refactor: Introduce pre-consume quota and unify relay handlers

This commit introduces a major architectural refactoring to improve quota management, centralize logging, and streamline the relay handling logic.

Key changes:
- **Pre-consume Quota:** Implements a new mechanism to check and reserve user quota *before* making the request to the upstream provider. This ensures more accurate quota deduction and prevents users from exceeding their limits due to concurrent requests.

- **Unified Relay Handlers:** Refactors the relay logic to use generic handlers (e.g., `ChatHandler`, `ImageHandler`) instead of provider-specific implementations. This significantly reduces code duplication and simplifies adding new channels.

- **Centralized Logger:** A new dedicated `logger` package is introduced, and all system logging calls are migrated to use it, moving this responsibility out of the `common` package.

- **Code Reorganization:** DTOs are generalized (e.g., `dalle.go` -> `openai_image.go`) and utility code is moved to more appropriate packages (e.g., `common/http.go` -> `service/http.go`) for better code structure.

relay/channel/openai/helper.go

@@ -7,6 +7,7 @@ import (
 	"net/http"
 	"one-api/common"
 	"one-api/dto"
+	"one-api/logger"
 	relaycommon "one-api/relay/common"
 	relayconstant "one-api/relay/constant"
 	"one-api/relay/helper"
@@ -50,7 +51,7 @@ func handleClaudeFormat(c *gin.Context, data string, info *relaycommon.RelayInfo
 func handleGeminiFormat(c *gin.Context, data string, info *relaycommon.RelayInfo) error {
 	var streamResponse dto.ChatCompletionsStreamResponse
 	if err := common.Unmarshal(common.StringToByteSlice(data), &streamResponse); err != nil {
-		common.LogError(c, "failed to unmarshal stream response: "+err.Error())
+		logger.LogError(c, "failed to unmarshal stream response: "+err.Error())
 		return err
 	}
 
@@ -63,7 +64,7 @@ func handleGeminiFormat(c *gin.Context, data string, info *relaycommon.RelayInfo
 
 	geminiResponseStr, err := common.Marshal(geminiResponse)
 	if err != nil {
-		common.LogError(c, "failed to marshal gemini response: "+err.Error())
+		logger.LogError(c, "failed to marshal gemini response: "+err.Error())
 		return err
 	}
 
@@ -110,14 +111,14 @@ func processChatCompletions(streamResp string, streamItems []string, responseTex
 	var streamResponses []dto.ChatCompletionsStreamResponse
 	if err := json.Unmarshal(common.StringToByteSlice(streamResp), &streamResponses); err != nil {
 		// 一次性解析失败，逐个解析
-		common.SysError("error unmarshalling stream response: " + err.Error())
+		logger.SysError("error unmarshalling stream response: " + err.Error())
 		for _, item := range streamItems {
 			var streamResponse dto.ChatCompletionsStreamResponse
 			if err := json.Unmarshal(common.StringToByteSlice(item), &streamResponse); err != nil {
 				return err
 			}
 			if err := ProcessStreamResponse(streamResponse, responseTextBuilder, toolCount); err != nil {
-				common.SysError("error processing stream response: " + err.Error())
+				logger.SysError("error processing stream response: " + err.Error())
 			}
 		}
 		return nil
@@ -146,7 +147,7 @@ func processCompletions(streamResp string, streamItems []string, responseTextBui
 	var streamResponses []dto.CompletionsStreamResponse
 	if err := json.Unmarshal(common.StringToByteSlice(streamResp), &streamResponses); err != nil {
 		// 一次性解析失败，逐个解析
-		common.SysError("error unmarshalling stream response: " + err.Error())
+		logger.SysError("error unmarshalling stream response: " + err.Error())
 		for _, item := range streamItems {
 			var streamResponse dto.CompletionsStreamResponse
 			if err := json.Unmarshal(common.StringToByteSlice(item), &streamResponse); err != nil {
@@ -213,7 +214,7 @@ func HandleFinalResponse(c *gin.Context, info *relaycommon.RelayInfo, lastStream
 		info.ClaudeConvertInfo.Done = true
 		var streamResponse dto.ChatCompletionsStreamResponse
 		if err := common.Unmarshal(common.StringToByteSlice(lastStreamData), &streamResponse); err != nil {
-			common.SysError("error unmarshalling stream response: " + err.Error())
+			logger.SysError("error unmarshalling stream response: " + err.Error())
 			return
 		}
 
