# This file was auto-generated by Fern from our API Definition. import typing import urllib.parse from json.decoder import JSONDecodeError from ...core.api_error import ApiError from ...core.client_wrapper import AsyncClientWrapper, SyncClientWrapper from ...core.jsonable_encoder import jsonable_encoder from ...core.remove_none_from_dict import remove_none_from_dict from ...errors.unprocessable_entity_error import UnprocessableEntityError from ...types.chat_data import ChatData from ...types.cloud_document import CloudDocument from ...types.cloud_document_create import CloudDocumentCreate from ...types.configured_transformation_item import ConfiguredTransformationItem from ...types.data_sink_create import DataSinkCreate from ...types.eval_execution_params import EvalExecutionParams from ...types.http_validation_error import HttpValidationError from ...types.input_message import InputMessage from ...types.llama_parse_parameters import LlamaParseParameters from ...types.managed_ingestion_status_response import ManagedIngestionStatusResponse from ...types.metadata_filters import MetadataFilters from ...types.paginated_list_cloud_documents_response import PaginatedListCloudDocumentsResponse from ...types.paginated_list_pipeline_files_response import PaginatedListPipelineFilesResponse from ...types.pipeline import Pipeline from ...types.pipeline_create import PipelineCreate from ...types.pipeline_data_source import PipelineDataSource from ...types.pipeline_data_source_create import PipelineDataSourceCreate from ...types.pipeline_deployment import PipelineDeployment from ...types.pipeline_file import PipelineFile from ...types.pipeline_file_create import PipelineFileCreate from ...types.pipeline_type import PipelineType from ...types.playground_session import PlaygroundSession from ...types.preset_retrieval_params import PresetRetrievalParams from ...types.retrieval_mode import RetrievalMode from ...types.retrieve_results import RetrieveResults from ...types.text_node import TextNode from .types.pipeline_file_update_custom_metadata_value import PipelineFileUpdateCustomMetadataValue from .types.pipeline_update_embedding_config import PipelineUpdateEmbeddingConfig from .types.pipeline_update_transform_config import PipelineUpdateTransformConfig try: import pydantic if pydantic.__version__.startswith("1."): raise ImportError import pydantic.v1 as pydantic # type: ignore except ImportError: import pydantic # type: ignore # this is used as the default value for optional parameters OMIT = typing.cast(typing.Any, ...) class PipelinesClient: def __init__(self, *, client_wrapper: SyncClientWrapper): self._client_wrapper = client_wrapper def search_pipelines( self, *, project_id: typing.Optional[str] = None, project_name: typing.Optional[str] = None, pipeline_name: typing.Optional[str] = None, pipeline_type: typing.Optional[PipelineType] = None, ) -> typing.List[Pipeline]: """ Search for pipelines by various parameters. Parameters: - project_id: typing.Optional[str]. - project_name: typing.Optional[str]. - pipeline_name: typing.Optional[str]. - pipeline_type: typing.Optional[PipelineType]. --- from llama_cloud import PipelineType from llama_cloud.client import LlamaCloud client = LlamaCloud( token="YOUR_TOKEN", ) client.pipelines.search_pipelines( pipeline_type=PipelineType.PLAYGROUND, ) """ _response = self._client_wrapper.httpx_client.request( "GET", urllib.parse.urljoin(f"{self._client_wrapper.get_base_url()}/", "api/v1/pipelines"), params=remove_none_from_dict( { "project_id": project_id, "project_name": project_name, "pipeline_name": pipeline_name, "pipeline_type": pipeline_type, } ), headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return pydantic.parse_obj_as(typing.List[Pipeline], _response.json()) # type: ignore if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json) def create_pipeline( self, *, project_id: typing.Optional[str] = None, organization_id: typing.Optional[str] = None, request: PipelineCreate, ) -> Pipeline: """ Create a new pipeline for a project. Parameters: - project_id: typing.Optional[str]. - organization_id: typing.Optional[str]. - request: PipelineCreate. """ _response = self._client_wrapper.httpx_client.request( "POST", urllib.parse.urljoin(f"{self._client_wrapper.get_base_url()}/", "api/v1/pipelines"), params=remove_none_from_dict({"project_id": project_id, "organization_id": organization_id}), json=jsonable_encoder(request), headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return pydantic.parse_obj_as(Pipeline, _response.json()) # type: ignore if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json) def upsert_pipeline( self, *, project_id: typing.Optional[str] = None, organization_id: typing.Optional[str] = None, request: PipelineCreate, ) -> Pipeline: """ Upsert a pipeline for a project. Updates if a pipeline with the same name and project_id already exists. Otherwise, creates a new pipeline. Parameters: - project_id: typing.Optional[str]. - organization_id: typing.Optional[str]. - request: PipelineCreate. """ _response = self._client_wrapper.httpx_client.request( "PUT", urllib.parse.urljoin(f"{self._client_wrapper.get_base_url()}/", "api/v1/pipelines"), params=remove_none_from_dict({"project_id": project_id, "organization_id": organization_id}), json=jsonable_encoder(request), headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return pydantic.parse_obj_as(Pipeline, _response.json()) # type: ignore if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json) def get_pipeline(self, pipeline_id: str) -> Pipeline: """ Get a pipeline by ID for a given project. Parameters: - pipeline_id: str. """ _response = self._client_wrapper.httpx_client.request( "GET", urllib.parse.urljoin(f"{self._client_wrapper.get_base_url()}/", f"api/v1/pipelines/{pipeline_id}"), headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return pydantic.parse_obj_as(Pipeline, _response.json()) # type: ignore if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json) def update_existing_pipeline( self, pipeline_id: str, *, embedding_config: typing.Optional[PipelineUpdateEmbeddingConfig] = OMIT, transform_config: typing.Optional[PipelineUpdateTransformConfig] = OMIT, configured_transformations: typing.Optional[typing.List[ConfiguredTransformationItem]] = OMIT, data_sink_id: typing.Optional[str] = OMIT, embedding_model_config_id: typing.Optional[str] = OMIT, data_sink: typing.Optional[DataSinkCreate] = OMIT, preset_retrieval_parameters: typing.Optional[PresetRetrievalParams] = OMIT, eval_parameters: typing.Optional[EvalExecutionParams] = OMIT, llama_parse_parameters: typing.Optional[LlamaParseParameters] = OMIT, name: typing.Optional[str] = OMIT, managed_pipeline_id: typing.Optional[str] = OMIT, ) -> Pipeline: """ Update an existing pipeline for a project. Parameters: - pipeline_id: str. - embedding_config: typing.Optional[PipelineUpdateEmbeddingConfig]. - transform_config: typing.Optional[PipelineUpdateTransformConfig]. Configuration for the transformation. - configured_transformations: typing.Optional[typing.List[ConfiguredTransformationItem]]. - data_sink_id: typing.Optional[str]. - embedding_model_config_id: typing.Optional[str]. - data_sink: typing.Optional[DataSinkCreate]. - preset_retrieval_parameters: typing.Optional[PresetRetrievalParams]. - eval_parameters: typing.Optional[EvalExecutionParams]. - llama_parse_parameters: typing.Optional[LlamaParseParameters]. - name: typing.Optional[str]. - managed_pipeline_id: typing.Optional[str]. """ _request: typing.Dict[str, typing.Any] = {} if embedding_config is not OMIT: _request["embedding_config"] = embedding_config if transform_config is not OMIT: _request["transform_config"] = transform_config if configured_transformations is not OMIT: _request["configured_transformations"] = configured_transformations if data_sink_id is not OMIT: _request["data_sink_id"] = data_sink_id if embedding_model_config_id is not OMIT: _request["embedding_model_config_id"] = embedding_model_config_id if data_sink is not OMIT: _request["data_sink"] = data_sink if preset_retrieval_parameters is not OMIT: _request["preset_retrieval_parameters"] = preset_retrieval_parameters if eval_parameters is not OMIT: _request["eval_parameters"] = eval_parameters if llama_parse_parameters is not OMIT: _request["llama_parse_parameters"] = llama_parse_parameters if name is not OMIT: _request["name"] = name if managed_pipeline_id is not OMIT: _request["managed_pipeline_id"] = managed_pipeline_id _response = self._client_wrapper.httpx_client.request( "PUT", urllib.parse.urljoin(f"{self._client_wrapper.get_base_url()}/", f"api/v1/pipelines/{pipeline_id}"), json=jsonable_encoder(_request), headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return pydantic.parse_obj_as(Pipeline, _response.json()) # type: ignore if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json) def delete_pipeline(self, pipeline_id: str) -> None: """ Delete a pipeline by ID. Parameters: - pipeline_id: str. --- from llama_cloud.client import LlamaCloud client = LlamaCloud( token="YOUR_TOKEN", ) client.pipelines.delete_pipeline( pipeline_id="string", ) """ _response = self._client_wrapper.httpx_client.request( "DELETE", urllib.parse.urljoin(f"{self._client_wrapper.get_base_url()}/", f"api/v1/pipelines/{pipeline_id}"), headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json) def get_pipeline_status(self, pipeline_id: str) -> ManagedIngestionStatusResponse: """ Get the status of a pipeline by ID. Parameters: - pipeline_id: str. --- from llama_cloud.client import LlamaCloud client = LlamaCloud( token="YOUR_TOKEN", ) client.pipelines.get_pipeline_status( pipeline_id="string", ) """ _response = self._client_wrapper.httpx_client.request( "GET", urllib.parse.urljoin(f"{self._client_wrapper.get_base_url()}/", f"api/v1/pipelines/{pipeline_id}/status"), headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return pydantic.parse_obj_as(ManagedIngestionStatusResponse, _response.json()) # type: ignore if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json) def sync_pipeline(self, pipeline_id: str) -> Pipeline: """ Run ingestion for the pipeline by incrementally updating the data-sink with upstream changes from data-sources & files. Parameters: - pipeline_id: str. """ _response = self._client_wrapper.httpx_client.request( "POST", urllib.parse.urljoin(f"{self._client_wrapper.get_base_url()}/", f"api/v1/pipelines/{pipeline_id}/sync"), headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return pydantic.parse_obj_as(Pipeline, _response.json()) # type: ignore if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json) def cancel_pipeline_sync(self, pipeline_id: str) -> Pipeline: """ Parameters: - pipeline_id: str. """ _response = self._client_wrapper.httpx_client.request( "POST", urllib.parse.urljoin( f"{self._client_wrapper.get_base_url()}/", f"api/v1/pipelines/{pipeline_id}/sync/cancel" ), headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return pydantic.parse_obj_as(Pipeline, _response.json()) # type: ignore if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json) def copy_pipeline(self, pipeline_id: str) -> Pipeline: """ Copy a pipeline by ID. Parameters: - pipeline_id: str. """ _response = self._client_wrapper.httpx_client.request( "POST", urllib.parse.urljoin(f"{self._client_wrapper.get_base_url()}/", f"api/v1/pipelines/{pipeline_id}/copy"), headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return pydantic.parse_obj_as(Pipeline, _response.json()) # type: ignore if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json) def list_pipeline_files( self, pipeline_id: str, *, data_source_id: typing.Optional[str] = None, only_manually_uploaded: typing.Optional[bool] = None, ) -> typing.List[PipelineFile]: """ Get files for a pipeline. Parameters: - pipeline_id: str. - data_source_id: typing.Optional[str]. - only_manually_uploaded: typing.Optional[bool]. --- from llama_cloud.client import LlamaCloud client = LlamaCloud( token="YOUR_TOKEN", ) client.pipelines.list_pipeline_files( pipeline_id="string", ) """ _response = self._client_wrapper.httpx_client.request( "GET", urllib.parse.urljoin(f"{self._client_wrapper.get_base_url()}/", f"api/v1/pipelines/{pipeline_id}/files"), params=remove_none_from_dict( {"data_source_id": data_source_id, "only_manually_uploaded": only_manually_uploaded} ), headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return pydantic.parse_obj_as(typing.List[PipelineFile], _response.json()) # type: ignore if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json) def add_files_to_pipeline_api( self, pipeline_id: str, *, request: typing.List[PipelineFileCreate] ) -> typing.List[PipelineFile]: """ Add files to a pipeline. Parameters: - pipeline_id: str. - request: typing.List[PipelineFileCreate]. --- from llama_cloud.client import LlamaCloud client = LlamaCloud( token="YOUR_TOKEN", ) client.pipelines.add_files_to_pipeline_api( pipeline_id="string", request=[], ) """ _response = self._client_wrapper.httpx_client.request( "PUT", urllib.parse.urljoin(f"{self._client_wrapper.get_base_url()}/", f"api/v1/pipelines/{pipeline_id}/files"), json=jsonable_encoder(request), headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return pydantic.parse_obj_as(typing.List[PipelineFile], _response.json()) # type: ignore if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json) def list_pipeline_files_2( self, pipeline_id: str, *, data_source_id: typing.Optional[str] = None, only_manually_uploaded: typing.Optional[bool] = None, limit: typing.Optional[int] = None, offset: typing.Optional[int] = None, ) -> PaginatedListPipelineFilesResponse: """ Get files for a pipeline. Parameters: - pipeline_id: str. - data_source_id: typing.Optional[str]. - only_manually_uploaded: typing.Optional[bool]. - limit: typing.Optional[int]. - offset: typing.Optional[int]. --- from llama_cloud.client import LlamaCloud client = LlamaCloud( token="YOUR_TOKEN", ) client.pipelines.list_pipeline_files_2( pipeline_id="string", ) """ _response = self._client_wrapper.httpx_client.request( "GET", urllib.parse.urljoin(f"{self._client_wrapper.get_base_url()}/", f"api/v1/pipelines/{pipeline_id}/files2"), params=remove_none_from_dict( { "data_source_id": data_source_id, "only_manually_uploaded": only_manually_uploaded, "limit": limit, "offset": offset, } ), headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return pydantic.parse_obj_as(PaginatedListPipelineFilesResponse, _response.json()) # type: ignore if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json) def get_pipeline_file_status(self, file_id: str, pipeline_id: str) -> ManagedIngestionStatusResponse: """ Get status of a file for a pipeline. Parameters: - file_id: str. - pipeline_id: str. --- from llama_cloud.client import LlamaCloud client = LlamaCloud( token="YOUR_TOKEN", ) client.pipelines.get_pipeline_file_status( file_id="string", pipeline_id="string", ) """ _response = self._client_wrapper.httpx_client.request( "GET", urllib.parse.urljoin( f"{self._client_wrapper.get_base_url()}/", f"api/v1/pipelines/{pipeline_id}/files/{file_id}/status" ), headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return pydantic.parse_obj_as(ManagedIngestionStatusResponse, _response.json()) # type: ignore if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json) def update_pipeline_file( self, file_id: str, pipeline_id: str, *, custom_metadata: typing.Optional[ typing.Dict[str, typing.Optional[PipelineFileUpdateCustomMetadataValue]] ] = OMIT, ) -> PipelineFile: """ Update a file for a pipeline. Parameters: - file_id: str. - pipeline_id: str. - custom_metadata: typing.Optional[typing.Dict[str, typing.Optional[PipelineFileUpdateCustomMetadataValue]]]. --- from llama_cloud.client import LlamaCloud client = LlamaCloud( token="YOUR_TOKEN", ) client.pipelines.update_pipeline_file( file_id="string", pipeline_id="string", ) """ _request: typing.Dict[str, typing.Any] = {} if custom_metadata is not OMIT: _request["custom_metadata"] = custom_metadata _response = self._client_wrapper.httpx_client.request( "PUT", urllib.parse.urljoin( f"{self._client_wrapper.get_base_url()}/", f"api/v1/pipelines/{pipeline_id}/files/{file_id}" ), json=jsonable_encoder(_request), headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return pydantic.parse_obj_as(PipelineFile, _response.json()) # type: ignore if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json) def delete_pipeline_file(self, file_id: str, pipeline_id: str) -> None: """ Delete a file from a pipeline. Parameters: - file_id: str. - pipeline_id: str. --- from llama_cloud.client import LlamaCloud client = LlamaCloud( token="YOUR_TOKEN", ) client.pipelines.delete_pipeline_file( file_id="string", pipeline_id="string", ) """ _response = self._client_wrapper.httpx_client.request( "DELETE", urllib.parse.urljoin( f"{self._client_wrapper.get_base_url()}/", f"api/v1/pipelines/{pipeline_id}/files/{file_id}" ), headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json) def import_pipeline_metadata(self, pipeline_id: str, *, upload_file: typing.IO) -> typing.Dict[str, str]: """ Import metadata for a pipeline. Parameters: - pipeline_id: str. - upload_file: typing.IO. """ _response = self._client_wrapper.httpx_client.request( "PUT", urllib.parse.urljoin(f"{self._client_wrapper.get_base_url()}/", f"api/v1/pipelines/{pipeline_id}/metadata"), data=jsonable_encoder({}), files={"upload_file": upload_file}, headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return pydantic.parse_obj_as(typing.Dict[str, str], _response.json()) # type: ignore if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json) def delete_pipeline_files_metadata(self, pipeline_id: str) -> None: """ Delete metadata for all files in a pipeline. Parameters: - pipeline_id: str. --- from llama_cloud.client import LlamaCloud client = LlamaCloud( token="YOUR_TOKEN", ) client.pipelines.delete_pipeline_files_metadata( pipeline_id="string", ) """ _response = self._client_wrapper.httpx_client.request( "DELETE", urllib.parse.urljoin(f"{self._client_wrapper.get_base_url()}/", f"api/v1/pipelines/{pipeline_id}/metadata"), headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json) def