feat: Add ContextKeyLocalCountTokens and update ResponseText2Usage to use context in multiple channels

relay/channel/gemini/relay-gemini-native.go

@@ -3,7 +3,6 @@ package gemini
 import (
 	"io"
 	"net/http"
-	"strings"
 
 	"github.com/QuantumNous/new-api/common"
 	"github.com/QuantumNous/new-api/dto"
@@ -13,8 +12,6 @@ import (
 	"github.com/QuantumNous/new-api/service"
 	"github.com/QuantumNous/new-api/types"
 
-	"github.com/pkg/errors"
-
 	"github.com/gin-gonic/gin"
 )
 
@@ -97,80 +94,15 @@ func NativeGeminiEmbeddingHandler(c *gin.Context, resp *http.Response, info *rel
 }
 
 func GeminiTextGenerationStreamHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *http.Response) (*dto.Usage, *types.NewAPIError) {
-	var usage = &dto.Usage{}
-	var imageCount int
-
 	helper.SetEventStreamHeaders(c)
 
-	responseText := strings.Builder{}
-
-	helper.StreamScannerHandler(c, resp, info, func(data string) bool {
-		var geminiResponse dto.GeminiChatResponse
-		err := common.UnmarshalJsonStr(data, &geminiResponse)
-		if err != nil {
-			logger.LogError(c, "error unmarshalling stream response: "+err.Error())
-			return false
-		}
-
-		// 统计图片数量
-		for _, candidate := range geminiResponse.Candidates {
-			for _, part := range candidate.Content.Parts {
-				if part.InlineData != nil && part.InlineData.MimeType != "" {
-					imageCount++
-				}
-				if part.Text != "" {
-					responseText.WriteString(part.Text)
-				}
-			}
-		}
-
-		// 更新使用量统计
-		if geminiResponse.UsageMetadata.TotalTokenCount != 0 {
-			usage.PromptTokens = geminiResponse.UsageMetadata.PromptTokenCount
-			usage.CompletionTokens = geminiResponse.UsageMetadata.CandidatesTokenCount + geminiResponse.UsageMetadata.ThoughtsTokenCount
-			usage.TotalTokens = geminiResponse.UsageMetadata.TotalTokenCount
-			usage.CompletionTokenDetails.ReasoningTokens = geminiResponse.UsageMetadata.ThoughtsTokenCount
-			for _, detail := range geminiResponse.UsageMetadata.PromptTokensDetails {
-				if detail.Modality == "AUDIO" {
-					usage.PromptTokensDetails.AudioTokens = detail.TokenCount
-				} else if detail.Modality == "TEXT" {
-					usage.PromptTokensDetails.TextTokens = detail.TokenCount
-				}
-			}
-		}
-
+	return geminiStreamHandler(c, info, resp, func(data string, geminiResponse *dto.GeminiChatResponse) bool {
 		// 直接发送 GeminiChatResponse 响应
-		err = helper.StringData(c, data)
+		err := helper.StringData(c, data)
 		if err != nil {
 			logger.LogError(c, err.Error())
 		}
 		info.SendResponseCount++
 		return true
 	})
-
-	if info.SendResponseCount == 0 {
-		return nil, types.NewOpenAIError(errors.New("no response received from Gemini API"), types.ErrorCodeEmptyResponse, http.StatusInternalServerError)
-	}
-
-	if imageCount != 0 {
-		if usage.CompletionTokens == 0 {
-			usage.CompletionTokens = imageCount * 258
-		}
-	}
-
-	// 如果usage.CompletionTokens为0，则使用本地统计的completion tokens
-	if usage.CompletionTokens == 0 {
-		str := responseText.String()
-		if len(str) > 0 {
-			usage = service.ResponseText2Usage(responseText.String(), info.UpstreamModelName, info.PromptTokens)
-		} else {
-			// 空补全，不需要使用量
-			usage = &dto.Usage{}
-		}
-	}
-
-	// 移除流式响应结尾的[Done]，因为Gemini API没有发送Done的行为
-	//helper.Done(c)
-
-	return usage, nil
 }