gen_ai_hub.document_grounding.models package

Models subpackage for Document Grounding API.

This subpackage contains Pydantic model definitions for all Document Grounding API requests and responses. Models are organized by API domain:

• pipeline: Models for Pipeline API (document vectorization pipelines)

• retrieval: Models for Retrieval API (content retrieval from repositories)

• vector: Models for Vector API (vector collection management and search)

These models provide type-safe data structures for interacting with the Document Grounding APIs and ensure proper validation of request/response data.

Submodules

gen_ai_hub.document_grounding.models.pipeline module

Pydantic models for Pipeline API.

This module defines data models for the Pipeline API, which manages document vectorization pipelines from various data sources (Microsoft SharePoint, AWS S3, SFTP).

Model categories:

• Pipeline configuration models (create/get requests and responses)

• Pipeline execution models (tracking pipeline runs)

• Document models (tracking document processing status)

• Search and metadata models (filtering pipelines by metadata)

• Trigger models (manual pipeline execution)

All models use Pydantic for validation and serialization.

class BasePipelineResponse

Bases: BaseModel

id: str

metadata: MetaData | None

model_config: ClassVar[ConfigDict] = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

type: str

class CommonConfiguration

Bases: BaseModel

destination: str

model_config: ClassVar[ConfigDict] = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class DataRepositoryMetadataItem

Bases: BaseModel

key: str

model_config: ClassVar[ConfigDict] = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

value: List[str]

class Document

Bases: BaseModel

absoluteUrl: str | None

createdTimestamp: datetime | None

downloadLocation: str | None

id: str

lastUpdatedTimestamp: datetime | None

metadataId: str | None

model_config: ClassVar[ConfigDict] = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

status: DocumentStatus | None

title: str | None

viewLocation: str | None

class DocumentStatus

Bases: str, Enum

__new__(value)

DEINDEXED = 'DEINDEXED'

FAILED = 'FAILED'

FAILED_TO_BE_RETRIED = 'FAILED_TO_BE_RETRIED'

INDEXED = 'INDEXED'

REINDEXED = 'REINDEXED'

TO_BE_PROCESSED = 'TO_BE_PROCESSED'

TO_BE_SCHEDULED = 'TO_BE_SCHEDULED'

class DocumentsStatusResponse

Bases: BaseModel

count: int | None

model_config: ClassVar[ConfigDict] = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

resources: List[Document]

class GetPipelineExecutionsResponse

Bases: BaseModel

count: int | None

model_config: ClassVar[ConfigDict] = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

resources: List[PipelineExecution]

class GetPipelineStatusResponse

Bases: BaseModel

lastStarted: str | None

model_config: ClassVar[ConfigDict] = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

status: str | None

class GetPipelinesResponse

Bases: BaseModel

count: int | None

model_config: ClassVar[ConfigDict] = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

resources: List[Annotated[MSSharePointPipelineGetResponse | S3PipelineGetResponse | SFTPPipelineGetResponse, FieldInfo(annotation=NoneType, required=True, discriminator='type')]]

class MSSharePointConfiguration

Bases: BaseModel

destination: str

model_config: ClassVar[ConfigDict] = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

sharePoint: SharePointConfig

class MSSharePointConfigurationGetResponse

Bases: BaseModel

destination: str

model_config: ClassVar[ConfigDict] = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

sharePoint: SharePointConfig

class MSSharePointPipelineCreateRequest

Bases: BaseModel

configuration: MSSharePointConfiguration

metadata: MetaData | None

model_config: ClassVar[ConfigDict] = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

type: Literal['MSSharePoint']

class MSSharePointPipelineGetResponse

Bases: BasePipelineResponse

configuration: MSSharePointConfigurationGetResponse

model_config: ClassVar[ConfigDict] = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

type: Literal['MSSharePoint']

class ManualPipelineTrigger

Bases: BaseModel

metadataOnly: bool | None

model_config: ClassVar[ConfigDict] = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

pipelineId: str

class MetaData

Bases: BaseModel

destination: str

model_config: ClassVar[ConfigDict] = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class PipelineExecution

Bases: BaseModel

createdAt: datetime | None

id: str

model_config: ClassVar[ConfigDict] = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

modifiedAt: datetime | None

status: PipelineExecutionStatus | None

class PipelineExecutionStatus

Bases: str, Enum

__new__(value)

FINISHED = 'FINISHED'

FINISHED_WITH_ERRORS = 'FINISHEDWITHERRORS'

INPROGRESS = 'INPROGRESS'

NEW = 'NEW'

TIMEOUT = 'TIMEOUT'

