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

16
internal/modules/dnd/validate/locationoccurrences/registry/validator.go
View File

@@ -16,7 +16,7 @@ import (
const (
Key = "extract/dnd/location-occurrences/registry"
ReasonCode = "invalid_location_occurrence_registry"
policy = "dnd.location_occurrences.validator.registry.v1"
policy = "dnd.location_occurrences.validator.registry.v2"
)
type Options struct{}
@@ -76,27 +76,33 @@ func (v *Validator) Validate(_ context.Context, req contracts.TypedValidationReq
return contracts.ValidationResult{}, fmt.Errorf("resolve location registry: %w", err)
}
if !registry.Bound() {
return rejection([]string{"location registry reference is required"}), nil
var corrections diagnostics.Corrections
corrections.Add("registry", "Use only proper contextual location names from the supplied location registry; omit an occurrence that cannot be matched unambiguously.", "")
return rejection([]string{"location registry reference is required"}, corrections), nil
}
issues := make([]string, 0)
var corrections diagnostics.Corrections
for index, occurrence := range req.Value.Occurrences {
record := fmt.Sprintf("Affected %s occurrence for location %s %s.", diagnostics.Quote(string(occurrence.Kind)), diagnostics.Quote(occurrence.Name), diagnostics.SourceRange(occurrence.SourceRefs))
location, ok := registry.Lookup(occurrence.LocationID)
if !ok {
issues = append(issues, fmt.Sprintf("occurrences[%d].location_id is not in the location registry: %s", index, diagnostics.Quote(occurrence.LocationID)))
corrections.Add("registry", "Use only proper contextual location names from the supplied location registry; omit an occurrence that cannot be matched unambiguously.", record)
continue
}
if occurrence.Name != location.Name {
issues = append(issues, fmt.Sprintf("occurrences[%d] does not match registry location %s", index, diagnostics.Quote(occurrence.LocationID)))
corrections.Add("canonical-name", "Use the exact proper location name supplied by the registry.", record+" Use registry name "+diagnostics.Quote(location.Name)+".")
}
}
if len(issues) == 0 {
return contracts.ValidationResult{Approved: true}, nil
}
return rejection(issues), nil
return rejection(issues, corrections), nil
}
func rejection(issues []string) contracts.ValidationResult {
return contracts.ValidationResult{Approved: false, ReasonCode: ReasonCode, Message: diagnostics.Aggregate("invalid location occurrence registry", issues), CorrectionGuidance: "Return location occurrences using proper contextual location names that match a location in the supplied registry; omit occurrences that cannot be matched unambiguously."}
func rejection(issues []string, corrections diagnostics.Corrections) contracts.ValidationResult {
return contracts.ValidationResult{Approved: false, ReasonCode: ReasonCode, Message: diagnostics.Aggregate("invalid location occurrence registry", issues), CorrectionGuidance: corrections.Guidance("Correct every rejected location occurrence and return the complete replacement occurrence list")}
}
func Spec() pipeline.ValidatorSpec {

27
internal/modules/dnd/validate/locationoccurrences/shape/validator.go
View File

