dsProject/dsSchoolBuddy/ElasticSearch/Utils/EsSearchUtil.py

import json
import logging
import warnings
import hashlib
import time

import requests

from Config.Config import ES_CONFIG
from ElasticSearch.Utils.ElasticsearchConnectionPool import ElasticsearchConnectionPool
from langchain_core.documents import Document
from langchain_text_splitters import RecursiveCharacterTextSplitter
from langchain_openai import OpenAIEmbeddings
from pydantic import SecretStr
from Config import Config
from typing import List, Tuple, Dict
# 初始化日志
logger = logging.getLogger(__name__)
logger.setLevel(logging.INFO)

class EsSearchUtil:
    def __init__(self, es_config):
        """
                初始化Elasticsearch搜索工具
                :param es_config: Elasticsearch配置字典，包含hosts, username, password, index_name等
                """
        # 抑制HTTPS相关警告
        warnings.filterwarnings('ignore', message='Connecting to .* using TLS with verify_certs=False is insecure')
        warnings.filterwarnings('ignore', message='Unverified HTTPS request is being made to host')

        self.es_config = es_config

        # 初始化连接池
        self.es_pool = ElasticsearchConnectionPool(
            hosts=es_config['hosts'],
            basic_auth=es_config['basic_auth'],
            verify_certs=es_config.get('verify_certs', False),
            max_connections=50
        )

        self.index_name = es_config['index_name']
        logger.info(f"EsSearchUtil初始化成功，索引名称: {self.index_name}")
        
    def rebuild_mapping(self):
        """
        重建Elasticsearch索引和mapping结构
        
        返回:
            bool: 操作是否成功
        """
        try:
            # 从连接池获取连接
            conn = self.es_pool.get_connection()
            
            # 定义mapping结构
            mapping = {
                "mappings": {
                    "properties": {
                        "embedding": {
                            "type": "dense_vector",
                            "dims": Config.EMBED_DIM,  # embedding维度为Config.EMBED_DIM
                            "index": True,
                            "similarity": "l2_norm"  # 使用L2距离
                        },
                        "user_input": {"type": "text"},
                        "tags": {
                            "type": "object",
                            "properties": {
                                "tags": {"type": "keyword"},
                                "full_content": {"type": "text"}
                            }
                        }
                    }
                }
            }
            
            # 检查索引是否存在，存在则删除
            if conn.indices.exists(index=self.index_name):
                conn.indices.delete(index=self.index_name)
                logger.info(f"删除已存在的索引 '{self.index_name}'")
                print(f"删除已存在的索引 '{self.index_name}'")
            
            # 创建索引和mapping
            conn.indices.create(index=self.index_name, body=mapping)
            logger.info(f"索引 '{self.index_name}' 创建成功，mapping结构已设置")
            print(f"索引 '{self.index_name}' 创建成功，mapping结构已设置。")
            
            return True
        except Exception as e:
            logger.error(f"重建mapping失败: {str(e)}")
            print(f"重建mapping失败: {e}")
            return False
        finally:
            # 释放连接回连接池
            self.es_pool.release_connection(conn)

    def text_search(self, query, size=10):
        # 从连接池获取连接
        conn = self.es_pool.get_connection()
        try:
            # 使用连接执行搜索
            result = conn.search(
                index=self.es_config['index_name'],
                query={"match": {"user_input": query}},
                size=size
            )
            return result
        except Exception as e:
            logger.error(f"文本搜索失败: {str(e)}")
            raise
        finally:
            # 释放连接回连接池
            self.es_pool.release_connection(conn)

    def select_all_data(self, size=1000):
        """
        查询索引中的所有数据
        
        参数:
            size: 返回的最大结果数量，默认1000
        
        返回:
            dict: 查询结果
        """
        # 从连接池获取连接
        conn = self.es_pool.get_connection()
        try:
            # 构建查询条件 - 匹配所有文档
            query = {
                "query": {
                    "match_all": {}
                },
                "size": size
            }
            
            # 执行查询
            response = conn.search(index=self.es_config['index_name'], body=query)
            return response
        except Exception as e:
            logger.error(f"查询所有数据失败: {str(e)}")
            raise
        finally:
            # 释放连接回连接池
            self.es_pool.release_connection(conn)

    def split_text_into_chunks(self,text: str, chunk_size: int = 200, chunk_overlap: int = 0) -> list:
        """
        将文本切割成块

        参数:
            text: 要切割的文本
            chunk_size: 每个块的大小
            chunk_overlap: 块之间的重叠大小