@@ -227,7 +228,7 @@ func HandleFinalResponse(c *gin.Context, info *relaycommon.RelayInfo, lastStream
 	case relaycommon.RelayFormatGemini:
 		var streamResponse dto.ChatCompletionsStreamResponse
 		if err := common.Unmarshal(common.StringToByteSlice(lastStreamData), &streamResponse); err != nil {
-			common.SysError("error unmarshalling stream response: " + err.Error())
+			logger.SysError("error unmarshalling stream response: " + err.Error())
 			return
 		}
 
@@ -245,7 +246,7 @@ func HandleFinalResponse(c *gin.Context, info *relaycommon.RelayInfo, lastStream
 
 		geminiResponseStr, err := common.Marshal(geminiResponse)
 		if err != nil {
-			common.SysError("error marshalling gemini response: " + err.Error())
+			logger.SysError("error marshalling gemini response: " + err.Error())
 			return
 		}
 

relay/channel/openai/relay-openai.go

@@ -10,6 +10,7 @@ import (
 	"one-api/common"
 	"one-api/constant"
 	"one-api/dto"
+	"one-api/logger"
 	relaycommon "one-api/relay/common"
 	"one-api/relay/helper"
 	"one-api/service"
@@ -108,11 +109,11 @@ func sendStreamData(c *gin.Context, info *relaycommon.RelayInfo, data string, fo
 
 func OaiStreamHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *http.Response) (*dto.Usage, *types.NewAPIError) {
 	if resp == nil || resp.Body == nil {
-		common.LogError(c, "invalid response or response body")
+		logger.LogError(c, "invalid response or response body")
 		return nil, types.NewOpenAIError(fmt.Errorf("invalid response"), types.ErrorCodeBadResponse, http.StatusInternalServerError)
 	}
 
-	defer common.CloseResponseBodyGracefully(resp)
+	defer service.CloseResponseBodyGracefully(resp)
 
 	model := info.UpstreamModelName
 	var responseId string
@@ -129,7 +130,7 @@ func OaiStreamHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *http.Re
 		if lastStreamData != "" {
 			err := HandleStreamFormat(c, info, lastStreamData, info.ChannelSetting.ForceFormat, info.ChannelSetting.ThinkingToContent)
 			if err != nil {
-				common.SysError("error handling stream format: " + err.Error())
+				logger.SysError("error handling stream format: " + err.Error())
 			}
 		}
 		if len(data) > 0 {
@@ -143,7 +144,7 @@ func OaiStreamHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *http.Re
 	shouldSendLastResp := true
 	if err := handleLastResponse(lastStreamData, &responseId, &createAt, &systemFingerprint, &model, &usage,
 		&containStreamUsage, info, &shouldSendLastResp); err != nil {
-		common.LogError(c, fmt.Sprintf("error handling last response: %s, lastStreamData: [%s]", err.Error(), lastStreamData))
+		logger.LogError(c, fmt.Sprintf("error handling last response: %s, lastStreamData: [%s]", err.Error(), lastStreamData))
 	}
 
 	if info.RelayFormat == relaycommon.RelayFormatOpenAI {
@@ -154,7 +155,7 @@ func OaiStreamHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *http.Re
 
 	// 处理token计算
 	if err := processTokens(info.RelayMode, streamItems, &responseTextBuilder, &toolCount); err != nil {
-		common.LogError(c, "error processing tokens: "+err.Error())
+		logger.LogError(c, "error processing tokens: "+err.Error())
 	}
 
 	if !containStreamUsage {
@@ -173,7 +174,7 @@ func OaiStreamHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *http.Re
 }
 
 func OpenaiHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *http.Response) (*dto.Usage, *types.NewAPIError) {
-	defer common.CloseResponseBodyGracefully(resp)
+	defer service.CloseResponseBodyGracefully(resp)
 
 	var simpleResponse dto.OpenAITextResponse
 	responseBody, err := io.ReadAll(resp.Body)
@@ -235,7 +236,7 @@ func OpenaiHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *http.Respo
 		responseBody = geminiRespStr
 	}
 
