seekrai 0.4.2__py3-none-any.whl → 0.5.0__py3-none-any.whl

seekrai/resources/vectordb.py

@@ -0,0 +1,482 @@
+from typing import List, Optional
+
+from seekrai.abstract import api_requestor
+from seekrai.resources.resource_base import ResourceBase
+from seekrai.seekrflow_response import SeekrFlowResponse
+from seekrai.types import (
+    SeekrFlowRequest,
+)
+from seekrai.types.vectordb import (
+    VectorDatabaseCreate,
+    VectorDatabaseFileList,
+    VectorDatabaseIngestionList,
+    VectorDatabaseIngestionRequest,
+    VectorDatabaseIngestionResponse,
+    VectorDatabaseList,
+    VectorDatabaseResponse,
+)
+
+
+class VectorDatabase(ResourceBase):
+    def create(
+        self,
+        name: str,
+        model: str,
+        description: Optional[str] = None,
+    ) -> VectorDatabaseResponse:
+        """
+        Create a new vector database.
+
+        Args:
+            name (str): Name of the vector database.
+            model (str): Model used to generate the vectors.
+            dimension (int): Dimension of the vectors.
+            description (Optional[str], optional): Optional description. Defaults to None.
+
+        Returns:
+            VectorDatabaseResponse: Object containing information about the created vector database.
+        """
+        requestor = api_requestor.APIRequestor(
+            client=self._client,
+        )
+
+        parameter_payload = VectorDatabaseCreate(
+            name=name, model=model, description=description
+        ).model_dump()
+
+        response, _, _ = requestor.request(
+            options=SeekrFlowRequest(
+                method="POST",
+                url="flow/vectordb",
+                params=parameter_payload,
+            ),
+            stream=False,
+        )
+
+        assert isinstance(response, SeekrFlowResponse)
+        return VectorDatabaseResponse(**response.data)
+
+    def list(self) -> VectorDatabaseList:
+        """
+        Lists all vector databases for the user.
+
+        Returns:
+            VectorDatabaseList: Object containing a list of vector databases.
+        """
+        requestor = api_requestor.APIRequestor(
+            client=self._client,
+        )
+
+        response, _, _ = requestor.request(
+            options=SeekrFlowRequest(
+                method="GET",
+                url="flow/vectordb",
+            ),
+            stream=False,
+        )
+
+        assert isinstance(response, SeekrFlowResponse)
+        return VectorDatabaseList(**response.data)
+
+    def retrieve(self, database_id: str) -> VectorDatabaseResponse:
+        """
+        Retrieves vector database details.
+
+        Args:
+            database_id (str): Vector database ID to retrieve.
+
+        Returns:
+            VectorDatabaseResponse: Object containing information about the vector database.
+        """
+        requestor = api_requestor.APIRequestor(
+            client=self._client,
+        )
+
+        response, _, _ = requestor.request(
+            options=SeekrFlowRequest(
+                method="GET",
+                url=f"flow/vectordb/{database_id}",
+            ),
+            stream=False,
+        )
+
+        assert isinstance(response, SeekrFlowResponse)
+        return VectorDatabaseResponse(**response.data)
+
+    def create_ingestion_job(
+        self,
+        database_id: str,
+        files: List[str],
+        method: str,
+        token_count: int = 800,
+        overlap_tokens: int = 100,
+    ) -> VectorDatabaseIngestionResponse:
+        """
+        Start an ingestion job for the specified files.
+
+        Args:
+            database_id (str): ID of the vector database to ingest into.
+            files (List[str]): List of file IDs to ingest.
+            method (str): Method to use for ingestion.
+
+        Returns:
+            VectorDatabaseIngestionResponse: Object containing information about the created ingestion job.
+        """
+        requestor = api_requestor.APIRequestor(
+            client=self._client,
+        )
+
+        parameter_payload = VectorDatabaseIngestionRequest(
+            file_ids=files,
+            method=method,
+            token_count=token_count,
+            overlap_tokens=overlap_tokens,
+        ).model_dump()
+
+        response, _, _ = requestor.request(
+            options=SeekrFlowRequest(
+                method="POST",
+                url=f"flow/vectordb/{database_id}/ingestion",
+                params=parameter_payload,
+            ),
+            stream=False,
+        )
+
+        assert isinstance(response, SeekrFlowResponse)
+        return VectorDatabaseIngestionResponse(**response.data)
+
+    def list_ingestion_jobs(self, database_id: str) -> VectorDatabaseIngestionList:
+        """
+        Lists ingestion job history for a specific vector database.