        返回:
            list: 文本块列表
        """
        # 创建文档对象
        docs = [Document(page_content=text, metadata={"source": "simulated_document"})]

        # 切割文档
        text_splitter = RecursiveCharacterTextSplitter(
            chunk_size=chunk_size, chunk_overlap=chunk_overlap, add_start_index=True
        )
        all_splits = text_splitter.split_documents(docs)
        print(f"切割后的文档块数量：{len(all_splits)}")

        return [split.page_content for split in all_splits]

    def insert_long_text_to_es(self,long_text: str, tags: list = None) -> bool:
        """
        将长文本切割后向量化并插入到Elasticsearch，基于文本内容哈希实现去重

        参数:
            long_text: 要插入的长文本
            tags: 可选的标签列表

        返回:
            bool: 插入是否成功
        """
        try:
            # 1. 创建EsSearchUtil实例以使用连接池
            search_util = EsSearchUtil(Config.ES_CONFIG)

            # 2. 从连接池获取连接
            conn = search_util.es_pool.get_connection()

            # 3. 检查索引是否存在，不存在则创建
            index_name = Config.ES_CONFIG['index_name']
            if not conn.indices.exists(index=index_name):
                # 定义mapping结构
                mapping = {
                    "mappings": {
                        "properties": {
                            "embedding": {
                                "type": "dense_vector",
                                "dims": 1024,  # 根据实际embedding维度调整
                                "index": True,
                                "similarity": "l2_norm"
                            },
                            "user_input": {"type": "text"},
                            "tags": {
                                "type": "object",
                                "properties": {
                                    "tags": {"type": "keyword"},
                                    "full_content": {"type": "text"}
                                }
                            },
                            "timestamp": {"type": "date"}
                        }
                    }
                }
                conn.indices.create(index=index_name, body=mapping)
                print(f"索引 '{index_name}' 创建成功")

            # 4. 切割文本
            text_chunks = self.split_text_into_chunks(long_text)

            # 5. 准备标签
            if tags is None:
                tags = ["general_text"]

            # 6. 获取当前时间
            timestamp = time.strftime("%Y-%m-%d %H:%M:%S", time.localtime())

            # 7. 创建嵌入模型
            embeddings = OpenAIEmbeddings(
                model=Config.EMBED_MODEL_NAME,
                base_url=Config.EMBED_BASE_URL,
                api_key=SecretStr(Config.EMBED_API_KEY)
            )

            # 8. 为每个文本块生成向量并插入
            for i, chunk in enumerate(text_chunks):
                # 生成文本块的哈希值作为文档ID
                doc_id = hashlib.md5(chunk.encode('utf-8')).hexdigest()

                # 检查文档是否已存在
                if conn.exists(index=index_name, id=doc_id):
                    print(f"文档块 {i+1} 已存在，跳过插入: {doc_id}")
                    continue

                # 生成文本块的嵌入向量
                embedding = embeddings.embed_documents([chunk])[0]

                # 准备文档数据
                doc = {
                    'tags': {"tags": tags, "full_content": long_text},
                    'user_input': chunk,
                    'timestamp': timestamp,
                    'embedding': embedding
                }

                # 插入数据到Elasticsearch
                conn.index(index=index_name, id=doc_id, document=doc)
                print(f"文档块 {i+1} 插入成功: {doc_id}")

            return True
        except Exception as e:
            print(f"插入数据失败: {e}")
            return False
        finally:
            # 确保释放连接回连接池
            if 'conn' in locals() and 'search_util' in locals():
                search_util.es_pool.release_connection(conn)

    def get_query_embedding(self, query: str) -> list:
        """
        将查询文本转换为向量

        参数:
            query: 查询文本

        返回:
            list: 向量表示
        """
        # 创建嵌入模型
        embeddings = OpenAIEmbeddings(
            model=Config.EMBED_MODEL_NAME,
            base_url=Config.EMBED_BASE_URL,
            api_key=SecretStr(Config.EMBED_API_KEY)
        )

        # 生成查询向量
        query_embedding = embeddings.embed_query(query)
        return query_embedding

    def rerank_results(self, query: str, results: List[Dict]) -> List[Tuple[Dict, float]]:
        """
        对搜索结果进行重排

        参数:
            query: 查询文本
            results: 搜索结果列表

        返回:
            list: 重排后的结果列表，每个元素是(文档, 分数)元组
        """
        if len(results) <= 1:
            return [(doc, 1.0) for doc in results]