UNKNOWN = 'UNKNOWN'

class PipelineIdResponse

Bases: BaseModel

model_config: ClassVar[ConfigDict] = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

pipelineId: str

class S3PipelineCreateRequest

Bases: BaseModel

configuration: CommonConfiguration

metadata: MetaData | None

model_config: ClassVar[ConfigDict] = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

type: Literal['S3']

class S3PipelineGetResponse

Bases: BasePipelineResponse

configuration: CommonConfiguration

model_config: ClassVar[ConfigDict] = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

type: Literal['S3']

class SFTPPipelineCreateRequest

Bases: BaseModel

configuration: CommonConfiguration

metadata: MetaData | None

model_config: ClassVar[ConfigDict] = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

type: Literal['SFTP']

class SFTPPipelineGetResponse

Bases: BasePipelineResponse

configuration: CommonConfiguration

model_config: ClassVar[ConfigDict] = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

type: Literal['SFTP']

class SearchPipelineData

Bases: BaseModel

model_config: ClassVar[ConfigDict] = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

pipelineId: str

class SearchPipelineRequest

Bases: BaseModel

dataRepositoryMetadata: List[DataRepositoryMetadataItem]

model_config: ClassVar[ConfigDict] = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class SearchPipelinesResponse

Bases: BaseModel

count: int | None

model_config: ClassVar[ConfigDict] = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

resources: List[SearchPipelineData]

class SharePointConfig

Bases: BaseModel

model_config: ClassVar[ConfigDict] = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

site: SharePointSite

class SharePointSite

Bases: BaseModel

includePaths: List[str] | None

model_config: ClassVar[ConfigDict] = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

name: str

gen_ai_hub.document_grounding.models.retrieval module

Pydantic models for Retrieval API.

This module defines data models for the Retrieval API, which enables querying and retrieving relevant content from configured data repositories (vector stores and external document sources).

Model categories:

• Data repository models (repository information and metadata)

• Chunk and document models (content structure)

• Search filter and configuration models (query parameters)

• Search input and result models (request/response structures)

The Retrieval API supports semantic search combined with metadata filtering for precise content retrieval across multiple repository types.

class DataRepositories

Bases: BaseModel

count: int | None

model_config: ClassVar[ConfigDict] = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

resources: List[DataRepository]

class DataRepository

Bases: BaseModel

id: str

metadata: List[RetrievalKeyValueListPair] | None

model_config: ClassVar[ConfigDict] = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

title: str

type: Literal['vector', 'help.sap.com'] | str

class DataRepositoryWithDocuments

Bases: BaseModel

documents: List[RetrievalDocument]

id: str

metadata: List[RetrievalKeyValueListPair] | None

model_config: ClassVar[ConfigDict] = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

title: str

class RetrievalChunk

Bases: BaseModel

content: str

id: str

metadata: List[RetrievalKeyValueListPair] | None

model_config: ClassVar[ConfigDict] = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class RetrievalDataRepositorySearchResult

Bases: BaseModel

dataRepository: DataRepositoryWithDocuments

model_config: ClassVar[ConfigDict] = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class RetrievalDocument

Bases: BaseModel

chunks: List[RetrievalChunk]

id: str

metadata: List[RetrievalDocumentKeyValueListPair] | None

model_config: ClassVar[ConfigDict] = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class RetrievalDocumentKeyValueListPair

Bases: RetrievalKeyValueListPair

matchMode: str | None

model_config: ClassVar[ConfigDict] = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class RetrievalKeyValueListPair

Bases: BaseModel

key: str

model_config: ClassVar[ConfigDict] = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

value: List[str]

class RetrievalPerFilterSearchResult

Bases: BaseModel

filterId: str

model_config: ClassVar[ConfigDict] = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

results: List[RetrievalDataRepositorySearchResult]

class RetrievalPerFilterSearchResultError

Bases: BaseModel

message: str

model_config: ClassVar[ConfigDict] = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class RetrievalPerFilterSearchResultWithError

Bases: BaseModel

error: RetrievalPerFilterSearchResultError

filterId: str

model_config: ClassVar[ConfigDict] = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class RetrievalSearchConfiguration

Bases: BaseModel

maxChunkCount: int | None

maxDocumentCount: int | None

model_config: ClassVar[ConfigDict] = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class RetrievalSearchDocumentKeyValueListPair

Bases: BaseModel

key: str

model_config: ClassVar[ConfigDict] = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

selectMode: List[str] | None

value: List[str]

class RetrievalSearchFilter

Bases: BaseModel

chunkMetadata: List[RetrievalKeyValueListPair] | None

dataRepositories: List[str] | None

dataRepositoryMetadata: List[RetrievalKeyValueListPair] | None

dataRepositoryType: Literal['vector', 'help.sap.com'] | str

documentMetadata: List[RetrievalSearchDocumentKeyValueListPair] | None

id: str

model_config: ClassVar[ConfigDict] = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

searchConfiguration: RetrievalSearchConfiguration | None

class RetrievalSearchInput

Bases: BaseModel

filters: List[RetrievalSearchFilter]

model_config: ClassVar[ConfigDict] = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

query: str

class RetrievalSearchResults

Bases: BaseModel

model_config: ClassVar[ConfigDict] = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

results: List[RetrievalPerFilterSearchResult | RetrievalPerFilterSearchResultWithError]

gen_ai_hub.document_grounding.models.vector module

Pydantic models for Vector API.

