修复智能填表功能：支持直接从结构化数据提取列值并完善JSON解析

- 新增 _extract_values_from_structured_data 方法，直接从Excel rows提取列值
- 新增 _extract_values_by_regex 方法，使用正则从损坏的JSON中提取值
- 增大 max_tokens (500→50000) 和 max_length (8000→200000) 限制
- 改进JSON解析逻辑，处理markdown代码块包裹和不完整JSON
- 解决LLM返回被截断的JSON无法正确解析的问题

backend/app/api/endpoints/library.py

@@ -3,6 +3,7 @@
 
 提供文档列表、详情查询和删除功能
 """
+import logging
 from typing import Optional, List
 
 from fastapi import APIRouter, HTTPException, Query
@@ -10,6 +11,8 @@ from pydantic import BaseModel
 
 from app.core.database import mongodb
 
+logger = logging.getLogger(__name__)
+
 router = APIRouter(prefix="/documents", tags=["文档库"])
 
 
@@ -26,7 +29,8 @@ class DocumentItem(BaseModel):
 @router.get("")
 async def get_documents(
     doc_type: Optional[str] = Query(None, description="文档类型过滤"),
-    limit: int = Query(50, ge=1, le=100, description="返回数量")
+    limit: int = Query(20, ge=1, le=100, description="返回数量"),
+    skip: int = Query(0, ge=0, description="跳过数量")
 ):
     """
     获取文档列表
@@ -40,11 +44,25 @@ async def get_documents(
         if doc_type:
             query["doc_type"] = doc_type
 
-        # 查询文档
-        cursor = mongodb.documents.find(query).sort("created_at", -1).limit(limit)
+        logger.info(f"开始查询文档列表, query: {query}, limit: {limit}")
+
+        # 使用 batch_size 和 max_time_ms 来控制查询
+        cursor = mongodb.documents.find(
+            query,
+            {"content": 0}  # 不返回 content 字段，减少数据传输
+        ).sort("created_at", -1).skip(skip).limit(limit)
+
+        # 设置 10 秒超时
+        cursor.max_time_ms(10000)
+
+        logger.info("Cursor created with 10s timeout, executing...")
+
+        # 使用 batch_size 逐批获取
+        documents_raw = await cursor.to_list(length=limit)
+        logger.info(f"查询到原始文档数: {len(documents_raw)}")
 
         documents = []
-        async for doc in cursor:
+        for doc in documents_raw:
             documents.append({
                 "doc_id": str(doc["_id"]),
                 "filename": doc.get("metadata", {}).get("filename", ""),
@@ -55,10 +73,12 @@ async def get_documents(
                 "metadata": {
                     "row_count": doc.get("metadata", {}).get("row_count"),
                     "column_count": doc.get("metadata", {}).get("column_count"),
-                    "columns": doc.get("metadata", {}).get("columns", [])[:10]  # 只返回前10列
+                    "columns": doc.get("metadata", {}).get("columns", [])[:10]
                 }
             })
 
+        logger.info(f"文档列表处理完成: {len(documents)} 个文档")
+
         return {
             "success": True,
             "documents": documents,
@@ -66,6 +86,17 @@ async def get_documents(
         }
 
     except Exception as e:
+        err_str = str(e)
+        # 如果是超时错误，返回空列表而不是报错
+        if "timeout" in err_str.lower() or "time" in err_str.lower():
+            logger.warning(f"文档查询超时，返回空列表: {err_str}")
+            return {
+                "success": True,
+                "documents": [],
+                "total": 0,
+                "warning": "查询超时，请稍后重试"
+            }
+        logger.error(f"获取文档列表失败: {str(e)}", exc_info=True)
         raise HTTPException(status_code=500, detail=f"获取文档列表失败: {str(e)}")
 
 