@@ -15,7 +15,7 @@ import (
const (
Key = "extract/dnd/location-occurrences/shape"
ReasonCode = "invalid_location_occurrence_shape"
policy = "dnd.location_occurrences.validator.shape.v1"
policy = "dnd.location_occurrences.validator.shape.v2"
)
type Options struct{}
@@ -34,41 +34,54 @@ func (v *Validator) CheckpointFingerprints() []pipeline.CheckpointFingerprint {
}
func (v *Validator) Validate(_ context.Context, req contracts.TypedValidationRequest[dnd.LocationOccurrenceList]) (contracts.ValidationResult, error) {
if err := Validate(req.Value); err != nil {
return contracts.ValidationResult{Approved: false, ReasonCode: ReasonCode, Message: err.Error(), CorrectionGuidance: "Return a complete location-occurrence list with every required field present, valid contextual location names, supported occurrence kinds, and valid source references."}, nil
issues, corrections := assess(req.Value)
if len(issues) != 0 {
return contracts.ValidationResult{
Approved: false,
ReasonCode: ReasonCode,
Message: diagnostics.Aggregate("invalid location occurrence shape", issues),
CorrectionGuidance: corrections.Guidance("Correct every rejected location occurrence and return the complete replacement occurrence list"),
}, nil
}
return contracts.ValidationResult{Approved: true}, nil
}
func Validate(value dnd.LocationOccurrenceList) error {
issues := issuesFor(value)
issues, _ := assess(value)
if len(issues) == 0 {
return nil
}
return fmt.Errorf("%s", diagnostics.Aggregate("invalid location occurrence shape", issues))
}
func issuesFor(value dnd.LocationOccurrenceList) []string {
func assess(value dnd.LocationOccurrenceList) ([]string, diagnostics.Corrections) {
var corrections diagnostics.Corrections
if value.Occurrences == nil {
return []string{"occurrences must be present"}
corrections.Add("list", "Return an `occurrences` array; use an empty array when the transcript establishes no location occurrences.", "")
return []string{"occurrences must be present"}, corrections
}
issues := make([]string, 0)
for index, occurrence := range value.Occurrences {
prefix := fmt.Sprintf("occurrences[%d]", index)
record := fmt.Sprintf("Affected %s occurrence for location %s %s.", diagnostics.Quote(string(occurrence.Kind)), diagnostics.Quote(strings.TrimSpace(occurrence.Name)), diagnostics.SourceRange(occurrence.SourceRefs))
if strings.TrimSpace(occurrence.LocationID) == "" {
issues = append(issues, prefix+".location_id must not be empty")
corrections.Add("name", "Select a nonblank proper location name from the supplied registry for every occurrence.", record)
}
if strings.TrimSpace(occurrence.Name) == "" {
issues = append(issues, prefix+".name must not be empty")
corrections.Add("name", "Select a nonblank proper location name from the supplied registry for every occurrence.", record)
}
if !validKind(occurrence.Kind) {
issues = append(issues, prefix+".kind is unsupported: "+diagnostics.Quote(string(occurrence.Kind)))
corrections.Add("kind", "Set `kind` to exactly one of `visited`, `planned`, `recalled`, or `mentioned`.", record)
}
if len(occurrence.SourceRefs) == 0 {
issues = append(issues, prefix+".source_refs must contain at least one reference")
corrections.Add("source-refs", "Provide at least one transcript source range that directly supports every location occurrence.", record)
}
}
return issues
return issues, corrections
}
func validKind(value dnd.LocationOccurrenceKind) bool {

28
internal/modules/dnd/validate/locationoccurrences/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"
occurrenceshape "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/validate/locationoccurrences/shape"
)
@@ -16,7 +17,7 @@ import (
const (
Key = "extract/dnd/location-occurrences/source_refs"
ReasonCode = "invalid_location_occurrence_source_refs"
policy = "dnd.location_occurrences.validator.source_refs.v1"
policy = "dnd.location_occurrences.validator.source_refs.v2"
)
type Options struct{}
@@ -42,35 +43,34 @@ 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
for occurrenceIndex, occurrence := range req.Value.Occurrences {
for refIndex, ref := range occurrence.SourceRefs {
record := fmt.Sprintf("Affected %s occurrence for location %s, citing %s.", diagnostics.Quote(string(occurrence.Kind)), diagnostics.Quote(occurrence.Name), diagnostics.SourceRefRange(ref))
if err := index.ValidateRef(ref); err != nil {
issues = append(issues, fmt.Sprintf("occurrences[%d].source_refs[%d]: %s", occurrenceIndex, refIndex, diagnostics.Truncate(err.Error())))
corrections.Add("valid-range", "Use positive source range endpoints that occur in the supplied transcript, with the earlier unit first.", 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("occurrences[%d].source_refs[%d]: source reference is outside the current extraction chunk", occurrenceIndex, 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 contracts.ValidationResult{Approved: false, ReasonCode: ReasonCode, Message: diagnostics.Aggregate("invalid location occurrence source references", issues), CorrectionGuidance: "Return location occurrences whose source references identify valid transcript ranges within the supplied extraction chunk and directly support each occurrence."}, nil
return contracts.ValidationResult{
Approved: false,
ReasonCode: ReasonCode,
Message: diagnostics.Aggregate("invalid location occurrence source references", issues),
CorrectionGuidance: corrections.Guidance("Correct every rejected location-occurrence citation and return the complete replacement occurrence list"),
}, nil
}
func chunkContainsRef(chunk *source.Chunk, ref source.SourceRef) bool {
if chunk == nil || ref.SourceID != chunk.SourceID {
return false
}
startFound, endFound := false, false
for _, unit := range chunk.Units {
startFound = startFound || unit.ID == ref.StartUnitID
endFound = endFound || unit.ID == ref.EndUnitID
}
return startFound && endFound
}
func Spec() pipeline.ValidatorSpec {
return pipeline.ValidatorSpec{Key: Key, ExecutionClass: contracts.ExecutionClassDeterministic}
}