agentrun-inner-test 0.0.62__py3-none-any.whl

agentrun/knowledgebase/api/data.py

@@ -0,0 +1,624 @@
+"""
+This file is auto generated by the code generation script.
+Do not modify this file manually.
+Use the `make codegen` command to regenerate.
+
+当前文件为自动生成的控制 API 客户端代码。请勿手动修改此文件。
+使用 `make codegen` 命令重新生成。
+
+source: agentrun/knowledgebase/api/__data_async_template.py
+
+KnowledgeBase 数据链路 API / KnowledgeBase Data API
+
+提供知识库检索功能的数据链路 API。
+Provides data API for knowledge base retrieval operations.
+
+根据不同的 provider 类型（ragflow / bailian）分发到不同的实现。
+Dispatches to different implementations based on provider type (ragflow / bailian).
+"""
+
+from abc import ABC, abstractmethod
+from typing import Any, Dict, List, Optional, Union
+
+from alibabacloud_bailian20231229 import models as bailian_models
+import httpx
+
+from agentrun.utils.config import Config
+from agentrun.utils.control_api import ControlAPI
+from agentrun.utils.data_api import DataAPI, ResourceType
+from agentrun.utils.log import logger
+
+from ..model import (
+    BailianProviderSettings,
+    BailianRetrieveSettings,
+    KnowledgeBaseProvider,
+    RagFlowProviderSettings,
+    RagFlowRetrieveSettings,
+)
+
+
+class KnowledgeBaseDataAPI(ABC):
+    """知识库数据链路 API 基类 / KnowledgeBase Data API Base Class
+
+    定义知识库检索的抽象接口，由具体的 provider 实现。
+    Defines abstract interface for knowledge base retrieval, implemented by specific providers.
+    """
+
+    def __init__(
+        self,
+        knowledge_base_name: str,
+        config: Optional[Config] = None,
+    ):
+        """初始化知识库数据链路 API / Initialize KnowledgeBase Data API
+
+        Args:
+            knowledge_base_name: 知识库名称 / Knowledge base name
+            config: 配置 / Configuration
+        """
+        self.knowledge_base_name = knowledge_base_name
+        self.config = Config.with_configs(config)
+
+    @abstractmethod
+    async def retrieve_async(
+        self,
+        query: str,
+        config: Optional[Config] = None,
+    ) -> Dict[str, Any]:
+        """检索知识库（异步）/ Retrieve from knowledge base (async)
+
+        Args:
+            query: 查询文本 / Query text
+            config: 配置 / Configuration
+
+        Returns:
+            Dict[str, Any]: 检索结果 / Retrieval results
+        """
+        raise NotImplementedError("Subclasses must implement retrieve_async")
+
+    @abstractmethod
+    def retrieve(
+        self,
+        query: str,
+        config: Optional[Config] = None,
+    ) -> Dict[str, Any]:
+        """检索知识库（同步）/ Retrieve from knowledge base (async)
+
+        Args:
+            query: 查询文本 / Query text
+            config: 配置 / Configuration
+
+        Returns:
+            Dict[str, Any]: 检索结果 / Retrieval results
+        """
+        raise NotImplementedError("Subclasses must implement retrieve")
+
+
+class RagFlowDataAPI(KnowledgeBaseDataAPI):
+    """RagFlow 知识库数据链路 API / RagFlow KnowledgeBase Data API
+
+    实现 RagFlow 知识库的检索逻辑。
+    Implements retrieval logic for RagFlow knowledge base.
