Implement songLyrics extension: getLyricsBySongId with real word/syllable cues

Adds getLyricsBySongId (versions 1 and 2 of the songLyrics extension, gated by the enhanced param): returns every stored structuredLyrics entry for a song, including line-level start times for synced lyrics and, only when a source actually provided word/syllable timing (an Enhanced LRC sidecar) and enhanced=true was requested, cueLine/cue data with the precomputed UTF-8 byte offsets. Plain and non-enhanced synced lyrics never get fabricated cue data. Removes the now-dead SetSongLyrics writer (superseded by database.InsertLyrics); SongLyricsPlainText keeps serving the legacy getLyrics endpoint unchanged. Adds a scanner integration test proving a sidecar .lrc (with an Enhanced LRC line) is discovered and indexed during a real scan. Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_014r98AtcBbevpscWLsfwVvv
2026-08-23 08:19:41 -05:00
parent e48f063f03
commit ec28053d36
5 changed files with 363 additions and 53 deletions
@@ -0,0 +1,94 @@
+package subsonic
+
+import (
+	"git.noroutes.net/NoRoutes/vibetunes/internal/database"
+)
+
+func init() {
+	registerExtension("songLyrics", 1, 2)
+	register("getLyricsBySongId", handleGetLyricsBySongID)
+}
+
+// handleGetLyricsBySongID implements the songLyrics extension. It never
+// fakes cue-level timing: cueLine data is only emitted for entries that
+// were actually parsed from an Enhanced-LRC-style source with word/
+// syllable cues, and only when the client asked for enhanced=true.
+func handleGetLyricsBySongID(rc *Context) *apiError {
+	id, aerr := rc.Params.Required("id")
+	if aerr != nil {
+		return aerr
+	}
+	enhanced := rc.Params.Bool("enhanced", false)
+
+	db := rc.Server.DB
+	if _, err := database.GetSong(db, id); err != nil {
+		return errNotFound("Song")
+	}
+
+	rows, err := database.GetSongStructuredLyrics(db, id)
+	if err != nil {
+		return newError(ErrGeneric, err.Error())
+	}
+
+	list := &LyricsList{}
+	for _, row := range rows {
+		kind := "main" // every entry VibeSonic currently stores/ingests is a primary lyric layer, never a translation or pronunciation.
+		if !enhanced && kind != "main" {
+			continue
+		}
+
+		sl := StructuredLyrics{
+			Lang:   row.Language,
+			Synced: row.Synced,
+			Line:   make([]Line, 0, len(row.Lines)),
+		}
+		if enhanced {
+			sl.Kind = kind
+		}
+
+		hasCues := false
+		for _, l := range row.Lines {
+			line := Line{Value: l.Text}
+			if row.Synced && l.StartMs != nil {
+				start := *l.StartMs
+				line.Start = &start
+			}
+			sl.Line = append(sl.Line, line)
+			if len(l.Words) > 0 {
+				hasCues = true
+			}
+		}
+
+		if enhanced && row.Synced && hasCues {
+			sl.CueLine = make([]CueLine, 0, len(row.Lines))
+			for i, l := range row.Lines {
+				if len(l.Words) == 0 {
+					continue
+				}
+				cl := CueLine{Index: i, Value: l.Text}
+				if l.StartMs != nil {
+					cl.Start = *l.StartMs
+				}
+				cl.Cue = make([]Cue, 0, len(l.Words))
+				for _, w := range l.Words {
+					cue := Cue{Start: w.StartMs, ByteStart: w.ByteStart, ByteEnd: w.ByteEnd, Value: w.Text}
+					if w.EndMs != nil {
+						end := *w.EndMs
+						cue.End = &end
+					}
+					cl.Cue = append(cl.Cue, cue)
+				}
+				if last := l.Words[len(l.Words)-1]; last.EndMs != nil {
+					cl.End = *last.EndMs
+				}
+				sl.CueLine = append(sl.CueLine, cl)
+			}
+		}
+
+		list.StructuredLyrics = append(list.StructuredLyrics, sl)
+	}
+
+	rc.Response = newResponse()
+	rc.Response.LyricsList = list
+	return nil
+}