refactor: clean up doubao tts code

relay/channel/volcengine/adaptor.go

@@ -24,7 +24,6 @@ import (
 )
 )
 
 
 const (
 const (
-	// Context keys for passing data between methods
 	contextKeyTTSRequest     = "volcengine_tts_request"
 	contextKeyTTSRequest     = "volcengine_tts_request"
 	contextKeyResponseFormat = "response_format"
 	contextKeyResponseFormat = "response_format"
 )
 )
@@ -76,27 +75,23 @@ func (a *Adaptor) ConvertAudioRequest(c *gin.Context, info *relaycommon.RelayInf
 		Request: VolcengineTTSReqInfo{
 		Request: VolcengineTTSReqInfo{
 			ReqID:     generateRequestID(),
 			ReqID:     generateRequestID(),
 			Text:      request.Input,
 			Text:      request.Input,
-			Operation: "submit", // default WebSocket uses "submit"
+			Operation: "submit",
 			Model:     info.OriginModelName,
 			Model:     info.OriginModelName,
 		},
 		},
 	}
 	}
 
 
-	// 同步扩展字段的厂商自定义metadata
 	if len(request.Metadata) > 0 {
 	if len(request.Metadata) > 0 {
 		if err = json.Unmarshal(request.Metadata, &volcRequest); err != nil {
 		if err = json.Unmarshal(request.Metadata, &volcRequest); err != nil {
 			return nil, fmt.Errorf("error unmarshalling metadata to volcengine request: %w", err)
 			return nil, fmt.Errorf("error unmarshalling metadata to volcengine request: %w", err)
 		}
 		}
 	}
 	}
 
 
-	// Store the request in context for WebSocket handler
 	c.Set(contextKeyTTSRequest, volcRequest)
 	c.Set(contextKeyTTSRequest, volcRequest)
-	// https://www.volcengine.com/docs/6561/1257584
-	// operation需要设置为submit才是流式返回
+
 	if volcRequest.Request.Operation == "submit" {
 	if volcRequest.Request.Operation == "submit" {
 		info.IsStream = true
 		info.IsStream = true
 	}
 	}
 
 
-	// Return nil as WebSocket doesn't use traditional request body
 	jsonData, err := json.Marshal(volcRequest)
 	jsonData, err := json.Marshal(volcRequest)
 	if err != nil {
 	if err != nil {
 		return nil, fmt.Errorf("error marshalling volcengine request: %w", err)
 		return nil, fmt.Errorf("error marshalling volcengine request: %w", err)
@@ -115,9 +110,8 @@ func (a *Adaptor) ConvertImageRequest(c *gin.Context, info *relaycommon.RelayInf
 		writer := multipart.NewWriter(&requestBody)
 		writer := multipart.NewWriter(&requestBody)
 
 
 		writer.WriteField("model", request.Model)
 		writer.WriteField("model", request.Model)
-		// 获取所有表单字段
+
 		formData := c.Request.PostForm
 		formData := c.Request.PostForm
-		// 遍历表单字段并打印输出
 		for key, values := range formData {
 		for key, values := range formData {
 			if key == "model" {
 			if key == "model" {
 				continue
 				continue
@@ -127,21 +121,16 @@ func (a *Adaptor) ConvertImageRequest(c *gin.Context, info *relaycommon.RelayInf
 			}
 			}
 		}
 		}
 
 
-		// Parse the multipart form to handle both single image and multiple images
-		if err := c.Request.ParseMultipartForm(32 << 20); err != nil { // 32MB max memory
+		if err := c.Request.ParseMultipartForm(32 << 20); err != nil {
 			return nil, errors.New("failed to parse multipart form")
 			return nil, errors.New("failed to parse multipart form")
 		}
 		}
 
 
 		if c.Request.MultipartForm != nil && c.Request.MultipartForm.File != nil {
 		if c.Request.MultipartForm != nil && c.Request.MultipartForm.File != nil {
-			// Check if "image" field exists in any form, including array notation
 			var imageFiles []*multipart.FileHeader
 			var imageFiles []*multipart.FileHeader
 			var exists bool
 			var exists bool
 
 
-			// First check for standard "image" field
 			if imageFiles, exists = c.Request.MultipartForm.File["image"]; !exists || len(imageFiles) == 0 {
 			if imageFiles, exists = c.Request.MultipartForm.File["image"]; !exists || len(imageFiles) == 0 {
-				// If not found, check for "image[]" field
 				if imageFiles, exists = c.Request.MultipartForm.File["image[]"]; !exists || len(imageFiles) == 0 {
 				if imageFiles, exists = c.Request.MultipartForm.File["image[]"]; !exists || len(imageFiles) == 0 {
-					// If still not found, iterate through all fields to find any that start with "image["
 					foundArrayImages := false
 					foundArrayImages := false
 					for fieldName, files := range c.Request.MultipartForm.File {
 					for fieldName, files := range c.Request.MultipartForm.File {
 						if strings.HasPrefix(fieldName, "image[") && len(files) > 0 {
 						if strings.HasPrefix(fieldName, "image[") && len(files) > 0 {
@@ -152,14 +141,12 @@ func (a *Adaptor) ConvertImageRequest(c *gin.Context, info *relaycommon.RelayInf
 						}
 						}
 					}
 					}
 
