame.rs in the Mova repository: what the code computes, what it never does, how to go back from its outputs, its invariants, cost and checks.

Facts

Implementation material. A Mova code model describes this code at one source revision: partial models, finite checks and their stated scope. It is not a reviewed general exposition, and its labels are shown as written, never raised.

Components

componentlabelwhere
approximantpartialapproximant
astpartial../syntax.json
compositionpartialcomposition
costpartialCost
descriptantpresentdescriptant
invariantpartialmodel.md
testsbounded checks partial scopetests

Obernants

The typed relations around the forward model: the behaviour complement (what the code never does), the reverse reconstruction (every input compatible with an output) and specialised views.

idtypelabelwhere
behavior_complementbehavior_complementpartial source derived modelbehavior_complement
reverse_reconstructionreverse_reconstructionpartial source derived modelreverse_reconstruction
intensifierintensifierpartial source derived modelintensifier
ground_negationground_negationpartial source derived modelground_negation
samenesssamenesspartial source derived modelsameness
failure_precedencefailure_precedencepartial source derived modelfailure_precedence

Model text

descriptant

world-stepgame-lexical

Reject clock; collect direction signs except intensifier right; reject conflicting axis signs; zero direction needs sameness. Ground chosen unique cue then has/have subject then a2; negate if ground is a1. Axis relation A(X)=None if (∃x>0)∧(∃x<0), else first(X) or0; lexical(s)=σ·(A(horizontal),A(vertical)), σ=+1 for ground=a2 else−1, after clock/zero-cue guards.

Modeling epoch — 2026-10-31

Read full assigned source before editing; prospective predictions retained in external wave29 receipts. This child binds only its stated half-open UTF-8 byte range; other APIs remain parent scope. Evidence is limited to named bounded fixtures; registration is partial modeling. Production behavior unchanged; only inline synthetic tests added to four assigned sources.

approximant

Forward relation

Reject clock; collect direction signs except intensifier right; reject conflicting axis signs; zero direction needs sameness. Ground chosen unique cue then has/have subject then a2; negate if ground is a1. Axis relation A(X)=None if (∃x>0)∧(∃x<0), else first(X) or0; lexical(s)=σ·(A(horizontal),A(vertical)), σ=+1 for ground=a2 else−1, after clock/zero-cue guards.

Domain and resource limits

Finite valid structures and allocation-success executions under the branch conditions in models.md; integer overflow is outside measured fixtures. Exact categorical relations have no supplied percent surrogate. Macro expansion, arbitrary model quality and external effects are not inferred from syntax.

invariant

Invariant

Each component in {-1,0,1}; repeated directions do not increase magnitude.

Cost

O(B) tokens, O(T²) worst prefix walks; no weights/model required.

composition

Composition

This scope contributes lexical to its parent. Its forward, complement and preimage are restricted to the bound byte interval. The parent syntax projection is referenced once through ../syntax.json; no independent semantic AST is claimed.

models

behavior_complement

Exclude Some for clock/conflict/no-direction-no-sameness. Multiple ground cues fall back rather than rejecting.

reverse_reconstruction

Some(0,-1) admits downward synonyms and A right under B; None combines several distinct failure fibers. Surface form unrecoverable. Pre(None)=Clock ∪ AxisConflict ∪ {no directions∧no sameness}; the guards are ordered operationally but the result forgets the reason.

intensifier

right under gives below; right of keeps horizontal.

ground_negation

B has A on its left gives B-relative-A right.

sameness

same/coincid*/overlap*/share*/where/identical permit zero.

failure_precedence

Clock before conflicting signs before sameness fallback.

tests

Bounded checks

Named checks associated with this source (not every fragment branch):

17 unique named tests in total passed twice debug and twice release; sorted named outcomes equal across all four runs. Four compiled controls cover bAbI identity, StepGame tie ordering, parser-command state removal and bench-command CPU selection; each exits101 at a runtime assertion, exact source restored then positive exit0. Other sources have finite adversarial fixtures but no separately compiled mutant.

CLI integration coverage is command construction and temporary cleanup only; legacy school/model/cadence paths are not executed. No speed benchmark, GPU, live teacher, datasets or held-out replay.

Limits:

External receipt root: $MOVA_DATA/runs/octopus-wave29-worldeval-2026-10-31.

This page in Markdown → https://md.groupants.elamur.ai/free/en/mova/world/src/stepgame.rs.groupant/lexical.groupant/