Elamur Groupants⌕
☰

math-suite-verdicts

The groupant beside math/src/suite.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
compositionpartialcomposition
descriptantpresentdescriptant
invariantpartialinvariant
testspartial bounded evidencetests

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
failure_precedencefailure_precedencepartial source derived modelfailure_precedence
representation_transportrepresentation_transportpartial source derived modelrepresentation_transport
resource_boundaryresource_boundarypartial source derived modelresource_boundary
domain_boundarydomain_boundarypartial source derived modeldomain_boundary

Model text

descriptant

Source-bound description

Trim expected, accept NU/ERR:* code, optional≈ then number/unit; scale before monomial. Exact output/nonapprox expectation exact equality else tolerance1e-6*max(abs(want),1e-12), plus unit equality. NU expected Hit else HonestNu. run requires3fields and category id prefix before '-'.

Epoch 2026-10-31: prospective model before bounded synthetic tests; production behavior unchanged. judge/run model-backed paths unexecuted.

approximant

Forward relation

Trim expected, accept NU/ERR:* code, optional≈ then number/unit; scale before monomial. Exact output/nonapprox expectation exact equality else tolerance1e-6*max(abs(want),1e-12), plus unit equality. NU expected Hit else HonestNu. run requires3fields and category id prefix before '-'.

Domain: Synthetic expected parsing only. Approx zero tolerance1e-18; no model loaded.

Executable Rust approximant unsupported: categorical/effect observations have no useful percentage metric yet. Future caps1/2/5/10% only; no automatic substitution.

invariant

Invariant and cost

Wrong unit excludes Hit; unexpected NU becomes HonestNu not Wrong. ERR:* parse acceptance does not ensure engine emits that code.

Ordering: Width→Want→understanding→verdict. Number parse before scale/monomial.

Cost: O(TSV+rows)+understanding/eval; unit sort O(u log u); unchecked usize counters.

composition

Composition and epoch memory

CLI slots use enum [Hit,HonestNu,Wrong]; monomial sort normalizes ordering, not dimensional equivalence.

judge/run model-backed paths unexecuted.

Partial syntax and finite tests do not establish all caller assumptions.

models

behavior_complement

Wrong unit excludes Hit; unexpected NU becomes HonestNu not Wrong. ERR:* parse acceptance does not ensure engine emits that code.

reverse_reconstruction

Totals admit all row sets with same category/verdict counts. Equivalent rational spellings lose formatting. HonestNu constrains error kind not unique question/model.

failure_precedence

Width→Want→understanding→verdict. Number parse before scale/monomial.

representation_transport

CLI slots use enum [Hit,HonestNu,Wrong]; monomial sort normalizes ordering, not dimensional equivalence.

resource_boundary

O(TSV+rows)+understanding/eval; unit sort O(u log u); unchecked usize counters.

domain_boundary

Synthetic expected parsing only. Approx zero tolerance1e-18; no model loaded.

tests

Evidence, 2026-10-31

Selected actual-source checks: cargo test -p math --lib suite::wave28. Exact commands, numeric statuses, normalized outcomes, compiled assertion controls and restoration receipts are in the external wave28 mathcontrol run.

judge/run model-backed paths unexecuted.

No live teacher/API, trained model, corpus/held-out, GPU, DB or expensive benchmark run. Parent syntax generated twice identically. Scope remains partial.

This page in Markdown → https://md.groupants.elamur.ai/free/en/mova/math/src/suite.rs.groupant/