+    """
+
+    def __init__(
+        self,
+        knowledge_base_name: str,
+        config: Optional[Config] = None,
+        provider_settings: Optional[RagFlowProviderSettings] = None,
+        retrieve_settings: Optional[RagFlowRetrieveSettings] = None,
+        credential_name: Optional[str] = None,
+    ):
+        """初始化 RagFlow 知识库数据链路 API / Initialize RagFlow KnowledgeBase Data API
+
+        Args:
+            knowledge_base_name: 知识库名称 / Knowledge base name
+            config: 配置 / Configuration
+            provider_settings: RagFlow 提供商设置 / RagFlow provider settings
+            retrieve_settings: RagFlow 检索设置 / RagFlow retrieve settings
+            credential_name: 凭证名称 / Credential name
+        """
+        super().__init__(knowledge_base_name, config)
+        self.provider_settings = provider_settings
+        self.retrieve_settings = retrieve_settings
+        self.credential_name = credential_name
+
+    async def _get_api_key_async(self, config: Optional[Config] = None) -> str:
+        """获取 API Key（异步）/ Get API Key (async)
+
+        Args:
+            config: 配置 / Configuration
+
+        Returns:
+            str: API Key
+
+        Raises:
+            ValueError: 凭证名称未设置或凭证不存在 / Credential name not set or credential not found
+        """
+        if not self.credential_name:
+            raise ValueError(
+                "credential_name is required for RagFlow retrieval"
+            )
+
+        from agentrun.credential import Credential
+
+        credential = await Credential.get_by_name_async(
+            self.credential_name, config=config
+        )
+        if not credential.credential_secret:
+            raise ValueError(
+                f"Credential '{self.credential_name}' has no secret configured"
+            )
+        return credential.credential_secret
+
+    def _get_api_key(self, config: Optional[Config] = None) -> str:
+        """获取 API Key（同步）/ Get API Key (async)
+
+        Args:
+            config: 配置 / Configuration
+
+        Returns:
+            str: API Key
+
+        Raises:
+            ValueError: 凭证名称未设置或凭证不存在 / Credential name not set or credential not found
+        """
+        if not self.credential_name:
+            raise ValueError(
+                "credential_name is required for RagFlow retrieval"
+            )
+
+        from agentrun.credential import Credential
+
+        credential = Credential.get_by_name(self.credential_name, config=config)
+        if not credential.credential_secret:
+            raise ValueError(
+                f"Credential '{self.credential_name}' has no secret configured"
+            )
+        return credential.credential_secret
+
+    def _build_request_body(self, query: str) -> Dict[str, Any]:
+        """构建请求体 / Build request body
+
+        Args:
+            query: 查询文本 / Query text
+
+        Returns:
+            Dict[str, Any]: 请求体 / Request body
+        """
+        if self.provider_settings is None:
+            raise ValueError(
+                "provider_settings is required for RagFlow retrieval"
+            )
+
+        body: Dict[str, Any] = {
+            "question": query,
+            "dataset_ids": self.provider_settings.dataset_ids,
+            "page": 1,
+            "page_size": 30,
+        }
+
+        # 添加检索设置 / Add retrieve settings
+        if self.retrieve_settings:
+            if self.retrieve_settings.similarity_threshold is not None:
+                body["similarity_threshold"] = (
+                    self.retrieve_settings.similarity_threshold
+                )
+            if self.retrieve_settings.vector_similarity_weight is not None:
+                body["vector_similarity_weight"] = (
+                    self.retrieve_settings.vector_similarity_weight
+                )
+            if self.retrieve_settings.cross_languages is not None:
+                body["cross_languages"] = self.retrieve_settings.cross_languages
+
+        return body
+
+    async def retrieve_async(
+        self,
+        query: str,
+        config: Optional[Config] = None,
+    ) -> Dict[str, Any]:
+        """RagFlow 检索（异步）/ RagFlow retrieval (async)
+
+        Args:
+            query: 查询文本 / Query text
+            config: 配置 / Configuration
+
+        Returns:
+            Dict[str, Any]: 检索结果 / Retrieval results
+        """
+        try:
+            if self.provider_settings is None:
+                raise ValueError(
+                    "provider_settings is required for RagFlow retrieval"
+                )
+
+            # 获取 API Key / Get API Key
+            api_key = await self._get_api_key_async(config)
+
+            # 构建请求 / Build request
+            base_url = self.provider_settings.base_url.rstrip("/")
+            url = f"{base_url}/api/v1/retrieval"
+            headers = {
