autogen_ext.tools.azure#

pydantic 模型 AzureAISearchConfig[source]#

基类: BaseModel

Azure AI 搜索工具的配置。

此类定义了 AzureAISearchTool 的配置参数。它提供了自定义搜索行为的选项，包括查询类型、字段选择、身份验证、重试策略和缓存策略。

注意

此类需要 autogen-ext 包的 azure 额外依赖。

pip install -U "autogen-ext[azure]"

示例

from azure.core.credentials import AzureKeyCredential
from autogen_ext.tools.azure import AzureAISearchConfig

config = AzureAISearchConfig(
    name="doc_search",
    endpoint="https://my-search.search.windows.net",
    index_name="my-index",
    credential=AzureKeyCredential("<your-key>"),
    query_type="vector",
    vector_fields=["embedding"],
)
有关更多详细信息，请参见
参数:

  • name (str) – 工具实例的名称，用于在代理的工具包中标识它。

  • description (Optional[str]) – 此工具的功能以及使用方法的易于理解的描述。

  • endpoint (str) – Azure AI 搜索服务的完整 URL，格式为“https://<service-name>.search.windows.net”。

  • index_name (str) – Azure AI 搜索服务中目标搜索索引的名称。索引必须预先创建并正确配置。

  • api_version (str) – 要使用的 Azure AI 搜索 REST API 版本。默认为“2023-11-01”。只有在需要来自不同 API 版本的特定功能时才更改。

  • credential (Union[AzureKeyCredential, TokenCredential]) – Azure 身份验证凭据：- AzureKeyCredential：用于 API 密钥身份验证（管理/查询密钥）- TokenCredential：用于 Azure AD 身份验证（例如，DefaultAzureCredential）

  • query_type (Literal["keyword", "fulltext", "vector", "semantic"]) – 要使用的搜索查询模式：- “keyword”：基本关键字搜索（默认）- “fulltext”：完整的 Lucene 查询语法- “vector”：向量相似度搜索- “semantic”：使用语义配置的语义搜索

  • search_fields (Optional[List[str]]) – 要在其中搜索的索引字段列表。如果未指定，则搜索所有可搜索字段。示例：[“title”, “content”]。

