Merge pull request #592 from Calcium-Ion/gzip

feat: support gzip

README.md

@@ -71,7 +71,7 @@
 - `STREAMING_TIMEOUT`：设置流式一次回复的超时时间，默认为 60 秒。
 - `DIFY_DEBUG`：设置 Dify 渠道是否输出工作流和节点信息到客户端，默认为 `true`。
 - `FORCE_STREAM_OPTION`：是否覆盖客户端stream_options参数，请求上游返回流模式usage，默认为 `true`，建议开启，不影响客户端传入stream_options参数返回结果。
-- `GET_MEDIA_TOKEN`：是统计图片token，默认为 `true`，关闭后将不再在本地计算图片token，可能会导致和上游计费不同，此项覆盖 `GET_MEDIA_TOKEN_NOT_STREAM` 选项作用。
+- `GET_MEDIA_TOKEN`：是否统计图片token，默认为 `true`，关闭后将不再在本地计算图片token，可能会导致和上游计费不同，此项覆盖 `GET_MEDIA_TOKEN_NOT_STREAM` 选项作用。
 - `GET_MEDIA_TOKEN_NOT_STREAM`：是否在非流（`stream=false`）情况下统计图片token，默认为 `true`。
 - `UPDATE_TASK`：是否更新异步任务（Midjourney、Suno），默认为 `true`，关闭后将不会更新任务进度。
 - `GEMINI_MODEL_MAP`：Gemini模型指定版本(v1/v1beta)，使用“模型:版本”指定，","分隔，例如：-e GEMINI_MODEL_MAP="gemini-1.5-pro-latest:v1beta,gemini-1.5-pro-001:v1beta"，为空则使用默认配置(v1beta)
@@ -136,6 +136,7 @@ docker run --name new-api -d --restart always -p 3000:3000 -e SQL_DSN="root:1234
 ![796df8d287b7b7bd7853b2497e7df511](https://github.com/user-attachments/assets/255b5e97-2d3a-4434-b4fa-e922ad88ff5a)
 
 ![image](https://github.com/Calcium-Ion/new-api/assets/61247483/ad0e7aae-0203-471c-9716-2d83768927d4)
+![image](https://github.com/user-attachments/assets/29f81de5-33fc-4fc5-a5ff-f9b54b653c7c)
 
 ![image](https://github.com/Calcium-Ion/new-api/assets/61247483/3ca0b282-00ff-4c96-bf9d-e29ef615c605)
 夜间模式  

common/constants.go

@@ -254,7 +254,7 @@ var ChannelBaseURLs = []string{
 	"https://open.bigmodel.cn",            // 16
 	"https://dashscope.aliyuncs.com",      // 17
 	"",                                    // 18
-	"https://ai.360.cn",                   // 19
+	"https://api.360.cn",                   // 19
 	"https://openrouter.ai/api",           // 20
 	"https://api.aiproxy.io",              // 21
 	"https://fastgpt.run/api/openapi",     // 22

common/email.go

@@ -9,12 +9,20 @@ import (
 	"time"
 )
 
-func generateMessageID() string {
+func generateMessageID() (string, error) {
+	split := strings.Split(SMTPAccount, "@")
+	if len(split) < 2 {
+		return "", fmt.Errorf("invalid SMTP account")
+	}
 	domain := strings.Split(SMTPAccount, "@")[1]
-	return fmt.Sprintf("<%d.%s@%s>", time.Now().UnixNano(), GetRandomString(12), domain)
+	return fmt.Sprintf("<%d.%s@%s>", time.Now().UnixNano(), GetRandomString(12), domain), nil
 }
 
 func SendEmail(subject string, receiver string, content string) error {
+	id, err2 := generateMessageID()
+	if err2 != nil {
+		return err2
+	}
 	if SMTPFrom == "" { // for compatibility
 		SMTPFrom = SMTPAccount
 	}
@@ -28,7 +36,7 @@ func SendEmail(subject string, receiver string, content string) error {
 		"Date: %s\r\n"+
 		"Message-ID: %s\r\n"+ // 添加 Message-ID 头
 		"Content-Type: text/html; charset=UTF-8\r\n\r\n%s\r\n",
-		receiver, SystemName, SMTPFrom, encodedSubject, time.Now().Format(time.RFC1123Z), generateMessageID(), content))
+		receiver, SystemName, SMTPFrom, encodedSubject, time.Now().Format(time.RFC1123Z), id, content))
 	auth := smtp.PlainAuth("", SMTPAccount, SMTPToken, SMTPServer)
 	addr := fmt.Sprintf("%s:%d", SMTPServer, SMTPPort)
 	to := strings.Split(receiver, ";")

