Docling

% pip install novastack-loaders-docling

DoclingLoader

Bases: BaseFileLoader

A document loader that uses the docling library to extract and structure content from various file types including PDF, DOCX, and HTML.

For more information, see Docling

Attributes:

Name	Type	Description
detached_tables	bool	If True, separates extracted tables from the main document text and treats them as individual documents. Default is False.
export_table_format	str	Format used when exporting tables. Applicable only if detached_tables is True. Choose between "markdown" or "html". Defaults to "markdown".
input_file	str	File path to load.

Example

from novastack.loaders.docling import DoclingLoader

docling_loader = DoclingLoader(input_file="path/to/file.pdf")
documents = docling_loader.load_data()

load_data

load_data() -> list[Document]

Loads data and returns a list of documents.