roundtrip-report-ranking

The groupant beside tools/roundtrip/src/report.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

component	label	where
approximant	partial	approximant
ast	partial	`../syntax.json`
cost	partial	invariant-and-cost
descriptant	present	descriptant
invariant	partial	invariant-and-cost
tests	bounded checks	tests

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.

id	type	label	where
`behavior_complement`	behavior_complement	partial source derived model	behavior_complement
`reverse_reconstruction`	reverse_reconstruction	partial source derived model	reverse_reconstruction
`zero_denominator`	zero_denominator	partial source derived model	zero_denominator
`equal_set_weight`	equal_set_weight	partial source derived model	equal_set_weight
`rank_ties`	rank_ties	partial source derived model	rank_ties
`format_loss`	format_loss	partial source derived model	format_loss

Model text

descriptant

Equal-set ranking

2026-10-31 epoch: source scope bytes [12491,13785); Mean costs divide by total set count including zero-sentence sets; Meta rows omitted and remaining keys sorted by descending total_cmp plus key.

No annotator/model loading, live corpus, full suite, teacher, GPU or persistent-learning evidence. No executable percentage surrogate: categorical codecs/effects need a useful metric before future caps 1/2/5/10%.

approximant

Equal-set ranking

Mean costs divide by total set count including zero-sentence sets; Meta rows omitted and remaining keys sorted by descending total_cmp plus key.

Domain: bounded synthetic values, valid source-specific coordinates, nonoverflowing counts, available allocation; fixed environment controls. Relations are source-derived partial models with finite checks, not universal proofs.

catalog

behavior_complement

Subject: roundtrip-report-ranking, source hash ad92d6e8baee2f83f1fb90c2fcb7ca638e9312a525c7e3c37857525dd9584899, transition observation.

For Equal-set ranking, excluded transitions violate the stated scoped rule: Mean costs divide by total set count including zero-sentence sets; Meta rows omitted and remaining keys sorted by descending total_cmp plus key. This is an inner source-derived exclusion under the domain, not a complement of arbitrary unseen runtime behavior.

Evidence: external wave32 rt_core receipt includes native-commands.json, exact named outcomes twice debug/twice release, raw status/log files, compiled production controls with byte restoration and paired syntax. Only named synthetic helpers execute; model_path/export and live CLI paths have representation/source evidence.