  • select_fields (Optional[List[str]]) – 要在搜索结果中返回的字段。如果未指定，则返回所有字段。用于优化响应大小。

  • vector_fields (Optional[List[str]]) – 向量搜索的向量字段名称。必须在搜索索引中配置为向量字段。向量搜索必需。

  • top (Optional[int]) – 要在搜索结果中返回的最大文档数。有助于控制响应大小和处理时间。

  • retry_enabled (bool) – 是否为瞬态错误启用重试策略。默认为 True。

  • retry_max_attempts (Optional[int]) – 失败请求的最大重试次数。默认为 3。

  • retry_mode (Literal["fixed", "exponential"]) – 重试退避策略：固定或指数。默认为“exponential”。

  • enable_caching (bool) – 是否启用搜索结果的客户端缓存。默认为 False。

  • cache_ttl_seconds (int) – 缓存的搜索结果的生存时间（以秒为单位）。默认为 300（5 分钟）。

  • filter (Optional[str]) – 用于优化搜索结果的 OData 筛选表达式。

显示 JSON 模式
{
   "title": "AzureAISearchConfig",
   "description": "Configuration for Azure AI Search tool.\n\nThis class defines the configuration parameters for :class:`AzureAISearchTool`.\nIt provides options for customizing search behavior including query types,\nfield selection, authentication, retry policies, and caching strategies.\n\n.. note::\n\n    This class requires the :code:`azure` extra for the :code:`autogen-ext` package.\n\n    .. code-block:: bash\n\n        pip install -U \"autogen-ext[azure]\"\n\nExample:\n    .. code-block:: python\n\n        from azure.core.credentials import AzureKeyCredential\n        from autogen_ext.tools.azure import AzureAISearchConfig\n\n        config = AzureAISearchConfig(\n            name=\"doc_search\",\n            endpoint=\"https://my-search.search.windows.net\",\n            index_name=\"my-index\",\n            credential=AzureKeyCredential(\"<your-key>\"),\n            query_type=\"vector\",\n            vector_fields=[\"embedding\"],\n        )\n\nFor more details, see:\n    * `Azure AI Search Overview <https://learn.microsoft.com/azure/search/search-what-is-azure-search>`_\n    * `Vector Search <https://learn.microsoft.com/azure/search/vector-search-overview>`_\n\nArgs:\n    name (str): Name for the tool instance, used to identify it in the agent's toolkit.\n    description (Optional[str]): Human-readable description of what this tool does and how to use it.\n    endpoint (str): The full URL of your Azure AI Search service, in the format\n        'https://<service-name>.search.windows.net'.\n    index_name (str): Name of the target search index in your Azure AI Search service.\n        The index must be pre-created and properly configured.\n    api_version (str): Azure AI Search REST API version to use. Defaults to '2023-11-01'.\n        Only change if you need specific features from a different API version.\n    credential (Union[AzureKeyCredential, TokenCredential]): Azure authentication credential:\n        - AzureKeyCredential: For API key authentication (admin/query key)\n        - TokenCredential: For Azure AD authentication (e.g., DefaultAzureCredential)\n    query_type (Literal[\"keyword\", \"fulltext\", \"vector\", \"semantic\"]): The search query mode to use:\n        - 'keyword': Basic keyword search (default)\n        - 'fulltext': Full Lucene query syntax\n        - 'vector': Vector similarity search\n        - 'semantic': Semantic search using semantic configuration\n    search_fields (Optional[List[str]]): List of index fields to search within. If not specified,\n        searches all searchable fields. Example: ['title', 'content'].\n    select_fields (Optional[List[str]]): Fields to return in search results. If not specified,\n        returns all fields. Use to optimize response size.\n    vector_fields (Optional[List[str]]): Vector field names for vector search. Must be configured\n        in your search index as vector fields. Required for vector search.\n    top (Optional[int]): Maximum number of documents to return in search results.\n        Helps control response size and processing time.\n    retry_enabled (bool): Whether to enable retry policy for transient errors. Defaults to True.\n    retry_max_attempts (Optional[int]): Maximum number of retry attempts for failed requests. Defaults to 3.\n    retry_mode (Literal[\"fixed\", \"exponential\"]): Retry backoff strategy: fixed or exponential. Defaults to \"exponential\".\n    enable_caching (bool): Whether to enable client-side caching of search results. Defaults to False.\n    cache_ttl_seconds (int): Time-to-live for cached search results in seconds. Defaults to 300 (5 minutes).\n    filter (Optional[str]): OData filter expression to refine search results.",
   "type": "object",
   "properties": {
      "name": {
         "description": "The name of the tool",
         "title": "Name",
         "type": "string"
      },
      "description": {
         "anyOf": [
            {
               "type": "string"
            },
            {
               "type": "null"
            }
         ],
         "default": null,
         "description": "A description of the tool",
         "title": "Description"
      },
      "endpoint": {
         "description": "The endpoint URL for your Azure AI Search service",
         "title": "Endpoint",
         "type": "string"
      },
