usecortex-ai 0.2.0__py3-none-any.whl → 0.2.2__py3-none-any.whl

usecortex_ai/document/raw_client.py

@@ -0,0 +1,312 @@
+# This file was auto-generated by Fern from our API Definition.
+
+import typing
+from json.decoder import JSONDecodeError
+
+from ..core.api_error import ApiError
+from ..core.client_wrapper import AsyncClientWrapper, SyncClientWrapper
+from ..core.http_response import AsyncHttpResponse, HttpResponse
+from ..core.pydantic_utilities import parse_obj_as
+from ..core.request_options import RequestOptions
+from ..errors.bad_request_error import BadRequestError
+from ..errors.forbidden_error import ForbiddenError
+from ..errors.internal_server_error import InternalServerError
+from ..errors.not_found_error import NotFoundError
+from ..errors.service_unavailable_error import ServiceUnavailableError
+from ..errors.unauthorized_error import UnauthorizedError
+from ..errors.unprocessable_entity_error import UnprocessableEntityError
+from ..types.actual_error_response import ActualErrorResponse
+from ..types.delete_sources import DeleteSources
+
+# this is used as the default value for optional parameters
+OMIT = typing.cast(typing.Any, ...)
+
+
+class RawDocumentClient:
+    def __init__(self, *, client_wrapper: SyncClientWrapper):
+        self._client_wrapper = client_wrapper
+
+    def delete(
+        self,
+        *,
+        tenant_id: str,
+        source_ids: typing.Sequence[str],
+        sub_tenant_id: typing.Optional[str] = OMIT,
+        request_options: typing.Optional[RequestOptions] = None,
+    ) -> HttpResponse[DeleteSources]:
+        """
+        Remove documents and content from your knowledge base.
+
+        This endpoint permanently deletes the specified sources from your knowledge base. Once deleted, the content will no longer be available for search or retrieval.
+
+        Use this carefully as the action cannot be undone. The system will confirm successful deletion of each source ID you specify.
+
+        Parameters
+        ----------
+        tenant_id : str
+            Unique identifier for the tenant/organization
+
+        source_ids : typing.Sequence[str]
+            List of source IDs to delete
+
+        sub_tenant_id : typing.Optional[str]
+            Optional sub-tenant identifier for organizing data within a tenant. If not provided, defaults to tenant_id
+
+        request_options : typing.Optional[RequestOptions]
+            Request-specific configuration.
+
+        Returns
+        -------
+        HttpResponse[DeleteSources]
+            Successful Response
+        """
+        _response = self._client_wrapper.httpx_client.request(
+            "document/delete",
+            method="DELETE",
+            json={
+                "tenant_id": tenant_id,
+                "source_ids": source_ids,
+                "sub_tenant_id": sub_tenant_id,
+            },
+            headers={
+                "content-type": "application/json",
+            },
+            request_options=request_options,
+            omit=OMIT,
+        )
+        try:
+            if 200 <= _response.status_code < 300:
+                _data = typing.cast(
+                    DeleteSources,
+                    parse_obj_as(
+                        type_=DeleteSources,  # type: ignore
+                        object_=_response.json(),
+                    ),
+                )
+                return HttpResponse(response=_response, data=_data)
+            if _response.status_code == 400:
+                raise BadRequestError(
+                    headers=dict(_response.headers),
+                    body=typing.cast(
+                        ActualErrorResponse,
+                        parse_obj_as(
+                            type_=ActualErrorResponse,  # type: ignore
+                            object_=_response.json(),
+                        ),
+                    ),
+                )
+            if _response.status_code == 401:
+                raise UnauthorizedError(
+                    headers=dict(_response.headers),
+                    body=typing.cast(
+                        ActualErrorResponse,
+                        parse_obj_as(
+                            type_=ActualErrorResponse,  # type: ignore
+                            object_=_response.json(),
+                        ),
+                    ),
+                )
+            if _response.status_code == 403:
+                raise ForbiddenError(
+                    headers=dict(_response.headers),
+                    body=typing.cast(
+                        ActualErrorResponse,
+                        parse_obj_as(
+                            type_=ActualErrorResponse,  # type: ignore
+                            object_=_response.json(),
+                        ),
+                    ),
+                )
+            if _response.status_code == 404:
+                raise NotFoundError(
+                    headers=dict(_response.headers),
+                    body=typing.cast(
+                        ActualErrorResponse,
+                        parse_obj_as(
+                            type_=ActualErrorResponse,  # type: ignore
+                            object_=_response.json(),
+                        ),
+                    ),
+                )
+            if _response.status_code == 422:
+                raise UnprocessableEntityError(
+                    headers=dict(_response.headers),
+                    body=typing.cast(
+                        typing.Optional[typing.Any],
+                        parse_obj_as(
+                            type_=typing.Optional[typing.Any],  # type: ignore
+                            object_=_response.json(),
+                        ),
+                    ),
+                )
+            if _response.status_code == 500:
+                raise InternalServerError(
+                    headers=dict(_response.headers),
+                    body=typing.cast(
+                        ActualErrorResponse,
+                        parse_obj_as(
+                            type_=ActualErrorResponse,  # type: ignore
+                            object_=_response.json(),
+                        ),
+                    ),
+                )
+            if _response.status_code == 503:
+                raise ServiceUnavailableError(
+                    headers=dict(_response.headers),
+                    body=typing.cast(
+                        ActualErrorResponse,