+
+        Args:
+            database_id (str): ID of the vector database.
+
+        Returns:
+            VectorDatabaseIngestionList: Object containing a list of ingestion jobs.
+        """
+        requestor = api_requestor.APIRequestor(
+            client=self._client,
+        )
+
+        response, _, _ = requestor.request(
+            options=SeekrFlowRequest(
+                method="GET",
+                url=f"flow/vectordb/{database_id}/ingestion",
+            ),
+            stream=False,
+        )
+
+        assert isinstance(response, SeekrFlowResponse)
+        return VectorDatabaseIngestionList(**response.data)
+
+    def retrieve_ingestion_job(
+        self, database_id: str, job_id: str
+    ) -> VectorDatabaseIngestionResponse:
+        """
+        Retrieves ingestion job details.
+
+        Args:
+            database_id (str): ID of the vector database.
+            job_id (str): Ingestion job ID to retrieve.
+
+        Returns:
+            VectorDatabaseIngestionResponse: Object containing information about the ingestion job.
+        """
+        requestor = api_requestor.APIRequestor(
+            client=self._client,
+        )
+
+        response, _, _ = requestor.request(
+            options=SeekrFlowRequest(
+                method="GET",
+                url=f"flow/vectordb/{database_id}/ingestion/{job_id}",
+            ),
+            stream=False,
+        )
+
+        assert isinstance(response, SeekrFlowResponse)
+        return VectorDatabaseIngestionResponse(**response.data)
+
+    def list_files(self, database_id: str) -> VectorDatabaseFileList:
+        """
+        Lists all files in a vector database.
+
+        Args:
+            database_id (str): ID of the vector database.
+
+        Returns:
+            VectorDatabaseFileList: Object containing a list of files in the vector database.
+        """
+        requestor = api_requestor.APIRequestor(
+            client=self._client,
+        )
+
+        response, _, _ = requestor.request(
+            options=SeekrFlowRequest(
+                method="GET",
+                url=f"flow/vectordb/{database_id}/files",
+            ),
+            stream=False,
+        )
+
+        assert isinstance(response, SeekrFlowResponse)
+        return VectorDatabaseFileList(**response.data)
+
+    def delete(self, database_id: str) -> None:
+        """
+        Delete a vector database.
+
+        Args:
+            database_id (str): ID of the vector database to delete.
+
+        Returns:
+            None
+        """
+        requestor = api_requestor.APIRequestor(
+            client=self._client,
+        )
+
+        response, _, _ = requestor.request(
+            options=SeekrFlowRequest(
+                method="DELETE",
+                url=f"flow/vectordb/{database_id}",
+            ),
+            stream=False,
+        )
+
+        # The endpoint returns 204 No Content
+        return None
+
+
+class AsyncVectorDatabase(ResourceBase):
+    async def create(
+        self,
+        name: str,
+        model: str,
+        description: Optional[str] = None,
+    ) -> VectorDatabaseResponse:
+        """
+        Create a new vector database asynchronously.
+
+        Args:
+            name (str): Name of the vector database.
+            model (str): Model used to generate the vectors.
+            description (Optional[str], optional): Optional description. Defaults to None.
+
+        Returns:
+            VectorDatabaseResponse: Object containing information about the created vector database.
+        """
+        requestor = api_requestor.APIRequestor(
+            client=self._client,
+        )
+
+        parameter_payload = VectorDatabaseCreate(
+            name=name, model=model, description=description
+        ).model_dump()
+
+        response, _, _ = await requestor.arequest(
+            options=SeekrFlowRequest(
+                method="POST",
+                url="flow/vectordb",
+                params=parameter_payload,
+            ),
+            stream=False,
+        )
+
+        assert isinstance(response, SeekrFlowResponse)
+        return VectorDatabaseResponse(**response.data)
+
+    async def list(self) -> VectorDatabaseList:
+        """
+        Lists all vector databases for the user asynchronously.
