Elamur Groupantscourt-absurd
The groupant beside yom-hadin/src/absurd.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 |
| composition | partial | descriptant |
| cost | partial | model.md |
| descriptant | present | descriptant |
| invariant | partial | Invariant |
| tests | finite native 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 |
severity_gap_threshold | severity_gap_threshold | partial source derived model | severity_gap_threshold |
reason_word_projection | reason_word_projection | partial source derived model | reason_word_projection |
status_counterfactual | status_counterfactual | partial source derived model | status_counterfactual |
flag_order_dedup | flag_order_dedup | partial source derived model | flag_order_dedup |
Parts
Model text
descriptant
Deterministic fixed-scale flags over rule paths and counterfactual cases.
Epoch 2026-10-31
Prospective review preceded test edits. Runtime logic unchanged; three sources gain tests. Synthetic witnesses do not validate legal conclusions.
Composition: Build lattice/honest path; inspect purpose via narrow reading and devices; inspect reachable severity gaps, unresolved principle ties, one-reading flips and six status swaps; sort descending score then kind/what and deduplicate.
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
Deterministic fixed-scale flags over rule paths and counterfactual cases.
Build lattice/honest path; inspect purpose via narrow reading and devices; inspect reachable severity gaps, unresolved principle ties, one-reading flips and six status swaps; sort descending score then kind/what and deduplicate.
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: Deterministic fixed-scale flags over rule paths and counterfactual cases.
Subject: yom-hadin/src/absurd.rs, SHA-256 31c4c2d448f44f622e075fc1cbeb3ace70372bb1782a048942a9bee026a085c7. 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
For admitted finite cases, excluded outputs include a status-only flag with equal compared severities, a disproportion flag with gap<3, or mutation of the caller case. This inner source-derived exclusion is about the fixed algorithm, not moral or legal validity.
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 flag output admits every case/jurisdiction whose actual lattice traces trigger the same criterion and rendered chain. Observing only kind/score forgets rules, witnesses and text, hence many compatible antecedents; no recovery of a unique real-world case.
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.
severity_gap_threshold
Given a reached winning verdict severity v>=1 and fixed harm h, gap abs(v-h) maps3->2,4->3,>=5->4; below3 no disproportion flag. Check harsh/fair synthetic fixture.
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.
reason_word_projection
reason_in_case splits semicolons then ': ', drops listed stopwords and accepts any Sure-covered verb/patient/argument value. Thus a single matching token suffices; it does not test all reason clauses.
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.
status_counterfactual
swap_status swaps agent/patient/colon argument values/persons in both directions, preserving verbs, places and text. With distinct words, applying twice is identity on the edited fields.
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.
flag_order_dedup
Output order sorts descending score, ascending kind, ascending what, then adjacent equal full Flag records dedup. Equal text with different chains can survive; no claim of one flag per semantic event.
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: Deterministic fixed-scale flags over rule paths and counterfactual cases.. 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
Flags refer to computed chains; score is a fixed heuristic in 2..4, empty flags produce lens score0. Input case and jurisdiction are cloned or borrowed unchanged.
Cost
Let L be lattice cost, T tipping cost, P points, D enabled devices, S present status pairs<=6. flags costs O((1+2D+S)L+T+P*E+F log F) plus strings; E is evaluate cost and F emitted flags.
Symbolic bounds, not timings. Resource exhaustion is not tested.
source-evidence
Source observation map
Actual flags and status swap; purpose/device flags not exhaustively covered.
- an_ordinary_proportionate_case_has_no_absurdity_and_a_disproportionate_one_has
- a_verdict_that_turns_on_status_alone_is_flagged
- wave31_codec_statistics_and_status_fibers
This page in Markdown → https://md.groupants.elamur.ai/free/en/mova/yom-hadin/src/absurd.rs.groupant/