Elamur Groupantsocrfix-text-command-alignment
The groupant beside tools/ocrfix/src/main.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 | symbolic source derived | invariant-and-cost |
| descriptant | present | descriptant |
| invariant | partial | invariant-and-cost |
| 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 |
|---|---|---|---|
ocrfix-text-command-alignment-behavior_complement | behavior_complement | partial source derived model | behavior_complement |
ocrfix-text-command-alignment-reverse_reconstruction | reverse_reconstruction | partial source derived model | reverse_reconstruction |
ocrfix-text-command-alignment-length_conservation | length_conservation | partial source derived model | length_conservation |
ocrfix-text-command-alignment-tie_priority | tie_priority | partial source derived model | tie_priority |
ocrfix-text-command-alignment-substitution_histogram | substitution_histogram | partial source derived model | substitution_histogram |
ocrfix-text-command-alignment-resource_boundary | resource_boundary | partial source derived model | resource_boundary |
Model text
descriptant
ocrfix-text-command-alignment
Epoch 2026-10-31
Full source read preceded changes. Prospective model predicted branch/order, resource and information-loss behavior. Tests exercise actual helpers on bounded synthetic inputs; command workflows remain unexecuted. Production source prefix is preserved; only inline tests added.
Initialize d[i,0]=i,d[0,j]=j; d[i,j]=min(diag+mismatch,up+1,left+1). Backtrace diagonal if equal, else deletion if equal, else insertion; accumulate gt->hyp substitution multiset.
Executable percentage approximation is unsupported: categorical parsing/effects have no useful certified metric here. Partial syntax excludes macro expansion and name resolution.
approximant
Forward relation
Initialize d[i,0]=i,d[0,j]=j; d[i,j]=min(diag+mismatch,up+1,left+1). Backtrace diagonal if equal, else deletion if equal, else insertion; accumulate gt->hyp substitution multiset.
Domain
Finite char slices with n,m and DP increments fitting u32 and allocation capacity.
models
Typed source models
All models are partial source-derived relations at this manifest snapshot. Scope is one call/trace under the stated domain; no universal semantic proof.
behavior_complement
Identity slices exclude positive distance; empty gt excludes substitutions/deletions; empty hyp excludes substitutions/insertions.
Domain: Finite char slices with n,m and DP increments fitting u32 and allocation capacity. Observation: s+i+d=d[n,m]; n-d+i=m; sum(pair counts)=s. Exact comparisons on bounded u32 DP cells; diagonal priority fixes an alignment, not necessarily unique. Assumptions: bounded resources and source branch semantics. Oracle: selected actual-helper tests in tests.md constrain only their finite fixtures; remaining branches are source-derived.
reverse_reconstruction
Counts and pair multiset admit many string pairs and placements. Given full DP and slices, chosen backtrace fixed, but minimal alternative alignments may exist.
Domain: Finite char slices with n,m and DP increments fitting u32 and allocation capacity. Observation: s+i+d=d[n,m]; n-d+i=m; sum(pair counts)=s. Exact comparisons on bounded u32 DP cells; diagonal priority fixes an alignment, not necessarily unique. Assumptions: bounded resources and source branch semantics. Oracle: selected actual-helper tests in tests.md constrain only their finite fixtures; remaining branches are source-derived.
length_conservation
Every diagonal consumes1 each; deletion gt only; insertion hyp only, so m-n=i-d.
Domain: Finite char slices with n,m and DP increments fitting u32 and allocation capacity. Observation: s+i+d=d[n,m]; n-d+i=m; sum(pair counts)=s. Exact comparisons on bounded u32 DP cells; diagonal priority fixes an alignment, not necessarily unique. Assumptions: bounded resources and source branch semantics. Oracle: selected actual-helper tests in tests.md constrain only their finite fixtures; remaining branches are source-derived.
tie_priority
At equal optimum diagonal beats deletion, deletion beats insertion; pair histogram is tie-policy dependent though distance is not.
Domain: Finite char slices with n,m and DP increments fitting u32 and allocation capacity. Observation: s+i+d=d[n,m]; n-d+i=m; sum(pair counts)=s. Exact comparisons on bounded u32 DP cells; diagonal priority fixes an alignment, not necessarily unique. Assumptions: bounded resources and source branch semantics. Oracle: selected actual-helper tests in tests.md constrain only their finite fixtures; remaining branches are source-derived.
substitution_histogram
Only unequal diagonal increments both s and directed pair(gt,hyp); equal diagonal consumes without record.
Domain: Finite char slices with n,m and DP increments fitting u32 and allocation capacity. Observation: s+i+d=d[n,m]; n-d+i=m; sum(pair counts)=s. Exact comparisons on bounded u32 DP cells; diagonal priority fixes an alignment, not necessarily unique. Assumptions: bounded resources and source branch semantics. Oracle: selected actual-helper tests in tests.md constrain only their finite fixtures; remaining branches are source-derived.
resource_boundary
Allocate(n+1)*(m+1) cells before filling; casts and increments require bounded lengths. Allocation failure/overflow are outside exact finite model.
Domain: Finite char slices with n,m and DP increments fitting u32 and allocation capacity. Observation: s+i+d=d[n,m]; n-d+i=m; sum(pair counts)=s. Exact comparisons on bounded u32 DP cells; diagonal priority fixes an alignment, not necessarily unique. Assumptions: bounded resources and source branch semantics. Oracle: selected actual-helper tests in tests.md constrain only their finite fixtures; remaining branches are source-derived.
tests
Evidence
Selected exact tests: ["tests::brackets_rejoined","tests::stream_and_cer","tests::channel_corrects_and_keeps","tests::negative_control_known_word_stays","wave33::channel_depth_alignment_and_masks"].
Each runs twice in debug and twice in release with normalized named outcomes compared. Compiled production-boundary mutants are recorded in the external wave33 text receipts with original/restored hashes and positive reruns. Existing negative assertions are separate from compiled mutants. No CLI, download, teacher, corpus or model execution. For this fragment, source-derived paths not reached by these tests remain unverified.
invariant-and-cost
Invariant and cost
Invariant
s+i+d=d[n,m]; n-d+i=m; sum(pair counts)=s. Exact comparisons on bounded u32 DP cells; diagonal priority fixes an alignment, not necessarily unique.
Cost
O((n+1)(m+1)) time and u32 memory, substitution map O(s log distinct); no linear-space claim.
Arithmetic and allocation premises are conditional. No timing or percentage surrogate claimed.
This page in Markdown → https://md.groupants.elamur.ai/free/en/mova/tools/ocrfix/src/main.rs.groupant/alignment.groupant/