feat: 增加支持对csv格式文件的上传和解析

management/server/routes/files/routes.py

@@ -1,11 +1,11 @@
-from flask import jsonify, request, send_file, current_app
 from io import BytesIO
-from .. import files_bp
 
-
+from flask import current_app, jsonify, request, send_file
-from services.files.service import get_files_list, get_file_info, download_file_from_minio, delete_file, batch_delete_files, handle_chunk_upload, merge_chunks, upload_files_to_server
+from services.files.service import batch_delete_files, delete_file, download_file_from_minio, get_file_info, get_files_list, handle_chunk_upload, merge_chunks, upload_files_to_server
 from services.files.utils import FileType
 
+from .. import files_bp
+
 UPLOAD_FOLDER = "/data/uploads"
 ALLOWED_EXTENSIONS = {"txt", "pdf", "png", "jpg", "jpeg", "gif", "doc", "docx", "xls", "xlsx"}
 

management/server/services/files/document_service.py

@@ -1,16 +1,18 @@
-from peewee import *
+from peewee import *  # noqa: F403
+
 from .base_service import BaseService
 from .models import Document
-from .utils import get_uuid, StatusEnum
+from .utils import StatusEnum, get_uuid
+
 
 class DocumentService(BaseService):
     model = Document
-    
+
     @classmethod
     def create_document(cls, kb_id: str, name: str, location: str, size: int, file_type: str, created_by: str = None, parser_id: str = None, parser_config: dict = None) -> Document:
         """
         创建文档记录
-        
+
         Args:
             kb_id: 知识库ID
             name: 文件名
@@ -20,34 +22,34 @@ class DocumentService(BaseService):
             created_by: 创建者ID
             parser_id: 解析器ID
             parser_config: 解析器配置
-            
+
         Returns:
             Document: 创建的文档对象
         """
         doc_id = get_uuid()
-        
+
         # 构建基本文档数据
         doc_data = {
-            'id': doc_id,
+            "id": doc_id,
-            'kb_id': kb_id,
+            "kb_id": kb_id,
-            'name': name,
+            "name": name,
-            'location': location,
+            "location": location,
-            'size': size,
+            "size": size,
-            'type': file_type,
+            "type": file_type,
-            'created_by': created_by or 'system',
+            "created_by": created_by or "system",
-            'parser_id': parser_id or '',
+            "parser_id": parser_id or "",
-            'parser_config': parser_config or {"pages": [[1, 1000000]]},
+            "parser_config": parser_config or {"pages": [[1, 1000000]]},
-            'source_type': 'local',
+            "source_type": "local",
-            'token_num': 0,
+            "token_num": 0,
-            'chunk_num': 0,
+            "chunk_num": 0,
-            'progress': 0,
+            "progress": 0,
-            'progress_msg': '',
+            "progress_msg": "",
-            'run': '0',  # 未开始解析
+            "run": "0",  # 未开始解析
-            'status': StatusEnum.VALID.value
+            "status": StatusEnum.VALID.value,
         }
-        
+
         return cls.insert(doc_data)
-    
+
     @classmethod
     def get_by_kb_id(cls, kb_id: str) -> list[Document]:
-        return cls.query(kb_id=kb_id)
+        return cls.query(kb_id=kb_id)

management/server/services/files/file_service.py

@@ -1,26 +1,20 @@
 from peewee import *  # noqa: F403
+
 from .base_service import BaseService
 from .models import File
 from .utils import FileType, get_uuid
 
+
 class FileService(BaseService):
     model = File
-    
+
     @classmethod
     def create_file(cls, parent_id: str, name: str, location: str, size: int, file_type: str) -> File:
-        return cls.insert({
+        return cls.insert({"parent_id": parent_id, "name": name, "location": location, "size": size, "type": file_type, "source_type": "knowledgebase"})
-            'parent_id': parent_id,
+
-            'name': name,
-            'location': location,
-            'size': size,
-            'type': file_type,
-            'source_type': 'knowledgebase'
-        })
-    
     @classmethod
     def get_parser(cls, file_type, filename, tenant_id):
         """获取适合文件类型的解析器ID"""
-        # 这里可能需要根据实际情况调整
         if file_type == FileType.PDF.value:
             return "pdf_parser"
         elif file_type == FileType.WORD.value:
@@ -40,7 +34,7 @@ class FileService(BaseService):
     def get_by_parent_id(cls, parent_id: str) -> list[File]:
         return cls.query(parent_id=parent_id)
 
