Add D&D location occurrence validators

This commit is contained in:
2026-08-04 00:33:39 +00:00
parent dd61a4efda
commit a168c13b85
14 changed files with 1034 additions and 3 deletions

View File

@@ -16,7 +16,7 @@ import (
const (
Key = "extract/dnd/locations/source_refs"
ReasonCode = "invalid_location_source_refs"
policy = "dnd.locations.validator.source_refs.v1"
policy = "dnd.locations.validator.source_refs.v2"
)
type Options struct{}
@@ -35,6 +35,9 @@ func (v *Validator) CheckpointFingerprints() []pipeline.CheckpointFingerprint {
}
func (v *Validator) Validate(_ context.Context, req contracts.TypedValidationRequest[dnd.LocationList]) (contracts.ValidationResult, error) {
if req.Stage == string(pipeline.StageExtract) && req.Chunk == nil {
return contracts.ValidationResult{}, fmt.Errorf("location source-reference validator requires the current extraction chunk")
}
if err := locationshape.Validate(req.Value); err != nil {
return contracts.ValidationResult{Approved: true}, nil
}
@@ -44,6 +47,10 @@ func (v *Validator) Validate(_ context.Context, req contracts.TypedValidationReq
for refIndex, ref := range location.SourceRefs {
if err := index.ValidateRef(ref); err != nil {
issues = append(issues, fmt.Sprintf("locations[%d].source_refs[%d]: %s", locationIndex, refIndex, diagnostics.Truncate(err.Error())))
continue
}
if req.Stage == string(pipeline.StageExtract) && !chunkContainsRef(req.Chunk, ref) {
issues = append(issues, fmt.Sprintf("locations[%d].source_refs[%d]: source reference is outside the current extraction chunk", locationIndex, refIndex))
}
}
}
@@ -53,6 +60,19 @@ func (v *Validator) Validate(_ context.Context, req contracts.TypedValidationReq
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
}
func Spec() pipeline.ValidatorSpec {
return pipeline.ValidatorSpec{Key: Key, ExecutionClass: contracts.ExecutionClassDeterministic}
}

View File

@@ -37,6 +37,30 @@ func TestValidatorApprovesValidReferencesAndDefersMalformedShape(t *testing.T) {
}
}
func TestValidatorRequiresCurrentChunkAndRejectsEvidenceOutsideIt(t *testing.T) {
doc := &source.SourceDocument{ID: "session", Units: []source.SourceUnit{{ID: 1}, {ID: 2}}}
value := validLocationList()
result, err := New(Options{}).Validate(context.Background(), contracts.TypedValidationRequest[dnd.LocationList]{
Stage: string(pipeline.StageExtract), Source: doc,
Chunk: &source.Chunk{SourceID: "session", Units: []source.SourceUnit{{ID: 1}}}, Value: value,
})
if err != nil || !result.Approved {
t.Fatalf("contained extraction evidence = %#v, %v", result, err)
}
value.Locations[0].SourceRefs[0].EndUnitID = 2
result, err = New(Options{}).Validate(context.Background(), contracts.TypedValidationRequest[dnd.LocationList]{
Stage: string(pipeline.StageExtract), Source: doc,
Chunk: &source.Chunk{SourceID: "session", Units: []source.SourceUnit{{ID: 1}}}, Value: value,
})
if err != nil || result.Approved || !strings.Contains(result.Message, "outside the current extraction chunk") {
t.Fatalf("out-of-chunk extraction evidence = %#v, %v", result, err)
}
_, err = New(Options{}).Validate(context.Background(), contracts.TypedValidationRequest[dnd.LocationList]{Stage: string(pipeline.StageExtract), Source: doc, Value: validLocationList()})
if err == nil || !strings.Contains(err.Error(), "requires the current extraction chunk") {
t.Fatalf("missing chunk error = %v", err)
}
}
func TestValidatorRegisters(t *testing.T) {
registry := pipeline.NewValidatorRegistry()
if err := Register(registry); err != nil {

View File

@@ -16,7 +16,8 @@ import (
const (
Key = "extract/dnd/locations/source_relatedness"
WarningReasonCode = "location_not_near_source"
policy = "dnd.locations.validator.source_relatedness.v1"
OmittedReasonCode = "location_relatedness_warnings_omitted"
policy = "dnd.locations.validator.source_relatedness.v2"
)
type Options struct{}
@@ -60,7 +61,7 @@ func (v *Validator) Validate(_ context.Context, req contracts.TypedValidationReq
Message: fmt.Sprintf("Location %s was not found in cited source text", diagnostics.Quote(location.Name)),
})
}
return contracts.ValidationResult{Approved: true, Warnings: warnings}, nil
return contracts.ValidationResult{Approved: true, Warnings: diagnostics.LimitWarnings(warnings, "locations", OmittedReasonCode)}, nil
}
func Spec() pipeline.ValidatorSpec {

View File

@@ -10,6 +10,7 @@ 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/diagnostics"
)
func TestValidatorUsesOnlyCitedTranscriptText(t *testing.T) {
@@ -54,6 +55,17 @@ func TestValidatorRegisters(t *testing.T) {
}
}
func TestValidatorBoundsWarnings(t *testing.T) {
locations := make([]dnd.Location, diagnostics.MaxWarnings+1)
for index := range locations {
locations[index] = dnd.Location{ID: "candidate", Name: "Missing", SourceRefs: []source.SourceRef{{SourceID: "session", StartUnitID: 1, EndUnitID: 1}}}
}
result, err := New(Options{}).Validate(context.Background(), request(document(), dnd.LocationList{Locations: locations}, contracts.ReferenceSet{}))
if err != nil || !result.Approved || len(result.Warnings) != diagnostics.MaxWarnings || result.Warnings[len(result.Warnings)-1].ReasonCode != OmittedReasonCode {
t.Fatalf("bounded warnings = %#v, %v", result, err)
}
}
func request(doc *source.SourceDocument, value dnd.LocationList, references contracts.ReferenceSet) contracts.TypedValidationRequest[dnd.LocationList] {
return contracts.TypedValidationRequest[dnd.LocationList]{Source: doc, References: references, Value: value}
}