feat: send word-by-word and translated lyrics to clients

internal/adapter/in/api/convert.go

@@ -126,9 +126,18 @@ func (s *server) nowPlaying(p *room.NowPlaying) *protocol.NowPlaying {
 		DurationMs:  p.Entry.Prep.Duration.Milliseconds(),
 		Gain:        p.Entry.Prep.Gain,
 		Lyric:       p.Entry.Prep.Lyric,
+		LyricWords:  optional(p.Entry.Prep.LyricWords),
+		LyricTrans:  optional(p.Entry.Prep.LyricTrans),
 	}
 }
 
+func optional(s string) *string {
+	if s == "" {
+		return nil
+	}
+	return &s
+}
+
 func (s *server) mediaURL(t track.Track) string {
 	if media.Stored(t.Source) {
 		return s.StorageURL + "/" + media.AudioObject(media.Key(t.ID))

internal/adapter/in/api/convert_test.go

@@ -71,6 +71,16 @@ func TestSnapshotDTO(t *testing.T) {
 	}
 }
 
+func TestNowPlayingCarriesWordAndTranslatedLyrics(t *testing.T) {
+	s := &server{}
+	entry := sunnyEntry
+	entry.Prep.LyricWords, entry.Prep.LyricTrans = "[1000,500](1000,500,0)故", "[00:01.00]The story"
+	got := s.nowPlaying(&room.NowPlaying{Entry: entry})
+	if got.LyricWords == nil || *got.LyricWords != entry.Prep.LyricWords || got.LyricTrans == nil || *got.LyricTrans != entry.Prep.LyricTrans {
+		t.Errorf("lyricWords = %v, lyricTrans = %v", got.LyricWords, got.LyricTrans)
+	}
+}
+
 func TestEventDTO(t *testing.T) {
 	s := &server{}
 	sentAt := time.UnixMilli(1760000000456)

internal/app/port/porttest/source.go

@@ -27,11 +27,13 @@ type Source struct {
 }
 
 type sourceSong struct {
-	id     string
-	meta   track.Meta
-	lyric  string
-	audio  []byte
-	reject string
+	id         string
+	meta       track.Meta
+	lyric      string
+	lyricWords string
+	lyricTrans string
+	audio      []byte
+	reject     string
 }
 
 type sourcePlaylist struct {
@@ -56,6 +58,16 @@ func (s *Source) Add(id string, m track.Meta, lyric string, audio []byte) {
 	s.songs = append(s.songs, sourceSong{id: id, meta: m, lyric: lyric, audio: audio})
 }
 
+func (s *Source) SetLyrics(id, words, trans string) {
+	s.mu.Lock()
+	defer s.mu.Unlock()
+	for i := range s.songs {
+		if s.songs[i].id == id {
+			s.songs[i].lyricWords, s.songs[i].lyricTrans = words, trans
+		}
+	}
+}
+
 func (s *Source) Reject(id, reason string) {
 	s.mu.Lock()
 	defer s.mu.Unlock()
@@ -164,7 +176,7 @@ func (s *Source) Resolve(_ context.Context, id string) (port.Resolved, error) {
 	case song.reject != "":
 		return port.Resolved{}, &port.RejectedError{Err: errors.New(song.reject)}
 	}
-	return port.Resolved{Meta: song.meta, Lyric: song.lyric, Audio: io.NopCloser(bytes.NewReader(song.audio))}, nil
+	return port.Resolved{Meta: song.meta, Lyric: song.lyric, LyricWords: song.lyricWords, LyricTrans: song.lyricTrans, Audio: io.NopCloser(bytes.NewReader(song.audio))}, nil
 }
 
 func (s *Source) PlayURL(_ context.Context, id string) (string, error) {

internal/app/prepare/run_test.go

@@ -69,6 +69,20 @@ func checkDecodeWarning(t *testing.T, logs *bytes.Buffer, tr track.Track, decode
 	}
 }
 
+func TestRunSavesWordAndTranslatedLyrics(t *testing.T) {
+	source := porttest.NewSource(track.NetEase)
+	source.Add("186016", meta, "[00:01.00]故事", []byte("不是音频"))
+	source.SetLyrics("186016", "[1000,500](1000,500,0)故", "[00:01.00]The story")
+	tracks, err := run(t, source, sunny)
+	if err != nil {
+		t.Fatal(err)
+	}
+	want := track.Prepared{Duration: meta.Duration, Gain: 1, Lyric: "[00:01.00]故事", LyricWords: "[1000,500](1000,500,0)故", LyricTrans: "[00:01.00]The story"}
+	if got, _ := tracks.row(sunny); got.Result != want {
+		t.Errorf("准备结果 = %+v，期望 %+v", got.Result, want)
+	}
+}
+
 func TestRunFallsBackWhenAudioCannotBeAnalyzed(t *testing.T) {
 	source := porttest.NewSource(track.NetEase)
 	source.Add("186016", meta, "", []byte("不是音频"))