-	common.IOCopyBytesGracefully(c, resp, responseBody)
+	service.IOCopyBytesGracefully(c, resp, responseBody)
 
 	return &simpleResponse.Usage, nil
 }
@@ -247,7 +248,7 @@ func OpenaiTTSHandler(c *gin.Context, resp *http.Response, info *relaycommon.Rel
 	// 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 common.CloseResponseBodyGracefully(resp)
+	defer service.CloseResponseBodyGracefully(resp)
 	usage := &dto.Usage{}
 	usage.PromptTokens = info.PromptTokens
 	usage.TotalTokens = info.PromptTokens
@@ -258,13 +259,13 @@ func OpenaiTTSHandler(c *gin.Context, resp *http.Response, info *relaycommon.Rel
 	c.Writer.WriteHeaderNow()
 	_, err := io.Copy(c.Writer, resp.Body)
 	if err != nil {
-		common.LogError(c, err.Error())
+		logger.LogError(c, err.Error())
 	}
 	return usage
 }
 
 func OpenaiSTTHandler(c *gin.Context, resp *http.Response, info *relaycommon.RelayInfo, responseFormat string) (*types.NewAPIError, *dto.Usage) {
-	defer common.CloseResponseBodyGracefully(resp)
+	defer service.CloseResponseBodyGracefully(resp)
 
 	// count tokens by audio file duration
 	audioTokens, err := countAudioTokens(c)
@@ -276,7 +277,7 @@ func OpenaiSTTHandler(c *gin.Context, resp *http.Response, info *relaycommon.Rel
 		return types.NewOpenAIError(err, types.ErrorCodeReadResponseBodyFailed, http.StatusInternalServerError), nil
 	}
 	// 写入新的 response body
-	common.IOCopyBytesGracefully(c, resp, responseBody)
+	service.IOCopyBytesGracefully(c, resp, responseBody)
 
 	usage := &dto.Usage{}
 	usage.PromptTokens = audioTokens
@@ -386,7 +387,7 @@ func OpenaiRealtimeHandler(c *gin.Context, info *relaycommon.RelayInfo) (*types.
 					errChan <- fmt.Errorf("error counting text token: %v", err)
 					return
 				}
-				common.LogInfo(c, fmt.Sprintf("type: %s, textToken: %d, audioToken: %d", realtimeEvent.Type, textToken, audioToken))
+				logger.LogInfo(c, fmt.Sprintf("type: %s, textToken: %d, audioToken: %d", realtimeEvent.Type, textToken, audioToken))
 				localUsage.TotalTokens += textToken + audioToken
 				localUsage.InputTokens += textToken + audioToken
 				localUsage.InputTokenDetails.TextTokens += textToken
@@ -459,7 +460,7 @@ func OpenaiRealtimeHandler(c *gin.Context, info *relaycommon.RelayInfo) (*types.
 							errChan <- fmt.Errorf("error counting text token: %v", err)
 							return
 						}
-						common.LogInfo(c, fmt.Sprintf("type: %s, textToken: %d, audioToken: %d", realtimeEvent.Type, textToken, audioToken))
+						logger.LogInfo(c, fmt.Sprintf("type: %s, textToken: %d, audioToken: %d", realtimeEvent.Type, textToken, audioToken))
 						localUsage.TotalTokens += textToken + audioToken
 						info.IsFirstRequest = false
 						localUsage.InputTokens += textToken + audioToken
@@ -474,9 +475,9 @@ func OpenaiRealtimeHandler(c *gin.Context, info *relaycommon.RelayInfo) (*types.
 						localUsage = &dto.RealtimeUsage{}
 						// print now usage
 					}
-					common.LogInfo(c, fmt.Sprintf("realtime streaming sumUsage: %v", sumUsage))
-					common.LogInfo(c, fmt.Sprintf("realtime streaming localUsage: %v", localUsage))
-					common.LogInfo(c, fmt.Sprintf("realtime streaming localUsage: %v", localUsage))
+					logger.LogInfo(c, fmt.Sprintf("realtime streaming sumUsage: %v", sumUsage))
+					logger.LogInfo(c, fmt.Sprintf("realtime streaming localUsage: %v", localUsage))
+					logger.LogInfo(c, fmt.Sprintf("realtime streaming localUsage: %v", localUsage))
 