+                "Content-Type": "application/json",
+                "Authorization": f"Bearer {api_key}",
+            }
+            body = self._build_request_body(query)
+
+            # 发送请求 / Send request
+            async with httpx.AsyncClient(
+                timeout=self.config.get_timeout()
+            ) as client:
+                response = await client.post(url, json=body, headers=headers)
+                response.raise_for_status()
+                result = response.json()
+                logger.debug(f"RagFlow retrieval result: {result}")
+
+            # 返回结果 / Return result
+            data = result.get("data", {})
+
+            if data == False:
+                raise Exception(f"RagFlow retrieval failed: {result}")
+
+            return {
+                "data": data,
+                "query": query,
+                "knowledge_base_name": self.knowledge_base_name,
+            }
+        except Exception as e:
+            logger.warning(
+                "Failed to retrieve from RagFlow knowledge base "
+                f"'{self.knowledge_base_name}': {e}"
+            )
+            return {
+                "data": f"Failed to retrieve: {e}",
+                "query": query,
+                "knowledge_base_name": self.knowledge_base_name,
+                "error": True,
+            }
+
+    def retrieve(
+        self,
+        query: str,
+        config: Optional[Config] = None,
+    ) -> Dict[str, Any]:
+        """RagFlow 检索（同步）/ RagFlow retrieval (async)
+
+        Args:
+            query: 查询文本 / Query text
+            config: 配置 / Configuration
+
+        Returns:
+            Dict[str, Any]: 检索结果 / Retrieval results
+        """
+        try:
+            if self.provider_settings is None:
+                raise ValueError(
+                    "provider_settings is required for RagFlow retrieval"
+                )
+
+            # 获取 API Key / Get API Key
+            api_key = self._get_api_key(config)
+
+            # 构建请求 / Build request
+            base_url = self.provider_settings.base_url.rstrip("/")
+            url = f"{base_url}/api/v1/retrieval"
+            headers = {
+                "Content-Type": "application/json",
+                "Authorization": f"Bearer {api_key}",
+            }
+            body = self._build_request_body(query)
+
+            # 发送请求 / Send request
+            with httpx.Client(timeout=self.config.get_timeout()) as client:
+                response = client.post(url, json=body, headers=headers)
+                response.raise_for_status()
+                result = response.json()
+                logger.debug(f"RagFlow retrieval result: {result}")
+
+            # 返回结果 / Return result
+            data = result.get("data", {})
+
+            if data == False:
+                raise Exception(f"RagFlow retrieval failed: {result}")
+
+            return {
+                "data": data,
+                "query": query,
+                "knowledge_base_name": self.knowledge_base_name,
+            }
+        except Exception as e:
+            logger.warning(
+                "Failed to retrieve from RagFlow knowledge base "
+                f"'{self.knowledge_base_name}': {e}"
+            )
+            return {
+                "data": f"Failed to retrieve: {e}",
+                "query": query,
+                "knowledge_base_name": self.knowledge_base_name,
+                "error": True,
+            }
+
+
+class BailianDataAPI(KnowledgeBaseDataAPI, ControlAPI):
+    """百炼知识库数据链路 API / Bailian KnowledgeBase Data API
+
+    实现百炼知识库的检索逻辑。
+    Implements retrieval logic for Bailian knowledge base.
+    """
+
+    def __init__(
+        self,
+        knowledge_base_name: str,
+        config: Optional[Config] = None,
+        provider_settings: Optional[BailianProviderSettings] = None,
+        retrieve_settings: Optional[BailianRetrieveSettings] = None,
+    ):
+        """初始化百炼知识库数据链路 API / Initialize Bailian KnowledgeBase Data API
+
+        Args:
+            knowledge_base_name: 知识库名称 / Knowledge base name
+            config: 配置 / Configuration
+            provider_settings: 百炼提供商设置 / Bailian provider settings
+            retrieve_settings: 百炼检索设置 / Bailian retrieve settings
+        """
+        KnowledgeBaseDataAPI.__init__(self, knowledge_base_name, config)
+        ControlAPI.__init__(self, config)
+        self.provider_settings = provider_settings
+        self.retrieve_settings = retrieve_settings
+
+    async def retrieve_async(
+        self,
+        query: str,
+        config: Optional[Config] = None,
+    ) -> Dict[str, Any]:
+        """百炼检索（异步）/ Bailian retrieval (async)
+
+        Args:
+            query: 查询文本 / Query text
+            config: 配置 / Configuration
+
+        Returns:
+            Dict[str, Any]: 检索结果 / Retrieval results