list_pipeline_data_sources(self, pipeline_id: str) -> typing.List[PipelineDataSource]: """ Get data sources for a pipeline. Parameters: - pipeline_id: str. --- from llama_cloud.client import LlamaCloud client = LlamaCloud( token="YOUR_TOKEN", ) client.pipelines.list_pipeline_data_sources( pipeline_id="string", ) """ _response = self._client_wrapper.httpx_client.request( "GET", urllib.parse.urljoin( f"{self._client_wrapper.get_base_url()}/", f"api/v1/pipelines/{pipeline_id}/data-sources" ), headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return pydantic.parse_obj_as(typing.List[PipelineDataSource], _response.json()) # type: ignore if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json) def add_data_sources_to_pipeline( self, pipeline_id: str, *, request: typing.List[PipelineDataSourceCreate] ) -> typing.List[PipelineDataSource]: """ Add data sources to a pipeline. Parameters: - pipeline_id: str. - request: typing.List[PipelineDataSourceCreate]. --- from llama_cloud.client import LlamaCloud client = LlamaCloud( token="YOUR_TOKEN", ) client.pipelines.add_data_sources_to_pipeline( pipeline_id="string", request=[], ) """ _response = self._client_wrapper.httpx_client.request( "PUT", urllib.parse.urljoin( f"{self._client_wrapper.get_base_url()}/", f"api/v1/pipelines/{pipeline_id}/data-sources" ), json=jsonable_encoder(request), headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return pydantic.parse_obj_as(typing.List[PipelineDataSource], _response.json()) # type: ignore if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json) def update_pipeline_data_source( self, data_source_id: str, pipeline_id: str, *, sync_interval: typing.Optional[float] = OMIT ) -> PipelineDataSource: """ Update the configuration of a data source in a pipeline. Parameters: - data_source_id: str. - pipeline_id: str. - sync_interval: typing.Optional[float]. --- from llama_cloud.client import LlamaCloud client = LlamaCloud( token="YOUR_TOKEN", ) client.pipelines.update_pipeline_data_source( data_source_id="string", pipeline_id="string", ) """ _request: typing.Dict[str, typing.Any] = {} if sync_interval is not OMIT: _request["sync_interval"] = sync_interval _response = self._client_wrapper.httpx_client.request( "PUT", urllib.parse.urljoin( f"{self._client_wrapper.get_base_url()}/", f"api/v1/pipelines/{pipeline_id}/data-sources/{data_source_id}", ), json=jsonable_encoder(_request), headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return pydantic.parse_obj_as(PipelineDataSource, _response.json()) # type: ignore if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json) def delete_pipeline_data_source(self, data_source_id: str, pipeline_id: str) -> None: """ Delete a data source from a pipeline. Parameters: - data_source_id: str. - pipeline_id: str. --- from llama_cloud.client import LlamaCloud client = LlamaCloud( token="YOUR_TOKEN", ) client.pipelines.delete_pipeline_data_source( data_source_id="string", pipeline_id="string", ) """ _response = self._client_wrapper.httpx_client.request( "DELETE", urllib.parse.urljoin( f"{self._client_wrapper.get_base_url()}/", f"api/v1/pipelines/{pipeline_id}/data-sources/{data_source_id}", ), headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json) def sync_pipeline_data_source(self, data_source_id: str, pipeline_id: str) -> Pipeline: """ Run ingestion for the pipeline data source by incrementally updating the data-sink with upstream changes from data-source. Parameters: - data_source_id: str. - pipeline_id: str. """ _response = self._client_wrapper.httpx_client.request( "POST", urllib.parse.urljoin( f"{self._client_wrapper.get_base_url()}/", f"api/v1/pipelines/{pipeline_id}/data-sources/{data_source_id}/sync", ), headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return pydantic.parse_obj_as(Pipeline, _response.json()) # type: ignore if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json) def get_pipeline_data_source_status(self, data_source_id: str, pipeline_id: str) -> ManagedIngestionStatusResponse: """ Get the status of a data source for a pipeline. Parameters: - data_source_id: str. - pipeline_id: str. --- from llama_cloud.client import LlamaCloud client = LlamaCloud( token="YOUR_TOKEN", ) client.pipelines.get_pipeline_data_source_status( data_source_id="string", pipeline_id="string", ) """ _response = self._client_wrapper.httpx_client.request( "GET", urllib.parse.urljoin( f"{self._client_wrapper.get_base_url()}/", f"api/v1/pipelines/{pipeline_id}/data-sources/{data_source_id}/status", ), headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return pydantic.parse_obj_as(ManagedIngestionStatusResponse, _response.json()) # type: ignore if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json) def run_search( self, pipeline_id: str, *, dense_similarity_top_k: typing.Optional[int] = OMIT, dense_similarity_cutoff: typing.Optional[float] = OMIT, sparse_similarity_top_k: typing.Optional[int] = OMIT, enable_reranking: typing.Optional[bool] = OMIT, rerank_top_n: typing.Optional[int] = OMIT, alpha: typing.Optional[float] = OMIT, search_filters: typing.Optional[MetadataFilters] = OMIT, files_top_k: typing.Optional[int] = OMIT, retrieval_mode: typing.Optional[RetrievalMode] = OMIT, retrieve_image_nodes: typing.Optional[bool] = OMIT, query: str, class_name: typing.Optional[str] = OMIT, ) -> RetrieveResults: """ Get retrieval results for a managed pipeline and a query Parameters: - pipeline_id: str. - dense_similarity_top_k: typing.Optional[int]. - dense_similarity_cutoff: typing.Optional[float]. - sparse_similarity_top_k: typing.Optional[int]. - enable_reranking: typing.Optional[bool]. - rerank_top_n: typing.Optional[int]. - alpha: typing.Optional[float]. - search_filters: typing.Optional[MetadataFilters]. - files_top_k: typing.Optional[int]. - retrieval_mode: typing.Optional[RetrievalMode]. The retrieval mode for the query. - retrieve_image_nodes: typing.Optional[bool]. Whether to retrieve image nodes. - query: str. The query to retrieve against. - class_name: typing.Optional[str]. --- from llama_cloud import FilterCondition, MetadataFilters, RetrievalMode from llama_cloud.client import LlamaCloud client = LlamaCloud( token="YOUR_TOKEN", ) client.pipelines.run_search( pipeline_id="string", search_filters=MetadataFilters( filters=[], condition=FilterCondition.AND, ), retrieval_mode=RetrievalMode.CHUNKS, query="string", ) """ _request: typing.Dict[str, typing.Any] = {"query": query} if dense_similarity_top_k is not OMIT: _request["dense_similarity_top_k"] = dense_similarity_top_k if dense_similarity_cutoff is not OMIT: _request["dense_similarity_cutoff"] = dense_similarity_cutoff if sparse_similarity_top_k is not OMIT: _request["sparse_similarity_top_k"] = sparse_similarity_top_k if enable_reranking is not OMIT: _request["enable_reranking"] = enable_reranking if rerank_top_n is not OMIT: _request["rerank_top_n"] = rerank_top_n if alpha is not OMIT: _request["alpha"] = alpha if search_filters is not OMIT: _request["search_filters"] = search_filters if files_top_k is not OMIT: _request["files_top_k"] = files_top_k if retrieval_mode is not OMIT: _request["retrieval_mode"] = retrieval_mode if retrieve_image_nodes is not OMIT: _request["retrieve_image_nodes"] = retrieve_image_nodes if class_name is not OMIT: _request["class_name"] = class_name _response = self._client_wrapper.httpx_client.request( "POST", urllib.parse.urljoin(f"{self._client_wrapper.get_base_url()}/", f"api/v1/pipelines/{pipeline_id}/retrieve"), json=jsonable_encoder(_request), headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return pydantic.parse_obj_as(RetrieveResults, _response.json()) # type: ignore if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json) def list_pipeline_jobs(self, pipeline_id: str) -> typing.List[PipelineDeployment]: """ Get jobs for a pipeline. Parameters: - pipeline_id: str. --- from llama_cloud.client import LlamaCloud client = LlamaCloud( token="YOUR_TOKEN", ) client.pipelines.list_pipeline_jobs( pipeline_id="string", ) """ _response = self._client_wrapper.httpx_client.request( "GET", urllib.parse.urljoin(f"{self._client_wrapper.get_base_url()}/", f"api/v1/pipelines/{pipeline_id}/jobs"), headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return pydantic.parse_obj_as(typing.List[PipelineDeployment], _response.json()) # type: ignore if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json) def get_pipeline_job(self, job_id: str, pipeline_id: str) -> PipelineDeployment: """ Get a job for a pipeline. Parameters: - job_id: str. - pipeline_id: str. --- from llama_cloud.client import LlamaCloud client = LlamaCloud( token="YOUR_TOKEN", ) client.pipelines.get_pipeline_job( job_id="string", pipeline_id="string", ) """ _response = self._client_wrapper.httpx_client.request( "GET", urllib.parse.urljoin( f"{self._client_wrapper.get_base_url()}/", f"api/v1/pipelines/{pipeline_id}/jobs/{job_id}" ), headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return pydantic.parse_obj_as(PipelineDeployment, _response.json()) # type: ignore if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json) def