This module defines data models for the Vector API, which provides management and search capabilities for vector-based document collections.

Model categories:

• Collection models (collection configuration and management)

• Document and chunk models (content structure with embeddings)

• Embedding configuration models (embedding model settings)

• Search models (semantic search requests and results)

• Status models (collection creation/deletion tracking)

The Vector API enables semantic search across document collections using vector embeddings for similarity-based retrieval.

class BaseDocument

Bases: BaseModel

chunks: List[TextOnlyBaseChunk]

metadata: List[VectorKeyValueListPair]

model_config: ClassVar[ConfigDict] = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class Collection

Bases: BaseModel

embeddingConfig: EmbeddingConfig

id: str

metadata: List[VectorKeyValueListPair] | None

model_config: ClassVar[ConfigDict] = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

title: str | None

class CollectionCreateRequest

Bases: BaseModel

embeddingConfig: EmbeddingConfig

metadata: List[VectorKeyValueListPair] | None

model_config: ClassVar[ConfigDict] = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

title: str | None

class CollectionCreatedResponse

Bases: BaseModel

collectionURL: str

model_config: ClassVar[ConfigDict] = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

status: Literal['CREATED']

class CollectionDeletedResponse

Bases: BaseModel

collectionURL: str

model_config: ClassVar[ConfigDict] = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

status: Literal['DELETED']

class CollectionPendingResponse

Bases: BaseModel

Location: str

model_config: ClassVar[ConfigDict] = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

status: Literal['PENDING']

class CollectionsListResponse

Bases: BaseModel

count: int | None

model_config: ClassVar[ConfigDict] = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

resources: List[Collection]

class Document

Bases: BaseDocument

id: str

model_config: ClassVar[ConfigDict] = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class DocumentOutput

Bases: BaseModel

chunks: List[VectorChunk]

id: str

metadata: List[VectorKeyValueListPair] | None

model_config: ClassVar[ConfigDict] = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class DocumentWithoutChunks

Bases: BaseModel

id: str

metadata: List[VectorKeyValueListPair]

model_config: ClassVar[ConfigDict] = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class DocumentsChunk

Bases: BaseModel

documents: List[DocumentOutput]

id: str

metadata: List[VectorKeyValueListPair] | None

model_config: ClassVar[ConfigDict] = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

title: str

class DocumentsCreateRequest

Bases: BaseModel

documents: List[BaseDocument]

model_config: ClassVar[ConfigDict] = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class DocumentsListResponse

Bases: BaseModel

documents: List[DocumentWithoutChunks]

model_config: ClassVar[ConfigDict] = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class DocumentsResponse

Bases: BaseModel

count: int | None

model_config: ClassVar[ConfigDict] = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

resources: List[DocumentWithoutChunks]

class DocumentsUpdateRequest

Bases: BaseModel

documents: List[Document]

model_config: ClassVar[ConfigDict] = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class EmbeddingConfig

Bases: BaseModel

modelName: str | None

model_config: ClassVar[ConfigDict] = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class TextOnlyBaseChunk

Bases: BaseModel

content: str

metadata: List[VectorKeyValueListPair] | None

model_config: ClassVar[ConfigDict] = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class TextSearchRequest

Bases: BaseModel

filters: List[VectorSearchFilter]

model_config: ClassVar[ConfigDict] = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

query: str

class VectorChunk

Bases: BaseModel

content: str

id: str

metadata: List[VectorKeyValueListPair] | None

model_config: ClassVar[ConfigDict] = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class VectorKeyValueListPair

Bases: BaseModel

key: str

model_config: ClassVar[ConfigDict] = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

value: List[str]

class VectorPerFilterSearchResult

Bases: BaseModel

filterId: str

model_config: ClassVar[ConfigDict] = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

results: List[DocumentsChunk]

class VectorSearchConfiguration

Bases: BaseModel

maxChunkCount: int | None

maxDocumentCount: int | None

model_config: ClassVar[ConfigDict] = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class VectorSearchDocumentKeyValueListPair

Bases: BaseModel

key: str

model_config: ClassVar[ConfigDict] = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

selectMode: List[str] | None

value: List[str]

class VectorSearchFilter

Bases: BaseModel

chunkMetadata: List[VectorKeyValueListPair] | None

collectionIds: List[str]

collectionMetadata: List[VectorKeyValueListPair] | None

configuration: VectorSearchConfiguration

documentMetadata: List[VectorSearchDocumentKeyValueListPair] | None

id: str

model_config: ClassVar[ConfigDict] = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class VectorSearchResults

Bases: BaseModel

model_config: ClassVar[ConfigDict] = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

results: List[VectorPerFilterSearchResult]