Elamur Groupantsanswer-span-why
The groupant beside world/src/span.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 source model | forward |
| ast | partial | ../../syntax.json |
| cost | partial source model | cost |
| descriptant | partial source model | epoch-memory |
| invariant | partial source model | invariant |
| tests | partial source model | checks |
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 |
|---|---|---|---|
answer-span-why-behavior-complement | behavior_complement | partial source model | behavior-complement |
answer-span-why-reverse-reconstruction | reverse_reconstruction | partial source model | reverse-reconstruction |
answer-span-why-chain-bound | resource_boundary | partial source model | chain-bound |
answer-span-why-reason-order | priority_relation | partial source model | reason-order |
Model text
Subject: world/src/span.rs, whole-file SHA-256 in manifest, baseline 4279da541. Domain: finite acyclic valid trees with consistent child lists, valid indices and terminating dependencies. Observation: a single call return, not semantic truth or reachability. Hold Tree, ans::lemmas and types::FEELING dependencies fixed; parser/model loading and external caller wiring are gaps. All entries are source-derived partial models. Exact-target definitions are not executable inverses; finite tests cannot establish universal correspondence.
forward
Bounded ancestor scan, ordered structural reason patterns, then token fallbacks.
invariant
WHY_MARK matches any mark word; When uses first mark word.
cost
At most three chain scans, subtree/clean plus O(n) lexical scan.
epoch-memory
2026-10-29: initial source-bound epoch. Preserve Num bypass, global Feel selection, early-return cleanup suppression, and verb_kids comment mismatch: code expands Xcomp only and checks no shared subject. No source changes or task-accuracy claim. Partial syntax has no semantic analysis or macro expansion.
checks
Structural inventory checks hashes, references, types and node bounds only. Proposed oracle: enumerate small annotated acyclic trees and compare literal expected returns, including None and labels; mutate priority or echo inequality to obtain a red control. Any executed bounded evidence is recorded separately in root evidence.md; catalog examples remain proposed until recorded there.
composition
Leaf scope shares immutable tree/question/helper contracts. Overlap with parent is not extra runtime work. Dependencies outside range remain opaque.
behavior-complement
U=valid Want::Why calls x optional output; R=Why branch. Inner U minus R excludes Some when p absent and later-rule outputs after earlier matched return. No nearby Advcl is insufficient exclusion because other patterns/fallbacks exist.
Assumptions/evidence/status: common domain above; source-derived partial relation at bound revision. Approximation and held-fixed inputs are stated in the relation. Oracle: finite literal fixtures exercising both membership and nonmembership against the actual function, preserving ambiguous preimages. No linguistic correctness or universal proof inferred.
reverse-reconstruction
advcl:because admits all chains with first qualifying outside-chain Advcl, stripped direct Mark/Cc then clean. words: because Y admits every earlier p, first qualifying post-p because/for/since with UPOS!=ADP and punctuation-bounded suffix. Deleted markers/context ambiguous.
Assumptions/evidence/status: common domain above; source-derived partial relation at bound revision. Approximation and held-fixed inputs are stated in the relation. Oracle: finite literal fixtures exercising both membership and nonmembership against the actual function, preserving ambiguous preimages. No linguistic correctness or universal proof inferred.
chain-bound
p plus two head attempts through Xcomp/Ccomp/Obj/Conj/Advcl/Nsubj/Obl yields at most three chain nodes. No further structural ancestor scan; changing distant words may still affect lexical fallback.
Assumptions/evidence/status: common domain above; source-derived partial relation at bound revision. Approximation and held-fixed inputs are stated in the relation. Oracle: finite literal fixtures exercising both membership and nonmembership against the actual function, preserving ambiguous preimages. No linguistic correctness or universal proof inferred.
reason-order
Per near-to-far node: WHY_MARK Advcl, later for/because Conj, later to-purpose, selected oblique cases, so-result head-minus-x, that-result head-minus-x. Then lexical X,so and because-Y. Ordered syntax heuristics are not causal entailment.
Assumptions/evidence/status: common domain above; source-derived partial relation at bound revision. Approximation and held-fixed inputs are stated in the relation. Oracle: finite literal fixtures exercising both membership and nonmembership against the actual function, preserving ambiguous preimages. No linguistic correctness or universal proof inferred.
This page in Markdown → https://md.groupants.elamur.ai/free/en/mova/world/src/span.rs.groupant/extraction.groupant/why.groupant/