get_playground_session(self, pipeline_id: str) -> PlaygroundSession: """ Get a playground session for a user and pipeline. Parameters: - pipeline_id: str. --- from llama_cloud.client import LlamaCloud client = LlamaCloud( token="YOUR_TOKEN", ) client.pipelines.get_playground_session( pipeline_id="string", ) """ _response = self._client_wrapper.httpx_client.request( "GET", urllib.parse.urljoin( f"{self._client_wrapper.get_base_url()}/", f"api/v1/pipelines/{pipeline_id}/playground-session" ), headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return pydantic.parse_obj_as(PlaygroundSession, _response.json()) # type: ignore if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json) def chat( self, pipeline_id: str, *, messages: typing.Optional[typing.List[InputMessage]] = OMIT, data: typing.Optional[ChatData] = OMIT, class_name: typing.Optional[str] = OMIT, ) -> typing.Any: """ Make a retrieval query + chat completion for a managed pipeline. Parameters: - pipeline_id: str. - messages: typing.Optional[typing.List[InputMessage]]. - data: typing.Optional[ChatData]. - class_name: typing.Optional[str]. --- from llama_cloud import ( ChatData, FilterCondition, LlmParameters, MetadataFilters, PresetRetrievalParams, RetrievalMode, SupportedLlmModelNames, ) from llama_cloud.client import LlamaCloud client = LlamaCloud( token="YOUR_TOKEN", ) client.pipelines.chat( pipeline_id="string", data=ChatData( retrieval_parameters=PresetRetrievalParams( search_filters=MetadataFilters( filters=[], condition=FilterCondition.AND, ), retrieval_mode=RetrievalMode.CHUNKS, ), llm_parameters=LlmParameters( model_name=SupportedLlmModelNames.GPT_4_O, ), ), ) """ _request: typing.Dict[str, typing.Any] = {} if messages is not OMIT: _request["messages"] = messages if data is not OMIT: _request["data"] = data if class_name is not OMIT: _request["class_name"] = class_name _response = self._client_wrapper.httpx_client.request( "POST", urllib.parse.urljoin(f"{self._client_wrapper.get_base_url()}/", f"api/v1/pipelines/{pipeline_id}/chat"), json=jsonable_encoder(_request), headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return pydantic.parse_obj_as(typing.Any, _response.json()) # type: ignore if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json) def list_pipeline_documents( self, pipeline_id: str, *, skip: typing.Optional[int] = None, limit: typing.Optional[int] = None, file_id: typing.Optional[str] = None, only_direct_upload: typing.Optional[bool] = None, only_api_data_source_documents: typing.Optional[bool] = None, ) -> typing.List[CloudDocument]: """ Return a list of documents for a pipeline. Parameters: - pipeline_id: str. - skip: typing.Optional[int]. - limit: typing.Optional[int]. - file_id: typing.Optional[str]. - only_direct_upload: typing.Optional[bool]. - only_api_data_source_documents: typing.Optional[bool]. --- from llama_cloud.client import LlamaCloud client = LlamaCloud( token="YOUR_TOKEN", ) client.pipelines.list_pipeline_documents( pipeline_id="string", ) """ _response = self._client_wrapper.httpx_client.request( "GET", urllib.parse.urljoin( f"{self._client_wrapper.get_base_url()}/", f"api/v1/pipelines/{pipeline_id}/documents" ), params=remove_none_from_dict( { "skip": skip, "limit": limit, "file_id": file_id, "only_direct_upload": only_direct_upload, "only_api_data_source_documents": only_api_data_source_documents, } ), headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return pydantic.parse_obj_as(typing.List[CloudDocument], _response.json()) # type: ignore if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json) def create_batch_pipeline_documents( self, pipeline_id: str, *, request: typing.List[CloudDocumentCreate] ) -> typing.List[CloudDocument]: """ Batch create documents for a pipeline. Parameters: - pipeline_id: str. - request: typing.List[CloudDocumentCreate]. --- from llama_cloud.client import LlamaCloud client = LlamaCloud( token="YOUR_TOKEN", ) client.pipelines.create_batch_pipeline_documents( pipeline_id="string", request=[], ) """ _response = self._client_wrapper.httpx_client.request( "POST", urllib.parse.urljoin( f"{self._client_wrapper.get_base_url()}/", f"api/v1/pipelines/{pipeline_id}/documents" ), json=jsonable_encoder(request), headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return pydantic.parse_obj_as(typing.List[CloudDocument], _response.json()) # type: ignore if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json) def upsert_batch_pipeline_documents( self, pipeline_id: str, *, request: typing.List[CloudDocumentCreate] ) -> typing.List[CloudDocument]: """ Batch create or update a document for a pipeline. Parameters: - pipeline_id: str. - request: typing.List[CloudDocumentCreate]. --- from llama_cloud.client import LlamaCloud client = LlamaCloud( token="YOUR_TOKEN", ) client.pipelines.upsert_batch_pipeline_documents( pipeline_id="string", request=[], ) """ _response = self._client_wrapper.httpx_client.request( "PUT", urllib.parse.urljoin( f"{self._client_wrapper.get_base_url()}/", f"api/v1/pipelines/{pipeline_id}/documents" ), json=jsonable_encoder(request), headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return pydantic.parse_obj_as(typing.List[CloudDocument], _response.json()) # type: ignore if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json) def paginated_list_pipeline_documents( self, pipeline_id: str, *, skip: typing.Optional[int] = None, limit: typing.Optional[int] = None, file_id: typing.Optional[str] = None, only_direct_upload: typing.Optional[bool] = None, only_api_data_source_documents: typing.Optional[bool] = None, ) -> PaginatedListCloudDocumentsResponse: """ Return a list of documents for a pipeline. Parameters: - pipeline_id: str. - skip: typing.Optional[int]. - limit: typing.Optional[int]. - file_id: typing.Optional[str]. - only_direct_upload: typing.Optional[bool]. - only_api_data_source_documents: typing.Optional[bool]. --- from llama_cloud.client import LlamaCloud client = LlamaCloud( token="YOUR_TOKEN", ) client.pipelines.paginated_list_pipeline_documents( pipeline_id="string", ) """ _response = self._client_wrapper.httpx_client.request( "GET", urllib.parse.urljoin( f"{self._client_wrapper.get_base_url()}/", f"api/v1/pipelines/{pipeline_id}/documents/paginated" ), params=remove_none_from_dict( { "skip": skip, "limit": limit, "file_id": file_id, "only_direct_upload": only_direct_upload, "only_api_data_source_documents": only_api_data_source_documents, } ), headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return pydantic.parse_obj_as(PaginatedListCloudDocumentsResponse, _response.json()) # type: ignore if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json) def get_pipeline_document(self, document_id: str, pipeline_id: str) -> CloudDocument: """ Return a single document for a pipeline. Parameters: - document_id: str. - pipeline_id: str. --- from llama_cloud.client import LlamaCloud client = LlamaCloud( token="YOUR_TOKEN", ) client.pipelines.get_pipeline_document( document_id="string", pipeline_id="string", ) """ _response = self._client_wrapper.httpx_client.request( "GET", urllib.parse.urljoin( f"{self._client_wrapper.get_base_url()}/", f"api/v1/pipelines/{pipeline_id}/documents/{document_id}" ), headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return pydantic.parse_obj_as(CloudDocument, _response.json()) # type: ignore if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json) def delete_pipeline_document(self, document_id: str, pipeline_id: str) -> None: """ Delete a document from a pipeline. Initiates an async job that will: 1. Delete vectors from the vector store 2. Delete the document from MongoDB after vectors are successfully deleted Parameters: - document_id: str. - pipeline_id: str. --- from llama_cloud.client import LlamaCloud client = LlamaCloud( token="YOUR_TOKEN", ) client.pipelines.delete_pipeline_document( document_id="string", pipeline_id="string", ) """ _response = self._client_wrapper.httpx_client.request( "DELETE", urllib.parse.urljoin( f"{self._client_wrapper.get_base_url()}/", f"api/v1/pipelines/{pipeline_id}/documents/{document_id}" ), headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json) def get_pipeline_document_status(self, document_id: str, pipeline_id: str) -> ManagedIngestionStatusResponse: """ Return a single document for a pipeline. Parameters: - document_id: str. - pipeline_id: str. --- from llama_cloud.client import LlamaCloud client = LlamaCloud( token="YOUR_TOKEN", ) client.pipelines.get_pipeline_document_status( document_id="string", pipeline_id="string", ) """ _response = self._client_wrapper.httpx_client.request( "GET", urllib.parse.urljoin( f"{self._client_wrapper.get_base_url()}/", f"api/v1/pipelines/{pipeline_id}/documents/{document_id}/status", ), headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return pydantic.parse_obj_as(ManagedIngestionStatusResponse, _response.json()) # type: ignore if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json) def list_pipeline_document_chunks(self, document_id: str, pipeline_id: str) -> typing.List[TextNode]: """ Return a list of chunks