+
+        Returns:
+            VectorDatabaseList: Object containing a list of vector databases.
+        """
+        requestor = api_requestor.APIRequestor(
+            client=self._client,
+        )
+
+        response, _, _ = await requestor.arequest(
+            options=SeekrFlowRequest(
+                method="GET",
+                url="flow/vectordb",
+            ),
+            stream=False,
+        )
+
+        assert isinstance(response, SeekrFlowResponse)
+        return VectorDatabaseList(**response.data)
+
+    async def retrieve(self, database_id: str) -> VectorDatabaseResponse:
+        """
+        Retrieves vector database details asynchronously.
+
+        Args:
+            database_id (str): Vector database ID to retrieve.
+
+        Returns:
+            VectorDatabaseResponse: Object containing information about the vector database.
+        """
+        requestor = api_requestor.APIRequestor(
+            client=self._client,
+        )
+
+        response, _, _ = await requestor.arequest(
+            options=SeekrFlowRequest(
+                method="GET",
+                url=f"flow/vectordb/{database_id}",
+            ),
+            stream=False,
+        )
+
+        assert isinstance(response, SeekrFlowResponse)
+        return VectorDatabaseResponse(**response.data)
+
+    async def create_ingestion_job(
+        self,
+        database_id: str,
+        files: List[str],
+        method: str,
+        token_count: int = 800,
+        overlap_tokens: int = 100,
+    ) -> VectorDatabaseIngestionResponse:
+        """
+        Start an ingestion job for the specified files asynchronously.
+
+        Args:
+            database_id (str): ID of the vector database to ingest into.
+            files (List[str]): List of file IDs to ingest.
+            method (str): Method to use for ingestion.
+
+        Returns:
+            VectorDatabaseIngestionResponse: Object containing information about the created ingestion job.
+        """
+        requestor = api_requestor.APIRequestor(
+            client=self._client,
+        )
+
+        parameter_payload = VectorDatabaseIngestionRequest(
+            file_ids=files,
+            method=method,
+            token_count=token_count,
+            overlap_tokens=overlap_tokens,
+        ).model_dump()
+
+        response, _, _ = await requestor.arequest(
+            options=SeekrFlowRequest(
+                method="POST",
+                url=f"flow/vectordb/{database_id}/ingestion",
+                params=parameter_payload,
+            ),
+            stream=False,
+        )
+
+        assert isinstance(response, SeekrFlowResponse)
+        return VectorDatabaseIngestionResponse(**response.data)
+
+    async def list_ingestion_jobs(
+        self, database_id: str
+    ) -> VectorDatabaseIngestionList:
+        """
+        Lists ingestion job history for a specific vector database asynchronously.
+
+        Args:
+            database_id (str): ID of the vector database.
+
+        Returns:
+            VectorDatabaseIngestionList: Object containing a list of ingestion jobs.
+        """
+        requestor = api_requestor.APIRequestor(
+            client=self._client,
+        )
+
+        response, _, _ = await requestor.arequest(
+            options=SeekrFlowRequest(
+                method="GET",
+                url=f"flow/vectordb/{database_id}/ingestion",
+            ),
+            stream=False,
+        )
+
+        assert isinstance(response, SeekrFlowResponse)
+        return VectorDatabaseIngestionList(**response.data)
+
+    async def retrieve_ingestion_job(
+        self, database_id: str, job_id: str
+    ) -> VectorDatabaseIngestionResponse:
+        """
+        Retrieves ingestion job details asynchronously.
+
+        Args:
+            database_id (str): ID of the vector database.
+            job_id (str): Ingestion job ID to retrieve.
+
+        Returns:
+            VectorDatabaseIngestionResponse: Object containing information about the ingestion job.
+        """
+        requestor = api_requestor.APIRequestor(
+            client=self._client,
+        )
+
+        response, _, _ = await requestor.arequest(
+            options=SeekrFlowRequest(
+                method="GET",
+                url=f"flow/vectordb/{database_id}/ingestion/{job_id}",
+            ),
+            stream=False,
+        )
+
+        assert isinstance(response, SeekrFlowResponse)
+        return VectorDatabaseIngestionResponse(**response.data)
+
+    async def list_files(self, database_id: str) -> VectorDatabaseFileList:
+        """
+        Lists all files in a vector database asynchronously.