+                        parse_obj_as(
+                            type_=ActualErrorResponse,  # type: ignore
+                            object_=_response.json(),
+                        ),
+                    ),
+                )
+            _response_json = _response.json()
+        except JSONDecodeError:
+            raise ApiError(status_code=_response.status_code, headers=dict(_response.headers), body=_response.text)
+        raise ApiError(status_code=_response.status_code, headers=dict(_response.headers), body=_response_json)
+
+
+class AsyncRawDocumentClient:
+    def __init__(self, *, client_wrapper: AsyncClientWrapper):
+        self._client_wrapper = client_wrapper
+
+    async def delete(
+        self,
+        *,
+        tenant_id: str,
+        source_ids: typing.Sequence[str],
+        sub_tenant_id: typing.Optional[str] = OMIT,
+        request_options: typing.Optional[RequestOptions] = None,
+    ) -> AsyncHttpResponse[DeleteSources]:
+        """
+        Remove documents and content from your knowledge base.
+
+        This endpoint permanently deletes the specified sources from your knowledge base. Once deleted, the content will no longer be available for search or retrieval.
+
+        Use this carefully as the action cannot be undone. The system will confirm successful deletion of each source ID you specify.
+
+        Parameters
+        ----------
+        tenant_id : str
+            Unique identifier for the tenant/organization
+
+        source_ids : typing.Sequence[str]
+            List of source IDs to delete
+
+        sub_tenant_id : typing.Optional[str]
+            Optional sub-tenant identifier for organizing data within a tenant. If not provided, defaults to tenant_id
+
+        request_options : typing.Optional[RequestOptions]
+            Request-specific configuration.
+
+        Returns
+        -------
+        AsyncHttpResponse[DeleteSources]
+            Successful Response
+        """
+        _response = await self._client_wrapper.httpx_client.request(
+            "document/delete",
+            method="DELETE",
+            json={
+                "tenant_id": tenant_id,
+                "source_ids": source_ids,
+                "sub_tenant_id": sub_tenant_id,
+            },
+            headers={
+                "content-type": "application/json",
+            },
+            request_options=request_options,
+            omit=OMIT,
+        )
+        try:
+            if 200 <= _response.status_code < 300:
+                _data = typing.cast(
+                    DeleteSources,
+                    parse_obj_as(
+                        type_=DeleteSources,  # type: ignore
+                        object_=_response.json(),
+                    ),
+                )
+                return AsyncHttpResponse(response=_response, data=_data)
+            if _response.status_code == 400:
+                raise BadRequestError(
+                    headers=dict(_response.headers),
+                    body=typing.cast(
+                        ActualErrorResponse,
+                        parse_obj_as(
+                            type_=ActualErrorResponse,  # type: ignore
+                            object_=_response.json(),
+                        ),
+                    ),
+                )
+            if _response.status_code == 401:
+                raise UnauthorizedError(
+                    headers=dict(_response.headers),
+                    body=typing.cast(
+                        ActualErrorResponse,
+                        parse_obj_as(
+                            type_=ActualErrorResponse,  # type: ignore
+                            object_=_response.json(),
+                        ),
+                    ),
+                )
+            if _response.status_code == 403:
+                raise ForbiddenError(
+                    headers=dict(_response.headers),
+                    body=typing.cast(
+                        ActualErrorResponse,
+                        parse_obj_as(
+                            type_=ActualErrorResponse,  # type: ignore
+                            object_=_response.json(),
+                        ),
+                    ),
+                )
+            if _response.status_code == 404:
+                raise NotFoundError(
+                    headers=dict(_response.headers),
+                    body=typing.cast(
+                        ActualErrorResponse,
+                        parse_obj_as(
+                            type_=ActualErrorResponse,  # type: ignore
+                            object_=_response.json(),
+                        ),
+                    ),
+                )
+            if _response.status_code == 422:
+                raise UnprocessableEntityError(
+                    headers=dict(_response.headers),
+                    body=typing.cast(
+                        typing.Optional[typing.Any],
+                        parse_obj_as(
+                            type_=typing.Optional[typing.Any],  # type: ignore
+                            object_=_response.json(),
+                        ),
+                    ),
+                )
+            if _response.status_code == 500:
+                raise InternalServerError(
+                    headers=dict(_response.headers),
+                    body=typing.cast(
+                        ActualErrorResponse,
+                        parse_obj_as(
+                            type_=ActualErrorResponse,  # type: ignore
+                            object_=_response.json(),
+                        ),
+                    ),
+                )
+            if _response.status_code == 503:
+                raise ServiceUnavailableError(
+                    headers=dict(_response.headers),
+                    body=typing.cast(
+                        ActualErrorResponse,
+                        parse_obj_as(
+                            type_=ActualErrorResponse,  # type: ignore
+                            object_=_response.json(),
+                        ),
+                    ),
+                )
+            _response_json = _response.json()
+        except JSONDecodeError:
+            raise ApiError(status_code=_response.status_code, headers=dict(_response.headers), body=_response.text)
+        raise ApiError(status_code=_response.status_code, headers=dict(_response.headers), body=_response_json)

