feat: 添加可配置的文件上传大小限制

新增 MAX_FILE_SIZE_MB 环境变量统一控制文件上传大小，默认 50MB

.env.example

@@ -152,4 +152,9 @@ COS_ENABLE_OLD_DOMAIN=true
 # NEO4J_USERNAME=neo4j
 
 # Neo4j的密码
-# NEO4J_PASSWORD=password
+# NEO4J_PASSWORD=password
+
+# ========== 文件上传大小限制 ==========
+# 统一的文件大小限制（MB），默认为50MB
+# 影响：单文件上传、gRPC消息大小、Nginx请求体大小
+# MAX_FILE_SIZE_MB=50

docker-compose.dev.yml

@@ -118,6 +118,8 @@ services:
       - MINIO_BUCKET_NAME=${MINIO_BUCKET_NAME:-}
       - MINIO_USE_SSL=${MINIO_USE_SSL:-}
       - WEB_PROXY=${WEB_PROXY:-}
+      # File size limit (in MB)
+      - MAX_FILE_SIZE_MB=${MAX_FILE_SIZE_MB:-50}
     healthcheck:
       test: ["CMD", "grpc_health_probe", "-addr=:50051"]
       interval: 30s

docker-compose.yml

@@ -5,6 +5,8 @@ services:
     container_name: WeKnora-frontend
     ports:
       - "${FRONTEND_PORT:-80}:80"
+    environment:
+      - MAX_FILE_SIZE_MB=${MAX_FILE_SIZE_MB:-50}
     depends_on:
       app:
         condition: service_healthy
@@ -96,6 +98,8 @@ services:
       - INIT_RERANK_MODEL_NAME=${INIT_RERANK_MODEL_NAME:-}
       - INIT_RERANK_MODEL_BASE_URL=${INIT_RERANK_MODEL_BASE_URL:-}
       - INIT_RERANK_MODEL_API_KEY=${INIT_RERANK_MODEL_API_KEY:-}
+      # File size limit (in MB)
+      - MAX_FILE_SIZE_MB=${MAX_FILE_SIZE_MB:-50}
     depends_on:
       redis:
         condition: service_started
@@ -137,6 +141,8 @@ services:
       - MINIO_USE_SSL=${MINIO_USE_SSL:-}
       - WEB_PROXY=${WEB_PROXY:-}
       - MINERU_ENDPOINT=${MINERU_ENDPOINT:-}
+      # File size limit (in MB)
+      - MAX_FILE_SIZE_MB=${MAX_FILE_SIZE_MB:-50}
     healthcheck:
       test: ["CMD", "grpc_health_probe", "-addr=:50051"]
       interval: 30s

docreader/client/client.go

@@ -4,6 +4,7 @@ import (
 	"fmt"
 	"log"
 	"os"
+	"strconv"
 	"time"
 
 	"github.com/Tencent/WeKnora/docreader/proto"
@@ -12,9 +13,16 @@ import (
 	"google.golang.org/grpc/resolver"
 )
 
-const (
-	maxMessageSize = 50 * 1024 * 1024 // 50MB
-)
+// getMaxMessageSize returns the maximum gRPC message size in bytes.
+// Default is 50MB, can be configured via MAX_FILE_SIZE_MB environment variable.
+func getMaxMessageSize() int {
+	if sizeStr := os.Getenv("MAX_FILE_SIZE_MB"); sizeStr != "" {
+		if size, err := strconv.Atoi(sizeStr); err == nil && size > 0 {
+			return size * 1024 * 1024
+		}
+	}
+	return 50 * 1024 * 1024 // default 50MB
+}
 
 // Logger is the default logger used by the client
 var Logger = log.New(os.Stdout, "[DocReader] ", log.LstdFlags|log.Lmicroseconds)
@@ -40,13 +48,14 @@ type Client struct {
 func NewClient(addr string) (*Client, error) {
 	Logger.Printf("INFO: Creating new DocReader client connecting to %s", addr)
 
-	// 设置消息大小限制
+	// 设置消息大小限制 (configurable via GRPC_MAX_MESSAGE_SIZE_MB)
+	maxMsgSize := getMaxMessageSize()
 	opts := []grpc.DialOption{
 		grpc.WithTransportCredentials(insecure.NewCredentials()),
 		grpc.WithDefaultServiceConfig(`{"loadBalancingPolicy":"round_robin"}`),
 		grpc.WithDefaultCallOptions(
-			grpc.MaxCallRecvMsgSize(maxMessageSize),
-			grpc.MaxCallSendMsgSize(maxMessageSize),
+			grpc.MaxCallRecvMsgSize(maxMsgSize),
+			grpc.MaxCallSendMsgSize(maxMsgSize),
 		),
 	}
 	resolver.SetDefaultScheme("dns")

docreader/main.py

@@ -57,8 +57,22 @@ logger.info("Initializing server logging")
 # Initialize request ID logging
 init_logging_request_id()
 