      "index_name": {
         "description": "The name of the search index to query",
         "title": "Index Name",
         "type": "string"
      },
      "api_version": {
         "default": "2023-11-01",
         "description": "API version to use",
         "title": "Api Version",
         "type": "string"
      },
      "credential": {
         "anyOf": [],
         "description": "The credential to use for authentication",
         "title": "Credential"
      },
      "query_type": {
         "default": "keyword",
         "description": "Type of query to perform (keyword for classic, fulltext for Lucene, vector for embedding, semantic for semantic/AI search)",
         "enum": [
            "keyword",
            "fulltext",
            "vector",
            "semantic"
         ],
         "title": "Query Type",
         "type": "string"
      },
      "search_fields": {
         "anyOf": [
            {
               "items": {
                  "type": "string"
               },
               "type": "array"
            },
            {
               "type": "null"
            }
         ],
         "default": null,
         "description": "Optional list of fields to search in",
         "title": "Search Fields"
      },
      "select_fields": {
         "anyOf": [
            {
               "items": {
                  "type": "string"
               },
               "type": "array"
            },
            {
               "type": "null"
            }
         ],
         "default": null,
         "description": "Optional list of fields to return in results",
         "title": "Select Fields"
      },
      "vector_fields": {
         "anyOf": [
            {
               "items": {
                  "type": "string"
               },
               "type": "array"
            },
            {
               "type": "null"
            }
         ],
         "default": null,
         "description": "Optional list of vector fields for vector search",
         "title": "Vector Fields"
      },
      "top": {
         "anyOf": [
            {
               "type": "integer"
            },
            {
               "type": "null"
            }
         ],
         "default": null,
         "description": "Optional number of results to return",
         "title": "Top"
      },
      "filter": {
         "anyOf": [
            {
               "type": "string"
            },
            {
               "type": "null"
            }
         ],
         "default": null,
         "description": "Optional OData filter expression to refine search results",
         "title": "Filter"
      },
      "retry_enabled": {
         "default": true,
         "description": "Whether to enable retry policy for transient errors",
         "title": "Retry Enabled",
         "type": "boolean"
      },
      "retry_max_attempts": {
         "anyOf": [
            {
               "type": "integer"
            },
            {
               "type": "null"
            }
         ],
         "default": 3,
         "description": "Maximum number of retry attempts for failed requests",
         "title": "Retry Max Attempts"
      },
      "retry_mode": {
         "default": "exponential",
         "description": "Retry backoff strategy: fixed or exponential",
         "enum": [
            "fixed",
            "exponential"
         ],
         "title": "Retry Mode",
         "type": "string"
      },
      "enable_caching": {
         "default": false,
         "description": "Whether to enable client-side caching of search results",
         "title": "Enable Caching",
         "type": "boolean"
      },
      "cache_ttl_seconds": {
         "default": 300,
         "description": "Time-to-live for cached search results in seconds",
         "title": "Cache Ttl Seconds",
         "type": "integer"
      },
      "embedding_provider": {
         "anyOf": [
            {
               "type": "string"
            },
            {
               "type": "null"
            }
         ],
         "default": null,
         "description": "Name of embedding provider to use (e.g., 'azure_openai', 'openai')",
         "title": "Embedding Provider"
      },
      "embedding_model": {
         "anyOf": [
            {
               "type": "string"
            },
            {
               "type": "null"
            }
         ],
         "default": null,
         "description": "Model name to use for generating embeddings",
         "title": "Embedding Model"
      },
      "embedding_dimension": {
         "anyOf": [
            {
               "type": "integer"
            },
            {
               "type": "null"
            }
         ],
         "default": null,
         "description": "Dimension of embedding vectors produced by the model",
         "title": "Embedding Dimension"
      }
   },
   "required": [
      "name",
      "endpoint",
      "index_name",
      "credential"
   ]
}