internal/bootstrap/contract_test.go

@@ -156,6 +156,7 @@ func contractSources() sources {
 		"[00:01.00]故事的小黄花", porttest.SilentWAV(269500*time.Millisecond))
 	netease.Add("185811", track.Meta{Title: "七里香", Artist: "周杰伦", Album: "七里香", CoverURL: "https://p1.music.126.net/qilixiang.jpg", PageURL: "https://music.163.com/#/song?id=185811", Duration: 299 * time.Second},
 		"", porttest.SilentWAV(299*time.Second))
+	netease.SetLyrics("185811", "[1000,500](1000,500,0)窗", "[00:01.00]Window")
 	netease.Add("1100", track.Meta{Title: "片头", Artist: "无名", PageURL: "https://music.163.com/#/song?id=1100", Duration: 15 * time.Second},
 		"", porttest.SilentWAV(15*time.Second))
 	netease.Add("404", track.Meta{Title: "不能播放的歌", Artist: "无名", PageURL: "https://music.163.com/#/song?id=404", Duration: time.Minute}, "", nil)

internal/bootstrap/testdata/ws/queue.txtar

@@ -126,7 +126,7 @@ Authorization: Bearer $alice
 {"kind": "event", "type": "playback.updated", "term": 1, "seq": 13, "data": {"nowPlaying": {
   "entry": {"id": "2", "track": {"source": "netease", "id": "185811"}, "meta": {"title": "七里香", "artist": "周杰伦", "album": "七里香", "coverUrl": "https://p1.music.126.net/qilixiang.jpg", "pageUrl": "https://music.163.com/#/song?id=185811", "durationMs": 299000},
     "by": {"id": "$aliceId", "name": "$aliceName", "avatarHash": ""}, "auto": false, "likes": 1, "status": "ready"},
-  "mediaUrl": "/media/netease/185811", "startedAtMs": 1, "durationMs": 299000, "gain": 1, "lyric": ""}}}
+  "mediaUrl": "/media/netease/185811", "startedAtMs": 1, "durationMs": 299000, "gain": 1, "lyric": "", "lyricWords": "[1000,500](1000,500,0)窗", "lyricTrans": "[00:01.00]Window"}}}
 -- receive alice --
 {"kind": "response", "type": "playback.voteSkip", "id": "v1", "data": {"votes": 1, "required": 1, "skipped": true}}
 

protocol/idl.json

@@ -171,7 +171,9 @@
         "startedAtMs": {"type": "integer"},
         "durationMs": {"type": "integer", "description": "准备结果中的时长"},
         "gain": {"type": "number", "description": "线性增益系数"},
-        "lyric": {"type": "string"}
+        "lyric": {"type": "string"},
+        "lyricWords": {"type": "string", "description": "逐字歌词，每行形如 [行起始ms,行时长ms](字起始ms,字时长ms,0)字，时间为绝对毫秒"},
+        "lyricTrans": {"type": "string", "description": "翻译歌词，行级 LRC，时间戳与原文的行起点对应"}
       },
       "required": ["entry", "mediaUrl", "startedAtMs", "durationMs", "gain", "lyric"]
     },

protocol/openapi.gen.json

@@ -631,6 +631,14 @@
           "lyric": {
             "type": "string"
           },
+          "lyricTrans": {
+            "description": "翻译歌词，行级 LRC，时间戳与原文的行起点对应",
+            "type": "string"
+          },
+          "lyricWords": {
+            "description": "逐字歌词，每行形如 [行起始ms,行时长ms](字起始ms,字时长ms,0)字，时间为绝对毫秒",
+            "type": "string"
+          },
           "mediaUrl": {
             "type": "string"
           },
@@ -1335,7 +1343,7 @@
   },
   "info": {
     "title": "Alisten",
-    "version": "09dbf7ea495db969"
+    "version": "473ee31f5ef912b4"
   },
   "openapi": "3.1.0",
   "paths": {

protocol/types.gen.go

@@ -4,7 +4,7 @@ package protocol
 
 import "encoding/json"
 
-const Version = "09dbf7ea495db969"
+const Version = "473ee31f5ef912b4"
 
 type APIKey struct {
 	Key string `json:"key"`
@@ -255,6 +255,8 @@ type NowPlaying struct {
 	Entry       Entry   `json:"entry"`
 	Gain        float64 `json:"gain"`
 	Lyric       string  `json:"lyric"`
+	LyricTrans  *string `json:"lyricTrans,omitempty"`
+	LyricWords  *string `json:"lyricWords,omitempty"`
 	MediaURL    string  `json:"mediaUrl"`
 	StartedAtMs int64   `json:"startedAtMs"`
 }

web/src/core/realtime/protocol.gen.ts

@@ -1,6 +1,6 @@
 // Code generated by protocolgen from protocol/idl.json; DO NOT EDIT.
 
-export const PROTOCOL_VERSION = '09dbf7ea495db969'
+export const PROTOCOL_VERSION = '473ee31f5ef912b4'
 
 export interface APIKey {
   key: string
@@ -67,6 +67,8 @@ export interface NowPlaying {
   entry: Entry
   gain: number
   lyric: string
+  lyricTrans?: string
+  lyricWords?: string
   mediaUrl: string
   startedAtMs: number
 }