usecortex_ai/embeddings/client.py

@@ -38,28 +38,20 @@ class EmbeddingsClient:
         request_options: typing.Optional[RequestOptions] = None,
     ) -> EmbeddingsDeleteData:
         """
-        Delete specific embedding chunks from indexed sources.
+        Delete embedding chunks by chunk ID.
 
-        This endpoint deletes specified embedding chunks from the Findr backend by sending
-        chunk IDs to the backend delete service.
-
-        Args:
-            request (EmbeddingsDeleteRequest): The delete request containing:
-                - chunk_ids (List[str]): List of chunk IDs to delete
-                - tenant_id (str): Tenant identifier for multi-tenancy
-                - sub_tenant_id (str, optional): Sub-tenant identifier, defaults to tenant_id
-            api_details (dict): Authentication details obtained from API key validation
-
-        Returns:
-            EmbeddingsDeleteData: Success response with deletion details
+        Use this to remove specific chunks from your embeddings index when they are no longer valid or should not appear in results.
 
         Parameters
         ----------
         chunk_ids : typing.Sequence[str]
+            The chunk IDs of the source you want to delete
 
         tenant_id : str
+            Unique identifier for the tenant/organization
 
         sub_tenant_id : typing.Optional[str]
+            Optional sub-tenant identifier for organizing data within a tenant. If not provided, defaults to tenant_id
 
         request_options : typing.Optional[RequestOptions]
             Request-specific configuration.
@@ -84,37 +76,31 @@ class EmbeddingsClient:
     def search(
         self,
         *,
-        embeddings: typing.Sequence[float],
         tenant_id: str,
+        embeddings: typing.Optional[typing.Sequence[float]] = OMIT,
         sub_tenant_id: typing.Optional[str] = OMIT,
         max_chunks: typing.Optional[int] = OMIT,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> EmbeddingsSearchData:
         """
-        Search for similar embedding chunks using vector similarity.
+        Find similar chunks using an embedding vector.
 
-        This endpoint performs semantic search by sending an embedding vector to the Findr backend
-        and returns a list of the most similar chunk IDs based on vector similarity.
+        Use this to retrieve the most similar chunk IDs to a single query embedding.
 
-        Args:
-            request (EmbeddingsSearchRequest): The search request containing:
-                - embeddings (List[float]): Single embedding vector for similarity search
-                - tenant_id (str): Tenant identifier for multi-tenancy
-                - sub_tenant_id (str, optional): Sub-tenant identifier, defaults to tenant_id
-                - max_chunks (int, optional): Maximum number of chunk IDs to return (default: 10)
-            api_details (dict): Authentication details obtained from API key validation
 
-        Returns:
-            EmbeddingsSearchData: List of chunk IDs with similarity scores
+        Expected outcome
+        - You receive the closest chunk IDs with optional similarity scores.
 