        # 准备重排请求数据
        rerank_data = {
            "model": Config.RERANK_MODEL,
            "query": query,
            "documents": [doc['_source']['user_input'] for doc in results],
            "top_n": len(results)
        }

        # 调用API进行重排
        headers = {
            "Content-Type": "application/json",
            "Authorization": f"Bearer {Config.RERANK_BINDING_API_KEY}"
        }

        try:
            response = requests.post(Config.RERANK_BASE_URL, headers=headers, data=json.dumps(rerank_data))
            response.raise_for_status()
            rerank_result = response.json()

            # 处理重排结果
            reranked_docs_with_scores = []
            if "results" in rerank_result:
                for item in rerank_result["results"]:
                    # 尝试获取index和relevance_score字段
                    doc_idx = item.get("index")
                    score = item.get("relevance_score", 0.0)
                    
                    # 如果找不到，尝试fallback到document和score字段
                    if doc_idx is None:
                        doc_idx = item.get("document")
                    if score == 0.0:
                        score = item.get("score", 0.0)

                    if doc_idx is not None and 0 <= doc_idx < len(results):
                        reranked_docs_with_scores.append((results[doc_idx], score))
                        logger.debug(f"重排结果: 文档索引={doc_idx}, 分数={score}")
                    else:
                        logger.warning(f"重排结果项索引无效: {doc_idx}")

            # 如果没有有效的重排结果，返回原始结果
            if not reranked_docs_with_scores:
                logger.warning("没有获取到有效的重排结果，返回原始结果")
                return [(doc, 1.0) for doc in results]

            return reranked_docs_with_scores
        except Exception as e:
            logger.error(f"重排失败: {str(e)}")
            return [(doc, 1.0) for doc in results]

    def search_by_vector(self, query_embedding: list, k: int = 10) -> dict:
        """
        在Elasticsearch中按向量搜索

        参数:
            query_embedding: 查询向量
            k: 返回结果数量

        返回:
            dict: 搜索结果
        """
        # 从连接池获取连接
        conn = self.es_pool.get_connection()
        try:
            # 构建向量搜索查询
            query = {
                "query": {
                    "script_score": {
                        "query": {
                            "bool": {
                                "should": [],
                                "minimum_should_match": 0
                            }
                        },
                        "script": {
                            "source": "double score = cosineSimilarity(params.query_vector, 'embedding'); return score >= 0 ? score : 0",
                            "params": {"query_vector": query_embedding}
                        }
                    }
                },
                "size": k
            }

            # 执行查询
            response = conn.search(index=self.es_config['index_name'], body=query)
            return response
        except Exception as e:
            logger.error(f"向量搜索失败: {str(e)}")
            raise
        finally:
            # 释放连接回连接池
            self.es_pool.release_connection(conn)

    def display_results(self, results: list, show_score: bool = True) -> None:
        """
        展示搜索结果

        参数:
            results: 搜索结果列表
            show_score: 是否显示分数
        """
        if not results:
            print("没有找到匹配的结果。")
            return

        print(f"找到 {len(results)} 条结果:\n")
        for i, (result, score) in enumerate(results, 1):
            print(f"结果 {i}:")
            print(f"内容: {result['_source']['user_input']}")
            if show_score:
                print(f"分数: {score:.4f}")
            print("---")

    def merge_results(self, keyword_results: List[Tuple[Dict, float]], vector_results: List[Tuple[Dict, float]]) -> List[Tuple[Dict, float, str]]:
        """
        合并关键字搜索和向量搜索结果

        参数:
            keyword_results: 关键字搜索结果列表，每个元素是(文档, 分数)元组
            vector_results: 向量搜索结果列表，每个元素是(文档, 分数)元组

        返回:
            list: 合并后的结果列表，每个元素是(文档, 分数, 来源)元组
        """
        # 标记结果来源并合并
        all_results = []
        for doc, score in keyword_results:
            all_results.append((doc, score, "关键字搜索"))
        for doc, score in vector_results:
            all_results.append((doc, score, "向量搜索"))

        # 去重并按分数排序
        unique_results = {}
        for doc, score, source in all_results:
            doc_id = doc['_id']
            if doc_id not in unique_results or score > unique_results[doc_id][1]:
                unique_results[doc_id] = (doc, score, source)

        # 按分数降序排序
        sorted_results = sorted(unique_results.values(), key=lambda x: x[1], reverse=True)
        return sorted_results