 
-					// If no image fields found at all
 					if !foundArrayImages && (len(imageFiles) == 0) {
 					if !foundArrayImages && (len(imageFiles) == 0) {
 						return nil, errors.New("image is required")
 						return nil, errors.New("image is required")
 					}
 					}
 				}
 				}
 			}
 			}
 
 
-			// Process all image files
 			for i, fileHeader := range imageFiles {
 			for i, fileHeader := range imageFiles {
 				file, err := fileHeader.Open()
 				file, err := fileHeader.Open()
 				if err != nil {
 				if err != nil {
@@ -167,16 +154,13 @@ func (a *Adaptor) ConvertImageRequest(c *gin.Context, info *relaycommon.RelayInf
 				}
 				}
 				defer file.Close()
 				defer file.Close()
 
 
-				// If multiple images, use image[] as the field name
 				fieldName := "image"
 				fieldName := "image"
 				if len(imageFiles) > 1 {
 				if len(imageFiles) > 1 {
 					fieldName = "image[]"
 					fieldName = "image[]"
 				}
 				}
 
 
-				// Determine MIME type based on file extension
 				mimeType := detectImageMimeType(fileHeader.Filename)
 				mimeType := detectImageMimeType(fileHeader.Filename)
 
 
-				// Create a form file with the appropriate content type
 				h := make(textproto.MIMEHeader)
 				h := make(textproto.MIMEHeader)
 				h.Set("Content-Disposition", fmt.Sprintf(`form-data; name="%s"; filename="%s"`, fieldName, fileHeader.Filename))
 				h.Set("Content-Disposition", fmt.Sprintf(`form-data; name="%s"; filename="%s"`, fieldName, fileHeader.Filename))
 				h.Set("Content-Type", mimeType)
 				h.Set("Content-Type", mimeType)
@@ -191,7 +175,6 @@ func (a *Adaptor) ConvertImageRequest(c *gin.Context, info *relaycommon.RelayInf
 				}
 				}
 			}
 			}
 
 
-			// Handle mask file if present
 			if maskFiles, exists := c.Request.MultipartForm.File["mask"]; exists && len(maskFiles) > 0 {
 			if maskFiles, exists := c.Request.MultipartForm.File["mask"]; exists && len(maskFiles) > 0 {
 				maskFile, err := maskFiles[0].Open()
 				maskFile, err := maskFiles[0].Open()
 				if err != nil {
 				if err != nil {
@@ -199,10 +182,8 @@ func (a *Adaptor) ConvertImageRequest(c *gin.Context, info *relaycommon.RelayInf
 				}
 				}
 				defer maskFile.Close()
 				defer maskFile.Close()
 
 
-				// Determine MIME type for mask file
 				mimeType := detectImageMimeType(maskFiles[0].Filename)
 				mimeType := detectImageMimeType(maskFiles[0].Filename)
 
 
-				// Create a form file with the appropriate content type
 				h := make(textproto.MIMEHeader)
 				h := make(textproto.MIMEHeader)
 				h.Set("Content-Disposition", fmt.Sprintf(`form-data; name="mask"; filename="%s"`, maskFiles[0].Filename))
 				h.Set("Content-Disposition", fmt.Sprintf(`form-data; name="mask"; filename="%s"`, maskFiles[0].Filename))
 				h.Set("Content-Type", mimeType)
 				h.Set("Content-Type", mimeType)
@@ -220,7 +201,6 @@ func (a *Adaptor) ConvertImageRequest(c *gin.Context, info *relaycommon.RelayInf
 			return nil, errors.New("no multipart form data found")
 			return nil, errors.New("no multipart form data found")
 		}
 		}
 
 
-		// 关闭 multipart 编写器以设置分界线
 		writer.Close()
 		writer.Close()
 		c.Request.Header.Set("Content-Type", writer.FormDataContentType())
 		c.Request.Header.Set("Content-Type", writer.FormDataContentType())
 		return bytes.NewReader(requestBody.Bytes()), nil
 		return bytes.NewReader(requestBody.Bytes()), nil
@@ -230,7 +210,6 @@ func (a *Adaptor) ConvertImageRequest(c *gin.Context, info *relaycommon.RelayInf
 	}
 	}
 }
 }
 
