Add output validation against a defined JSON schema

2026-04-27 21:27:19 -05:00
parent 8a95dba276
commit 6cb739be55
14 changed files with 638 additions and 30 deletions
--- a/internal/artifact/transcript.go
+++ b/internal/artifact/transcript.go
@@ -0,0 +1,68 @@
+package artifact
+
+import (
+	"gitea.maximumdirect.net/eric/seriatim/internal/config"
+	"gitea.maximumdirect.net/eric/seriatim/internal/model"
+	"gitea.maximumdirect.net/eric/seriatim/schema"
+)
+
+const (
+	ApplicationName = "seriatim"
+	Version         = "dev"
+)
+
+// FromMerged converts the internal merged transcript model into the public
+// serialized output contract.
+func FromMerged(cfg config.Config, merged model.MergedTranscript) schema.Transcript {
+	segments := make([]schema.Segment, len(merged.Segments))
+	for index, segment := range merged.Segments {
+		segments[index] = schema.Segment{
+			ID:                 segment.ID,
+			Source:             segment.Source,
+			SourceSegmentIndex: copyIntPtr(segment.SourceSegmentIndex),
+			SourceRef:          segment.SourceRef,
+			DerivedFrom:        append([]string(nil), segment.DerivedFrom...),
+			Speaker:            segment.Speaker,
+			Start:              segment.Start,
+			End:                segment.End,
+			Text:               segment.Text,
+			Categories:         append([]string(nil), segment.Categories...),
+			OverlapGroupID:     segment.OverlapGroupID,
+		}
+	}
+
+	overlapGroups := make([]schema.OverlapGroup, len(merged.OverlapGroups))
+	for index, group := range merged.OverlapGroups {
+		overlapGroups[index] = schema.OverlapGroup{
+			ID:         group.ID,
+			Start:      group.Start,
+			End:        group.End,
+			Segments:   append([]string(nil), group.Segments...),
+			Speakers:   append([]string(nil), group.Speakers...),
+			Class:      group.Class,
+			Resolution: group.Resolution,
+		}
+	}
+
+	return schema.Transcript{
+		Metadata: schema.Metadata{
+			Application:           ApplicationName,
+			Version:               Version,
+			InputReader:           cfg.InputReader,
+			InputFiles:            append([]string(nil), cfg.InputFiles...),
+			PreprocessingModules:  append([]string(nil), cfg.PreprocessingModules...),
+			PostprocessingModules: append([]string(nil), cfg.PostprocessingModules...),
+			OutputModules:         append([]string(nil), cfg.OutputModules...),
+		},
+		Segments:      segments,
+		OverlapGroups: overlapGroups,
+	}
+}
+
+func copyIntPtr(value *int) *int {
+	if value == nil {
+		return nil
+	}
+	copied := *value
+	return &copied
+}