-    @classmethod 
+    @classmethod
     def generate_bucket_name(cls):
         """生成随机存储桶名称"""
-        return f"kb-{get_uuid()}"
+        return f"kb-{get_uuid()}"

management/server/services/files/service.py

@@ -1,13 +1,14 @@
 import os
-import shutil
 import re
+import shutil
 import tempfile
-from dotenv import load_dotenv
 from datetime import datetime
 from pathlib import Path
-from database import get_db_connection, get_minio_client, get_redis_connection
-from .utils import FileType, FileSource, get_uuid
 
+from database import get_db_connection, get_minio_client, get_redis_connection
+from dotenv import load_dotenv
+
+from .utils import FileSource, FileType, get_uuid
 
 # 加载环境变量
 load_dotenv("../../docker/.env")
@@ -18,7 +19,7 @@ CHUNK_EXPIRY_SECONDS = 3600 * 24  # 分块24小时过期
 
 temp_dir = tempfile.gettempdir()
 UPLOAD_FOLDER = os.path.join(temp_dir, "uploads")
-ALLOWED_EXTENSIONS = {"pdf", "doc", "docx", "ppt", "pptx", "xls", "xlsx", "jpg", "jpeg", "png", "bmp", "txt", "md", "html"}
+ALLOWED_EXTENSIONS = {"pdf", "doc", "docx", "ppt", "pptx", "xls", "xlsx", "jpg", "jpeg", "png", "bmp", "txt", "md", "html", "csv"}
 
 
 def allowed_file(filename):
@@ -36,7 +37,7 @@ def filename_type(filename):
         return FileType.PDF.value
     elif ext in [".doc", ".docx"]:
         return FileType.WORD.value
-    elif ext in [".xls", ".xlsx"]:
+    elif ext in [".xls", ".xlsx", ".csv"]:
         return FileType.EXCEL.value
     elif ext in [".ppt", ".pptx"]:
         return FileType.PPT.value

management/server/services/files/utils.py

@@ -1,9 +1,9 @@
 import uuid
-from strenum import StrEnum
 from enum import Enum
 
+from strenum import StrEnum
+
 
-# 参考：api.db
 class FileType(StrEnum):
     FOLDER = "folder"
     PDF = "pdf"
@@ -15,15 +15,18 @@ class FileType(StrEnum):
     HTML = "html"
     OTHER = "other"
 
+
 class FileSource(StrEnum):
     LOCAL = ""
     KNOWLEDGEBASE = "knowledgebase"
     S3 = "s3"
 
+
 class StatusEnum(Enum):
     VALID = "1"
     INVALID = "0"
 
+
 # 参考：api.utils
 def get_uuid():
-    return uuid.uuid1().hex
+    return uuid.uuid1().hex

management/server/services/knowledgebases/document_parser.py

@@ -20,7 +20,7 @@ from magic_pdf.data.read_api import read_local_images, read_local_office
 from magic_pdf.model.doc_analyze_by_custom_model import doc_analyze
 
 from . import logger
-from .excel_parser import parse_excel
+from .excel_parser import parse_excel_file
 from .rag_tokenizer import RagTokenizer
 from .utils import _create_task_record, _update_document_progress, _update_kb_chunk_count, generate_uuid, get_bbox_from_block
 
@@ -196,7 +196,7 @@ def perform_parse(doc_id, doc_info, file_info, embedding_config, kb_info):
 
             update_progress(0.8, "提取内容")
             # 处理内容列表
-            content_list = parse_excel(temp_file_path)
+            content_list = parse_excel_file(temp_file_path)
 
         elif file_type.endswith("visual"):
             update_progress(0.3, "使用MinerU解析器")

management/server/services/knowledgebases/excel_parser.py

@@ -1,24 +1,69 @@
+import os
+
 import pandas as pd
 
 
-def parse_excel(file_path):
+def parse_excel_file(file_path):
-    all_sheets = pd.read_excel(file_path, sheet_name=None)
+    """
-
+    通用表格解析函数，支持 Excel (.xlsx/.xls) 和 CSV 文件
+    返回统一格式的数据块列表
+    """
     blocks = []
 
-    for sheet_name, df in all_sheets.items():
+    # 根据文件扩展名选择读取方式
-        df = df.ffill()
+    file_ext = os.path.splitext(file_path)[1].lower()
-        headers = df.columns.tolist()
 