 
-// detectImageMimeType determines the MIME type based on the file extension
 func detectImageMimeType(filename string) string {
 func detectImageMimeType(filename string) string {
 	ext := strings.ToLower(filepath.Ext(filename))
 	ext := strings.ToLower(filepath.Ext(filename))
 	switch ext {
 	switch ext {
@@ -241,11 +220,9 @@ func detectImageMimeType(filename string) string {
 	case ".webp":
 	case ".webp":
 		return "image/webp"
 		return "image/webp"
 	default:
 	default:
-		// Try to detect from extension if possible
 		if strings.HasPrefix(ext, ".jp") {
 		if strings.HasPrefix(ext, ".jp") {
 			return "image/jpeg"
 			return "image/jpeg"
 		}
 		}
-		// Default to png as a fallback
 		return "image/png"
 		return "image/png"
 	}
 	}
 }
 }
@@ -281,7 +258,6 @@ func (a *Adaptor) GetRequestURL(info *relaycommon.RelayInfo) (string, error) {
 		case constant.RelayModeRerank:
 		case constant.RelayModeRerank:
 			return fmt.Sprintf("%s/api/v3/rerank", baseUrl), nil
 			return fmt.Sprintf("%s/api/v3/rerank", baseUrl), nil
 		case constant.RelayModeAudioSpeech:
 		case constant.RelayModeAudioSpeech:
-			// 只有当 baseUrl 是火山默认的官方Url时才改为官方的的TTS接口，否则走透传的New接口
 			if baseUrl == channelconstant.ChannelBaseURLs[channelconstant.ChannelTypeVolcEngine] {
 			if baseUrl == channelconstant.ChannelBaseURLs[channelconstant.ChannelTypeVolcEngine] {
 				return "wss://openspeech.bytedance.com/api/v1/tts/ws_binary", nil
 				return "wss://openspeech.bytedance.com/api/v1/tts/ws_binary", nil
 			}
 			}
@@ -312,7 +288,7 @@ func (a *Adaptor) ConvertOpenAIRequest(c *gin.Context, info *relaycommon.RelayIn
 	if request == nil {
 	if request == nil {
 		return nil, errors.New("request is nil")
 		return nil, errors.New("request is nil")
 	}
 	}
-	// 适配 方舟deepseek混合模型 的 thinking 后缀
+
 	if strings.HasSuffix(info.UpstreamModelName, "-thinking") && strings.HasPrefix(info.UpstreamModelName, "deepseek") {
 	if strings.HasSuffix(info.UpstreamModelName, "-thinking") && strings.HasPrefix(info.UpstreamModelName, "deepseek") {
 		info.UpstreamModelName = strings.TrimSuffix(info.UpstreamModelName, "-thinking")
 		info.UpstreamModelName = strings.TrimSuffix(info.UpstreamModelName, "-thinking")
 		request.Model = info.UpstreamModelName
 		request.Model = info.UpstreamModelName
@@ -330,18 +306,16 @@ func (a *Adaptor) ConvertEmbeddingRequest(c *gin.Context, info *relaycommon.Rela
 }
 }
 
 
 func (a *Adaptor) ConvertOpenAIResponsesRequest(c *gin.Context, info *relaycommon.RelayInfo, request dto.OpenAIResponsesRequest) (any, error) {
 func (a *Adaptor) ConvertOpenAIResponsesRequest(c *gin.Context, info *relaycommon.RelayInfo, request dto.OpenAIResponsesRequest) (any, error) {
-	// TODO implement me
 	return nil, errors.New("not implemented")
 	return nil, errors.New("not implemented")
 }
 }
 
 
 func (a *Adaptor) DoRequest(c *gin.Context, info *relaycommon.RelayInfo, requestBody io.Reader) (any, error) {
 func (a *Adaptor) DoRequest(c *gin.Context, info *relaycommon.RelayInfo, requestBody io.Reader) (any, error) {
-	// For TTS with WebSocket, skip traditional HTTP request
 	if info.RelayMode == constant.RelayModeAudioSpeech {
 	if info.RelayMode == constant.RelayModeAudioSpeech {
 		baseUrl := info.ChannelBaseUrl
 		baseUrl := info.ChannelBaseUrl
 		if baseUrl == "" {
 		if baseUrl == "" {
 			baseUrl = channelconstant.ChannelBaseURLs[channelconstant.ChannelTypeVolcEngine]
 			baseUrl = channelconstant.ChannelBaseURLs[channelconstant.ChannelTypeVolcEngine]
 		}
 		}
-		// Only use WebSocket for official Volcengine endpoint
+
 		if baseUrl == channelconstant.ChannelBaseURLs[channelconstant.ChannelTypeVolcEngine] {
 		if baseUrl == channelconstant.ChannelBaseURLs[channelconstant.ChannelTypeVolcEngine] {
 			if info.IsStream {
 			if info.IsStream {
 				return nil, nil
 				return nil, nil

relay/channel/volcengine/tts.go

@@ -196,9 +196,7 @@ func generateRequestID() string {
 	return uuid.New().String()
 	return uuid.New().String()
 }
 }
 
