auto-coder 1.0.0__py3-none-any.whl → 2.0.0__py3-none-any.whl

autocoder/rag/tools/bochaai_sdk.py

@@ -0,0 +1,541 @@
+
+"""
+BochaAI SDK 封装
+
+该模块提供了 BochaAI API 的 Python SDK 封装，支持网页搜索和内容爬取功能。
+"""
+
+import json
+import time
+import requests
+from typing import Dict, Any, List, Optional, Union
+from dataclasses import dataclass, field
+from loguru import logger
+from urllib.parse import urlparse, urljoin
+
+
+@dataclass
+class BochaAIWebPage:
+    """BochaAI 网页搜索结果数据类"""
+    name: str
+    url: str
+    snippet: str
+    display_url: Optional[str] = None
+    summary: Optional[str] = None
+    site_name: Optional[str] = None
+    site_icon: Optional[str] = None
+    date_published: Optional[str] = None
+    date_last_crawled: Optional[str] = None
+    cached_page_url: Optional[str] = None
+    language: Optional[str] = None
+    is_family_friendly: Optional[bool] = None
+    is_navigational: Optional[bool] = None
+    
+    def to_dict(self) -> Dict[str, Any]:
+        """转换为字典格式"""
+        result = {
+            "name": self.name,
+            "url": self.url,
+            "snippet": self.snippet
+        }
+        
+        # 添加可选字段
+        optional_fields = [
+            "display_url", "summary", "site_name", "site_icon",
+            "date_published", "date_last_crawled", "cached_page_url",
+            "language", "is_family_friendly", "is_navigational"
+        ]
+        
+        for field_name in optional_fields:
+            value = getattr(self, field_name)
+            if value is not None:
+                result[field_name] = value
+                
+        return result
+
+
+@dataclass
+class BochaAIImage:
+    """BochaAI 图片搜索结果数据类"""
+    thumbnail_url: str
+    content_url: str
+    host_page_url: str
+    width: int
+    height: int
+    name: Optional[str] = None
+    web_search_url: Optional[str] = None
+    date_published: Optional[str] = None
+    content_size: Optional[str] = None
+    encoding_format: Optional[str] = None
+    host_page_display_url: Optional[str] = None
+    
+    def to_dict(self) -> Dict[str, Any]:
+        """转换为字典格式"""
+        return {
+            k: v for k, v in self.__dict__.items() if v is not None
+        }
+
+
+@dataclass
+class BochaAISearchResponse:
+    """BochaAI 搜索响应数据类"""
+    success: bool
+    query: str
+    total_matches: int = 0
+    webpages: List[BochaAIWebPage] = field(default_factory=list)
+    images: List[BochaAIImage] = field(default_factory=list)
+    error: Optional[str] = None
+    log_id: Optional[str] = None
+    
+    def to_dict(self) -> Dict[str, Any]:
+        """转换为字典格式"""
+        return {
+            "success": self.success,
+            "query": self.query,
+            "total_matches": self.total_matches,
+            "webpages": [page.to_dict() for page in self.webpages],
+            "images": [img.to_dict() for img in self.images],
+            "error": self.error,
+            "log_id": self.log_id
+        }
+
+
+class BochaAIClient:
+    """BochaAI API 客户端"""
+    
+    BASE_URL = "https://api.bochaai.com/v1"
+    DEFAULT_TIMEOUT = 30  # 默认超时时间（秒）
+    
+    def __init__(self, api_key: str, timeout: int = DEFAULT_TIMEOUT):
+        """
+        初始化 BochaAI 客户端
+        
+        Args:
+            api_key: BochaAI API 密钥
+            timeout: 请求超时时间（秒）
+        """
+        self.api_key = api_key
+        self.timeout = timeout
+        self.session = requests.Session()
+        self.session.headers.update({
+            "Authorization": f"Bearer {api_key}",
+            "Content-Type": "application/json"
+        })
+    
+    def search(
+        self,
+        query: str,
+        count: int = 10,
+        freshness: str = "noLimit",
+        summary: bool = False,
+        include: Optional[str] = None,
+        exclude: Optional[str] = None
+    ) -> BochaAISearchResponse:
+        """
+        执行网页搜索
+        
+        Args:
+            query: 搜索查询字符串
+            count: 返回结果数量 (1-50)
+            freshness: 时间范围过滤 (noLimit, oneDay, oneWeek, oneMonth, oneYear)
+            summary: 是否包含文本摘要
+            include: 指定搜索的网站范围，多个域名用|或,分隔
+            exclude: 排除搜索的网站范围，多个域名用|或,分隔
+            
+        Returns:
+            BochaAISearchResponse: 搜索响应对象
+        """
+        try:
+            # 准备请求数据
+            data = {
+                "query": query,
+                "count": min(max(count, 1), 50),  # 限制在 1-50 范围内
+                "freshness": freshness,
+                "summary": summary
+            }
+            
+            # 添加可选参数
+            if include:
+                data["include"] = include
+            if exclude:
+                data["exclude"] = exclude
+            
