import json import threading import time import traceback from datetime import datetime import mysql.connector import requests from database import DB_CONFIG, get_es_client from utils import generate_uuid # 解析相关模块 from .document_parser import _update_document_progress, perform_parse # 用于存储进行中的顺序批量任务状态 # 结构: { kb_id: {"status": "running/completed/failed", "total": N, "current": M, "message": "...", "start_time": timestamp} } SEQUENTIAL_BATCH_TASKS = {} class KnowledgebaseService: @classmethod def _get_db_connection(cls): """创建数据库连接""" return mysql.connector.connect(**DB_CONFIG) @classmethod def get_knowledgebase_list(cls, page=1, size=10, name="", sort_by="create_time", sort_order="desc"): """获取知识库列表""" conn = cls._get_db_connection() cursor = conn.cursor(dictionary=True) # 验证排序字段 valid_sort_fields = ["name", "create_time", "create_date"] if sort_by not in valid_sort_fields: sort_by = "create_time" # 构建排序子句 sort_clause = f"ORDER BY k.{sort_by} {sort_order.upper()}" query = """ SELECT k.id, k.name, k.description, k.create_date, k.update_date, k.doc_num, k.language, k.permission FROM knowledgebase k """ params = [] if name: query += " WHERE k.name LIKE %s" params.append(f"%{name}%") # 添加查询排序条件 query += f" {sort_clause}" query += " LIMIT %s OFFSET %s" params.extend([size, (page - 1) * size]) cursor.execute(query, params) results = cursor.fetchall() # 处理结果 for result in results: # 处理空描述 if not result.get("description"): result["description"] = "暂无描述" # 处理时间格式 if result.get("create_date"): if isinstance(result["create_date"], datetime): result["create_date"] = result["create_date"].strftime("%Y-%m-%d %H:%M:%S") elif isinstance(result["create_date"], str): try: # 尝试解析已有字符串格式 datetime.strptime(result["create_date"], "%Y-%m-%d %H:%M:%S") except ValueError: result["create_date"] = "" # 获取总数 count_query = "SELECT COUNT(*) as total FROM knowledgebase" if name: count_query += " WHERE name LIKE %s" cursor.execute(count_query, params[:1] if name else []) total = cursor.fetchone()["total"] cursor.close() conn.close() return {"list": results, "total": total} @classmethod def get_knowledgebase_detail(cls, kb_id): """获取知识库详情""" conn = cls._get_db_connection() cursor = conn.cursor(dictionary=True) query = """ SELECT k.id, k.name, k.description, k.create_date, k.update_date, k.doc_num, k.avatar FROM knowledgebase k WHERE k.id = %s """ cursor.execute(query, (kb_id,)) result = cursor.fetchone() if result: # 处理空描述 if not result.get("description"): result["description"] = "暂无描述" # 处理时间格式 if result.get("create_date"): if isinstance(result["create_date"], datetime): result["create_date"] = result["create_date"].strftime("%Y-%m-%d %H:%M:%S") elif isinstance(result["create_date"], str): try: datetime.strptime(result["create_date"], "%Y-%m-%d %H:%M:%S") except ValueError: result["create_date"] = "" cursor.close() conn.close() return result @classmethod def _check_name_exists(cls, name): """检查知识库名称是否已存在""" conn = cls._get_db_connection() cursor = conn.cursor() query = """ SELECT COUNT(*) as count FROM knowledgebase WHERE name = %s """ cursor.execute(query, (name,)) result = cursor.fetchone() cursor.close() conn.close() return result[0] > 0 @classmethod def create_knowledgebase(cls, **data): """创建知识库""" try: # 检查知识库名称是否已存在 exists = cls._check_name_exists(data["name"]) if exists: raise Exception("知识库名称已存在") conn = cls._get_db_connection() cursor = conn.cursor(dictionary=True) # 使用传入的 creator_id 作为 tenant_id 和 created_by tenant_id = data.get("creator_id") created_by = data.get("creator_id") if not tenant_id: # 如果没有提供 creator_id，则使用默认值 print("未提供 creator_id，尝试获取最早用户 ID") try: query_earliest_user = """ SELECT id FROM user WHERE create_time = (SELECT MIN(create_time) FROM user) LIMIT 1 """ cursor.execute(query_earliest_user) earliest_user = cursor.fetchone() if earliest_user: tenant_id = earliest_user["id"] created_by = earliest_user["id"] print(f"使用创建时间最早的用户ID作为tenant_id和created_by: {tenant_id}") else: # 如果找不到用户，使用默认值 tenant_id = "system" created_by = "system" print(f"未找到用户, 使用默认值作为tenant_id和created_by: {tenant_id}") except Exception as e: print(f"获取用户ID失败: {str(e)}，使用默认值") tenant_id = "system" created_by = "system" else: print(f"使用传入的 creator_id 作为 tenant_id 和 created_by: {tenant_id}") # --- 获取动态 embd_id --- dynamic_embd_id = None default_embd_id = "bge-m3" # Fallback default try: query_embedding_model = """ SELECT llm_name FROM tenant_llm WHERE model_type = 'embedding' ORDER BY create_time DESC LIMIT 1 """ cursor.execute(query_embedding_model) embedding_model = cursor.fetchone() if embedding_model and embedding_model.get("llm_name"): dynamic_embd_id = embedding_model["llm_name"] # 对硅基流动平台进行特异性处理 if dynamic_embd_id == "netease-youdao/bce-embedding-base_v1": dynamic_embd_id = "BAAI/bge-m3" print(f"动态获取到的 embedding 模型 ID: {dynamic_embd_id}") else: dynamic_embd_id = default_embd_id print(f"未在 tenant_llm 表中找到 embedding 模型, 使用默认值: {dynamic_embd_id}") except Exception as e: dynamic_embd_id = default_embd_id print(f"查询 embedding 模型失败: {str(e)}，使用默认值: {dynamic_embd_id}") traceback.print_exc() # Log the full traceback for debugging current_time = datetime.now() create_date = current_time.strftime("%Y-%m-%d %H:%M:%S") create_time = int(current_time.timestamp() * 1000) # 毫秒级时间戳 update_date = create_date update_time = create_time # 完整的字段列表 query = """ INSERT INTO knowledgebase ( id, create_time, create_date, update_time, update_date, avatar, tenant_id, name, language, description, embd_id, permission, created_by, doc_num, token_num, chunk_num, similarity_threshold, vector_similarity_weight, parser_id, parser_config, pagerank, status ) VALUES ( %s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s ) """ # 设置默认值 default_parser_config = json.dumps( { "layout_recognize": "MinerU", "chunk_token_num": 512, "delimiter": "\n!?;。；！？", "auto_keywords": 0, "auto_questions": 0, "html4excel": False, "raptor": {"use_raptor": False}, "graphrag": {"use_graphrag": False}, } ) kb_id = generate_uuid() cursor.execute( query, ( kb_id, # id create_time, # create_time create_date, # create_date update_time, # update_time update_date, # update_date None, # avatar tenant_id, # tenant_id data["name"], # name data.get("language", "Chinese"), # language data.get("description", ""), # description dynamic_embd_id, # embd_id data.get("permission", "me"), # permission created_by, # created_by - 使用内部获取的值 0, # doc_num 0, # token_num 0, # chunk_num 0.7, # similarity_threshold 0.3, # vector_similarity_weight "naive", # parser_id default_parser_config, # parser_config 0, # pagerank "1", # status ), ) conn.commit() cursor.close() conn.close() # 返回创建后的知识库详情 return cls.get_knowledgebase_detail(kb_id) except Exception as e: print(f"创建知识库失败: {str(e)}") raise Exception(f"创建知识库失败: {str(e)}") @classmethod def update_knowledgebase(cls, kb_id, **data): """更新知识库""" try: # 直接通过ID检查知识库是否存在 kb = cls.get_knowledgebase_detail(kb_id) if not kb: return None conn = cls._get_db_connection() cursor = conn.cursor() # 如果要更新名称，先检查名称是否已存在 if data.get("name") and data["name"] != kb["name"]: exists = cls._check_name_exists(data["name"]) if exists: raise Exception("知识库名称已存在") # 构建更新语句 update_fields = [] params = [] if data.get("name"): update_fields.append("name = %s") params.append(data["name"]) if "description" in data: update_fields.append("description = %s") params.append(data["description"]) if "permission" in data: update_fields.append("permission = %s") params.append(data["permission"]) if "avatar" in data and data["avatar"]: avatar_base64 = data["avatar"] # 拼接上前缀 full_avatar_url = f"data:image/png;base64,{avatar_base64}" update_fields.append("avatar = %s") params.append(full_avatar_url) # 更新时间 current_time = datetime.now() update_date = current_time.strftime("%Y-%m-%d %H:%M:%S") update_fields.append("update_date = %s") params.append(update_date) # 如果没有要更新的字段，直接返回 if not update_fields: return kb_id # 构建并执行更新语句 query = f""" UPDATE knowledgebase SET {", ".join(update_fields)} WHERE id = %s """ params.append(kb_id) cursor.execute(query, params) conn.commit() cursor.close() conn.close() # 返回更新后的知识库详情 return cls.get_knowledgebase_detail(kb_id) except Exception as e: print(f"更新知识库失败: {str(e)}") raise Exception(f"更新知识库失败: {str(e)}") @classmethod def delete_knowledgebase(cls, kb_id): """删除知识库""" try: conn = cls._get_db_connection() cursor = conn.cursor() # 先检查知识库是否存在 check_query = "SELECT id FROM knowledgebase WHERE id = %s" cursor.execute(check_query, (kb_id,)) if not cursor.fetchone(): raise Exception("知识库不存在") # 执行删除 delete_query = "DELETE FROM knowledgebase WHERE id = %s" cursor.execute(delete_query, (kb_id,)) conn.commit() cursor.close() conn.close() return True except Exception as e: print(f"删除知识库失败: {str(e)}") raise Exception(f"删除知识库失败: {str(e)}") @classmethod def batch_delete_knowledgebase(cls, kb_ids): """批量删除知识库""" try: conn = cls._get_db_connection() cursor = conn.cursor() # 检查所有ID是否存在 check_query = "SELECT id FROM knowledgebase WHERE id IN (%s)" % ",".join(["%s"] * len(kb_ids)) cursor.execute(check_query, kb_ids) existing_ids = [row[0] for row in cursor.fetchall()] if len(existing_ids) != len(kb_ids): missing_ids = set(kb_ids) - set(existing_ids) raise Exception(f"以下知识库不存在: {', '.join(missing_ids)}") # 执行批量删除 delete_query = "DELETE FROM knowledgebase WHERE id IN (%s)" % ",".join(["%s"] * len(kb_ids)) cursor.execute(delete_query, kb_ids) conn.commit() cursor.close() conn.close() return len(kb_ids) except Exception as e: print(f"批量删除知识库失败: {str(e)}") raise Exception(f"批量删除知识库失败: {str(e)}") @classmethod def get_knowledgebase_documents(cls, kb_id, page=1, size=10, name="", sort_by="create_time", sort_order="desc"): """获取知识库下的文档列表""" try: conn = cls._get_db_connection() cursor = conn.cursor(dictionary=True) # 先检查知识库是否存在 check_query = "SELECT id FROM knowledgebase WHERE id = %s" cursor.execute(check_query, (kb_id,)) if not cursor.fetchone(): raise Exception("知识库不存在") # 验证排序字段 valid_sort_fields = ["name", "size", "create_time", "create_date"] if sort_by not in valid_sort_fields: sort_by = "create_time" # 构建排序子句 sort_clause = f"ORDER BY d.{sort_by} {sort_order.upper()}" # 查询文档列表 query = """ SELECT d.id, d.name, d.chunk_num, d.create_date, d.status, d.run, d.progress, d.parser_id, d.parser_config, d.meta_fields FROM document d WHERE d.kb_id = %s """ params = [kb_id] if name: query += " AND d.name LIKE %s" params.append(f"%{name}%") # 添加查询排序条件 query += f" {sort_clause}" query += " LIMIT %s OFFSET %s" params.extend([size, (page - 1) * size]) cursor.execute(query, params) results = cursor.fetchall() # 处理日期时间格式 for result in results: if result.get("create_date"): result["create_date"] = result["create_date"].strftime("%Y-%m-%d %H:%M:%S") # 获取总数 count_query = "SELECT COUNT(*) as total FROM document WHERE kb_id = %s" count_params = [kb_id] if name: count_query += " AND name LIKE %s" count_params.append(f"%{name}%") cursor.execute(count_query, count_params) total = cursor.fetchone()["total"] cursor.close() conn.close() return {"list": results, "total": total} except Exception as e: print(f"获取知识库文档列表失败: {str(e)}") raise Exception(f"获取知识库文档列表失败: {str(e)}") @classmethod def add_documents_to_knowledgebase(cls, kb_id, file_ids, created_by=None): """添加文档到知识库""" try: print(f"[DEBUG] 开始添加文档，参数: kb_id={kb_id}, file_ids={file_ids}") # 如果没有传入created_by，则获取最早的用户ID if created_by is None: conn = cls._get_db_connection() cursor = conn.cursor(dictionary=True) # 查询创建时间最早的用户ID query_earliest_user = """ SELECT id FROM user WHERE create_time = (SELECT MIN(create_time) FROM user) LIMIT 1 """ cursor.execute(query_earliest_user) earliest_user = cursor.fetchone() if earliest_user: created_by = earliest_user["id"] print(f"使用创建时间最早的用户ID: {created_by}") else: created_by = "system" print("未找到用户, 使用默认用户ID: system") cursor.close() conn.close() # 检查知识库是否存在 kb = cls.get_knowledgebase_detail(kb_id) print(f"[DEBUG] 知识库检查结果: {kb}") if not kb: print(f"[ERROR] 知识库不存在: {kb_id}") raise Exception("知识库不存在") conn = cls._get_db_connection() cursor = conn.cursor() # 获取文件信息 file_query = """ SELECT id, name, location, size, type FROM file WHERE id IN (%s) """ % ",".join(["%s"] * len(file_ids)) print(f"[DEBUG] 执行文件查询SQL: {file_query}") print(f"[DEBUG] 查询参数: {file_ids}") try: cursor.execute(file_query, file_ids) files = cursor.fetchall() print(f"[DEBUG] 查询到的文件数据: {files}") except Exception as e: print(f"[ERROR] 文件查询失败: {str(e)}") raise if len(files) != len(file_ids): print(f"部分文件不存在: 期望={len(file_ids)}, 实际={len(files)}") raise Exception("部分文件不存在") # 添加文档记录 added_count = 0 for file in files: file_id = file[0] file_name = file[1] print(f"处理文件: id={file_id}, name={file_name}") file_location = file[2] file_size = file[3] file_type = file[4] # 检查文档是否已存在于知识库 check_query = """ SELECT COUNT(*) FROM document d JOIN file2document f2d ON d.id = f2d.document_id WHERE d.kb_id = %s AND f2d.file_id = %s """ cursor.execute(check_query, (kb_id, file_id)) exists = cursor.fetchone()[0] > 0 if exists: continue # 跳过已存在的文档 # 创建文档记录 doc_id = generate_uuid() current_datetime = datetime.now() create_time = int(current_datetime.timestamp() * 1000) # 毫秒级时间戳 current_date = current_datetime.strftime("%Y-%m-%d %H:%M:%S") # 格式化日期字符串 # 设置默认值 default_parser_id = "naive" default_parser_config = json.dumps( { "layout_recognize": "MinerU", "chunk_token_num": 512, "delimiter": "\n!?;。；！？", "auto_keywords": 0, "auto_questions": 0, "html4excel": False, "raptor": {"use_raptor": False}, "graphrag": {"use_graphrag": False}, } ) default_source_type = "local" # 插入document表 doc_query = """ INSERT INTO document ( id, create_time, create_date, update_time, update_date, thumbnail, kb_id, parser_id, parser_config, source_type, type, created_by, name, location, size, token_num, chunk_num, progress, progress_msg, process_begin_at, process_duation, meta_fields, run, status ) VALUES ( %s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s ) """ doc_params = [ doc_id, create_time, current_date, create_time, current_date, # ID和时间 None, kb_id, default_parser_id, default_parser_config, default_source_type, # thumbnail到source_type file_type, created_by, file_name, file_location, file_size, # type到size 0, 0, 0.0, None, None, # token_num到process_begin_at 0.0, None, "0", "1", # process_duation到status ] cursor.execute(doc_query, doc_params) # 创建文件到文档的映射 f2d_id = generate_uuid() f2d_query = """ INSERT INTO file2document ( id, create_time, create_date, update_time, update_date, file_id, document_id ) VALUES ( %s, %s, %s, %s, %s, %s, %s ) """ f2d_params = [f2d_id, create_time, current_date, create_time, current_date, file_id, doc_id] cursor.execute(f2d_query, f2d_params) added_count += 1 # 更新知识库文档数量 if added_count > 0: try: update_query = """ UPDATE knowledgebase SET doc_num = doc_num + %s, update_date = %s WHERE id = %s """ cursor.execute(update_query, (added_count, current_date, kb_id)) conn.commit() # 先提交更新操作 except Exception as e: print(f"[WARNING] 更新知识库文档数量失败，但文档已添加: {str(e)}") cursor.close() conn.close() return {"added_count": added_count} except Exception as e: print(f"[ERROR] 添加文档失败: {str(e)}") print(f"[ERROR] 错误类型: {type(e)}") import traceback print(f"[ERROR] 堆栈信息: {traceback.format_exc()}") raise Exception(f"添加文档到知识库失败: {str(e)}") @classmethod def delete_document(cls, doc_id): """删除文档""" try: conn = cls._get_db_connection() cursor = conn.cursor(dictionary=True) # 先检查文档是否存在 check_query = """ SELECT d.kb_id, kb.created_by