 
-// handleTTSWebSocketResponse handles streaming TTS response via WebSocket
 func handleTTSWebSocketResponse(c *gin.Context, requestURL string, volcRequest VolcengineTTSRequest, info *relaycommon.RelayInfo, encoding string) (usage any, err *types.NewAPIError) {
 func handleTTSWebSocketResponse(c *gin.Context, requestURL string, volcRequest VolcengineTTSRequest, info *relaycommon.RelayInfo, encoding string) (usage any, err *types.NewAPIError) {
-	// Parse API key for auth
 	_, token, parseErr := parseVolcengineAuth(info.ApiKey)
 	_, token, parseErr := parseVolcengineAuth(info.ApiKey)
 	if parseErr != nil {
 	if parseErr != nil {
 		return nil, types.NewErrorWithStatusCode(
 		return nil, types.NewErrorWithStatusCode(
@@ -208,11 +206,9 @@ func handleTTSWebSocketResponse(c *gin.Context, requestURL string, volcRequest V
 		)
 		)
 	}
 	}
 
 
-	// Setup WebSocket headers
 	header := http.Header{}
 	header := http.Header{}
 	header.Set("Authorization", fmt.Sprintf("Bearer;%s", token))
 	header.Set("Authorization", fmt.Sprintf("Bearer;%s", token))
 
 
-	// Dial WebSocket connection
 	conn, resp, dialErr := websocket.DefaultDialer.DialContext(context.Background(), requestURL, header)
 	conn, resp, dialErr := websocket.DefaultDialer.DialContext(context.Background(), requestURL, header)
 	if dialErr != nil {
 	if dialErr != nil {
 		if resp != nil {
 		if resp != nil {
@@ -239,7 +235,6 @@ func handleTTSWebSocketResponse(c *gin.Context, requestURL string, volcRequest V
 		)
 		)
 	}
 	}
 
 
-	// Send full client request
 	if sendErr := FullClientRequest(conn, payload); sendErr != nil {
 	if sendErr := FullClientRequest(conn, payload); sendErr != nil {
 		return nil, types.NewErrorWithStatusCode(
 		return nil, types.NewErrorWithStatusCode(
 			fmt.Errorf("failed to send request: %w", sendErr),
 			fmt.Errorf("failed to send request: %w", sendErr),
@@ -248,13 +243,10 @@ func handleTTSWebSocketResponse(c *gin.Context, requestURL string, volcRequest V
 		)
 		)
 	}
 	}
 
 
-	// Set response headers
 	contentType := getContentTypeByEncoding(encoding)
 	contentType := getContentTypeByEncoding(encoding)
 	c.Header("Content-Type", contentType)
 	c.Header("Content-Type", contentType)
 	c.Header("Transfer-Encoding", "chunked")
 	c.Header("Transfer-Encoding", "chunked")
 
 
-	// Stream audio data
-	var audioBuffer []byte
 	for {
 	for {
 		msg, recvErr := ReceiveMessage(conn)
 		msg, recvErr := ReceiveMessage(conn)
 		if recvErr != nil {
 		if recvErr != nil {
@@ -279,7 +271,6 @@ func handleTTSWebSocketResponse(c *gin.Context, requestURL string, volcRequest V
 			continue
 			continue
 		case MsgTypeAudioOnlyServer:
 		case MsgTypeAudioOnlyServer:
 			if len(msg.Payload) > 0 {
 			if len(msg.Payload) > 0 {
-				audioBuffer = append(audioBuffer, msg.Payload...)
 				if _, writeErr := c.Writer.Write(msg.Payload); writeErr != nil {
 				if _, writeErr := c.Writer.Write(msg.Payload); writeErr != nil {
 					return nil, types.NewErrorWithStatusCode(
 					return nil, types.NewErrorWithStatusCode(
 						fmt.Errorf("failed to write audio data: %w", writeErr),
 						fmt.Errorf("failed to write audio data: %w", writeErr),
@@ -287,7 +278,6 @@ func handleTTSWebSocketResponse(c *gin.Context, requestURL string, volcRequest V
 						http.StatusInternalServerError,
 						http.StatusInternalServerError,
 					)
 					)
 				}
 				}
-				//logger.Infof("write audio chunk size: %d", len(msg.Payload))
 				c.Writer.Flush()
 				c.Writer.Flush()
 			}
 			}