Classify source relatedness as data-quality advisories

This commit is contained in:
2026-08-27 15:49:53 +00:00
parent 1f1967c8d2
commit ccba2ce3f9
25 changed files with 232 additions and 189 deletions

View File

@@ -1,4 +1,4 @@
// Package sourcerelatedness warns when item-occurrence evidence does not mention its item.
// Package sourcerelatedness reports advisory findings when item-occurrence evidence does not mention its item.
package sourcerelatedness
import (
@@ -14,10 +14,9 @@ import (
)
const (
Key = "extract/dnd/item-occurrences/source_relatedness"
WarningReasonCode = "item_occurrence_source_unrelated"
OmittedReasonCode = "item_occurrence_relatedness_warnings_omitted"
policy = "dnd.item_occurrences.source_relatedness.v1"
Key = "extract/dnd/item-occurrences/source_relatedness"
ReasonCode = "item_occurrence_not_near_source"
policy = "dnd.item_occurrences.source_relatedness.v1"
)
type Options struct{}
@@ -54,14 +53,11 @@ func (v *Validator) Validate(_ context.Context, req contracts.TypedValidationReq
}
warnings = append(warnings, contracts.Warning{
Scope: fmt.Sprintf("occurrences[%d]", index),
ReasonCode: WarningReasonCode,
ReasonCode: ReasonCode,
Message: fmt.Sprintf("item occurrence name %s was not found in cited source text", diagnostics.Quote(occurrence.Name)),
})
}
return contracts.ValidationResult{
Approved: true,
Warnings: diagnostics.LimitWarnings(warnings, "item_occurrences", OmittedReasonCode),
}, nil
return diagnostics.DataQualityResult(warnings)
}
func Spec() pipeline.ValidatorSpec {

View File

@@ -10,7 +10,6 @@ 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 TestValidatorMatchesTokenSequencesAcrossRanges(t *testing.T) {
@@ -26,11 +25,11 @@ func TestValidatorMatchesTokenSequencesAcrossRanges(t *testing.T) {
"glossary": {Items: []contracts.ReferenceItem{{Content: []byte("missing-item")}}},
}}
result, err := New(Options{}).Validate(context.Background(), contracts.TypedValidationRequest[dnd.ItemOccurrenceList]{Source: doc, References: references, Value: value})
if err != nil || !result.Approved || len(result.Warnings) != 1 {
if err != nil || !result.Approved || len(result.Warnings) != 0 || len(result.Diagnostics) != 1 {
t.Fatalf("Validate() = %#v, %v", result, err)
}
if warning := result.Warnings[0]; warning.Scope != "occurrences[1]" || warning.ReasonCode != WarningReasonCode || !strings.Contains(warning.Message, "missing-item") {
t.Fatalf("warning = %#v", warning)
if diagnostic := result.Diagnostics[0]; diagnostic.Samples[0].Scope != "occurrences[1]" || diagnostic.ReasonCode != ReasonCode || !strings.Contains(diagnostic.Samples[0].Message, "missing-item") {
t.Fatalf("diagnostic = %#v", diagnostic)
}
}
@@ -41,14 +40,14 @@ func TestValidatorDefersMalformedAndUnreadableEvidence(t *testing.T) {
{Occurrences: []dnd.ItemOccurrence{{ItemID: "item", Name: "potion", Kind: dnd.ItemOccurrenceKindDiscovered, SourceRefs: []source.SourceRef{{SourceID: doc.ID, StartUnitID: 99, EndUnitID: 99}}}}},
} {
result, err := New(Options{}).Validate(context.Background(), contracts.TypedValidationRequest[dnd.ItemOccurrenceList]{Source: doc, Value: value})
if err != nil || !result.Approved || len(result.Warnings) != 0 {
if err != nil || !result.Approved || len(result.Diagnostics) != 0 {
t.Fatalf("deferred result = %#v, %v", result, err)
}
}
}
func TestValidatorBoundsWarningsAndRegistersPolicy(t *testing.T) {
count := diagnostics.MaxWarnings + 5
count := contracts.MaxDiagnosticSamples + 5
occurrences := make([]dnd.ItemOccurrence, count)
for index := range occurrences {
occurrences[index] = dnd.ItemOccurrence{ItemID: "item", Name: "missing item", Kind: dnd.ItemOccurrenceKindDiscovered, SourceRefs: []source.SourceRef{{SourceID: "session", StartUnitID: 1, EndUnitID: 1}}}
@@ -57,7 +56,7 @@ func TestValidatorBoundsWarningsAndRegistersPolicy(t *testing.T) {
Source: &source.SourceDocument{ID: "session", Units: []source.SourceUnit{{ID: 1, Text: "nothing useful"}}},
Value: dnd.ItemOccurrenceList{Occurrences: occurrences},
})
if err != nil || !result.Approved || len(result.Warnings) != diagnostics.MaxWarnings || result.Warnings[len(result.Warnings)-1].ReasonCode != OmittedReasonCode {
if err != nil || !result.Approved || len(result.Warnings) != 0 || len(result.Diagnostics) != 1 || result.Diagnostics[0].OccurrenceCount != count || len(result.Diagnostics[0].Samples) != contracts.MaxDiagnosticSamples {
t.Fatalf("Validate() = %#v, %v", result, err)
}
if got := New(Options{}).CheckpointFingerprints(); !reflect.DeepEqual(got, []pipeline.CheckpointFingerprint{{Name: "policy", Value: policy}}) {