 				} else if realtimeEvent.Type == dto.RealtimeEventTypeSessionUpdated || realtimeEvent.Type == dto.RealtimeEventTypeSessionCreated {
 					realtimeSession := realtimeEvent.Session
@@ -491,7 +492,7 @@ func OpenaiRealtimeHandler(c *gin.Context, info *relaycommon.RelayInfo) (*types.
 						errChan <- fmt.Errorf("error counting text token: %v", err)
 						return
 					}
-					common.LogInfo(c, fmt.Sprintf("type: %s, textToken: %d, audioToken: %d", realtimeEvent.Type, textToken, audioToken))
+					logger.LogInfo(c, fmt.Sprintf("type: %s, textToken: %d, audioToken: %d", realtimeEvent.Type, textToken, audioToken))
 					localUsage.TotalTokens += textToken + audioToken
 					localUsage.OutputTokens += textToken + audioToken
 					localUsage.OutputTokenDetails.TextTokens += textToken
@@ -517,7 +518,7 @@ func OpenaiRealtimeHandler(c *gin.Context, info *relaycommon.RelayInfo) (*types.
 	case <-targetClosed:
 	case err := <-errChan:
 		//return service.OpenAIErrorWrapper(err, "realtime_error", http.StatusInternalServerError), nil
-		common.LogError(c, "realtime error: "+err.Error())
+		logger.LogError(c, "realtime error: "+err.Error())
 	case <-c.Done():
 	}
 
@@ -553,7 +554,7 @@ func preConsumeUsage(ctx *gin.Context, info *relaycommon.RelayInfo, usage *dto.R
 }
 
 func OpenaiHandlerWithUsage(c *gin.Context, info *relaycommon.RelayInfo, resp *http.Response) (*dto.Usage, *types.NewAPIError) {
-	defer common.CloseResponseBodyGracefully(resp)
+	defer service.CloseResponseBodyGracefully(resp)
 
 	responseBody, err := io.ReadAll(resp.Body)
 	if err != nil {
@@ -567,7 +568,7 @@ func OpenaiHandlerWithUsage(c *gin.Context, info *relaycommon.RelayInfo, resp *h
 	}
 
 	// 写入新的 response body
-	common.IOCopyBytesGracefully(c, resp, responseBody)
+	service.IOCopyBytesGracefully(c, resp, responseBody)
 
 	// Once we've written to the client, we should not return errors anymore
 	// because the upstream has already consumed resources and returned content

relay/channel/openai/relay_responses.go

@@ -6,6 +6,7 @@ import (
 	"net/http"
 	"one-api/common"
 	"one-api/dto"
+	"one-api/logger"
 	relaycommon "one-api/relay/common"
 	"one-api/relay/helper"
 	"one-api/service"
@@ -16,7 +17,7 @@ import (
 )
 
 func OaiResponsesHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *http.Response) (*dto.Usage, *types.NewAPIError) {
-	defer common.CloseResponseBodyGracefully(resp)
+	defer service.CloseResponseBodyGracefully(resp)
 
 	// read response body
 	var responsesResponse dto.OpenAIResponsesResponse
@@ -33,7 +34,7 @@ func OaiResponsesHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *http
 	}
 
 	// 写入新的 response body
-	common.IOCopyBytesGracefully(c, resp, responseBody)
+	service.IOCopyBytesGracefully(c, resp, responseBody)
 
 	// compute usage
 	usage := dto.Usage{}
@@ -54,7 +55,7 @@ func OaiResponsesHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *http
 
 func OaiResponsesStreamHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *http.Response) (*dto.Usage, *types.NewAPIError) {
 	if resp == nil || resp.Body == nil {
-		common.LogError(c, "invalid response or response body")
+		logger.LogError(c, "invalid response or response body")
 		return nil, types.NewError(fmt.Errorf("invalid response"), types.ErrorCodeBadResponse)
 	}