+        """
+        try:
+            if self.provider_settings is None:
+                raise ValueError(
+                    "provider_settings is required for Bailian retrieval"
+                )
+
+            workspace_id = self.provider_settings.workspace_id
+            index_ids = self.provider_settings.index_ids
+
+            # 构建检索请求 / Build retrieve request
+            request_params: Dict[str, Any] = {
+                "query": query,
+            }
+
+            # 添加检索设置 / Add retrieve settings
+            if self.retrieve_settings:
+                if self.retrieve_settings.dense_similarity_top_k is not None:
+                    request_params["dense_similarity_top_k"] = (
+                        self.retrieve_settings.dense_similarity_top_k
+                    )
+                if self.retrieve_settings.sparse_similarity_top_k is not None:
+                    request_params["sparse_similarity_top_k"] = (
+                        self.retrieve_settings.sparse_similarity_top_k
+                    )
+                if self.retrieve_settings.rerank_min_score is not None:
+                    request_params["rerank_min_score"] = (
+                        self.retrieve_settings.rerank_min_score
+                    )
+                if self.retrieve_settings.rerank_top_n is not None:
+                    request_params["rerank_top_n"] = (
+                        self.retrieve_settings.rerank_top_n
+                    )
+
+            # 获取百炼客户端 / Get Bailian client
+            client = self._get_bailian_client(config)
+
+            # 对每个 index_id 进行检索并合并结果 / Retrieve from each index and merge results
+            all_nodes: List[Dict[str, Any]] = []
+            for index_id in index_ids:
+                request_params["index_id"] = index_id
+                request = bailian_models.RetrieveRequest(**request_params)
+                response = await client.retrieve_async(workspace_id, request)
+                logger.debug(f"Bailian retrieve response: {response}")
+
+                if (
+                    response.body
+                    and response.body.data
+                    and response.body.data.nodes
+                ):
+                    for node in response.body.data.nodes:
+                        all_nodes.append({
+                            "text": (
+                                node.text if hasattr(node, "text") else None
+                            ),
+                            "score": (
+                                node.score if hasattr(node, "score") else None
+                            ),
+                            "metadata": (
+                                node.metadata
+                                if hasattr(node, "metadata")
+                                else None
+                            ),
+                        })
+
+            return {
+                "data": all_nodes,
+                "query": query,
+                "knowledge_base_name": self.knowledge_base_name,
+            }
+        except Exception as e:
+            logger.warning(
+                "Failed to retrieve from Bailian knowledge base "
+                f"'{self.knowledge_base_name}': {e}"
+            )
+            return {
+                "data": f"Failed to retrieve: {e}",
+                "query": query,
+                "knowledge_base_name": self.knowledge_base_name,
+                "error": True,
+            }
+
+    def retrieve(
+        self,
+        query: str,
+        config: Optional[Config] = None,
+    ) -> Dict[str, Any]:
+        """百炼检索（同步）/ Bailian retrieval (async)
+
+        Args:
+            query: 查询文本 / Query text
+            config: 配置 / Configuration
+
+        Returns:
+            Dict[str, Any]: 检索结果 / Retrieval results
+        """
+        try:
+            if self.provider_settings is None:
+                raise ValueError(
+                    "provider_settings is required for Bailian retrieval"
+                )
+
+            workspace_id = self.provider_settings.workspace_id
+            index_ids = self.provider_settings.index_ids
+
+            # 构建检索请求 / Build retrieve request
+            request_params: Dict[str, Any] = {
+                "query": query,
+            }
+
+            # 添加检索设置 / Add retrieve settings
+            if self.retrieve_settings:
+                if self.retrieve_settings.dense_similarity_top_k is not None:
+                    request_params["dense_similarity_top_k"] = (
+                        self.retrieve_settings.dense_similarity_top_k
+                    )
+                if self.retrieve_settings.sparse_similarity_top_k is not None:
+                    request_params["sparse_similarity_top_k"] = (
+                        self.retrieve_settings.sparse_similarity_top_k
+                    )