字段:

  • api_version (str)

  • cache_ttl_seconds (int)

  • credential (azure.core.credentials.AzureKeyCredential | azure.core.credentials.TokenCredential)

  • description (str | None)

  • embedding_dimension (int | None)

  • embedding_model (str | None)

  • embedding_provider (str | None)

  • enable_caching (bool)

  • endpoint (str)

  • filter (str | None)

  • index_name (str)

  • name (str)

  • query_type (Literal['keyword', 'fulltext', 'vector', 'semantic'])

  • retry_enabled (bool)

  • retry_max_attempts (int | None)

  • retry_mode (Literal['fixed', 'exponential'])

  • search_fields (List[str] | None)

  • select_fields (List[str] | None)

  • top (int | None)

  • vector_fields (List[str] | None)

field api_version: str = '2023-11-01'#

要使用的 API 版本

field cache_ttl_seconds: int = 300#

缓存的搜索结果的生存时间（以秒为单位）

field credential: AzureKeyCredential | TokenCredential [Required]#

用于身份验证的凭据

field description: str | None = None#

工具的描述

field embedding_dimension: int | None = None#

模型生成的嵌入向量的维度

field embedding_model: str | None = None#

用于生成嵌入的模型名称

field embedding_provider: str | None = None#

要使用的嵌入提供程序的名称（例如，“azure_openai”、“openai”）

field enable_caching: bool = False#

是否启用搜索结果的客户端缓存

field endpoint: str [Required]#

Azure AI 搜索服务的终结点 URL

field filter: str | None = None#

可选的 OData 筛选器表达式，用于优化搜索结果

field index_name: str [Required]#

要查询的搜索索引的名称

field name: str [Required]#

工具的名称

field query_type: Literal['keyword', 'fulltext', 'vector', 'semantic'] = 'keyword'#

要执行的查询类型（keyword 代表经典查询，fulltext 代表 Lucene 查询，vector 代表嵌入查询，semantic 代表语义/AI 搜索）

field retry_enabled: bool = True#

是否为瞬态错误启用重试策略

field retry_max_attempts: int | None = 3#

失败请求的最大重试次数

field retry_mode: Literal['fixed', 'exponential'] = 'exponential'#

重试退避策略：fixed (固定) 或 exponential (指数)

field search_fields: List[str] | None = None#

要在其中搜索的可选字段列表

field select_fields: List[str] | None = None#

要在结果中返回的可选字段列表

field top: int | None = None#

要返回的可选结果数

field vector_fields: List[str] | None = None#

用于向量搜索的可选向量字段列表

model_dump(**kwargs: Any) → Dict[str, Any][source]#

自定义 model_dump 以处理凭据。

classmethod validate_credentials(data: Any) → Any[source]#

封装一个类方法、静态方法、属性或未绑定的函数，并充当一个描述符，允许我们从类的属性中检测装饰的项。

这个类的 __get__ 返回被封装项的 __get__ 结果，这使得它对于类方法和静态方法是透明的。

wrapped#

需要被封装的装饰器。

decorator_info#

装饰器信息。

shim#

用于封装 V1 风格函数的包装函数。

class AzureAISearchTool(name: str, endpoint: str, index_name: str, credential: AzureKeyCredential | TokenCredential | Dict[str, str], query_type: Literal['keyword', 'fulltext', 'vector', 'semantic'], search_fields: List[str] | None = None, select_fields: List[str] | None = None, vector_fields: List[str] | None = None, filter: str | None = None, top: int | None = 5, **kwargs: Any)[source]#

基类: BaseAzureAISearchTool

用于查询 Azure 搜索索引的 Azure AI 搜索工具。

该工具提供了一个简化的界面，用于使用各种搜索方法查询 Azure AI 搜索索引。 该工具支持四种主要的搜索类型

  1. 关键词搜索：使用 Azure 文本分析的传统文本搜索

  2. 全文搜索：使用特定于语言的分析器的增强型文本搜索

  3. 向量搜索：使用向量嵌入的语义相似性搜索

  4. 混合搜索：结合了全文和向量搜索，以获得全面的结果

你应该使用工厂方法来创建特定搜索类型的实例： - create_keyword_search() - create_full_text_search() - create_vector_search() - create_hybrid_search()

classmethod create_full_text_search(name: str, endpoint: str, index_name: str, credential: AzureKeyCredential | TokenCredential | Dict[str, str], search_fields: List[str] | None = None, select_fields: List[str] | None = None, filter: str | None = None, top: int | None = 5, **kwargs: Any) → AzureAISearchTool[source]#

工厂方法，用于创建全文搜索工具。

全文搜索使用高级文本分析（词干提取、词形还原等）来提供比基本关键字搜索更全面的文本匹配。

参数:

  • name (str) – 工具的名称

  • endpoint (str) – 你的 Azure AI 搜索服务的 URL

  • index_name (str) – 搜索索引的名称

  • credential (Union[AzureKeyCredential, TokenCredential, Dict[str, str]]) – 身份验证凭据

  • search_fields (Optional[List[str]]) – 要在其中搜索的字段

  • select_fields (Optional[List[str]]) – 要包含在结果中的字段

  • filter (Optional[str]) – 用于过滤结果的 OData 筛选表达式

  • top (Optional[int]) – 要返回的最大结果数

  • **kwargs (Any) – 其他配置选项

返回:

一个已初始化的全文搜索工具

使用示例
# type: ignore
# Example of using full-text search with Azure AI Search
from autogen_ext.tools.azure import AzureAISearchTool
from azure.core.credentials import AzureKeyCredential

# Create a full-text search tool
full_text_search = AzureAISearchTool.create_full_text_search(
    name="document_search",
    endpoint="https://your-search-service.search.windows.net",
    index_name="your-index",
    credential=AzureKeyCredential("your-api-key"),
    search_fields=["title", "content"],
    select_fields=["title", "content", "category", "url"],
    top=10,
)

# The search tool can be used with an Agent
# assistant = Agent("assistant", tools=[full_text_search])

工厂方法，用于创建混合搜索工具（文本 + 向量）。

混合搜索将文本搜索（全文或语义）与向量相似性搜索相结合，以提供更全面的结果。 这是混合（文本 + 向量）搜索的推荐入口点。如果提供了 semantic_config_name，则 query_type 将为“semantic”，否则为“fulltext”。

参数:

  • name (str) – 工具的名称

  • endpoint (str) – 你的 Azure AI 搜索服务的 URL

  • index_name (str) – 搜索索引的名称

  • credential (Union[AzureKeyCredential, TokenCredential, Dict[str, str]]) – 身份验证凭据

  • vector_fields (List[str]) – 包含用于相似性搜索的向量嵌入的字段

  • search_fields (Optional[List[str]]) – 要在其中进行文本搜索的字段

  • select_fields (Optional[List[str]]) – 要包含在结果中的字段

  • filter (Optional[str]) – 用于过滤结果的 OData 筛选表达式

  • top (Optional[int]) – 要返回的最大结果数

  • **kwargs (Any) – 其他配置选项

返回:

一个已初始化的混合搜索工具

使用示例
# type: ignore
# Example of using hybrid search with Azure AI Search
from autogen_ext.tools.azure import AzureAISearchTool
from azure.core.credentials import AzureKeyCredential

# Create a hybrid search tool
hybrid_search = AzureAISearchTool.create_hybrid_search(
    name="hybrid_search",
    endpoint="https://your-search-service.search.windows.net",
    index_name="your-index",
    credential=AzureKeyCredential("your-api-key"),
    vector_fields=["embedding_field"],
    search_fields=["title", "content"],
    select_fields=["title", "content", "url", "date"],
    top=10,
)

# The search tool can be used with an Agent
# assistant = Agent("researcher", tools=[hybrid_search])

警告

如果您设置 query_type="semantic"，您还必须提供有效的 semantic_config_name。 如果没有，该工具将默认为配置名称 "semantic"。

用于创建关键词搜索工具的工厂方法。

关键词搜索执行传统的基于文本的搜索，适合查找包含特定术语或与查询完全匹配的文档。

参数:

  • name (str) – 工具的名称

  • endpoint (str) – 你的 Azure AI 搜索服务的 URL

  • index_name (str) – 搜索索引的名称

  • credential (Union[AzureKeyCredential, TokenCredential, Dict[str, str]]) – 身份验证凭据

  • search_fields (Optional[List[str]]) – 要在其中进行文本搜索的字段

  • select_fields (Optional[List[str]]) – 要包含在结果中的字段

  • filter (Optional[str]) – 用于过滤结果的 OData 筛选表达式

  • top (Optional[int]) – 要返回的最大结果数

  • **kwargs (Any) – 其他配置选项

返回:

一个已初始化的关键词搜索工具

使用示例
# type: ignore
# Example of using keyword search with Azure AI Search
from autogen_ext.tools.azure import AzureAISearchTool
from azure.core.credentials import AzureKeyCredential

# Create a keyword search tool
keyword_search = AzureAISearchTool.create_keyword_search(
    name="keyword_search",
    endpoint="https://your-service.search.windows.net",
    index_name="your-index",
    credential=AzureKeyCredential("your-api-key"),
    search_fields=["title", "content"],
    select_fields=["id", "title", "content", "category"],
    top=10,
)

