Elamur Groupantsbibliography-value-tools
The groupant beside models/bibliography-value-tools.groupant 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 | forward |
| ast | not materialized | |
| composition | partial | composition |
| cost | symbolic source derived | cost |
| descriptant | present | descriptant |
| invariant | partial | invariant |
| tests | see source local evidence | 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 |
|---|---|---|---|
behavior-complement | behavior_complement | partial source derived model | behavior-complement |
reverse-reconstruction | reverse_reconstruction | partial source derived model | reverse-reconstruction |
manifest-transport | representation_transport | partial source derived model | manifest-transport |
admission-vs-merge | schema_refinement | partial source derived model | admission-vs-merge |
verification-gap | effect_boundary | partial source derived model | verification-gap |
independent-value | tagged_path | partial source derived model | independent-value |
Model text
descriptant
bibliography-value-tools: wave33
sources::bib public namespace is a concrete call edge. bib row's paper/open or open-nc/null metadata fields deserialize into Resource and FileRef. run_ids may fail before write; plan loads all before target effects. Merge accepts broader Value shapes than load: merge success is not sufficient load validation. Value main never calls these helpers; composition is tagged independent paths, not sequential execution.
This source-bound partial group organizes existing Mova code. Native evidence and excluded effects are recorded in source sidecars and the root wave33 report. Registration is not a universal proof or whole-system runtime claim.
model
bibliography-value-tools
forward
Bibliography parse -> Entry -> row JSONL -> sources Resource/load -> plan/check is an actual representation and effect join. The library namespace transports those same public symbols. Merge is an alternate manifest transform with weaker schema validation. Value is an independent tagged subprocess/ablation/report path, not a bibliography consumer.
invariant
Synthetic safe-path domain only. Bibliography ID authority exact HTTP(S) host; row status threshold feeds OPEN admission, nullable file sizes remain unknown. The query-free canonical-path fixture yields id.pdf; host admission alone does not canonicalize PDF URLs. Queue omission by size does not imply verify success. Value metric zip requires external same-schema premise.
cost
Sum reached stages with shared JSON/output bytes: bibliography scanning and rule matching, CLI quadratic naming/subset scans, O(file bytes) verification and bounded buffer. Value child builds/bench opaque and unmeasured; no joint speed or accuracy bound.
composition
sources::bib public namespace is a concrete call edge. bib row's paper/open or open-nc/null metadata fields deserialize into Resource and FileRef. run_ids may fail before write; plan loads all before target effects. Merge accepts broader Value shapes than load: merge success is not sufficient load validation. Value main never calls these helpers; composition is tagged independent paths, not sequential execution.
domain
Bounded UTF-8/BibTeX/JSONL, safe confined fixture paths, representable arithmetic, available memory; actual-helper subprocess fixtures have tiny bounded output. No external network, corpus, model, DB, benchmark or main ablation workflow.
limitations
No concurrent fixture-file mutation. Source-derived unexecuted branches remain conditional; namespace test is compile/representation evidence. Topical bib run shares canonical anthology_id admission for JSONL while retaining matched foreign entries in abstract listings. Partial syntax excludes macro expansion and semantic proof.
checks
- 17 exact target-qualified tests, each twice debug/twice release
- bibliography_row_load_plan_check_join actual bounded JSONL/queue/check interaction
- 3 compiled behavior mutants rejected and restored green
- old anthology_id host-boundary failure then narrow repair green
- 4 independently paired equal syntax projections
catalog
behavior-complement
Missing-ID bib run excludes destination write; filtered resource excludes queue admission; generated queue never invokes curl. Value failed subprocess excludes Ok stdout. Source-derived inner exclusion across declared tagged paths.
reverse-reconstruction
Given queue/output, recover all compatible row/Entry/field normalization antecedents; shadowed duplicate bibliography and skipped/already-present resource metadata remain ambiguous. Value report preimage independent and many-to-one.
manifest-transport
bib row status open/open-nc lies within OPEN; files null size -> parse_size(None); URL suffix .pdf derives id.pdf on safe canonical path. Actual-helper join fixture checks these interfaces.
admission-vs-merge
merge(Value) accepts missing/nonarray files and repeated within-row URLs; load(Resource) schema/duplicate checks can reject. Therefore merge success cannot transport validity unconditionally.
verification-gap
plan accepts present matching-size bytes or unknown-size bytes without hashing; check adds digest/magic constraints. Same-size non-PDF may disappear from queue yet fail verify.
independent-value
Value's sh/bench/report owns independent environment and subprocess traces; no caller edge to bibliography exists. Costs and evidence remain separate instead of invented cross-product proof.
This page in Markdown → https://md.groupants.elamur.ai/free/en/mova/models/bibliography-value-tools.groupant/