+
+        Args:
+            database_id (str): ID of the vector database.
+
+        Returns:
+            VectorDatabaseFileList: Object containing a list of files in the vector database.
+        """
+        requestor = api_requestor.APIRequestor(
+            client=self._client,
+        )
+
+        response, _, _ = await requestor.arequest(
+            options=SeekrFlowRequest(
+                method="GET",
+                url=f"flow/vectordb/{database_id}/files",
+            ),
+            stream=False,
+        )
+
+        assert isinstance(response, SeekrFlowResponse)
+        return VectorDatabaseFileList(**response.data)
+
+    async def delete(self, database_id: str) -> None:
+        """
+        Delete a vector database asynchronously.
+
+        Args:
+            database_id (str): ID of the vector database to delete.
+
+        Returns:
+            None
+        """
+        requestor = api_requestor.APIRequestor(
+            client=self._client,
+        )
+
+        response, _, _ = await requestor.arequest(
+            options=SeekrFlowRequest(
+                method="DELETE",
+                url=f"flow/vectordb/{database_id}",
+            ),
+            stream=False,
+        )
+
+        # The endpoint returns 204 No Content
+        return None

seekrai/types/__init__.py

@@ -1,4 +1,49 @@
 from seekrai.types.abstract import SeekrFlowClient
+from seekrai.types.agents import (
+    Agent,
+    AgentDeleteResponse,
+    AgentStatus,
+    CreateAgentRequest,
+    Env,
+    EnvConfig,
+    FileSearch,
+    FileSearchEnv,
+    InputFile,
+    InputImage,
+    InputMessage,
+    InputText,
+    MessageUpdateRequest,
+    OutputGuardrail,
+    OutputMessage,
+    OutputText,
+    Run,
+    RunRequest,
+    RunResponse,
+    RunStatus,
+    RunStep,
+    RunStepUsage,
+    RunUsage,
+    StreamEndNodeChunk,
+    StreamFinalResultEventChunk,
+    StreamingToolRequest,
+    StreamingToolResponse,
+    StreamNodeHeaderChunk,
+    StreamPartStartEventChunk,
+    StreamReasoningChunk,
+    StreamTextChunk,
+    StreamTextDeltaChunk,
+    StreamToolCallPartDeltaChunk,
+    StreamUserPromptChunk,
+    Thread,
+    ThreadCreateRequest,
+    ThreadMessage,
+    ThreadMessageContentType,
+    ThreadStatus,
+    Tool,
+    ToolBase,
+    ToolType,
+)
+from seekrai.types.agents.tools.schemas import FileSearch, FileSearchEnv
 from seekrai.types.alignment import (
     AlignmentEstimationRequest,
     AlignmentEstimationResponse,
@@ -107,4 +152,46 @@ __all__ = [
     "GetDeploymentsResponse",
     "HardwareType",
     "NewDeploymentRequest",
+    "MessageUpdateRequest",
+    "ThreadCreateRequest",
+    "ThreadStatus",
+    "Thread",
+    "StreamReasoningChunk",
+    "StreamTextChunk",
+    "StreamingToolRequest",
+    "StreamingToolResponse",
+    "StreamNodeHeaderChunk",
+    "StreamUserPromptChunk",
+    "StreamPartStartEventChunk",
+    "StreamToolCallPartDeltaChunk",
+    "StreamFinalResultEventChunk",
+    "StreamEndNodeChunk",
+    "StreamTextDeltaChunk",
+    "InputText",
+    "InputMessage",
+    "InputFile",
+    "InputImage",
+    "OutputMessage",
+    "OutputText",
+    "OutputGuardrail",
+    "ThreadMessage",
+    "ThreadMessageContentType",
+    "RunResponse",
+    "Run",
+    "RunRequest",
+    "RunStep",