common/model-ratio.go

@@ -150,6 +150,7 @@ var defaultModelRatio = map[string]float64{
 	"360gpt-turbo":                   0.0858, // ¥0.0012 / 1k tokens
 	"360gpt-turbo-responsibility-8k": 0.8572, // ¥0.012 / 1k tokens
 	"360gpt-pro":                     0.8572, // ¥0.012 / 1k tokens
+	"360gpt2-pro":                    0.8572, // ¥0.012 / 1k tokens
 	"embedding-bert-512-v1":          0.0715, // ¥0.001 / 1k tokens
 	"embedding_s1_v1":                0.0715, // ¥0.001 / 1k tokens
 	"semantic_similarity_s1_v1":      0.0715, // ¥0.001 / 1k tokens

docker-compose.yml

@@ -14,8 +14,8 @@ services:
     environment:
       - SQL_DSN=root:123456@tcp(mysql:3306)/new-api  # Point to the mysql service
       - REDIS_CONN_STRING=redis://redis
-      - SESSION_SECRET=random_string  # 修改为随机字符串
       - TZ=Asia/Shanghai
+    #      - SESSION_SECRET=random_string  # 多机部署时设置，必须修改这个随机字符串！！！！！！！
     #      - NODE_TYPE=slave  # Uncomment for slave node in multi-node deployment
     #      - SYNC_FREQUENCY=60  # Uncomment if regular database syncing is needed
     #      - FRONTEND_BASE_URL=https://openai.justsong.cn  # Uncomment for multi-node deployment with front-end URL

middleware/gzip.go

@@ -0,0 +1,27 @@
+package middleware
+
+import (
+	"compress/gzip"
+	"github.com/gin-gonic/gin"
+	"io"
+	"net/http"
+)
+
+func GzipDecodeMiddleware() gin.HandlerFunc {
+	return func(c *gin.Context) {
+		if c.GetHeader("Content-Encoding") == "gzip" {
+			gzipReader, err := gzip.NewReader(c.Request.Body)
+			if err != nil {
+				c.AbortWithStatus(http.StatusBadRequest)
+				return
+			}
+			defer gzipReader.Close()
+
+			// Replace the request body with the decompressed data
+			c.Request.Body = io.NopCloser(gzipReader)
+		}
+
+		// Continue processing the request
+		c.Next()
+	}
+}

relay/channel/ai360/constants.go

@@ -4,6 +4,7 @@ var ModelList = []string{
 	"360gpt-turbo",
 	"360gpt-turbo-responsibility-8k",
 	"360gpt-pro",
+	"360gpt2-pro",
 	"360GPT_S2_V9",
 	"embedding-bert-512-v1",
 	"embedding_s1_v1",

relay/channel/openai/adaptor.go

@@ -32,11 +32,15 @@ func (a *Adaptor) Init(info *relaycommon.RelayInfo) {
 
 func (a *Adaptor) GetRequestURL(info *relaycommon.RelayInfo) (string, error) {
 	if info.RelayMode == constant.RelayModeRealtime {
-		// trim https
-		baseUrl := strings.TrimPrefix(info.BaseUrl, "https://")
-		baseUrl = strings.TrimPrefix(baseUrl, "http://")
-		baseUrl = "wss://" + baseUrl
-		info.BaseUrl = baseUrl
+		if strings.HasPrefix(info.BaseUrl, "https://") {
+			baseUrl := strings.TrimPrefix(info.BaseUrl, "https://")
+			baseUrl = "wss://" + baseUrl
+			info.BaseUrl = baseUrl
+		} else if strings.HasPrefix(info.BaseUrl, "http://") {
+			baseUrl := strings.TrimPrefix(info.BaseUrl, "http://")
+			baseUrl = "ws://" + baseUrl
+			info.BaseUrl = baseUrl
+		}
 	}
 	switch info.ChannelType {
 	case common.ChannelTypeAzure:

relay/relay-text.go

@@ -2,11 +2,9 @@ package relay
 
 import (
 	"bytes"
-	"context"
 	"encoding/json"
 	"errors"
 	"fmt"
-	"github.com/bytedance/sonic"
 	"io"
 	"math"
 	"net/http"
@@ -20,6 +18,8 @@ import (
 	"strings"
 	"time"
 
+	"github.com/bytedance/sonic"
+
 	"github.com/gin-gonic/gin"
 )
 
@@ -36,7 +36,7 @@ func getAndValidateTextRequest(c *gin.Context, relayInfo *relaycommon.RelayInfo)
 		textRequest.Model = c.Param("model")
 	}
 