+            logger.info(f"BochaAI 搜索请求: query={query}, count={count}, summary={summary}")
+            
+            # 发送请求
+            response = self.session.post(
+                f"{self.BASE_URL}/web-search",
+                json=data,
+                timeout=self.timeout
+            )
+            
+            # 检查响应状态
+            if response.status_code != 200:
+                error_msg = f"API 请求失败，状态码: {response.status_code}"
+                try:
+                    error_data = response.json()
+                    if "message" in error_data:
+                        error_msg = f"{error_msg}, 错误信息: {error_data['message']}"
+                    log_id = error_data.get("log_id", "")
+                except:
+                    error_msg = f"{error_msg}, 响应内容: {response.text}"
+                    log_id = ""
+                
+                logger.error(error_msg)
+                return BochaAISearchResponse(
+                    success=False,
+                    query=query,
+                    error=error_msg,
+                    log_id=log_id
+                )
+            
+            # 解析响应
+            result = response.json()
+            
+            # 检查响应格式
+            if result.get("code") != 200:
+                error_msg = result.get("msg", "未知错误")
+                return BochaAISearchResponse(
+                    success=False,
+                    query=query,
+                    error=error_msg,
+                    log_id=result.get("log_id")
+                )
+            
+            data = result.get("data", {})
+            
+            # 解析网页搜索结果
+            webpages = []
+            web_pages_data = data.get("webPages", {})
+            for item in web_pages_data.get("value", []):
+                webpage = BochaAIWebPage(
+                    name=item.get("name", ""),
+                    url=item.get("url", ""),
+                    snippet=item.get("snippet", ""),
+                    display_url=item.get("displayUrl"),
+                    summary=item.get("summary"),
+                    site_name=item.get("siteName"),
+                    site_icon=item.get("siteIcon"),
+                    date_published=item.get("datePublished"),
+                    date_last_crawled=item.get("dateLastCrawled"),
+                    cached_page_url=item.get("cachedPageUrl"),
+                    language=item.get("language"),
+                    is_family_friendly=item.get("isFamilyFriendly"),
+                    is_navigational=item.get("isNavigational")
+                )
+                webpages.append(webpage)
+            
+            # 解析图片搜索结果
+            images = []
+            images_data = data.get("images", {})
+            for item in images_data.get("value", []):
+                image = BochaAIImage(
+                    thumbnail_url=item.get("thumbnailUrl", ""),
+                    content_url=item.get("contentUrl", ""),
+                    host_page_url=item.get("hostPageUrl", ""),
+                    width=item.get("width", 0),
+                    height=item.get("height", 0),
+                    name=item.get("name"),
+                    web_search_url=item.get("webSearchUrl"),
+                    date_published=item.get("datePublished"),
+                    content_size=item.get("contentSize"),
+                    encoding_format=item.get("encodingFormat"),
+                    host_page_display_url=item.get("hostPageDisplayUrl")
+                )
+                images.append(image)
+            
+            total_matches = web_pages_data.get("totalEstimatedMatches", 0)
+            
+            return BochaAISearchResponse(
+                success=True,
+                query=query,
+                total_matches=total_matches,
+                webpages=webpages,
+                images=images,
+                log_id=result.get("log_id")
+            )
+            
+        except requests.exceptions.Timeout:
+            error_msg = f"搜索请求超时（{self.timeout}秒）"
+            logger.error(error_msg)
+            return BochaAISearchResponse(
+                success=False,
+                query=query,
+                error=error_msg
+            )
+        except requests.exceptions.RequestException as e:
+            error_msg = f"网络请求失败: {str(e)}"
+            logger.error(error_msg)
+            return BochaAISearchResponse(
+                success=False,
+                query=query,
+                error=error_msg
+            )
+        except Exception as e:
+            error_msg = f"搜索失败: {str(e)}"
+            logger.error(error_msg)
+            return BochaAISearchResponse(
+                success=False,
+                query=query,