-        for _, row in df.iterrows():
+    try:
-            html_table = "<html><body><table><tr>{}</tr><tr>{}</tr></table></body></html>".format("".join(f"<td>{col}</td>" for col in headers), "".join(f"<td>{row[col]}</td>" for col in headers))
+        if file_ext in (".xlsx", ".xls"):
-            block = {"type": "table", "img_path": "", "table_caption": [f"Sheet: {sheet_name}"], "table_footnote": [], "table_body": f"{html_table}", "page_idx": 0}
+            # 处理Excel文件（多sheet）
-            blocks.append(block)
+            all_sheets = pd.read_excel(file_path, sheet_name=None)
+            for sheet_name, df in all_sheets.items():
+                blocks.extend(_process_dataframe(df, sheet_name))
+
+        elif file_ext == ".csv":
+            # 处理CSV文件（单sheet）
+            df = pd.read_csv(file_path)
+            blocks.extend(_process_dataframe(df, "CSV"))
+
+        else:
+            raise ValueError(f"Unsupported file format: {file_ext}")
+
+    except Exception as e:
+        raise ValueError(f"Failed to parse file {file_path}: {str(e)}")
+
+    return blocks
+
+
+def _process_dataframe(df, sheet_name):
+    """处理单个DataFrame，生成统一格式的数据块"""
+    df = df.ffill()
+    headers = df.columns.tolist()
+    blocks = []
+
+    for _, row in df.iterrows():
+        html_table = "<html><body><table><tr>{}</tr><tr>{}</tr></table></body></html>".format("".join(f"<td>{col}</td>" for col in headers), "".join(f"<td>{row[col]}</td>" for col in headers))
+
+        block = {"type": "table", "img_path": "", "table_caption": "", "table_footnote": [], "table_body": html_table, "page_idx": 0}
+        blocks.append(block)
 
     return blocks
 
 
 if __name__ == "__main__":
-    file_path = "test_excel.xls"
+    # 测试示例
-    parse_excel_result = parse_excel(file_path)
+    excel_path = "test.xlsx"
-    print(parse_excel_result)
+    csv_path = "test.csv"
+
+    try:
+        # 测试Excel解析
+        excel_blocks = parse_excel_file(excel_path)
+        print(f"Excel解析结果（共{len(excel_blocks)}条）:")
+        print(excel_blocks[:1])  # 打印第一条示例
+
+        # 测试CSV解析
+        csv_blocks = parse_excel_file(csv_path)
+        print(f"\nCSV解析结果（共{len(csv_blocks)}条）:")
+        print(csv_blocks[:1])
+
+    except Exception as e:
+        print(f"错误: {str(e)}")

management/server/utils.py

@@ -1,8 +1,9 @@
-import uuid
 import base64
-from flask import jsonify
+import uuid
-from Cryptodome.PublicKey import RSA
+
 from Cryptodome.Cipher import PKCS1_v1_5
+from Cryptodome.PublicKey import RSA
+from flask import jsonify
 from werkzeug.security import generate_password_hash
 
 
@@ -10,6 +11,7 @@ from werkzeug.security import generate_password_hash
 def generate_uuid():
     return str(uuid.uuid4()).replace("-", "")
 
+
 # RSA 加密密码
 def rsa_psw(password: str) -> str:
     pub_key = """-----BEGIN PUBLIC KEY-----
@@ -21,26 +23,22 @@ def rsa_psw(password: str) -> str:
     encrypted_data = cipher.encrypt(base64.b64encode(password.encode()))
     return base64.b64encode(encrypted_data).decode()
 
+
 # 加密密码
 def encrypt_password(raw_password: str) -> str:
     base64_password = base64.b64encode(raw_password.encode()).decode()
     return generate_password_hash(base64_password)
 
+
 # 标准响应格式
 def success_response(data=None, message="操作成功", code=0):
-    return jsonify({
+    return jsonify({"code": code, "message": message, "data": data})
-        "code": code,
+
-        "message": message,
-        "data": data
-    })
 
 # 错误响应格式
 def error_response(message="操作失败", code=500, details=None):
     """标准错误响应格式"""
-    response = {
+    response = {"code": code, "message": message}
-        "code": code,
-        "message": message
-    }
     if details:
         response["details"] = details
-    return jsonify(response), code if code >= 400 else 500
+    return jsonify(response), code if code >= 400 else 500