-# Set max message size to 50MB
-MAX_MESSAGE_LENGTH = 50 * 1024 * 1024
+
+def get_max_message_length() -> int:
+    """Get max gRPC message length from environment variable.
+    Default is 50MB, can be configured via MAX_FILE_SIZE_MB.
+    """
+    try:
+        size_mb = int(os.environ.get("MAX_FILE_SIZE_MB", "50"))
+        if size_mb > 0:
+            return size_mb * 1024 * 1024
+    except ValueError:
+        pass
+    return 50 * 1024 * 1024  # default 50MB
+
+
+# Set max message size (default 50MB, configurable via MAX_FILE_SIZE_MB)
+MAX_MESSAGE_LENGTH = get_max_message_length()
 
 
 parser = Parser()

frontend/Dockerfile

@@ -27,11 +27,14 @@ FROM nginx:stable-alpine as production-stage
 # 复制构建产物到nginx服务目录
 COPY --from=build-stage /app/dist /usr/share/nginx/html
 
-# 复制nginx配置文件
-COPY nginx.conf /etc/nginx/conf.d/default.conf
+# 复制nginx配置模板文件
+COPY nginx.conf /etc/nginx/templates/default.conf.template
+
+# 设置默认环境变量（MB）
+ENV MAX_FILE_SIZE_MB=50
 
 # 暴露端口
 EXPOSE 80
 
-# 启动nginx
-CMD ["nginx", "-g", "daemon off;"] 
+# 启动时将 MAX_FILE_SIZE_MB 转换为带单位的 MAX_FILE_SIZE，然后替换到 nginx 配置
+CMD ["/bin/sh", "-c", "export MAX_FILE_SIZE=${MAX_FILE_SIZE_MB}M && envsubst '${MAX_FILE_SIZE}' < /etc/nginx/templates/default.conf.template > /etc/nginx/conf.d/default.conf && nginx -g 'daemon off;'"] 

frontend/nginx.conf

@@ -1,7 +1,8 @@
 server {
     listen 80;
     server_name localhost;
-    client_max_body_size 50M;
+    # Default 50M, configured via MAX_FILE_SIZE_MB env var
+    client_max_body_size ${MAX_FILE_SIZE};
     
     # 安全头配置
     add_header X-Frame-Options "SAMEORIGIN" always;

internal/handler/initialization.go

@@ -1823,10 +1823,11 @@ func (h *InitializationHandler) TestMultimodalFunction(c *gin.Context) {
 		return
 	}
 
-	// 验证文件大小 (10MB)
-	if header.Size > 10*1024*1024 {
+	// 验证文件大小 (default 50MB, configurable via MAX_FILE_SIZE_MB)
+	maxSize := utils.GetMaxFileSize()
+	if header.Size > maxSize {
 		logger.Error(ctx, "File size too large")
-		c.Error(errors.NewBadRequestError("图片文件大小不能超过10MB"))
+		c.Error(errors.NewBadRequestError(fmt.Sprintf("图片文件大小不能超过%dMB", utils.GetMaxFileSizeMB())))
 		return
 	}
 	logger.Infof(ctx, "Processing image: %s", utils.SanitizeForLog(header.Filename))

internal/handler/knowledge.go

@@ -118,6 +118,14 @@ func (h *KnowledgeHandler) CreateKnowledgeFromFile(c *gin.Context) {
 		return
 	}
 
+	// Validate file size (configurable via MAX_FILE_SIZE_MB)
+	maxSize := secutils.GetMaxFileSize()
+	if file.Size > maxSize {
+		logger.Error(ctx, "File size too large")
+		c.Error(errors.NewBadRequestError(fmt.Sprintf("文件大小不能超过%dMB", secutils.GetMaxFileSizeMB())))
+		return
+	}
+
 	// Get custom filename if provided (for folder uploads with path)
 	customFileName := c.PostForm("fileName")
 	customFileName = secutils.SanitizeForLog(customFileName)

internal/utils/filesize.go

@@ -0,0 +1,27 @@
+package utils
+
+import (
+	"os"
+	"strconv"
+)
+
+// GetMaxFileSize returns the maximum file upload size in bytes.
+// Default is 50MB, can be configured via MAX_FILE_SIZE_MB environment variable.
+func GetMaxFileSize() int64 {
+	if sizeStr := os.Getenv("MAX_FILE_SIZE_MB"); sizeStr != "" {
+		if size, err := strconv.ParseInt(sizeStr, 10, 64); err == nil && size > 0 {
+			return size * 1024 * 1024
+		}
+	}
+	return 50 * 1024 * 1024 // default 50MB
+}
+
+// GetMaxFileSizeMB returns the maximum file upload size in MB.
+func GetMaxFileSizeMB() int64 {
+	if sizeStr := os.Getenv("MAX_FILE_SIZE_MB"); sizeStr != "" {
+		if size, err := strconv.ParseInt(sizeStr, 10, 64); err == nil && size > 0 {
+			return size
+		}
+	}
+	return 50 // default 50MB
+}