# The search tool can be used with an Agent
# assistant = Agent("assistant", tools=[keyword_search])

用于创建向量搜索工具的工厂方法。

向量搜索使用嵌入向量来查找语义相似的内容，即使使用不同的术语，也能发现相关信息。

参数:

  • name (str) – 工具的名称

  • endpoint (str) – 你的 Azure AI 搜索服务的 URL

  • index_name (str) – 搜索索引的名称

  • credential (Union[AzureKeyCredential, TokenCredential, Dict[str, str]]) – 身份验证凭据

  • vector_fields (List[str]) – 包含用于相似性搜索的向量嵌入的字段

  • select_fields (Optional[List[str]]) – 要包含在结果中的字段

  • filter (Optional[str]) – 用于过滤结果的 OData 筛选表达式

  • top (Optional[int]) – 要返回的最大结果数

  • **kwargs (Any) – 其他配置选项

返回:

一个已初始化的向量搜索工具

使用示例
# type: ignore
# Example of using vector search with Azure AI Search
from autogen_ext.tools.azure import AzureAISearchTool
from azure.core.credentials import AzureKeyCredential

# Create a vector search tool
vector_search = AzureAISearchTool.create_vector_search(
    name="vector_search",
    endpoint="https://your-search-service.search.windows.net",
    index_name="your-index",
    credential=AzureKeyCredential("your-api-key"),
    vector_fields=["embedding"],
    select_fields=["title", "content", "url"],
    top=5,
)

# The search tool can be used with an Agent
# assistant = Agent("assistant", tools=[vector_search])
classmethod load_component(model: ComponentModel | Dict[str, Any], expected: None = None) → AzureAISearchTool[source]#
classmethod load_component(model: ComponentModel | Dict[str, Any], expected: Type[ExpectedType]) → ExpectedType

从组件模型加载组件。

参数:

  • model – 组件模型或包含配置的字典

  • expected – 可选的预期返回类型

返回:

已初始化的 AzureAISearchTool 实例

class BaseAzureAISearchTool(name: str, endpoint: str, index_name: str, credential: AzureKeyCredential | TokenCredential | Dict[str, str], description: str | None = None, api_version: str = '2023-11-01', query_type: Literal['keyword', 'fulltext', 'vector', 'semantic'] = 'keyword', search_fields: List[str] | None = None, select_fields: List[str] | None = None, vector_fields: List[str] | None = None, top: int | None = None, filter: str | None = None, semantic_config_name: str | None = None, enable_caching: bool = False, cache_ttl_seconds: int = 300)[source]#

基类：BaseTool[SearchQuery, SearchResults], ABC

Azure AI 搜索工具的抽象基类。

此类定义了所有 Azure AI 搜索工具的通用接口和功能。它处理配置管理、客户端初始化以及子类必须实现的抽象方法。

search_config#

搜索服务的配置参数。

注意

这是一个抽象基类，不应直接实例化。 使用 AzureAISearchTool 中的具体实现或工厂方法。

async close() → None[source]#

如果需要，显式关闭 Azure SearchClient（用于长时间运行的应用程序/测试中的清理）。

dump_component() → ComponentModel[source]#

将工具序列化为组件模型。

返回:

ComponentModel – 工具的序列化表示

classmethod load_component(model: ComponentModel | Dict[str, Any], expected: None = None) → BaseAzureAISearchTool[source]#
classmethod load_component(model: ComponentModel | Dict[str, Any], expected: Type[ExpectedType]) → ExpectedType

从组件模型加载工具。

参数:

  • model (Union[ComponentModel, Dict[str, Any]]) – 组件配置。

  • expected (Optional[Type[ExpectedType]]) – 用于反序列化的可选组件类。

返回:

Union[BaseAzureAISearchTool, ExpectedType] – 工具的一个实例。

引发:

ValueError – 如果组件配置无效。

return_value_as_string(value: SearchResults) → str[source]#

将搜索结果转换为字符串表示形式。

此方法用于格式化搜索结果，使其适合向用户显示或供语言模型使用。

参数:

value (List[SearchResult]) – 要转换的搜索结果。

返回:

str – 搜索结果的格式化字符串表示形式。

async run(args: str | Dict[str, Any] | SearchQuery, cancellation_token: CancellationToken | None = None) → SearchResults[source]#

