feat：LLM 分析文件并生成结构化建议

app/api/files.py

@@ -3,13 +3,17 @@
 from typing import Annotated, Literal, get_args
 
 from fastapi import APIRouter, Depends, File, Form, Query, Request, UploadFile
+from fastapi.responses import JSONResponse
+from starlette.types import ASGIApp, Receive, Scope, Send
 
-from app.common.response import ApiResponse, success_response
+from app.common.response import ApiResponse, error_content, success_response
 from app.core.auth import get_current_user
+from app.datafiles.base import MAX_TABLE_ROWS
 from app.datafiles.service import (
     MAX_FILES_PER_CONVERSATION,
     PARSE_CONCURRENCY,
     DataFileService,
+    file_too_large,
 )
 from app.schemas.base import NonBlankString
 from app.schemas.data_file import (
@@ -31,18 +35,55 @@ from app.schemas.user import CurrentUser
 limits_router = APIRouter(prefix="/files", tags=["DataFile"])
 router = APIRouter(prefix="/files", tags=["DataFile"])
 
-# 文件选择框的过滤条件；后端按文件内容识别格式，不看扩展名
-# 翻页取表数据：单页上限，和一张表最多的行数（100 万）
+# 翻页取表数据：单页上限，offset 不超过一张表最多的行数
 MAX_PAGE_ROWS = 500
-MAX_OFFSET = 1_000_000
+MAX_OFFSET = MAX_TABLE_ROWS
 
+# 文件选择框的过滤条件；后端按文件内容识别格式，不看扩展名
 ACCEPT = ".csv,.tsv,.txt,.xlsx,.xls,.xlsb,.json,.jsonl,.ndjson,.parquet"
 
+# 上传的请求体比文件本身多出 multipart 的分隔行、字段名和会话 ID
+UPLOAD_OVERHEAD_BYTES = 64 * 1024
+
 
 def _service(request: Request) -> DataFileService:
     return request.app.state.data_file_service
 
 
+class UploadSizeLimit:
+    """上传接口在读取请求体之前按 Content-Length 拒绝过大的请求
+
+    FastAPI 会先把整个 multipart 请求体收完、写进临时文件，再调用接口函数；
+    只在 DataFileService 里边写边查大小的话，超大的上传已经占满了带宽和磁盘
+    """
+
+    def __init__(self, app: ASGIApp, path: str):
+        self.app = app
+        self.path = path
+
+    async def __call__(self, scope: Scope, receive: Receive, send: Send) -> None:
+        if (
+            scope["type"] == "http"
+            and scope["method"] == "POST"
+            and scope["path"] == self.path
+            and (rejection := self._reject(scope)) is not None
+        ):
+            await rejection(scope, receive, send)
+            return
+        await self.app(scope, receive, send)
+
+    @staticmethod
+    def _reject(scope: Scope) -> JSONResponse | None:
+        max_mb = scope["app"].state.data_file_service.max_mb
+        length = dict(scope["headers"]).get(b"content-length")
+        if length is None or not length.isdigit():
+            return JSONResponse(error_content(411, "缺少 Content-Length"), 411)
+        if int(length) > max_mb * 1024 * 1024 + UPLOAD_OVERHEAD_BYTES:
+            error = file_too_large(max_mb)
+            return JSONResponse(error_content(error.code, error.msg), error.code)
+        return None
+
+
 def _detail(record: DataFileRecord) -> DataFileDetail:
     # 按接口模型重新校验，去掉 owner_id、sha256、列概况等只在后端用的字段
     return DataFileDetail.model_validate(record.model_dump())