+                error=error_msg
+            )
+    
+    def crawl_page(
+        self,
+        url: str,
+        headers: Optional[Dict[str, str]] = None
+    ) -> Dict[str, Any]:
+        """
+        爬取单个网页内容（使用 requests）
+        
+        Args:
+            url: 要爬取的网页 URL
+            headers: 自定义请求头
+            
+        Returns:
+            Dict: 包含网页内容的字典
+        """
+        try:
+            logger.info(f"BochaAI 爬取页面: {url}")
+            
+            # 设置默认请求头
+            default_headers = {
+                "User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36"
+            }
+            
+            if headers:
+                default_headers.update(headers)
+            
+            # 发送请求
+            response = requests.get(
+                url,
+                headers=default_headers,
+                timeout=self.timeout,
+                allow_redirects=True
+            )
+            
+            # 检查响应状态
+            if response.status_code != 200:
+                error_msg = f"页面请求失败，状态码: {response.status_code}"
+                logger.error(error_msg)
+                return {
+                    "success": False,
+                    "url": url,
+                    "error": error_msg,
+                    "status_code": response.status_code
+                }
+            
+            # 获取内容
+            content = response.text
+            
+            # 尝试提取标题
+            title = ""
+            import re
+            title_match = re.search(r'<title[^>]*>([^<]+)</title>', content, re.IGNORECASE)
+            if title_match:
+                title = title_match.group(1).strip()
+            
+            return {
+                "success": True,
+                "url": url,
+                "title": title,
+                "content": content,
+                "content_type": response.headers.get("Content-Type", ""),
+                "content_length": len(content),
+                "status_code": response.status_code
+            }
+            
+        except requests.exceptions.Timeout:
+            error_msg = f"页面请求超时（{self.timeout}秒）"
+            logger.error(error_msg)
+            return {
+                "success": False,
+                "url": url,
+                "error": error_msg
+            }
+        except requests.exceptions.RequestException as e:
+            error_msg = f"网络请求失败: {str(e)}"
+            logger.error(error_msg)
+            return {
+                "success": False,
+                "url": url,
+                "error": error_msg
+            }
+        except Exception as e:
+            error_msg = f"爬取失败: {str(e)}"
+            logger.error(error_msg)
+            return {
+                "success": False,
+                "url": url,
+                "error": error_msg
+            }
+    
+    def crawl(
+        self,
+        url: str,
+        limit: int = 10,
+        max_depth: Optional[int] = None,
+        exclude_paths: Optional[List[str]] = None,
+        include_paths: Optional[List[str]] = None,
+        allow_subdomains: bool = False
+    ) -> List[Dict[str, Any]]:
+        """
+        爬取网站内容（使用搜索和爬取组合实现）
+        
+        Args:
+            url: 起始 URL
+            limit: 最大爬取页面数
+            max_depth: 最大爬取深度（暂未实现）
+            exclude_paths: 排除的路径列表
+            include_paths: 包含的路径列表
+            allow_subdomains: 是否允许子域名
+            
+        Returns:
+            List[Dict]: 爬取的页面内容列表
+        """
+        try:
+            # 解析起始 URL
+            parsed_url = urlparse(url)
+            domain = parsed_url.netloc
+            
+            # 如果只爬取一个页面
+            if limit == 1:
+                result = self.crawl_page(url)
+                if result.get("success"):
+                    return [{
+                        "url": url,
+                        "title": result.get("title", ""),
+                        "content": result.get("content", ""),
+                        "markdown": result.get("content", ""),  # 简单处理，实际应该转换为 markdown
+                        "links": [],
+                        "metadata": {
+                            "source": "bochaai",
+                            "content_type": result.get("content_type", ""),
+                            "status_code": result.get("status_code", 200)
+                        }
+                    }]
+                else:
+                    return []
+            
+            # 多页爬取：先搜索该域名下的相关页面
+            search_query = f"site:{domain}"
+            if parsed_url.path and parsed_url.path != "/":
+                # 如果有特定路径，添加到搜索中
+                path_keywords = parsed_url.path.strip("/").replace("/", " ")
+                search_query += f" {path_keywords}"