针对 Azure AI Search 索引执行搜索。

参数:

  • args – 搜索查询文本或 SearchQuery 对象

  • cancellation_token – 用于取消操作的可选令牌

返回:

搜索结果

property schema: ToolSchema#

返回工具的 schema。

pydantic model SearchQuery[source]#

基类: BaseModel

搜索查询参数。

这种简化的界面只需要一个搜索查询字符串。所有其他参数（top、filters、vector fields 等）都在工具创建期间指定，而不是在查询时指定，从而使语言模型更容易生成结构化输出。

参数:

query (str) – 搜索查询文本。

显示 JSON 模式
{
   "title": "SearchQuery",
   "description": "Search query parameters.\n\nThis simplified interface only requires a search query string.\nAll other parameters (top, filters, vector fields, etc.) are specified during tool creation\nrather than at query time, making it easier for language models to generate structured output.\n\nArgs:\n    query (str): The search query text.",
   "type": "object",
   "properties": {
      "query": {
         "description": "Search query text",
         "title": "Query",
         "type": "string"
      }
   },
   "required": [
      "query"
   ]
}

字段:

  • query (str)

field query: str [Required]#

搜索查询文本

pydantic model SearchResult[source]#

基类: BaseModel

搜索结果。

参数:

  • score (float) – 搜索分数。

  • content (Dict[str, Any]) – 文档内容。

  • metadata (Dict[str, Any]) – 关于文档的附加元数据。

显示 JSON 模式
{
   "title": "SearchResult",
   "description": "Search result.\n\nArgs:\n    score (float): The search score.\n    content (Dict[str, Any]): The document content.\n    metadata (Dict[str, Any]): Additional metadata about the document.",
   "type": "object",
   "properties": {
      "score": {
         "description": "The search score",
         "title": "Score",
         "type": "number"
      },
      "content": {
         "description": "The document content",
         "title": "Content",
         "type": "object"
      },
      "metadata": {
         "description": "Additional metadata about the document",
         "title": "Metadata",
         "type": "object"
      }
   },
   "required": [
      "score",
      "content",
      "metadata"
   ]
}

字段:

  • content (Dict[str, Any])

  • metadata (Dict[str, Any])

  • score (float)

field content: Dict[str, Any] [Required]#

文档内容

field metadata: Dict[str, Any] [Required]#

关于文档的附加元数据

field score: float [Required]#

搜索分数

pydantic model SearchResults[source]#

基类: BaseModel

搜索结果的容器。

参数:

results (List[SearchResult]) – 搜索结果列表。

显示 JSON 模式
{
   "title": "SearchResults",
   "description": "Container for search results.\n\nArgs:\n    results (List[SearchResult]): List of search results.",
   "type": "object",
   "properties": {
      "results": {
         "description": "List of search results",
         "items": {
            "$ref": "#/$defs/SearchResult"
         },
         "title": "Results",
         "type": "array"
      }
   },
   "$defs": {
      "SearchResult": {
         "description": "Search result.\n\nArgs:\n    score (float): The search score.\n    content (Dict[str, Any]): The document content.\n    metadata (Dict[str, Any]): Additional metadata about the document.",
         "properties": {
            "score": {
               "description": "The search score",
               "title": "Score",
               "type": "number"
            },
            "content": {
               "description": "The document content",
               "title": "Content",
               "type": "object"
            },
            "metadata": {
               "description": "Additional metadata about the document",
               "title": "Metadata",
               "type": "object"
            }
         },
         "required": [
            "score",
            "content",
            "metadata"
         ],
         "title": "SearchResult",
         "type": "object"
      }
   },
   "required": [
      "results"
   ]
}

字段:

  • results (List[autogen_ext.tools.azure._ai_search.SearchResult])

field results: List[SearchResult] [Required]#

搜索结果列表

本页内容
在 GitHub 上编辑
显示源代码