for a pipeline document. Parameters: - document_id: str. - pipeline_id: str. --- from llama_cloud.client import LlamaCloud client = LlamaCloud( token="YOUR_TOKEN", ) client.pipelines.list_pipeline_document_chunks( document_id="string", pipeline_id="string", ) """ _response = self._client_wrapper.httpx_client.request( "GET", urllib.parse.urljoin( f"{self._client_wrapper.get_base_url()}/", f"api/v1/pipelines/{pipeline_id}/documents/{document_id}/chunks", ), headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return pydantic.parse_obj_as(typing.List[TextNode], _response.json()) # type: ignore if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json) class AsyncPipelinesClient: def __init__(self, *, client_wrapper: AsyncClientWrapper): self._client_wrapper = client_wrapper async def search_pipelines( self, *, project_id: typing.Optional[str] = None, project_name: typing.Optional[str] = None, pipeline_name: typing.Optional[str] = None, pipeline_type: typing.Optional[PipelineType] = None, ) -> typing.List[Pipeline]: """ Search for pipelines by various parameters. Parameters: - project_id: typing.Optional[str]. - project_name: typing.Optional[str]. - pipeline_name: typing.Optional[str]. - pipeline_type: typing.Optional[PipelineType]. --- from llama_cloud import PipelineType from llama_cloud.client import AsyncLlamaCloud client = AsyncLlamaCloud( token="YOUR_TOKEN", ) await client.pipelines.search_pipelines( pipeline_type=PipelineType.PLAYGROUND, ) """ _response = await self._client_wrapper.httpx_client.request( "GET", urllib.parse.urljoin(f"{self._client_wrapper.get_base_url()}/", "api/v1/pipelines"), params=remove_none_from_dict( { "project_id": project_id, "project_name": project_name, "pipeline_name": pipeline_name, "pipeline_type": pipeline_type, } ), headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return pydantic.parse_obj_as(typing.List[Pipeline], _response.json()) # type: ignore if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json) async def create_pipeline( self, *, project_id: typing.Optional[str] = None, organization_id: typing.Optional[str] = None, request: PipelineCreate, ) -> Pipeline: """ Create a new pipeline for a project. Parameters: - project_id: typing.Optional[str]. - organization_id: typing.Optional[str]. - request: PipelineCreate. """ _response = await self._client_wrapper.httpx_client.request( "POST", urllib.parse.urljoin(f"{self._client_wrapper.get_base_url()}/", "api/v1/pipelines"), params=remove_none_from_dict({"project_id": project_id, "organization_id": organization_id}), json=jsonable_encoder(request), headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return pydantic.parse_obj_as(Pipeline, _response.json()) # type: ignore if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json) async def upsert_pipeline( self, *, project_id: typing.Optional[str] = None, organization_id: typing.Optional[str] = None, request: PipelineCreate, ) -> Pipeline: """ Upsert a pipeline for a project. Updates if a pipeline with the same name and project_id already exists. Otherwise, creates a new pipeline. Parameters: - project_id: typing.Optional[str]. - organization_id: typing.Optional[str]. - request: PipelineCreate. """ _response = await self._client_wrapper.httpx_client.request( "PUT", urllib.parse.urljoin(f"{self._client_wrapper.get_base_url()}/", "api/v1/pipelines"), params=remove_none_from_dict({"project_id": project_id, "organization_id": organization_id}), json=jsonable_encoder(request), headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return pydantic.parse_obj_as(Pipeline, _response.json()) # type: ignore if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json) async def get_pipeline(self, pipeline_id: str) -> Pipeline: """ Get a pipeline by ID for a given project. Parameters: - pipeline_id: str. """ _response = await self._client_wrapper.httpx_client.request( "GET", urllib.parse.urljoin(f"{self._client_wrapper.get_base_url()}/", f"api/v1/pipelines/{pipeline_id}"), headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return pydantic.parse_obj_as(Pipeline, _response.json()) # type: ignore if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json) async def update_existing_pipeline( self, pipeline_id: str, *, embedding_config: typing.Optional[PipelineUpdateEmbeddingConfig] = OMIT, transform_config: typing.Optional[PipelineUpdateTransformConfig] = OMIT, configured_transformations: typing.Optional[typing.List[ConfiguredTransformationItem]] = OMIT, data_sink_id: typing.Optional[str] = OMIT, embedding_model_config_id: typing.Optional[str] = OMIT, data_sink: typing.Optional[DataSinkCreate] = OMIT, preset_retrieval_parameters: typing.Optional[PresetRetrievalParams] = OMIT, eval_parameters: typing.Optional[EvalExecutionParams] = OMIT, llama_parse_parameters: typing.Optional[LlamaParseParameters] = OMIT, name: typing.Optional[str] = OMIT, managed_pipeline_id: typing.Optional[str] = OMIT, ) -> Pipeline: """ Update an existing pipeline for a project. Parameters: - pipeline_id: str. - embedding_config: typing.Optional[PipelineUpdateEmbeddingConfig]. - transform_config: typing.Optional[PipelineUpdateTransformConfig]. Configuration for the transformation. - configured_transformations: typing.Optional[typing.List[ConfiguredTransformationItem]]. - data_sink_id: typing.Optional[str]. - embedding_model_config_id: typing.Optional[str]. - data_sink: typing.Optional[DataSinkCreate]. - preset_retrieval_parameters: typing.Optional[PresetRetrievalParams]. - eval_parameters: typing.Optional[EvalExecutionParams]. - llama_parse_parameters: typing.Optional[LlamaParseParameters]. - name: typing.Optional[str]. - managed_pipeline_id: typing.Optional[str]. """ _request: typing.Dict[str, typing.Any] = {} if embedding_config is not OMIT: _request["embedding_config"] = embedding_config if transform_config is not OMIT: _request["transform_config"] = transform_config if configured_transformations is not OMIT: _request["configured_transformations"] = configured_transformations if data_sink_id is not OMIT: _request["data_sink_id"] = data_sink_id if embedding_model_config_id is not OMIT: _request["embedding_model_config_id"] = embedding_model_config_id if data_sink is not OMIT: _request["data_sink"] = data_sink if preset_retrieval_parameters is not OMIT: _request["preset_retrieval_parameters"] = preset_retrieval_parameters if eval_parameters is not OMIT: _request["eval_parameters"] = eval_parameters if llama_parse_parameters is not OMIT: _request["llama_parse_parameters"] = llama_parse_parameters if name is not OMIT: _request["name"] = name if managed_pipeline_id is not OMIT: _request["managed_pipeline_id"] = managed_pipeline_id _response = await self._client_wrapper.httpx_client.request( "PUT", urllib.parse.urljoin(f"{self._client_wrapper.get_base_url()}/", f"api/v1/pipelines/{pipeline_id}"), json=jsonable_encoder(_request), headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return pydantic.parse_obj_as(Pipeline, _response.json()) # type: ignore if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json) async def delete_pipeline(self, pipeline_id: str) -> None: """ Delete a pipeline by ID. Parameters: - pipeline_id: str. --- from llama_cloud.client import AsyncLlamaCloud client = AsyncLlamaCloud( token="YOUR_TOKEN", ) await client.pipelines.delete_pipeline( pipeline_id="string", ) """ _response = await self._client_wrapper.httpx_client.request( "DELETE", urllib.parse.urljoin(f"{self._client_wrapper.get_base_url()}/", f"api/v1/pipelines/{pipeline_id}"), headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json) async def get_pipeline_status(self, pipeline_id: str) -> ManagedIngestionStatusResponse: """ Get the status of a pipeline by ID. Parameters: - pipeline_id: str. --- from llama_cloud.client import AsyncLlamaCloud client = AsyncLlamaCloud( token="YOUR_TOKEN", ) await client.pipelines.get_pipeline_status( pipeline_id="string", ) """ _response = await self._client_wrapper.httpx_client.request( "GET", urllib.parse.urljoin(f"{self._client_wrapper.get_base_url()}/", f"api/v1/pipelines/{pipeline_id}/status"), headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return pydantic.parse_obj_as(ManagedIngestionStatusResponse, _response.json()) # type: ignore if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json) async def sync_pipeline(self, pipeline_id: str) -> Pipeline: """ Run ingestion for the pipeline by incrementally updating the data-sink with upstream changes from data-sources & files. Parameters: - pipeline_id: str. """ _response = await self._client_wrapper.httpx_client.request( "POST", urllib.parse.urljoin(f"{self._client_wrapper.get_base_url()}/", f"api/v1/pipelines/{pipeline_id}/sync"), headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return pydantic.parse_obj_as(Pipeline, _response.json()) # type: ignore if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json) async def cancel_pipeline_sync(self, pipeline_id: str) -> Pipeline: """ Parameters: - pipeline_id: str. """ _response = await