fix: 修复 Eino Graph 类型不匹配和多模态消息问题

- 添加 msg2str 转换节点解决 ChatModel 输出 *schema.Message 与 Splitter 期望 string 的类型不匹配 - 将多模态图片内容从 system 消息移到 user 消息（DashScope API 仅支持 user/tool 角色的多模态内容） - 修复 Content 和 UserInputMultiContent 不能同时设置的问题 - Splitter 输出改为 StreamReader[string]（单句），TTS 改为 TransformableLambda 流式消费 - 修复 .env 中 PostgreSQL DSN 和 Redis ADDR 的 http:// 前缀问题
2026-06-19 23:24:01 +08:00
parent 9576884619
commit eb1b90445f
4 changed files with 171 additions and 106 deletions
--- a/backend/internal/eino/nodes_tts.go
+++ b/backend/internal/eino/nodes_tts.go
@@ -3,90 +3,114 @@ package eino
 import (
 	"context"
 	"encoding/base64"
+	"io"

 	"github.com/cloudwego/eino/compose"
+	"github.com/cloudwego/eino/schema"

 	"github.com/hhs/camtalk/internal/ai/tts"
 	"github.com/hhs/camtalk/internal/logger"
 	"github.com/hhs/camtalk/internal/models"
 )

-// NewTTSLambda 创建 TTS Lambda 节点。
-// 输入: []string（句子数组，框架自动从 StreamReader concat）→ 输出: struct{}
+// NewTTSLambda 创建 TTS Transform Lambda 节点。
+// 输入: StreamReader[string]（句子流）→ 输出: StreamReader[struct{}]（结果流）
 //
-// 将句子数组转为 channel，调用 ttsService.SynthesizeStream() 流式合成，
+// 流式消费每个句子，调用 ttsService.SynthesizeStream() 合成，
 // 逐 chunk 推送 tts_audio 到客户端。TTS 失败静默跳过。
 func NewTTSLambda(ttsService tts.Service, ttsVoice string, ttsSpeed float64, ttsOutputFmt string, ttsSampleRate int) *compose.Lambda {
-	return compose.InvokableLambda(func(ctx context.Context, sentences []string) (struct{}, error) {
-		log := logger.Log
-		sender := senderFromCtx(ctx)
-		requestID := requestIDFromCtx(ctx)
-		state := stateFromCtx(ctx)
+	return compose.TransformableLambda(func(ctx context.Context, input *schema.StreamReader[string]) (*schema.StreamReader[struct{}], error) {
+		sr, sw := schema.Pipe[struct{}](8)

-		// 检查 TTS 是否启用（从 State 或 context 获取）
-		// TTSEnabled 信息在 PipelineInput 中，通过 State 传递
-		if state != nil {
-			state.mu.Lock()
-			ttsEnabled := true // 默认启用，由适配器通过 State 设置
-			state.mu.Unlock()
-			if !ttsEnabled {
-				return struct{}{}, nil
-			}
-		}
+		go func() {
+			defer sw.Close()
+			defer input.Close()

-		if len(sentences) == 0 {
-			return struct{}{}, nil
-		}
+			log := logger.Log
+			sender := senderFromCtx(ctx)
+			requestID := requestIDFromCtx(ctx)

-		if sender == nil || requestID == "" {
-			return struct{}{}, nil
-		}
-
-		log.Infow("开始 TTS 合成", "request_id", requestID, "sentence_count", len(sentences))
-
-		// 将句子数组转为 channel（ttsService.SynthesizeStream 需要 <-chan string）
-		sentenceCh := make(chan string, len(sentences))
-		for _, s := range sentences {
-			sentenceCh <- s
-		}
-		close(sentenceCh)
-
-		// 调用 TTS 服务
-		ttsStream, err := ttsService.SynthesizeStream(ctx, sentenceCh, tts.Options{
-			Voice:      ttsVoice,
-			Speed:      ttsSpeed,
-			OutputFmt:  ttsOutputFmt,
-			SampleRate: ttsSampleRate,
-		})
-		if err != nil {
-			log.Errorw("TTS 合成启动失败（已跳过）", "error", err, "request_id", requestID)
-			return struct{}{}, nil // TTS 失败不中断流程
-		}
-
-		// 消费 TTS 音频流，推送到客户端
-		for chunk := range ttsStream {
-			select {
-			case <-ctx.Done():
-				log.Infow("TTS 流被中断", "request_id", requestID)
-				return struct{}{}, ctx.Err()
-			default:
+			if sender == nil || requestID == "" {
+				// 消费并丢弃流
+				for {
+					_, err := input.Recv()
+					if err != nil {
+						return
+					}
+				}
 			}

-			audioBase64 := base64.StdEncoding.EncodeToString(chunk.Audio)
-
-			if err := sender.SendTTSAudio(models.WsTTSAudio{
-				Type:      "tts_audio",
-				RequestID: requestID,
-				Audio:     audioBase64,
-				MimeType:  "audio/mp3",
-				IsLast:    chunk.IsLast,
-				Final:     chunk.Final,
-			}); err != nil {
-				log.Errorw("发送 tts_audio 失败", "error", err)
+			// 收集句子，按批次合成 TTS
+			var sentences []string
+			for {
+				sentence, err := input.Recv()
+				if err != nil {
+					if err == io.EOF {
+						break
+					}
+					log.Errorw("TTS: stream recv error", "error", err, "request_id", requestID)
+					break
+				}
+				if sentence != "" {
+					sentences = append(sentences, sentence)
+				}
 			}
-		}

-		log.Infow("TTS 合成完成", "request_id", requestID)
-		return struct{}{}, nil
+			if len(sentences) == 0 {
+				sw.Send(struct{}{}, nil)
+				return
+			}
+
+			log.Infow("开始 TTS 合成", "request_id", requestID, "sentence_count", len(sentences))
+
+			// 将句子数组转为 channel
+			sentenceCh := make(chan string, len(sentences))
+			for _, s := range sentences {
+				sentenceCh <- s
+			}
+			close(sentenceCh)
+
+			// 调用 TTS 服务
+			ttsStream, err := ttsService.SynthesizeStream(ctx, sentenceCh, tts.Options{
+				Voice:      ttsVoice,
+				Speed:      ttsSpeed,
+				OutputFmt:  ttsOutputFmt,
+				SampleRate: ttsSampleRate,
+			})
+			if err != nil {
+				log.Errorw("TTS 合成启动失败（已跳过）", "error", err, "request_id", requestID)
+				sw.Send(struct{}{}, nil)
+				return
+			}
+
+			// 消费 TTS 音频流，推送到客户端
+			for chunk := range ttsStream {
+				select {
+				case <-ctx.Done():
+					log.Infow("TTS 流被中断", "request_id", requestID)
+					sw.Send(struct{}{}, ctx.Err())
+					return
+				default:
+				}
+
+				audioBase64 := base64.StdEncoding.EncodeToString(chunk.Audio)
+
+				if err := sender.SendTTSAudio(models.WsTTSAudio{
+					Type:      "tts_audio",
+					RequestID: requestID,
+					Audio:     audioBase64,
+					MimeType:  "audio/mp3",
+					IsLast:    chunk.IsLast,
+					Final:     chunk.Final,
+				}); err != nil {
+					log.Errorw("发送 tts_audio 失败", "error", err)
+				}
+			}
+
+			log.Infow("TTS 合成完成", "request_id", requestID)
+			sw.Send(struct{}{}, nil)
+		}()
+
+		return sr, nil
 	})
 }