DifyService

Initialize self. See help(type(self)) for accurate signature.

Create metadata for a specific document in a dataset. Parameters

dataset_id : str Knowledge Base ID body : DifyCreateMetadataRequest Metadata to create Raises

requests.exceptions.HTTPError If the API request fails

Delete a document from a dataset.

Parameters

dataset_id : str Knowledge Base ID document_id : str Document ID to delete

Raises

requests.exceptions.HTTPError If the API request fails

Get all documents in a dataset.

Parameters

dataset_id : str Knowledge Base ID

Returns

List[DifyDatasetDocument] List of documents in the dataset

Raises

requests.exceptions.HTTPError If the API request fails

Get the base URL of the Dify API before the first '/'. Returns: str: Base URL of the Dify API

Get metadata for a specific dataset. Parameters

dataset_id : str Knowledge Base ID Raises

requests.exceptions.HTTPError If the API request fails

Get metadata for a specific dataset. Parameters

dataset_id : str Knowledge Base ID metadata_name : str Metadata name to get Raises

requests.exceptions.HTTPError If the API request fails

Get chunks from a specific document in a knowledge base.

This method retrieves chunks from a specific document within a dataset.

Parameters

dataset_id : str Knowledge Base ID document_id : str Document ID to get chunks from keyword : Optional[str], optional Search keyword to filter chunks, by default None status : str, optional Search status filter, by default "completed" page : int, optional Page number for pagination, by default 1 limit : int, optional Number of items to return per page (1-100), by default 20

Returns

DifyDocumentChunksResponse Response containing the document chunks and pagination info

Raises

requests.exceptions.HTTPError If the API request fails

Get information about a document's uploaded file and optionally download it.

This method retrieves information about the uploaded file associated with a document and can optionally download the file to a temporary directory.

Parameters

dataset_id : str Knowledge Base ID document_id : str Document ID to get the file information for download_to_temp : bool, optional Whether to download the file to a temporary directory, by default True

Returns

Download url

Raises

requests.exceptions.HTTPError If the API request fails RuntimeError If file download fails

Get a list of documents in a dataset.

Parameters

dataset_id : str Knowledge Base ID page : int Page number for pagination limit : int Number of items to return per page (1-100)

Returns

DifyGetDocumentsResponse Response containing the list of documents and pagination info

Raises

requests.exceptions.HTTPError If the API request fails

Get or create the access right metadata in Dify.

Retrieve chunks from a Knowledge Base (dataset).

This method retrieves relevant chunks from the specified dataset based on the query and search parameters.

Parameters

dataset_id : str Knowledge Base ID query : str Query keyword to search for search_method : Literal['keyword_search', 'semantic_search', 'full_text_search', 'hybrid_search'], optional Search method to use, by default 'semantic_search' reranking_enable : bool, optional Whether to enable reranking, by default False reranking_provider_name : Optional[str], optional Rerank model provider, required if reranking is enabled reranking_model_name : Optional[str], optional Rerank model name, required if reranking is enabled weights : Optional[float], optional Semantic search weight setting in hybrid search mode top_k : Optional[int], optional Number of results to return score_threshold_enabled : bool, optional Whether to enable score threshold, by default False score_threshold : Optional[float], optional Score threshold value, used if score_threshold_enabled is True

Returns

DifyChunksResponse Response containing query info and retrieved chunks

Raises

requests.exceptions.HTTPError If the API request fails

Call the Dify API chat endpoint with streaming response.

Args: query: The user's message conversation_id: Optional ID for continuing a conversation user: Optional user identifier inputs: Optional dictionary of input variables files: Optional list of files to include

Returns: Generator that yields response chunks (string or objects)

Update a document in a dataset. Parameters

doc_path : str Path to the document file dataset_id : str Knowledge Base ID document_id : str Document ID to update options : DifyUpdateDocumentOptions Options for updating the document Returns

DifySendDocumentResponse Response containing the updated document information Raises

requests.exceptions.HTTPError If the API request fails

Update metadata for a specific document in a dataset.

Parameters

dataset_id : str Knowledge Base ID document_id : str Document ID to update metadata for metadata : Dict[str, Any] Metadata to update

Raises

requests.exceptions.HTTPError If the API request fails