AS tenant_id -- 获取 tenant_id (knowledgebase的创建者) FROM document d JOIN knowledgebase kb ON d.kb_id = kb.id -- JOIN knowledgebase 表 WHERE d.id = %s """ cursor.execute(check_query, (doc_id,)) doc_data = cursor.fetchone() if not doc_data: print(f"[INFO] 文档 {doc_id} 在数据库中未找到。") return False kb_id = doc_data["kb_id"] # 删除文件到文档的映射 f2d_query = "DELETE FROM file2document WHERE document_id = %s" cursor.execute(f2d_query, (doc_id,)) # 删除文档 doc_query = "DELETE FROM document WHERE id = %s" cursor.execute(doc_query, (doc_id,)) # 更新知识库文档数量 update_query = """ UPDATE knowledgebase SET doc_num = doc_num - 1, update_date = %s WHERE id = %s """ current_date = datetime.now().strftime("%Y-%m-%d %H:%M:%S") cursor.execute(update_query, (current_date, kb_id)) conn.commit() cursor.close() conn.close() es_client = get_es_client() tenant_id_for_cleanup = doc_data["tenant_id"] # 删除 Elasticsearch 中的相关文档块 if es_client and tenant_id_for_cleanup: es_index_name = f"ragflow_{tenant_id_for_cleanup}" try: if es_client.indices.exists(index=es_index_name): query_body = {"query": {"term": {"doc_id": doc_id}}} resp = es_client.delete_by_query( index=es_index_name, body=query_body, refresh=True, # 确保立即生效 ignore_unavailable=True, # 如果索引在此期间被删除 ) deleted_count = resp.get("deleted", 0) print(f"[ES-SUCCESS] 从索引 {es_index_name} 中删除 {deleted_count} 个与 doc_id {doc_id} 相关的块。") else: print(f"[ES-INFO] 索引 {es_index_name} 不存在，跳过 ES 清理 for doc_id {doc_id}。") except Exception as es_err: print(f"[ES-ERROR] 清理 ES 块 for doc_id {doc_id} (index {es_index_name}) 失败: {str(es_err)}") return True except Exception as e: print(f"[ERROR] 删除文档失败: {str(e)}") raise Exception(f"删除文档失败: {str(e)}") @classmethod def parse_document(cls, doc_id): """解析文档""" conn = None cursor = None try: # 获取文档和文件信息 conn = cls._get_db_connection() cursor = conn.cursor(dictionary=True) # 查询文档信息 doc_query = """ SELECT d.id, d.name, d.location, d.type, d.kb_id, d.parser_id, d.parser_config, d.created_by FROM document d WHERE d.id = %s """ cursor.execute(doc_query, (doc_id,)) doc_info = cursor.fetchone() if not doc_info: raise Exception("文档不存在") # 获取关联的文件信息 (主要是 parent_id 作为 bucket_name) f2d_query = "SELECT file_id FROM file2document WHERE document_id = %s" cursor.execute(f2d_query, (doc_id,)) f2d_result = cursor.fetchone() if not f2d_result: raise Exception("无法找到文件到文档的映射关系") file_id = f2d_result["file_id"] file_query = "SELECT parent_id FROM file WHERE id = %s" cursor.execute(file_query, (file_id,)) file_info = cursor.fetchone() if not file_info: raise Exception("无法找到文件记录") # 获取知识库创建人信息 # 根据doc_id查询document这张表，得到kb_id kb_id_query = "SELECT kb_id FROM document WHERE id = %s" cursor.execute(kb_id_query, (doc_id,)) kb_id = cursor.fetchone() # 根据kb_id查询knowledgebase这张表，得到created_by kb_query = "SELECT created_by FROM knowledgebase WHERE id = %s" cursor.execute(kb_query, (kb_id["kb_id"],)) kb_info = cursor.fetchone() cursor.close() conn.close() conn = None # 确保连接已关闭 # 更新文档状态为处理中 (使用 parser 模块的函数) _update_document_progress(doc_id, status="2", run="1", progress=0.0, message="开始解析") # 调用后台解析函数 embedding_config = cls.get_system_embedding_config() parse_result = perform_parse(doc_id, doc_info, file_info, embedding_config, kb_info) # 返回解析结果 return parse_result except Exception as e: print(f"文档解析启动或执行过程中出错 (Doc ID: {doc_id}): {str(e)}") # 确保在异常时更新状态为失败 try: _update_document_progress(doc_id, status="1", run="0", message=f"解析失败: {str(e)}") except Exception as update_err: print(f"更新文档失败状态时出错 (Doc ID: {doc_id}): {str(update_err)}") # raise Exception(f"文档解析失败: {str(e)}") return {"success": False, "error": f"文档解析失败: {str(e)}"} finally: if cursor: cursor.close() if conn: conn.close() @classmethod def async_parse_document(cls, doc_id): """异步解析文档""" try: # 启动后台线程执行同步的 parse_document 方法 thread = threading.Thread(target=cls.parse_document, args=(doc_id,)) thread.daemon = True # 设置为守护线程，主程序退出时线程也退出 thread.start() # 立即返回，表示任务已提交 return { "task_id": doc_id, # 使用 doc_id 作为任务标识符 "status": "processing", "message": "文档解析任务已提交到后台处理", } except Exception as e: print(f"启动异步解析任务失败 (Doc ID: {doc_id}): {str(e)}") try: _update_document_progress(doc_id, status="1", run="0", message=f"启动解析失败: {str(e)}") except Exception as update_err: print(f"更新文档启动失败状态时出错 (Doc ID: {doc_id}): {str(update_err)}") raise Exception(f"启动异步解析任务失败: {str(e)}") @classmethod def get_document_parse_progress(cls, doc_id): """获取文档解析进度""" conn = None cursor = None try: conn = cls._get_db_connection() cursor = conn.cursor(dictionary=True) query = """ SELECT progress, progress_msg, status, run FROM document WHERE id = %s """ cursor.execute(query, (doc_id,)) result = cursor.fetchone() if not result: return {"error": "文档不存在"} # 确保 progress 是浮点数 progress_value = 0.0 if result.get("progress") is not None: try: progress_value = float(result["progress"]) except (ValueError, TypeError): progress_value = 0.0 # 或记录错误 return { "progress": progress_value, "message": result.get("progress_msg", ""), "status": result.get("status", "0"), "running": result.get("run", "0"), } except Exception as e: print(f"获取文档进度失败 (Doc ID: {doc_id}): {str(e)}") return {"error": f"获取进度失败: {str(e)}"} finally: if cursor: cursor.close() if conn: conn.close() # --- 获取最早用户 ID --- @classmethod def _get_earliest_user_tenant_id(cls): """获取创建时间最早的用户的 ID (作为 tenant_id)""" conn = None cursor = None try: conn = cls._get_db_connection() cursor = conn.cursor() query = "SELECT id FROM user ORDER BY create_time ASC LIMIT 1" cursor.execute(query) result = cursor.fetchone() if result: return result[0] # 返回用户 ID else: print("警告: 数据库中没有用户！") return None except Exception as e: print(f"查询最早用户时出错: {e}") traceback.print_exc() return None finally: if cursor: cursor.close() if conn and conn.is_connected(): conn.close() # --- 测试 Embedding 连接 --- @classmethod def _test_embedding_connection(cls, base_url, model_name, api_key): """ 测试与自定义 Embedding 模型的连接 (使用 requests)。 """ print(f"开始测试连接: base_url={base_url}, model_name={model_name}") try: headers = {"Content-Type": "application/json"} if api_key: headers["Authorization"] = f"Bearer {api_key}" payload = {"input": ["Test connection"], "model": model_name} if not base_url.startswith(("http://", "https://")): base_url = "http://" + base_url if not base_url.endswith("/"): base_url += "/" # --- URL 拼接优化 --- endpoint_segment = "embeddings" full_endpoint_path = "v1/embeddings" # 移除末尾斜杠以方便判断 normalized_base_url = base_url.rstrip("/") if normalized_base_url.endswith("/v1"): # 如果 base_url 已经是 http://host/v1 形式 current_test_url = normalized_base_url + "/" + endpoint_segment elif normalized_base_url.endswith("/embeddings"): # 如果 base_url 已经是 http://host/embeddings 形式(比如硅基流动API，无需再进行处理) current_test_url = normalized_base_url else: # 如果 base_url 是 http://host 或 http://host/api 形式 current_test_url = normalized_base_url + "/" + full_endpoint_path # --- 结束 URL 拼接优化 --- print(f"尝试请求 URL: {current_test_url}") try: response = requests.post(current_test_url, headers=headers, json=payload, timeout=15) print(f"请求 {current_test_url} 返回状态码: {response.status_code}") if response.status_code == 200: res_json = response.json() if ( "data" in res_json and isinstance(res_json["data"], list) and len(res_json["data"]) > 0 and "embedding" in res_json["data"][0] and len(res_json["data"][0]["embedding"]) > 0 ) or (isinstance(res_json, list) and len(res_json) > 0 and isinstance(res_json[0], list) and len(res_json[0]) > 0): print(f"连接测试成功: {current_test_url}") return True, "连接成功" else: print(f"连接成功但响应格式不正确于 {current_test_url}") except Exception as json_e: print(f"解析 JSON 响应失败于 {current_test_url}: {json_e}") return False, "连接失败: 响应错误" except Exception as e: print(f"连接测试发生未知错误: {str(e)}") traceback.print_exc() return False, f"测试时发生未知错误: {str(e)}" # --- 获取系统 Embedding 配置 --- @classmethod def get_system_embedding_config(cls): """获取系统级（最早用户）的 Embedding 配置""" conn = None cursor = None try: conn = cls._get_db_connection() cursor = conn.cursor(dictionary=True) # 使用字典游标方便访问列名 # 1. 