+    "RunUsage",
+    "RunStatus",
+    "RunStepUsage",
+    "Agent",
+    "AgentStatus",
+    "CreateAgentRequest",
+    "AgentDeleteResponse",
+    "ToolBase",
+    "ToolType",
+    "EnvConfig",
+    "Env",
+    "Tool",
+    "FileSearch",
+    "FileSearchEnv",
 ]

seekrai/types/agents/__init__.py

@@ -0,0 +1,89 @@
+from seekrai.types.agents.agent import (
+    Agent,
+    AgentDeleteResponse,
+    AgentStatus,
+    CreateAgentRequest,
+)
+from seekrai.types.agents.runs import (
+    Run,
+    RunRequest,
+    RunResponse,
+    RunStatus,
+    RunStep,
+    RunStepUsage,
+    RunUsage,
+)
+from seekrai.types.agents.threads import (
+    InputFile,
+    InputImage,
+    InputMessage,
+    InputText,
+    MessageUpdateRequest,
+    OutputGuardrail,
+    OutputMessage,
+    OutputText,
+    StreamEndNodeChunk,
+    StreamFinalResultEventChunk,
+    StreamingToolRequest,
+    StreamingToolResponse,
+    StreamNodeHeaderChunk,
+    StreamPartStartEventChunk,
+    StreamReasoningChunk,
+    StreamTextChunk,
+    StreamTextDeltaChunk,
+    StreamToolCallPartDeltaChunk,
+    StreamUserPromptChunk,
+    Thread,
+    ThreadCreateRequest,
+    ThreadMessage,
+    ThreadMessageContentType,
+    ThreadStatus,
+)
+from seekrai.types.agents.tools import Env, EnvConfig, Tool, ToolBase, ToolType
+from seekrai.types.agents.tools.schemas import FileSearch, FileSearchEnv
+
+
+__all__ = [
+    "RunResponse",
+    "Run",
+    "RunRequest",
+    "RunStep",
+    "RunUsage",
+    "RunStatus",
+    "RunStepUsage",
+    "MessageUpdateRequest",
+    "ThreadCreateRequest",
+    "ThreadStatus",
+    "Thread",
+    "StreamReasoningChunk",
+    "StreamTextChunk",
+    "StreamingToolRequest",
+    "StreamingToolResponse",
+    "StreamNodeHeaderChunk",
+    "StreamUserPromptChunk",
+    "StreamPartStartEventChunk",
+    "StreamToolCallPartDeltaChunk",
+    "StreamFinalResultEventChunk",
+    "StreamEndNodeChunk",
+    "StreamTextDeltaChunk",
+    "InputText",
+    "InputMessage",
+    "InputFile",
+    "InputImage",
+    "OutputMessage",
+    "OutputText",
+    "OutputGuardrail",
+    "ThreadMessage",
+    "ThreadMessageContentType",
+    "Agent",
+    "AgentStatus",
+    "CreateAgentRequest",
+    "AgentDeleteResponse",
+    "ToolBase",
+    "ToolType",
+    "EnvConfig",
+    "Env",
+    "Tool",
+    "FileSearch",
+    "FileSearchEnv",
+]

seekrai/types/agents/agent.py

@@ -0,0 +1,42 @@
+import enum
+from datetime import datetime
+from typing import Optional
+
+from pydantic import BaseModel, ConfigDict, Field
+
+from seekrai.types.agents.tools.tool_types import Tool
+
+
+class AgentStatus(str, enum.Enum):
+    INACTIVE = "Inactive"
+    PENDING = "Pending"
+    ACTIVE = "Active"
+    FAILED = "Failed"
+
+
+class CreateAgentRequest(BaseModel):
+    name: str
+    instructions: str
+    tools: list[Tool]
+    model_id: str
+
+
+class Agent(BaseModel):
+    model_config = ConfigDict(from_attributes=True)
+
+    id: str
+    name: str
+    instructions: str
+    status: AgentStatus
+    model_id: str
+    user_id: str
+    tools: list[Tool]
+    created_at: datetime
+    updated_at: datetime
+    last_deployed_at: Optional[datetime] = None
+    active_duration: int = Field(default=0, ge=0)
+
+
+class AgentDeleteResponse(BaseModel):
+    id: str
+    deleted: bool