         Parameters
         ----------
-        embeddings : typing.Sequence[float]
-            Single embedding vector for search
-
         tenant_id : str
+            Unique identifier for the tenant/organization
+
+        embeddings : typing.Optional[typing.Sequence[float]]
+            The embedding vector for search
 
         sub_tenant_id : typing.Optional[str]
+            Optional sub-tenant identifier for organizing data within a tenant. If not provided, defaults to tenant_id
 
         max_chunks : typing.Optional[int]
 
@@ -131,11 +117,11 @@ class EmbeddingsClient:
         from usecortex-ai import CortexAI
 
         client = CortexAI(token="YOUR_TOKEN", )
-        client.embeddings.search(embeddings=[1.1], tenant_id='tenant_id', )
+        client.embeddings.search(tenant_id='tenant_id', )
         """
         _response = self._raw_client.search(
-            embeddings=embeddings,
             tenant_id=tenant_id,
+            embeddings=embeddings,
             sub_tenant_id=sub_tenant_id,
             max_chunks=max_chunks,
             request_options=request_options,
@@ -151,20 +137,20 @@ class EmbeddingsClient:
         request_options: typing.Optional[RequestOptions] = None,
     ) -> EmbeddingsGetData:
         """
-        Get embeddings based on chunk IDs.
+        Retrieve embeddings for specific chunk IDs.
 
-        This endpoint returns embeddings for a list of chunk IDs.
-
-        Returns:
-            EmbeddingsGetData: Embeddings data for the requested chunk IDs
+        Use this when you already know the chunk IDs and need their corresponding embeddings.
 
         Parameters
         ----------
         chunk_ids : typing.Sequence[str]
+            The chunk IDs of the source you want to get embeddings for
 
         tenant_id : str
+            Unique identifier for the tenant/organization
 
         sub_tenant_id : typing.Optional[str]
+            Optional sub-tenant identifier for organizing data within a tenant. If not provided, defaults to tenant_id
 
         request_options : typing.Optional[RequestOptions]
             Request-specific configuration.
@@ -190,16 +176,14 @@ class EmbeddingsClient:
         self, *, tenant_id: str, request_options: typing.Optional[RequestOptions] = None
     ) -> EmbeddingsCreateCollectionData:
         """
-        Create an embeddings collection for the given tenant in Findr.
-
-        sub_tenant_id is set to be the same as tenant_id as per requirements.
+        Create an embeddings collection for a tenant.
 
-        Returns:
-            EmbeddingsCreateCollectionData: Success response with collection details
+        Use this to initialize storage so you can index and query embeddings for the tenant.
 
         Parameters
         ----------
         tenant_id : str
+            Unique identifier for the tenant/organization
 
         request_options : typing.Optional[RequestOptions]
             Request-specific configuration.
@@ -244,28 +228,20 @@ class AsyncEmbeddingsClient:
         request_options: typing.Optional[RequestOptions] = None,
     ) -> EmbeddingsDeleteData:
         """
-        Delete specific embedding chunks from indexed sources.
+        Delete embedding chunks by chunk ID.
 
-        This endpoint deletes specified embedding chunks from the Findr backend by sending
-        chunk IDs to the backend delete service.
-
-        Args:
-            request (EmbeddingsDeleteRequest): The delete request containing:
-                - chunk_ids (List[str]): List of chunk IDs to delete
-                - tenant_id (str): Tenant identifier for multi-tenancy
-                - sub_tenant_id (str, optional): Sub-tenant identifier, defaults to tenant_id
-            api_details (dict): Authentication details obtained from API key validation
-
-        Returns:
-            EmbeddingsDeleteData: Success response with deletion details
+        Use this to remove specific chunks from your embeddings index when they are no longer valid or should not appear in results.
 