self._client_wrapper.httpx_client.request( "POST", urllib.parse.urljoin( f"{self._client_wrapper.get_base_url()}/", f"api/v1/pipelines/{pipeline_id}/sync/cancel" ), headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return pydantic.parse_obj_as(Pipeline, _response.json()) # type: ignore if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json) async def copy_pipeline(self, pipeline_id: str) -> Pipeline: """ Copy a pipeline by ID. Parameters: - pipeline_id: str. """ _response = await self._client_wrapper.httpx_client.request( "POST", urllib.parse.urljoin(f"{self._client_wrapper.get_base_url()}/", f"api/v1/pipelines/{pipeline_id}/copy"), headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return pydantic.parse_obj_as(Pipeline, _response.json()) # type: ignore if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json) async def list_pipeline_files( self, pipeline_id: str, *, data_source_id: typing.Optional[str] = None, only_manually_uploaded: typing.Optional[bool] = None, ) -> typing.List[PipelineFile]: """ Get files for a pipeline. Parameters: - pipeline_id: str. - data_source_id: typing.Optional[str]. - only_manually_uploaded: typing.Optional[bool]. --- from llama_cloud.client import AsyncLlamaCloud client = AsyncLlamaCloud( token="YOUR_TOKEN", ) await client.pipelines.list_pipeline_files( pipeline_id="string", ) """ _response = await self._client_wrapper.httpx_client.request( "GET", urllib.parse.urljoin(f"{self._client_wrapper.get_base_url()}/", f"api/v1/pipelines/{pipeline_id}/files"), params=remove_none_from_dict( {"data_source_id": data_source_id, "only_manually_uploaded": only_manually_uploaded} ), headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return pydantic.parse_obj_as(typing.List[PipelineFile], _response.json()) # type: ignore if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json) async def add_files_to_pipeline_api( self, pipeline_id: str, *, request: typing.List[PipelineFileCreate] ) -> typing.List[PipelineFile]: """ Add files to a pipeline. Parameters: - pipeline_id: str. - request: typing.List[PipelineFileCreate]. --- from llama_cloud.client import AsyncLlamaCloud client = AsyncLlamaCloud( token="YOUR_TOKEN", ) await client.pipelines.add_files_to_pipeline_api( pipeline_id="string", request=[], ) """ _response = await self._client_wrapper.httpx_client.request( "PUT", urllib.parse.urljoin(f"{self._client_wrapper.get_base_url()}/", f"api/v1/pipelines/{pipeline_id}/files"), json=jsonable_encoder(request), headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return pydantic.parse_obj_as(typing.List[PipelineFile], _response.json()) # type: ignore if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json) async def list_pipeline_files_2( self, pipeline_id: str, *, data_source_id: typing.Optional[str] = None, only_manually_uploaded: typing.Optional[bool] = None, limit: typing.Optional[int] = None, offset: typing.Optional[int] = None, ) -> PaginatedListPipelineFilesResponse: """ Get files for a pipeline. Parameters: - pipeline_id: str. - data_source_id: typing.Optional[str]. - only_manually_uploaded: typing.Optional[bool]. - limit: typing.Optional[int]. - offset: typing.Optional[int]. --- from llama_cloud.client import AsyncLlamaCloud client = AsyncLlamaCloud( token="YOUR_TOKEN", ) await client.pipelines.list_pipeline_files_2( pipeline_id="string", ) """ _response = await self._client_wrapper.httpx_client.request( "GET", urllib.parse.urljoin(f"{self._client_wrapper.get_base_url()}/", f"api/v1/pipelines/{pipeline_id}/files2"), params=remove_none_from_dict( { "data_source_id": data_source_id, "only_manually_uploaded": only_manually_uploaded, "limit": limit, "offset": offset, } ), headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return pydantic.parse_obj_as(PaginatedListPipelineFilesResponse, _response.json()) # type: ignore if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json) async def get_pipeline_file_status(self, file_id: str, pipeline_id: str) -> ManagedIngestionStatusResponse: """ Get status of a file for a pipeline. Parameters: - file_id: str. - pipeline_id: str. --- from llama_cloud.client import AsyncLlamaCloud client = AsyncLlamaCloud( token="YOUR_TOKEN", ) await client.pipelines.get_pipeline_file_status( file_id="string", pipeline_id="string", ) """ _response = await self._client_wrapper.httpx_client.request( "GET", urllib.parse.urljoin( f"{self._client_wrapper.get_base_url()}/", f"api/v1/pipelines/{pipeline_id}/files/{file_id}/status" ), headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return pydantic.parse_obj_as(ManagedIngestionStatusResponse, _response.json()) # type: ignore if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json) async def update_pipeline_file( self, file_id: str, pipeline_id: str, *, custom_metadata: typing.Optional[ typing.Dict[str, typing.Optional[PipelineFileUpdateCustomMetadataValue]] ] = OMIT, ) -> PipelineFile: """ Update a file for a pipeline. Parameters: - file_id: str. - pipeline_id: str. - custom_metadata: typing.Optional[typing.Dict[str, typing.Optional[PipelineFileUpdateCustomMetadataValue]]]. --- from llama_cloud.client import AsyncLlamaCloud client = AsyncLlamaCloud( token="YOUR_TOKEN", ) await client.pipelines.update_pipeline_file( file_id="string", pipeline_id="string", ) """ _request: typing.Dict[str, typing.Any] = {} if custom_metadata is not OMIT: _request["custom_metadata"] = custom_metadata _response = await self._client_wrapper.httpx_client.request( "PUT", urllib.parse.urljoin( f"{self._client_wrapper.get_base_url()}/", f"api/v1/pipelines/{pipeline_id}/files/{file_id}" ), json=jsonable_encoder(_request), headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return pydantic.parse_obj_as(PipelineFile, _response.json()) # type: ignore if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json) async def delete_pipeline_file(self, file_id: str, pipeline_id: str) -> None: """ Delete a file from a pipeline. Parameters: - file_id: str. - pipeline_id: str. --- from llama_cloud.client import AsyncLlamaCloud client = AsyncLlamaCloud( token="YOUR_TOKEN", ) await client.pipelines.delete_pipeline_file( file_id="string", pipeline_id="string", ) """ _response = await self._client_wrapper.httpx_client.request( "DELETE", urllib.parse.urljoin( f"{self._client_wrapper.get_base_url()}/", f"api/v1/pipelines/{pipeline_id}/files/{file_id}" ), headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json) async def import_pipeline_metadata(self, pipeline_id: str, *, upload_file: typing.IO) -> typing.Dict[str, str]: """ Import metadata for a pipeline. Parameters: - pipeline_id: str. - upload_file: typing.IO. """ _response = await self._client_wrapper.httpx_client.request( "PUT", urllib.parse.urljoin(f"{self._client_wrapper.get_base_url()}/", f"api/v1/pipelines/{pipeline_id}/metadata"), data=jsonable_encoder({}), files={"upload_file": upload_file}, headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return pydantic.parse_obj_as(typing.Dict[str, str], _response.json()) # type: ignore if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json) async def delete_pipeline_files_metadata(self, pipeline_id: str) -> None: """ Delete metadata for all files in a pipeline. Parameters: - pipeline_id: str. --- from llama_cloud.client import AsyncLlamaCloud client = AsyncLlamaCloud( token="YOUR_TOKEN", ) await client.pipelines.delete_pipeline_files_metadata( pipeline_id="string", ) """ _response = await self._client_wrapper.httpx_client.request( "DELETE", urllib.parse.urljoin(f"{self._client_wrapper.get_base_url()}/", f"api/v1/pipelines/{pipeline_id}/metadata"), headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json) async def list_pipeline_data_sources(self, pipeline_id: str) -> typing.List[PipelineDataSource]: """ Get data sources for a pipeline. Parameters: - pipeline_id: str. --- from llama_cloud.client import AsyncLlamaCloud client = AsyncLlamaCloud( token="YOUR_TOKEN", ) await client.pipelines.list_pipeline_data_sources( pipeline_id="string", ) """ _response = await self._client_wrapper.httpx_client.request( "GET", urllib.parse.urljoin( f"{self._client_wrapper.get_base_url()}/", f"api/v1/pipelines/{pipeline_id}/data-sources" ), headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return pydantic.parse_obj_as(typing.List[PipelineDataSource], _response.json()) # type: ignore if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json) async def add_data_sources_to_pipeline( self, pipeline_id: str, *, request: typing.List[PipelineDataSourceCreate] ) -> typing.List[PipelineDataSource]: """ Add data sources to a pipeline. Parameters: - pipeline_id: str. - request: typing.List[PipelineDataSourceCreate]. --- from llama_cloud.client import AsyncLlamaCloud client = AsyncLlamaCloud( token="YOUR_TOKEN", ) await client.pipelines.add_data_sources_to_pipeline( pipeline_id="string", request=[], ) """ _response = await self._client_wrapper.httpx_client.request( "PUT", urllib.parse.urljoin( f"{self._client_wrapper.get_base_url()}/", f"api/v1/pipelines/{pipeline_id}/data-sources" ), json=jsonable_encoder(request), headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return pydantic.parse_obj_as(typing.List[PipelineDataSource], _response.json()) # type: ignore if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json) async def update_pipeline_data_source( self, data_source_id: str, pipeline_id: str, *, sync_interval: typing.Optional[float] = OMIT ) -> PipelineDataSource: """ Update the configuration of a data source in a pipeline. Parameters: - data_source_id: str. - pipeline_id: str. - sync_interval: typing.Optional[float]. --- from llama_cloud.client import AsyncLlamaCloud client = AsyncLlamaCloud( token="YOUR_TOKEN", ) await client.pipelines.update_pipeline_data_source( data_source_id="string", pipeline_id="string", ) """ _request: typing.Dict[str, typing.Any] = {} if sync_interval is not OMIT: _request["sync_interval"] = sync_interval _response = await self._client_wrapper.httpx_client.request( "PUT", urllib.parse.urljoin( f"{self._client_wrapper.get_base_url()}/", f"api/v1/pipelines/{pipeline_id}/data-sources/{data_source_id}", ), json=jsonable_encoder(_request), headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return pydantic.parse_obj_as(PipelineDataSource, _response.json()) # type: ignore if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json) async def delete_pipeline_data_source(self, data_source_id: str, pipeline_id: str) -> None: """ Delete a data source from a pipeline. Parameters: - data_source_id: str. - pipeline_id: str. --- from llama_cloud.client import AsyncLlamaCloud client = AsyncLlamaCloud( token="YOUR_TOKEN", ) await client.pipelines.delete_pipeline_data_source( data_source_id="string", pipeline_id="string", ) """ _response = await self._client_wrapper.httpx_client.request( "DELETE", urllib.parse.urljoin( f"{self._client_wrapper.get_base_url()}/", f"api/v1/pipelines/{pipeline_id}/data-sources/{data_source_id}", ), headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json) async def sync_pipeline_data_source(self, data_source_id: str, pipeline_id: str) -> Pipeline: """ Run ingestion for the pipeline data source by incrementally updating the data-sink with upstream changes from data-source. Parameters: - data_source_id: str. - pipeline_id: str. """ _response = await self._client_wrapper.httpx_client.request( "POST", urllib.parse.urljoin( f"{self._client_wrapper.get_base_url()}/", f"api/v1/pipelines/{pipeline_id}/data-sources/{data_source_id}/sync", ), headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return pydantic.parse_obj_as(Pipeline, _response.json()) # type: ignore if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json) async def get_pipeline_data_source_status( self, data_source_id: str, pipeline_id: str ) -> ManagedIngestionStatusResponse: """ Get the status of a data source for a pipeline. Parameters: - data_source_id: str. - pipeline_id: str. --- from llama_cloud.client import AsyncLlamaCloud client = AsyncLlamaCloud( token="YOUR_TOKEN", ) await client.pipelines.get_pipeline_data_source_status( data_source_id="string", pipeline_id="string", ) """ _response = await self._client_wrapper.httpx_client.request( "GET", urllib.parse.urljoin( f"{self._client_wrapper.get_base_url()}/", f"api/v1/pipelines/{pipeline_id}/data-sources/{data_source_id}/status", ), headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return pydantic.parse_obj_as(ManagedIngestionStatusResponse, _response.json()) # type: ignore if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json) async def run_search( self, pipeline_id: str, *, dense_similarity_top_k: typing.Optional[int] = OMIT, dense_similarity_cutoff: typing.Optional[float] = OMIT, sparse_similarity_top_k: typing.Optional[int] = OMIT, enable_reranking: typing.Optional[bool] = OMIT, rerank_top_n: typing.Optional[int] = OMIT, alpha: typing.Optional[float] = OMIT, search_filters: typing.Optional[MetadataFilters] = OMIT, files_top_k: typing.Optional[int] = OMIT, retrieval_mode: typing.Optional[RetrievalMode] = OMIT, retrieve_image_nodes: typing.Optional[bool] = OMIT, query: str, class_name: typing.Optional[str] = OMIT, ) -> RetrieveResults: """ Get retrieval results for a managed pipeline and a query Parameters: - pipeline_id: str. - dense_similarity_top_k: typing.Optional[int]. - dense_similarity_cutoff: typing.Optional[float]. - sparse_similarity_top_k: typing.Optional[int]. - enable_reranking: typing.Optional[bool]. - rerank_top_n: typing.Optional[int]. - alpha: typing.Optional[float]. - search_filters: typing.Optional[MetadataFilters]. - files_top_k: typing.Optional[int]. - retrieval_mode: typing.Optional[RetrievalMode]. The retrieval mode for the query. - retrieve_image_nodes: typing.Optional[bool]. Whether to retrieve image nodes. - query: str. The query to retrieve against. - class_name: typing.Optional[str]. --- from llama_cloud import FilterCondition, MetadataFilters, RetrievalMode from llama_cloud.client import AsyncLlamaCloud client = AsyncLlamaCloud( token="YOUR_TOKEN", ) await client.pipelines.run_search( pipeline_id="string", search_filters=MetadataFilters( filters=[], condition=FilterCondition.AND, ), retrieval_mode=RetrievalMode.CHUNKS, query="string", ) """ _request: typing.Dict[str, typing.Any] = {"query": query} if dense_similarity_top_k is not OMIT: _request["dense_similarity_top_k"] = dense_similarity_top_k if dense_similarity_cutoff is not OMIT: _request["dense_similarity_cutoff"] = dense_similarity_cutoff if sparse_similarity_top_k is not OMIT: _request["sparse_similarity_top_k"] = sparse_similarity_top_k if enable_reranking is not OMIT: _request["enable_reranking"] = enable_reranking if rerank_top_n is not OMIT: _request["rerank_top_n"] = rerank_top_n if alpha is not OMIT: _request["alpha"] = alpha if search_filters is not OMIT: _request["search_filters"] = search_filters if files_top_k is not OMIT: _request["files_top_k"] = files_top_k if retrieval_mode is not OMIT: _request["retrieval_mode"] = retrieval_mode if retrieve_image_nodes is not OMIT: _request["retrieve_image_nodes"] = retrieve_image_nodes if class_name is not OMIT: _request["class_name"] = class_name _response = await self._client_wrapper.httpx_client.request( "POST", urllib.parse.urljoin(f"{self._client_wrapper.get_base_url()}/", f"api/v1/pipelines/{pipeline_id}/retrieve"), json=jsonable_encoder(_request), headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return pydantic.parse_obj_as(RetrieveResults, _response.json()) # type: ignore if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json) async def list_pipeline_jobs(self, pipeline_id: str) -> typing.List[PipelineDeployment]: """ Get jobs for a pipeline. Parameters: - pipeline_id: str. --- from llama_cloud.client import AsyncLlamaCloud client = AsyncLlamaCloud( token="YOUR_TOKEN", ) await client.pipelines.list_pipeline_jobs( pipeline_id="string", ) """ _response = await self._client_wrapper.httpx_client.request( "GET", urllib.parse.urljoin(f"{self._client_wrapper.get_base_url()}/", f"api/v1/pipelines/{pipeline_id}/jobs"), headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return pydantic.parse_obj_as(typing.List[PipelineDeployment], _response.json()) # type: ignore if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json) async def get_pipeline_job(self, job_id: str, pipeline_id: str) -> PipelineDeployment: """ Get a job for a pipeline. Parameters: - job_id: str. - pipeline_id: str. --- from llama_cloud.client import AsyncLlamaCloud client = AsyncLlamaCloud( token="YOUR_TOKEN", ) await client.pipelines.get_pipeline_job( job_id="string", pipeline_id="string", ) """ _response = await self._client_wrapper.httpx_client.request( "GET", urllib.parse.urljoin( f"{self._client_wrapper.get_base_url()}/", f"api/v1/pipelines/{pipeline_id}/jobs/{job_id}" ), headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return pydantic.parse_obj_as(PipelineDeployment, _response.json()) # type: ignore if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json) async def get_playground_session(self, pipeline_id: str) -> PlaygroundSession: """ Get a playground session for a user and pipeline. Parameters: - pipeline_id: str. --- from llama_cloud.client import AsyncLlamaCloud client = AsyncLlamaCloud( token="YOUR_TOKEN", ) await client.pipelines.get_playground_session( pipeline_id="string", ) """ _response = await self._client_wrapper.httpx_client.request( "GET", urllib.parse.urljoin( f"{self._client_wrapper.get_base_url()}/", f"api/v1/pipelines/{pipeline_id}/playground-session" ), headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return pydantic.parse_obj_as(PlaygroundSession, _response.json()) # type: ignore if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json) async def chat( self, pipeline_id: str, *, messages: typing.Optional[typing.List[InputMessage]] = OMIT, data: typing.Optional[ChatData] = OMIT, class_name: typing.Optional[str] = OMIT, ) -> typing.Any: """ Make a retrieval query + chat completion for a managed pipeline. Parameters: - pipeline_id: str. - messages: typing.Optional[typing.List[InputMessage]]. - data: typing.Optional[ChatData]. - class_name: typing.Optional[str]. --- from llama_cloud import ( ChatData, FilterCondition, LlmParameters, MetadataFilters, PresetRetrievalParams, RetrievalMode, SupportedLlmModelNames, ) from llama_cloud.client import AsyncLlamaCloud client = AsyncLlamaCloud( token="YOUR_TOKEN", ) await client.pipelines.chat( pipeline_id="string", data=ChatData( retrieval_parameters=PresetRetrievalParams( search_filters=MetadataFilters( filters=[], condition=FilterCondition.AND, ), retrieval_mode=RetrievalMode.CHUNKS, ), llm_parameters=LlmParameters( model_name=SupportedLlmModelNames.GPT_4_O, ), ), ) """ _request: typing.Dict[str, typing.Any] = {} if messages is not OMIT: _request["messages"] = messages if data is not OMIT: _request["data"] = data if class_name is not OMIT: _request["class_name"] = class_name _response = await self._client_wrapper.httpx_client.request( "POST", urllib.parse.urljoin(f"{self._client_wrapper.get_base_url()}/", f"api/v1/pipelines/{pipeline_id}/chat"), json=jsonable_encoder(_request), headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return pydantic.parse_obj_as(typing.Any, _response.json()) # type: ignore if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json) async def list_pipeline_documents( self, pipeline_id: str, *, skip: typing.Optional[int] = None, limit: typing.Optional[int] = None, file_id: typing.Optional[str] = None, only_direct_upload: typing.Optional[bool] = None, only_api_data_source_documents: typing.Optional[bool] = None, ) -> typing.List[CloudDocument]: """ Return a list of documents for a pipeline. Parameters: - pipeline_id: str. - skip: typing.Optional[int]. - limit: typing.Optional[int]. - file_id: typing.Optional[str]. - only_direct_upload: typing.Optional[bool]. - only_api_data_source_documents: typing.Optional[bool]. --- from llama_cloud.client import AsyncLlamaCloud client = AsyncLlamaCloud( token="YOUR_TOKEN", ) await client.pipelines.list_pipeline_documents( pipeline_id="string", ) """ _response = await self._client_wrapper.httpx_client.request( "GET", urllib.parse.urljoin( f"{self._client_wrapper.get_base_url()}/", f"api/v1/pipelines/{pipeline_id}/documents" ), params=remove_none_from_dict( { "skip": skip, "limit": limit, "file_id": file_id, "only_direct_upload": only_direct_upload, "only_api_data_source_documents": only_api_data_source_documents, } ), headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return pydantic.parse_obj_as(typing.List[CloudDocument], _response.json()) # type: ignore if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json) async def create_batch_pipeline_documents( self, pipeline_id: str, *, request: typing.List[CloudDocumentCreate] ) -> typing.List[CloudDocument]: """ Batch create documents for a pipeline. Parameters: - pipeline_id: str. - request: typing.List[CloudDocumentCreate]. --- from llama_cloud.client import AsyncLlamaCloud client = AsyncLlamaCloud( token="YOUR_TOKEN", ) await client.pipelines.create_batch_pipeline_documents( pipeline_id="string", request=[], ) """ _response = await self._client_wrapper.httpx_client.request( "POST", urllib.parse.urljoin( f"{self._client_wrapper.get_base_url()}/", f"api/v1/pipelines/{pipeline_id}/documents" ), json=jsonable_encoder(request), headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return pydantic.parse_obj_as(typing.List[CloudDocument], _response.json()) # type: ignore if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json) async def upsert_batch_pipeline_documents( self, pipeline_id: str, *, request: typing.List[CloudDocumentCreate] ) -> typing.List[CloudDocument]: """ Batch create or update a document for a pipeline. Parameters: - pipeline_id: str. - request: typing.List[CloudDocumentCreate]. --- from llama_cloud.client import AsyncLlamaCloud client = AsyncLlamaCloud( token="YOUR_TOKEN", ) await client.pipelines.upsert_batch_pipeline_documents( pipeline_id="string", request=[], ) """ _response = await self._client_wrapper.httpx_client.request( "PUT", urllib.parse.urljoin( f"{self._client_wrapper.get_base_url()}/", f"api/v1/pipelines/{pipeline_id}/documents" ), json=jsonable_encoder(request), headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return pydantic.parse_obj_as(typing.List[CloudDocument], _response.json()) # type: ignore if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json) async def paginated_list_pipeline_documents( self, pipeline_id: str, *, skip: typing.Optional[int] = None, limit: typing.Optional[int] = None, file_id: typing.Optional[str] = None, only_direct_upload: typing.Optional[bool] = None, only_api_data_source_documents: typing.Optional[bool] = None, ) -> PaginatedListCloudDocumentsResponse: """ Return a list of documents for a pipeline. Parameters: - pipeline_id: str. - skip: typing.Optional[int]. - limit: typing.Optional[int]. - file_id: typing.Optional[str]. - only_direct_upload: typing.Optional[bool]. - only_api_data_source_documents: typing.Optional[bool]. --- from llama_cloud.client import AsyncLlamaCloud client = AsyncLlamaCloud( token="YOUR_TOKEN", ) await client.pipelines.paginated_list_pipeline_documents( pipeline_id="string", ) """ _response = await self._client_wrapper.httpx_client.request( "GET", urllib.parse.urljoin( f"{self._client_wrapper.get_base_url()}/", f"api/v1/pipelines/{pipeline_id}/documents/paginated" ), params=remove_none_from_dict( { "skip": skip, "limit": limit, "file_id": file_id, "only_direct_upload": only_direct_upload, "only_api_data_source_documents": only_api_data_source_documents, } ), headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return pydantic.parse_obj_as(PaginatedListCloudDocumentsResponse, _response.json()) # type: ignore if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json) async def get_pipeline_document(self, document_id: str, pipeline_id: str) -> CloudDocument: """ Return a single document for a pipeline. Parameters: - document_id: str. - pipeline_id: str. --- from llama_cloud.client import AsyncLlamaCloud client = AsyncLlamaCloud( token="YOUR_TOKEN", ) await client.pipelines.get_pipeline_document( document_id="string", pipeline_id="string", ) """ _response = await self._client_wrapper.httpx_client.request( "GET", urllib.parse.urljoin( f"{self._client_wrapper.get_base_url()}/", f"api/v1/pipelines/{pipeline_id}/documents/{document_id}" ), headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return pydantic.parse_obj_as(CloudDocument, _response.json()) # type: ignore if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json) async def delete_pipeline_document(self, document_id: str, pipeline_id: str) -> None: """ Delete a document from a pipeline. Initiates an async job that will: 1. Delete vectors from the vector store 2. Delete the document from MongoDB after vectors are successfully deleted Parameters: - document_id: str. - pipeline_id: str. --- from llama_cloud.client import AsyncLlamaCloud client = AsyncLlamaCloud( token="YOUR_TOKEN", ) await client.pipelines.delete_pipeline_document( document_id="string", pipeline_id="string", ) """ _response = await self._client_wrapper.httpx_client.request( "DELETE", urllib.parse.urljoin( f"{self._client_wrapper.get_base_url()}/", f"api/v1/pipelines/{pipeline_id}/documents/{document_id}" ), headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json) async def get_pipeline_document_status(self, document_id: str, pipeline_id: str) -> ManagedIngestionStatusResponse: """ Return a single document for a pipeline. Parameters: - document_id: str. - pipeline_id: str. --- from llama_cloud.client import AsyncLlamaCloud client = AsyncLlamaCloud( token="YOUR_TOKEN", ) await client.pipelines.get_pipeline_document_status( document_id="string", pipeline_id="string", ) """ _response = await self._client_wrapper.httpx_client.request( "GET", urllib.parse.urljoin( f"{self._client_wrapper.get_base_url()}/", f"api/v1/pipelines/{pipeline_id}/documents/{document_id}/status", ), headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return pydantic.parse_obj_as(ManagedIngestionStatusResponse, _response.json()) # type: ignore if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json) async def list_pipeline_document_chunks(self, document_id: str, pipeline_id: str) -> typing.List[TextNode]: """ Return a list of chunks for a pipeline document. Parameters: - document_id: str. - pipeline_id: str. --- from llama_cloud.client import AsyncLlamaCloud client = AsyncLlamaCloud( token="YOUR_TOKEN", ) await client.pipelines.list_pipeline_document_chunks( document_id="string", pipeline_id="string", ) """ _response = await self._client_wrapper.httpx_client.request( "GET", urllib.parse.urljoin( f"{self._client_wrapper.get_base_url()}/", f"api/v1/pipelines/{pipeline_id}/documents/{document_id}/chunks", ), headers=self._client_wrapper.get_headers(), timeout=60, ) if 200 <= _response.status_code < 300: return pydantic.parse_obj_as(typing.List[TextNode], _response.json()) # type: ignore if _response.status_code == 422: raise UnprocessableEntityError(pydantic.parse_obj_as(HttpValidationError, _response.json())) # type: ignore try: _response_json = _response.json() except JSONDecodeError: raise ApiError(status_code=_response.status_code, body=_response.text) raise ApiError(status_code=_response.status_code, body=_response_json)