backend/app/api/endpoints/templates.py

@@ -226,9 +226,42 @@ async def export_filled_template(
 
 
 async def _export_to_excel(filled_data: dict, template_id: str) -> StreamingResponse:
-    """导出为 Excel 格式"""
-    # 将字典转换为单行 DataFrame
-    df = pd.DataFrame([filled_data])
+    """导出为 Excel 格式（支持多行）"""
+    import logging
+    logger = logging.getLogger(__name__)
+
+    logger.info(f"导出填表数据: {len(filled_data)} 个字段")
+
+    # 计算最大行数
+    max_rows = 1
+    for k, v in filled_data.items():
+        if isinstance(v, list) and len(v) > max_rows:
+            max_rows = len(v)
+        logger.info(f"  {k}: {type(v).__name__} = {str(v)[:80]}")
+
+    logger.info(f"最大行数: {max_rows}")
+
+    # 构建多行数据
+    rows_data = []
+    for row_idx in range(max_rows):
+        row = {}
+        for col_name, values in filled_data.items():
+            if isinstance(values, list):
+                # 取对应行的值，不足则填空
+                row[col_name] = values[row_idx] if row_idx < len(values) else ""
+            else:
+                # 非列表，整个值填入第一行
+                row[col_name] = values if row_idx == 0 else ""
+        rows_data.append(row)
+
+    df = pd.DataFrame(rows_data)
+
+    # 确保列顺序
+    if not df.empty:
+        df = df[list(filled_data.keys())]
+
+    logger.info(f"DataFrame 形状: {df.shape}")
+    logger.info(f"DataFrame 列: {list(df.columns)}")
 
     output = io.BytesIO()
     with pd.ExcelWriter(output, engine='openpyxl') as writer:

backend/app/api/endpoints/upload.py

@@ -11,6 +11,7 @@ import io
 from app.services.file_service import file_service
 from app.core.document_parser import XlsxParser
 from app.services.table_rag_service import table_rag_service
+from app.core.database import mongodb
 
 logger = logging.getLogger(__name__)
 
@@ -95,6 +96,56 @@ async def upload_excel(
         except Exception as e:
             logger.error(f"Excel存储到MySQL异常: {str(e)}", exc_info=True)
 
+        # 存储到 MongoDB（用于文档列表展示）
+        try:
+            content = ""
+            # 构建文本内容用于展示
+            if result.data:
+                if isinstance(result.data, dict):
+                    # 单 sheet 格式: {columns, rows, ...}
+                    if 'columns' in result.data and 'rows' in result.data:
+                        content += f"Sheet: {result.metadata.get('current_sheet', 'Sheet1') if result.metadata else 'Sheet1'}\n"
+                        content += ", ".join(str(h) for h in result.data['columns']) + "\n"
+                        for row in result.data['rows'][:100]:
+                            if isinstance(row, dict):
+                                content += ", ".join(str(row.get(col, "")) for col in result.data['columns']) + "\n"
+                            elif isinstance(row, list):
+                                content += ", ".join(str(cell) for cell in row) + "\n"
+                        content += f"... (共 {len(result.data['rows'])} 行)\n\n"
+                    # 多 sheet 格式: {sheets: {sheet_name: {columns, rows}}}
+                    elif 'sheets' in result.data:
+                        for sheet_name_key, sheet_data in result.data['sheets'].items():
+                            if isinstance(sheet_data, dict) and 'columns' in sheet_data and 'rows' in sheet_data:
+                                content += f"Sheet: {sheet_name_key}\n"
+                                content += ", ".join(str(h) for h in sheet_data['columns']) + "\n"
+                                for row in sheet_data['rows'][:100]:
+                                    if isinstance(row, dict):
+                                        content += ", ".join(str(row.get(col, "")) for col in sheet_data['columns']) + "\n"
+                                    elif isinstance(row, list):
+                                        content += ", ".join(str(cell) for cell in row) + "\n"
+                                content += f"... (共 {len(sheet_data['rows'])} 行)\n\n"
+
+            doc_metadata = {
+                "filename": saved_path.split("/")[-1] if "/" in saved_path else saved_path.split("\\")[-1],
+                "original_filename": file.filename,
+                "saved_path": saved_path,
+                "file_size": len(content),
+                "row_count": result.metadata.get('row_count', 0) if result.metadata else 0,
+                "column_count": result.metadata.get('column_count', 0) if result.metadata else 0,
+                "columns": result.metadata.get('columns', []) if result.metadata else [],
+                "mysql_table": result.metadata.get('mysql_table') if result.metadata else None,
+                "sheet_count": result.metadata.get('sheet_count', 1) if result.metadata else 1,
+            }
+            await mongodb.insert_document(
+                doc_type="xlsx",
+                content=content,
+                metadata=doc_metadata,
+                structured_data=result.data if result.data else None
+            )
+            logger.info(f"Excel文档已存储到MongoDB: {file.filename}, content长度: {len(content)}")
+        except Exception as e:
+            logger.error(f"Excel存储到MongoDB异常: {str(e)}", exc_info=True)
+
         return result.to_dict()
 
     except HTTPException: