refactor: token counter logic

relay/channel/openai/relay-openai.go

@@ -8,7 +8,6 @@ import (
 	"math"
 	"mime/multipart"
 	"net/http"
-	"path/filepath"
 	"one-api/common"
 	"one-api/constant"
 	"one-api/dto"
@@ -16,6 +15,7 @@ import (
 	"one-api/relay/helper"
 	"one-api/service"
 	"os"
+	"path/filepath"
 	"strings"
 
 	"github.com/bytedance/gopkg/util/gopool"
@@ -181,7 +181,7 @@ func OaiStreamHandler(c *gin.Context, resp *http.Response, info *relaycommon.Rel
 	}
 
 	if !containStreamUsage {
-		usage, _ = service.ResponseText2Usage(responseTextBuilder.String(), info.UpstreamModelName, info.PromptTokens)
+		usage = service.ResponseText2Usage(responseTextBuilder.String(), info.UpstreamModelName, info.PromptTokens)
 		usage.CompletionTokens += toolCount * 7
 	} else {
 		if info.ChannelType == common.ChannelTypeDeepSeek {
@@ -216,7 +216,7 @@ func OpenaiHandler(c *gin.Context, resp *http.Response, info *relaycommon.RelayI
 			StatusCode: resp.StatusCode,
 		}, nil
 	}
-	
+
 	forceFormat := false
 	if forceFmt, ok := info.ChannelSetting[constant.ForceFormat].(bool); ok {
 		forceFormat = forceFmt
@@ -225,7 +225,7 @@ func OpenaiHandler(c *gin.Context, resp *http.Response, info *relaycommon.RelayI
 	if simpleResponse.Usage.TotalTokens == 0 || (simpleResponse.Usage.PromptTokens == 0 && simpleResponse.Usage.CompletionTokens == 0) {
 		completionTokens := 0
 		for _, choice := range simpleResponse.Choices {
-			ctkm, _ := service.CountTextToken(choice.Message.StringContent()+choice.Message.ReasoningContent+choice.Message.Reasoning, info.UpstreamModelName)
+			ctkm := service.CountTextToken(choice.Message.StringContent()+choice.Message.ReasoningContent+choice.Message.Reasoning, info.UpstreamModelName)
 			completionTokens += ctkm
 		}
 		simpleResponse.Usage = dto.Usage{
@@ -276,9 +276,9 @@ func OpenaiHandler(c *gin.Context, resp *http.Response, info *relaycommon.RelayI
 func OpenaiTTSHandler(c *gin.Context, resp *http.Response, info *relaycommon.RelayInfo) (*dto.OpenAIErrorWithStatusCode, *dto.Usage) {
 	// the status code has been judged before, if there is a body reading failure,
 	// it should be regarded as a non-recoverable error, so it should not return err for external retry.
-	// Analogous to nginx's load balancing, it will only retry if it can't be requested or 
-	// if the upstream returns a specific status code, once the upstream has already written the header, 
-	// the subsequent failure of the response body should be regarded as a non-recoverable error, 
+	// Analogous to nginx's load balancing, it will only retry if it can't be requested or
+	// if the upstream returns a specific status code, once the upstream has already written the header,
+	// the subsequent failure of the response body should be regarded as a non-recoverable error,
 	// and can be terminated directly.
 	defer resp.Body.Close()
 	usage := &dto.Usage{}
@@ -346,12 +346,12 @@ func countAudioTokens(c *gin.Context) (int, error) {
 	if err = c.ShouldBind(&reqBody); err != nil {
 		return 0, errors.WithStack(err)
 	}
-  ext := filepath.Ext(reqBody.File.Filename) // 获取文件扩展名
+	ext := filepath.Ext(reqBody.File.Filename) // 获取文件扩展名
 	reqFp, err := reqBody.File.Open()
 	if err != nil {
 		return 0, errors.WithStack(err)
 	}
-  defer reqFp.Close()
+	defer reqFp.Close()
 
 	tmpFp, err := os.CreateTemp("", "audio-*"+ext)
 	if err != nil {

relay/channel/openai/relay_responses.go

@@ -110,7 +110,7 @@ func OaiResponsesStreamHandler(c *gin.Context, resp *http.Response, info *relayc
 		tempStr := responseTextBuilder.String()
 		if len(tempStr) > 0 {
 			// 非正常结束，使用输出文本的 token 数量
-			completionTokens, _ := service.CountTextToken(tempStr, info.UpstreamModelName)
+			completionTokens := service.CountTextToken(tempStr, info.UpstreamModelName)
 			usage.CompletionTokens = completionTokens
 		}
 	}