         Parameters
         ----------
         chunk_ids : typing.Sequence[str]
+            The chunk IDs of the source you want to delete
 
         tenant_id : str
+            Unique identifier for the tenant/organization
 
         sub_tenant_id : typing.Optional[str]
+            Optional sub-tenant identifier for organizing data within a tenant. If not provided, defaults to tenant_id
 
         request_options : typing.Optional[RequestOptions]
             Request-specific configuration.
@@ -294,37 +270,31 @@ class AsyncEmbeddingsClient:
     async def search(
         self,
         *,
-        embeddings: typing.Sequence[float],
         tenant_id: str,
+        embeddings: typing.Optional[typing.Sequence[float]] = OMIT,
         sub_tenant_id: typing.Optional[str] = OMIT,
         max_chunks: typing.Optional[int] = OMIT,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> EmbeddingsSearchData:
         """
-        Search for similar embedding chunks using vector similarity.
+        Find similar chunks using an embedding vector.
 
-        This endpoint performs semantic search by sending an embedding vector to the Findr backend
-        and returns a list of the most similar chunk IDs based on vector similarity.
+        Use this to retrieve the most similar chunk IDs to a single query embedding.
 
-        Args:
-            request (EmbeddingsSearchRequest): The search request containing:
-                - embeddings (List[float]): Single embedding vector for similarity search
-                - tenant_id (str): Tenant identifier for multi-tenancy
-                - sub_tenant_id (str, optional): Sub-tenant identifier, defaults to tenant_id
-                - max_chunks (int, optional): Maximum number of chunk IDs to return (default: 10)
-            api_details (dict): Authentication details obtained from API key validation
 
-        Returns:
-            EmbeddingsSearchData: List of chunk IDs with similarity scores
+        Expected outcome
+        - You receive the closest chunk IDs with optional similarity scores.
 
         Parameters
         ----------
-        embeddings : typing.Sequence[float]
-            Single embedding vector for search
-
         tenant_id : str
+            Unique identifier for the tenant/organization
+
+        embeddings : typing.Optional[typing.Sequence[float]]
+            The embedding vector for search
 
         sub_tenant_id : typing.Optional[str]
+            Optional sub-tenant identifier for organizing data within a tenant. If not provided, defaults to tenant_id
 
         max_chunks : typing.Optional[int]
 
@@ -344,12 +314,12 @@ class AsyncEmbeddingsClient:
 
         client = AsyncCortexAI(token="YOUR_TOKEN", )
         async def main() -> None:
-            await client.embeddings.search(embeddings=[1.1], tenant_id='tenant_id', )
+            await client.embeddings.search(tenant_id='tenant_id', )
         asyncio.run(main())
         """
         _response = await self._raw_client.search(
-            embeddings=embeddings,
             tenant_id=tenant_id,
+            embeddings=embeddings,
             sub_tenant_id=sub_tenant_id,
             max_chunks=max_chunks,
             request_options=request_options,
@@ -365,20 +335,20 @@ class AsyncEmbeddingsClient:
         request_options: typing.Optional[RequestOptions] = None,
     ) -> EmbeddingsGetData:
         """
-        Get embeddings based on chunk IDs.
+        Retrieve embeddings for specific chunk IDs.
 
-        This endpoint returns embeddings for a list of chunk IDs.
-
-        Returns:
-            EmbeddingsGetData: Embeddings data for the requested chunk IDs
+        Use this when you already know the chunk IDs and need their corresponding embeddings.
 
         Parameters
         ----------
         chunk_ids : typing.Sequence[str]
+            The chunk IDs of the source you want to get embeddings for
 
         tenant_id : str
+            Unique identifier for the tenant/organization
 
         sub_tenant_id : typing.Optional[str]
+            Optional sub-tenant identifier for organizing data within a tenant. If not provided, defaults to tenant_id
 
         request_options : typing.Optional[RequestOptions]
             Request-specific configuration.
@@ -408,16 +378,14 @@ class AsyncEmbeddingsClient:
         self, *, tenant_id: str, request_options: typing.Optional[RequestOptions] = None
     ) -> EmbeddingsCreateCollectionData:
         """
-        Create an embeddings collection for the given tenant in Findr.
-
-        sub_tenant_id is set to be the same as tenant_id as per requirements.
+        Create an embeddings collection for a tenant.
 
-        Returns:
-            EmbeddingsCreateCollectionData: Success response with collection details
+        Use this to initialize storage so you can index and query embeddings for the tenant.
 
         Parameters
         ----------
         tenant_id : str
+            Unique identifier for the tenant/organization
 
         request_options : typing.Optional[RequestOptions]
             Request-specific configuration.