eric/notarius
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Improve D&D validation reliability

Browse Source
This commit is contained in:
Eric Rakestraw
2026-08-29 01:24:45 +00:00
parent 4da9360d74
commit 917d150279
55 changed files with 1300 additions and 565 deletions
Download Patch File Download Diff File Expand all files Collapse all files

39
internal/modules/dnd/validate/locationregistry/source_refs/validator.go
View File

@@ -9,6 +9,7 @@ import (
"gitea.maximumdirect.net/eric/notarius/internal/framework/contracts"
"gitea.maximumdirect.net/eric/notarius/internal/framework/pipeline"
"gitea.maximumdirect.net/eric/notarius/internal/modules/dnd"
"gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/shared"
"gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/shared/diagnostics"
locationshape "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/validate/locationregistry/shape"
)
@@ -16,7 +17,7 @@ import (
const (
Key = "extract/dnd/location-registry/source_refs"
ReasonCode = "invalid_location_source_refs"
policy = "dnd.location_registry.validator.source_refs.v2"
policy = "dnd.location_registry.validator.source_refs.v3"
)
type Options struct{}
@@ -42,35 +43,37 @@ func (v *Validator) Validate(_ context.Context, req contracts.TypedValidationReq
return contracts.ValidationResult{Approved: true}, nil
}
index := source.NewDocumentIndex(req.Source)
coverage := shared.NewChunkCoverage(req.Chunk)
issues := make([]string, 0)
var corrections diagnostics.Corrections
normalization := req.Stage == string(pipeline.StageNormalize)
validRangeRule := "Use positive source range endpoints that occur in the supplied transcript, with the earlier unit first."
guidancePrefix := "Correct every rejected location citation and return the complete replacement location registry"
if normalization {
validRangeRule = "Revise the duplicate-group proposals so every selected canonical location preserves valid transcript evidence; do not reproduce application IDs."
guidancePrefix = "Correct the duplicate-group proposals and return the complete replacement proposal response"
}
for locationIndex, location := range req.Value.Locations {
for refIndex, ref := range location.SourceRefs {
record := fmt.Sprintf("Affected location %s, citing %s.", diagnostics.Quote(location.Name), diagnostics.SourceRefRange(ref))
if err := index.ValidateRef(ref); err != nil {
issues = append(issues, fmt.Sprintf("locations[%d].source_refs[%d]: %s", locationIndex, refIndex, diagnostics.Truncate(err.Error())))
corrections.Add("valid-range", validRangeRule, record)
continue
}
if req.Stage == string(pipeline.StageExtract) && !chunkContainsRef(req.Chunk, ref) {
if req.Stage == string(pipeline.StageExtract) && !coverage.Contains(index, req.Source, ref) {
issues = append(issues, fmt.Sprintf("locations[%d].source_refs[%d]: source reference is outside the current extraction chunk", locationIndex, refIndex))
corrections.Add("chunk-range", "Use only source ranges wholly contained in the supplied extraction chunk.", record)
}
}
}
if len(issues) == 0 {
return contracts.ValidationResult{Approved: true}, nil
}
return rejection(diagnostics.Aggregate("invalid location source references", issues)), nil
}
func chunkContainsRef(chunk *source.Chunk, ref source.SourceRef) bool {
if chunk == nil || ref.SourceID != chunk.SourceID {
return false
}
startFound := false
endFound := false
for _, unit := range chunk.Units {
startFound = startFound || unit.ID == ref.StartUnitID
endFound = endFound || unit.ID == ref.EndUnitID
}
return startFound && endFound
return rejection(
diagnostics.Aggregate("invalid location source references", issues),
corrections.Guidance(guidancePrefix),
), nil
}
func Spec() pipeline.ValidatorSpec {
@@ -96,6 +99,6 @@ func DecodeOptions(options map[string]any) (Options, error) {
func validateOptions(options map[string]any) error { _, err := DecodeOptions(options); return err }
func rejection(message string) contracts.ValidationResult {
return contracts.ValidationResult{Approved: false, ReasonCode: ReasonCode, Message: message, CorrectionGuidance: "Return registry locations whose source references identify valid transcript ranges within the supplied extraction chunk and directly support each proper location name."}
func rejection(message, guidance string) contracts.ValidationResult {
return contracts.ValidationResult{Approved: false, ReasonCode: ReasonCode, Message: message, CorrectionGuidance: guidance}
}