+                if self.retrieve_settings.rerank_min_score is not None:
+                    request_params["rerank_min_score"] = (
+                        self.retrieve_settings.rerank_min_score
+                    )
+                if self.retrieve_settings.rerank_top_n is not None:
+                    request_params["rerank_top_n"] = (
+                        self.retrieve_settings.rerank_top_n
+                    )
+
+            # 获取百炼客户端 / Get Bailian client
+            client = self._get_bailian_client(config)
+
+            # 对每个 index_id 进行检索并合并结果 / Retrieve from each index and merge results
+            all_nodes: List[Dict[str, Any]] = []
+            for index_id in index_ids:
+                request_params["index_id"] = index_id
+                request = bailian_models.RetrieveRequest(**request_params)
+                response = client.retrieve(workspace_id, request)
+                logger.debug(f"Bailian retrieve response: {response}")
+
+                if (
+                    response.body
+                    and response.body.data
+                    and response.body.data.nodes
+                ):
+                    for node in response.body.data.nodes:
+                        all_nodes.append({
+                            "text": (
+                                node.text if hasattr(node, "text") else None
+                            ),
+                            "score": (
+                                node.score if hasattr(node, "score") else None
+                            ),
+                            "metadata": (
+                                node.metadata
+                                if hasattr(node, "metadata")
+                                else None
+                            ),
+                        })
+
+            return {
+                "data": all_nodes,
+                "query": query,
+                "knowledge_base_name": self.knowledge_base_name,
+            }
+        except Exception as e:
+            logger.warning(
+                "Failed to retrieve from Bailian knowledge base "
+                f"'{self.knowledge_base_name}': {e}"
+            )
+            return {
+                "data": f"Failed to retrieve: {e}",
+                "query": query,
+                "knowledge_base_name": self.knowledge_base_name,
+                "error": True,
+            }
+
+
+def get_data_api(
+    provider: KnowledgeBaseProvider,
+    knowledge_base_name: str,
+    config: Optional[Config] = None,
+    provider_settings: Optional[
+        Union[RagFlowProviderSettings, BailianProviderSettings]
+    ] = None,
+    retrieve_settings: Optional[
+        Union[RagFlowRetrieveSettings, BailianRetrieveSettings]
+    ] = None,
+    credential_name: Optional[str] = None,
+) -> KnowledgeBaseDataAPI:
+    """根据 provider 类型获取对应的数据链路 API / Get data API by provider type
+
+    Args:
+        provider: 提供商类型 / Provider type
+        knowledge_base_name: 知识库名称 / Knowledge base name
+        config: 配置 / Configuration
+        provider_settings: 提供商设置 / Provider settings
+        retrieve_settings: 检索设置 / Retrieve settings
+        credential_name: 凭证名称（RagFlow 需要）/ Credential name (required for RagFlow)
+
+    Returns:
+        KnowledgeBaseDataAPI: 对应的数据链路 API 实例 / Corresponding data API instance
+
+    Raises:
+        ValueError: 不支持的 provider 类型 / Unsupported provider type
+    """
+    if provider == KnowledgeBaseProvider.RAGFLOW or provider == "ragflow":
+        ragflow_provider_settings = (
+            provider_settings
+            if isinstance(provider_settings, RagFlowProviderSettings)
+            else None
+        )
+        ragflow_retrieve_settings = (
+            retrieve_settings
+            if isinstance(retrieve_settings, RagFlowRetrieveSettings)
+            else None
+        )
+        return RagFlowDataAPI(
+            knowledge_base_name,
+            config,
+            provider_settings=ragflow_provider_settings,
+            retrieve_settings=ragflow_retrieve_settings,
+            credential_name=credential_name,
+        )
+    elif provider == KnowledgeBaseProvider.BAILIAN or provider == "bailian":
+        bailian_provider_settings = (
+            provider_settings
+            if isinstance(provider_settings, BailianProviderSettings)
+            else None
+        )
+        bailian_retrieve_settings = (
+            retrieve_settings
+            if isinstance(retrieve_settings, BailianRetrieveSettings)
+            else None
+        )
+        return BailianDataAPI(
+            knowledge_base_name,
+            config,
+            provider_settings=bailian_provider_settings,
+            retrieve_settings=bailian_retrieve_settings,
+        )
+    else:
+        raise ValueError(f"Unsupported provider type: {provider}")