-	if textRequest.MaxTokens < 0 || textRequest.MaxTokens > math.MaxInt32/2 {
+	if textRequest.MaxTokens > math.MaxInt32/2 {
 		return nil, errors.New("max_tokens is invalid")
 	}
 	if textRequest.Model == "" {
@@ -48,12 +48,12 @@ func getAndValidateTextRequest(c *gin.Context, relayInfo *relaycommon.RelayInfo)
 			return nil, errors.New("field prompt is required")
 		}
 	case relayconstant.RelayModeChatCompletions:
-		if textRequest.Messages == nil || len(textRequest.Messages) == 0 {
+		if len(textRequest.Messages) == 0 {
 			return nil, errors.New("field messages is required")
 		}
 	case relayconstant.RelayModeEmbeddings:
 	case relayconstant.RelayModeModerations:
-		if textRequest.Input == "" || textRequest.Input == nil {
+		if textRequest.Input == nil || textRequest.Input == "" {
 			return nil, errors.New("field input is required")
 		}
 	case relayconstant.RelayModeEdits:
@@ -264,7 +264,7 @@ func preConsumeQuota(c *gin.Context, preConsumedQuota int, relayInfo *relaycommo
 		return 0, 0, service.OpenAIErrorWrapperLocal(errors.New("user quota is not enough"), "insufficient_user_quota", http.StatusForbidden)
 	}
 	if userQuota-preConsumedQuota < 0 {
-		return 0, 0, service.OpenAIErrorWrapperLocal(errors.New(fmt.Sprintf("chat pre-consumed quota failed, user quota: %d, need quota: %d", userQuota, preConsumedQuota)), "insufficient_user_quota", http.StatusBadRequest)
+		return 0, 0, service.OpenAIErrorWrapperLocal(fmt.Errorf("chat pre-consumed quota failed, user quota: %d, need quota: %d", userQuota, preConsumedQuota), "insufficient_user_quota", http.StatusBadRequest)
 	}
 	err = model.CacheDecreaseUserQuota(relayInfo.UserId, preConsumedQuota)
 	if err != nil {
@@ -298,13 +298,14 @@ func preConsumeQuota(c *gin.Context, preConsumedQuota int, relayInfo *relaycommo
 
 func returnPreConsumedQuota(c *gin.Context, relayInfo *relaycommon.RelayInfo, userQuota int, preConsumedQuota int) {
 	if preConsumedQuota != 0 {
-		go func(ctx context.Context) {
-			// return pre-consumed quota
-			err := model.PostConsumeTokenQuota(relayInfo, userQuota, -preConsumedQuota, 0, false)
+		go func() {
+			relayInfoCopy := *relayInfo
+
+			err := model.PostConsumeTokenQuota(&relayInfoCopy, userQuota, -preConsumedQuota, 0, false)
 			if err != nil {
 				common.SysError("error return pre-consumed quota: " + err.Error())
 			}
-		}(c)
+		}()
 	}
 }
 

router/relay-router.go

@@ -9,6 +9,7 @@ import (
 
 func SetRelayRouter(router *gin.Engine) {
 	router.Use(middleware.CORS())
+	router.Use(middleware.GzipDecodeMiddleware())
 	// https://platform.openai.com/docs/api-reference/introduction
 	modelsRouter := router.Group("/v1/models")
 	modelsRouter.Use(middleware.TokenAuth())

service/quota.go

@@ -53,7 +53,7 @@ func PreWssConsumeQuota(ctx *gin.Context, relayInfo *relaycommon.RelayInfo, usag
 		return errors.New(fmt.Sprintf("用户额度不足，剩余额度为 %d", userQuota))
 	}
 
-	if token.RemainQuota < quota {
+	if !token.UnlimitedQuota && token.RemainQuota < quota {
 		return errors.New(fmt.Sprintf("令牌额度不足，剩余额度为 %d", token.RemainQuota))
 	}
 

web/src/helpers/api.js

@@ -6,7 +6,8 @@ export let API = axios.create({
     ? import.meta.env.VITE_REACT_APP_SERVER_URL
     : '',
   headers: {
-    'New-API-User': getUserIdFromLocalStorage()
+    'New-API-User': getUserIdFromLocalStorage(),
+    'Cache-Control': 'no-store'
   }
 });
 
@@ -16,7 +17,8 @@ export function updateAPI() {
       ? import.meta.env.VITE_REACT_APP_SERVER_URL
       : '',
     headers: {
-      'New-API-User': getUserIdFromLocalStorage()
+      'New-API-User': getUserIdFromLocalStorage(),
+      'Cache-Control': 'no-store'
     }
   });
 }

web/src/pages/Channel/EditTagModal.js

@@ -136,9 +136,9 @@ const EditTagModal = (props) => {
     if (inputs.models.length > 0) {
       data.models = inputs.models.join(',');
     }
-    data.newTag = inputs.newTag;
+    data.new_tag = inputs.new_tag;
     // check have any change
-    if (data.model_mapping === undefined && data.groups === undefined && data.models === undefined && data.newTag === undefined) {
+    if (data.model_mapping === undefined && data.groups === undefined && data.models === undefined && data.new_tag === undefined) {
       showWarning('没有任何修改！');
       setLoading(false);
       return;