+            
+            logger.info(f"开始爬取: {url}, 限制: {limit} 页")
+            
+            # 执行搜索获取相关页面
+            search_result = self.search(
+                query=search_query,
+                count=min(limit * 2, 50),  # 搜索更多结果以便筛选
+                summary=False
+            )
+            
+            if not search_result.success:
+                logger.error(f"爬取失败: {search_result.error}")
+                # 至少尝试爬取起始页面
+                result = self.crawl_page(url)
+                if result.get("success"):
+                    return [{
+                        "url": url,
+                        "title": result.get("title", ""),
+                        "content": result.get("content", ""),
+                        "markdown": result.get("content", ""),
+                        "links": [],
+                        "metadata": {"source": "bochaai"}
+                    }]
+                return []
+            
+            crawled_pages = []
+            urls_to_crawl = [url]  # 起始URL
+            
+            # 添加搜索结果中的URL
+            for webpage in search_result.webpages:
+                if webpage.url and webpage.url not in urls_to_crawl:
+                    # 检查是否应该包含这个URL
+                    should_include = True
+                    
+                    # 检查排除路径
+                    if exclude_paths:
+                        for exclude_path in exclude_paths:
+                            if exclude_path in webpage.url:
+                                should_include = False
+                                break
+                    
+                    # 检查包含路径
+                    if include_paths and should_include:
+                        should_include = False
+                        for include_path in include_paths:
+                            if include_path in webpage.url:
+                                should_include = True
+                                break
+                    
+                    # 检查子域名
+                    if not allow_subdomains and should_include:
+                        webpage_domain = urlparse(webpage.url).netloc
+                        if webpage_domain != domain:
+                            should_include = False
+                    
+                    if should_include:
+                        urls_to_crawl.append(webpage.url)
+            
+            # 限制爬取数量
+            urls_to_crawl = urls_to_crawl[:limit]
+            
+            # 爬取每个页面的内容
+            for idx, page_url in enumerate(urls_to_crawl):
+                logger.info(f"爬取页面 {idx + 1}/{len(urls_to_crawl)}: {page_url}")
+                
+                result = self.crawl_page(page_url)
+                
+                if result.get("success"):
+                    # 从搜索结果中查找对应的信息
+                    page_info = None
+                    for webpage in search_result.webpages:
+                        if webpage.url == page_url:
+                            page_info = webpage
+                            break
+                    
+                    page_result = {
+                        "url": page_url,
+                        "title": page_info.name if page_info else result.get("title", ""),
+                        "content": result.get("content", ""),
+                        "markdown": result.get("content", ""),  # 简单处理
+                        "links": [],
+                        "metadata": {
+                            "position": idx + 1,
+                            "source": "bochaai",
+                            "content_type": result.get("content_type", ""),
+                            "status_code": result.get("status_code", 200)
+                        }
+                    }
+                    
+                    # 添加额外的元数据
+                    if page_info:
+                        if page_info.date_published:
+                            page_result["metadata"]["date_published"] = page_info.date_published
+                        if page_info.site_name:
+                            page_result["metadata"]["site_name"] = page_info.site_name
+                    
+                    crawled_pages.append(page_result)
+                
+                # 添加小延迟避免请求过快
+                if idx < len(urls_to_crawl) - 1:
+                    time.sleep(0.5)
+            
+            return crawled_pages
+            
+        except Exception as e:
+            logger.error(f"爬取失败: {str(e)}")
+            return []
+    
+    def __enter__(self):
+        """上下文管理器入口"""
+        return self
+    
+    def __exit__(self, exc_type, exc_val, exc_tb):
+        """上下文管理器退出，关闭会话"""
+        self.session.close()
+