找到最早创建的用户ID query_earliest_user = """ SELECT id FROM user ORDER BY create_time ASC LIMIT 1 """ cursor.execute(query_earliest_user) earliest_user = cursor.fetchone() if not earliest_user: # 如果没有用户，返回空配置 return {"llm_name": "", "api_key": "", "api_base": ""} earliest_user_id = earliest_user["id"] # 2. 根据最早用户ID查询 tenant_llm 表中 model_type 为 embedding 的配置 query_embedding_config = """ SELECT llm_name, api_key, api_base FROM tenant_llm WHERE tenant_id = %s AND model_type = 'embedding' ORDER BY create_time DESC # 如果一个用户可能有多个embedding配置，取最早的 LIMIT 1 """ cursor.execute(query_embedding_config, (earliest_user_id,)) config = cursor.fetchone() if config: llm_name = config.get("llm_name", "") api_key = config.get("api_key", "") api_base = config.get("api_base", "") # 对模型名称进行处理 (可选，根据需要保留或移除) if llm_name and "___" in llm_name: llm_name = llm_name.split("___")[0] # (对硅基流动平台进行特异性处理) if llm_name == "netease-youdao/bce-embedding-base_v1": llm_name = "BAAI/bge-m3" # 如果 API 基础地址为空字符串，设置为硅基流动嵌入模型的 API 地址 if api_base == "": api_base = "https://api.siliconflow.cn/v1/embeddings" # 如果有配置，返回 return {"llm_name": llm_name, "api_key": api_key, "api_base": api_base} else: # 如果最早的用户没有 embedding 配置，返回空 return {"llm_name": "", "api_key": "", "api_base": ""} except Exception as e: print(f"获取系统 Embedding 配置时出错: {e}") traceback.print_exc() # 保持原有的异常处理逻辑，向上抛出，让调用者处理 raise Exception(f"获取配置时数据库出错: {e}") finally: if cursor: cursor.close() if conn and conn.is_connected(): conn.close() # --- 设置系统 Embedding 配置 --- @classmethod def set_system_embedding_config(cls, llm_name, api_base, api_key): """设置系统级（最早用户）的 Embedding 配置""" tenant_id = cls._get_earliest_user_tenant_id() if not tenant_id: raise Exception("无法找到系统基础用户") print(f"开始设置系统 Embedding 配置: {llm_name}, {api_base}, {api_key}") # 执行连接测试 is_connected, message = cls._test_embedding_connection(base_url=api_base, model_name=llm_name, api_key=api_key) if not is_connected: # 返回具体的测试失败原因给调用者（路由层）处理 return False, f"连接测试失败: {message}" return True, f"连接成功: {message}" # 顺序批量解析 (核心逻辑，在后台线程运行) @classmethod def _run_sequential_batch_parse(cls, kb_id): """顺序执行批量解析，并在 SEQUENTIAL_BATCH_TASKS 中更新状态""" global SEQUENTIAL_BATCH_TASKS task_info = SEQUENTIAL_BATCH_TASKS.get(kb_id) if not task_info: print(f"[Seq Batch ERROR] Task info for KB {kb_id} not found at start.") return # 理论上不应发生 conn = None cursor = None parsed_count = 0 failed_count = 0 total_count = 0 try: conn = cls._get_db_connection() cursor = conn.cursor(dictionary=True) # 查询需要解析的文档 query = """ SELECT id, name FROM document WHERE kb_id = %s AND run != '3' """ cursor.execute(query, (kb_id,)) documents_to_parse = cursor.fetchall() total_count = len(documents_to_parse) # 更新任务总数 task_info["total"] = total_count task_info["status"] = "running" task_info["message"] = f"共找到 {total_count} 个文档待解析。" task_info["start_time"] = time.time() start_time = time.time() SEQUENTIAL_BATCH_TASKS[kb_id] = task_info # 更新字典 if not documents_to_parse: task_info["status"] = "completed" task_info["message"] = "没有需要解析的文档。" SEQUENTIAL_BATCH_TASKS[kb_id] = task_info print(f"[Seq Batch] KB {kb_id}: 没有需要解析的文档。") return print(f"[Seq Batch] KB {kb_id}: 开始顺序解析 {total_count} 个文档...") # 按顺序解析每个文档 for i, doc in enumerate(documents_to_parse): doc_id = doc["id"] doc_name = doc["name"] # 更新当前进度 task_info["current"] = i + 1 task_info["message"] = f"正在解析: {doc_name} ({i + 1}/{total_count})" SEQUENTIAL_BATCH_TASKS[kb_id] = task_info print(f"[Seq Batch] KB {kb_id}: ({i + 1}/{total_count}) Parsing {doc_name} (ID: {doc_id})...") try: # 调用同步的 parse_document 方法 # 这个方法内部会处理单个文档的状态更新 (run, status) result = cls.parse_document(doc_id) if result and result.get("success"): parsed_count += 1 print(f"[Seq Batch] KB {kb_id}: Document {doc_id} parsed successfully.") else: failed_count += 1 error_msg = result.get("message", "未知错误") if result else "未知错误" print(f"[Seq Batch] KB {kb_id}: Document {doc_id} parsing failed: {error_msg}") except Exception as e: failed_count += 1 print(f"[Seq Batch ERROR] KB {kb_id}: Error calling parse_document for {doc_id}: {str(e)}") traceback.print_exc() # 更新文档状态为失败 try: _update_document_progress(doc_id, status="1", run="0", progress=0.0, message=f"批量任务中解析失败: {str(e)[:255]}") except Exception as update_err: print(f"[Service-ERROR] 更新文档 {doc_id} 失败状态时出错: {str(update_err)}") # 任务完成 end_time = time.time() duration = round(end_time - task_info.get("start_time", start_time), 2) final_message = f"批量顺序解析完成。总计 {total_count} 个，成功 {parsed_count} 个，失败 {failed_count} 个。耗时 {duration} 秒。" task_info["status"] = "completed" task_info["message"] = final_message task_info["current"] = total_count SEQUENTIAL_BATCH_TASKS[kb_id] = task_info print(f"[Seq Batch] KB {kb_id}: {final_message}") except Exception as e: # 任务执行中发生严重错误 error_message = f"批量顺序解析过程中发生严重错误: {str(e)}" print(f"[Seq Batch ERROR] KB {kb_id}: {error_message}") traceback.print_exc() task_info["status"] = "failed" task_info["message"] = error_message SEQUENTIAL_BATCH_TASKS[kb_id] = task_info finally: if cursor: cursor.close() if conn and conn.is_connected(): conn.close() # 启动顺序批量解析 (异步请求) @classmethod def start_sequential_batch_parse_async(cls, kb_id): """异步启动知识库的顺序批量解析任务""" global SEQUENTIAL_BATCH_TASKS if kb_id in SEQUENTIAL_BATCH_TASKS and SEQUENTIAL_BATCH_TASKS[kb_id].get("status") == "running": return {"success": False, "message": "该知识库的批量解析任务已在运行中。"} # 初始化任务状态 start_time = time.time() SEQUENTIAL_BATCH_TASKS[kb_id] = {"status": "starting", "total": 0, "current": 0, "message": "任务准备启动...", "start_time": start_time} try: # 启动后台线程执行顺序解析逻辑 thread = threading.Thread(target=cls._run_sequential_batch_parse, args=(kb_id,)) thread.daemon = True thread.start() print(f"[Seq Batch] KB {kb_id}: 已启动后台顺序解析线程。") return {"success": True, "message": "顺序批量解析任务已启动。"} except Exception as e: error_message = f"启动顺序批量解析任务失败: {str(e)}" print(f"[Seq Batch ERROR] KB {kb_id}: {error_message}") traceback.print_exc() # 更新任务状态为失败 SEQUENTIAL_BATCH_TASKS[kb_id] = {"status": "failed", "total": 0, "current": 0, "message": error_message, "start_time": start_time} return {"success": False, "message": error_message} # 获取顺序批量解析进度 @classmethod def get_sequential_batch_parse_progress(cls, kb_id): """获取指定知识库的顺序批量解析任务进度""" global SEQUENTIAL_BATCH_TASKS task_info = SEQUENTIAL_BATCH_TASKS.get(kb_id) if not task_info: return {"status": "not_found", "message": "未找到该知识库的批量解析任务记录。"} # 返回当前任务状态 return task_info # 获取知识库所有文档状态 (用于刷新列表) @classmethod def get_knowledgebase_parse_progress(cls, kb_id): """获取指定知识库下所有文档的解析进度和状态 (保持不变)""" conn = None cursor = None try: conn = cls._get_db_connection() cursor = conn.cursor(dictionary=True) query = """ SELECT id, name, progress, progress_msg, status, run FROM document WHERE kb_id = %s ORDER BY create_date DESC -- 或者其他排序方式 """ cursor.execute(query, (kb_id,)) documents_status = cursor.fetchall() # 处理 progress 确保是浮点数 for doc in documents_status: progress_value = 0.0 if doc.get("progress") is not None: try: progress_value = float(doc["progress"]) except (ValueError, TypeError): progress_value = 0.0 doc["progress"] = progress_value # 确保其他字段存在，给予默认值 doc["progress_msg"] = doc.get("progress_msg", "") doc["status"] = doc.get("status", "0") doc["run"] = doc.get("run", "0") return {"documents": documents_status} except Exception as e: print(f"获取知识库 {kb_id} 文档进度失败: {str(e)}") traceback.print_exc() return {"error": f"获取知识库文档进度失败: {str(e)}"} finally: if cursor: cursor.close() if conn and conn.is_connected(): conn.close()