Elamur GroupantsโŒ•
โ˜ฐ

court-native-fixtures

The groupant beside yom-hadin/tests/court.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
astpartialsyntax.json
costpartialmodel.md
descriptantpresentdescriptant
invariantpartialInvariant
testsfinite native checkstests

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
independent_count_oracleindependent_count_oraclepartial source derived modelindependent_count_oracle
fixture_interventionfixture_interventionpartial source derived modelfixture_intervention
repeatability_observationrepeatability_observationpartial source derived modelrepeatability_observation
source_mutation_controlsource_mutation_controlpartial source derived modelsource_mutation_control

Model text

descriptant

Tiny hand-built rule/case integration witnesses using actual crate APIs.

Epoch 2026-10-31

Prospective review preceded test edits. Runtime logic unchanged; three sources gain tests. Synthetic witnesses do not validate legal conclusions.

Composition: Construct deterministic Clauses/Rules/Jurisdictions/Facts without parser models. Existing tests cover empty match, path count changes, broad reading, conflict resolution, report determinism, flag filtering, stats, devices and absurd criteria. Wave31 adds explicit count enumeration, tie/zero-tip boundaries, postprocessing/metrics, codec/status fibers and lookup order.

Retained limits: partial syntax, opaque macros; no live corpus, teacher, parser model, production filesystem or frozen gold. Complete semantics and executable approximants remain open.

approximant

Forward model

Tiny hand-built rule/case integration witnesses using actual crate APIs.

Construct deterministic Clauses/Rules/Jurisdictions/Facts without parser models. Existing tests cover empty match, path count changes, broad reading, conflict resolution, report determinism, flag filtering, stats, devices and absurd criteria. Wave31 adds explicit count enumeration, tie/zero-tip boundaries, postprocessing/metrics, codec/status fibers and lookup order.

Not an executable substitute. Percentage approximation is unsupported without a useful metric; future profiles limited to 1/2/5/10% require actual-reference comparison.

models

Typed models: Tiny hand-built rule/case integration witnesses using actual crate APIs.

Subject: yom-hadin/tests/court.rs, SHA-256 2b4c2a041025b717d7518564a83ef2b98d6f3b97cf66a1298361fabc9cf5df41. Domain: bounded finite synthetic rules/cases; valid indices and successful allocation unless a section narrows this. Each relation observes implementation, not legal authority. Status: partial source-derived model, finite evidence mapped in tests.md.

behavior_complement

The selected test traces cannot query teacher, parse a corpus or write production rules because fixtures call in-memory helpers/report only. A passing test does not exclude bugs outside asserted observations.

Operator: stated relation on this source projection. Assumptions: named domain and unchanged dependencies. Evidence: tests.md maps finite observations; other branches remain source review only.

reverse_reconstruction

A named pass constrains implementation only at executed fixture/assertion points; all implementations agreeing there are compatible. Test result alone cannot reconstruct the fixture values or certify whole APIs.

Operator: stated relation on this source projection. Assumptions: named domain and unchanged dependencies. Evidence: tests.md maps finite observations; other branches remain source review only.

independent_count_oracle

wave31 enumerates every mixed-radix assignment and calls evaluate, aggregates verdicts, compares count() output and total. It tests count/evaluate correspondence, not independent correctness of evaluate semantics.

Operator: stated relation on this source projection. Assumptions: named domain and unchanged dependencies. Evidence: tests.md maps finite observations; other branches remain source review only.

fixture_intervention

Existing shuffled-rule and removed-condition comparisons alter fixture inputs, not compiled production source. They demonstrate sensitivity but are counted separately from compiled source mutants.

Operator: stated relation on this source projection. Assumptions: named domain and unchanged dependencies. Evidence: tests.md maps finite observations; other branches remain source review only.

repeatability_observation

Exact selected names run twice debug and twice release, sorted named outcomes compared. Report determinism is also asserted in-process; elapsed timings are excluded and no throughput gain inferred.

Operator: stated relation on this source projection. Assumptions: named domain and unchanged dependencies. Evidence: tests.md maps finite observations; other branches remain source review only.

source_mutation_control

Mutating count weight or precedence tie in actual court source and modal tie in report should compile but fail chosen assertion. Restore exact bytes/hash then rerun; compile failure alone is not a successful negative control.

Operator: stated relation on this source projection. Assumptions: named domain and unchanged dependencies. Evidence: tests.md maps finite observations; other branches remain source review only.

tests

Checks and evidence

External run label: octopus-wave31-court-2026-10-31; exact commands in native-commands.json, per-source observations in source-test-map.json, numeric exits and named outcomes in raw logs.

19 exact selected native tests twice debug and twice release execute actual linked code on bounded hand-built cases. CLI covers argument helpers only. Full extraction/model initialization, actual jurisdiction loading, protocol disk errors, resource exhaustion and malformed public lattices are unexecuted. Module declarations have compile evidence only.

Three compiled source-behavior mutants: count weight, precedence tie, modal tie; each requires compile success, named assertion failure, exact byte/hash restoration and positive rerun. Legacy fixture interventions count separately. Final observations: external evidence-summary.md.

This node observes: Tiny hand-built rule/case integration witnesses using actual crate APIs.. Unmapped models remain source-derived/prospective, not verified by the whole test count.

The exact per-source names and limits are retained in source-evidence.md.

invariant-and-cost

Invariant and cost

Invariant

Assertions execute actual linked implementation; fixtures contain at most a few rules/points and explicit count enumeration<=128. No corpus or frozen gold is loaded. Passing finite cases is not a universal gate.

Cost

Fixture allocation linear in tiny literal sizes, court algorithm costs remain as modeled. Explicit assignment oracle visits product(option sizes)<=128; reports/tips use small bounded depth.

Symbolic bounds, not timings. Resource exhaustion is not tested.

source-evidence

Source observation map

Actual test harness using linked source, explicit enumeration is independent of count but shares evaluate semantics.

This page in Markdown โ†’ https://md.groupants.elamur.ai/free/en/mova/yom-hadin/tests/court.rs.groupant/