Elamur Groupantskg-textgen-realization
The groupant beside kg/src/textgen.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 | composition |
| cost | partial | cost |
| descriptant | present | descriptant |
| invariant | partial | invariant |
| tests | executed 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 |
conditional_priority | conditional_priority | partial source derived model | conditional_priority |
pronoun_pool | pronoun_pool | partial source derived model | pronoun_pool |
modifier_order | modifier_order | partial source derived model | modifier_order |
punctuation | punctuation | partial source derived model | punctuation |
Model text
descriptant
kg-textgen-realization
Form pool choice, recursive modifier chunks and final surface text.
Modeling epoch — 2026-10-31
Source read before tests and models; retained prospective notes accompany the parent. Current evidence uses bounded invented inputs only. This is partial modeling, not proof or corpus accuracy.
approximant
Forward relation
Strip NEG/IMPLICIT synthesis flags, choose lexicographic frequency winner first-on-tie, restrict н-pronouns after prep; build adjectives/head/nonadj children; order chunks and add dash/case/punctuation.
Executable approximant gap
Exact categorical structure, identifiers, parser decisions and effects have no useful percentage metric established here. No executable surrogate, automatic substitution or 1/2/5/10% guarantee is claimed.
invariant
Invariant and domain
Used modifier flag prevents duplicate attachment; depth<=4,1..12 heads admitted. Orphan modifiers can remain unused.
Cost and resources
Variant scans plus member scans per chunk, bounded recursion then DP; retained generated strings.
cost
Cost and resources
Variant scans plus member scans per chunk, bounded recursion then DP; retained generated strings.
composition
Composition
Form pool choice, recursive modifier chunks and final surface text.
Strip NEG/IMPLICIT synthesis flags, choose lexicographic frequency winner first-on-tie, restrict н-pronouns after prep; build adjectives/head/nonadj children; order chunks and add dash/case/punctuation.
The parent invokes scoped transformations through the actual functions named in their source ranges. Children partition useful observations, not a complete semantic proof; boundaries can omit helpers, tests and effects.
models
behavior_complement
No synth/depth5/bad head count→None; no н pool→ordinary variants remain allowed.
reverse_reconstruction
Text admits multiple morphological variants/count histories/Facts; implicit heads and original word coordinates do not appear in output.
conditional_priority
A conditional1/global0 beats conditional0/global8; zero-zero tie chooses dictionary first.
pronoun_pool
After prep+PRON+person3+any н, every non-н option excluded irrespective of frequency.
modifier_order
PRON adjectives first, ordinary adjectives next, head then recursive nonadjectives by member order.
punctuation
Gram0 noun subject/predicate pair gets dash unless pronoun; IMPR ! else period; initial Unicode char uppercased.
tests
Evidence
Parent source checks (a fragment inherits only the behavior they actually exercise):
- textgen::tests::subject_verb_object
- textgen::wave29::variant_priority_and_head_bounds
Selected kg tests: 15 named outcomes, twice debug/twice release, equal sorted outcomes. Standalone actual global/build.rs: 3 named outcomes per run, twice debug/twice optimized; build_child is normally no-op, parent fixture invokes it against isolated synthetic seeds. Eight compiled source mutations (one per source) each compiled successfully and exited101 on the focused assertion; exact source restoration followed by positive rerun.
This fragment's untested branches remain source-derived. No live graph, DB, corpus, teacher, GPU or held-out evaluation was used. Command specs and numeric exits are retained in the external wave29 graphlang receipt. Strict kg Clippy reports existing source lints; see receipt.
This page in Markdown → https://md.groupants.elamur.ai/free/en/mova/kg/src/textgen.rs.groupant/realization.groupant/