autogen.agents.experimental.document_agent.docling_parse_docs

docling_parse_docs

docling_parse_docs(
    input_file_path: Annotated[Path | str, 'Path to the input file or directory'],
    output_dir_path: Annotated[Path | str | None, 'Path to the output directory'] = None,
    output_formats: Annotated[list[str] | None, 'List of output formats (markdown, json)'] = None,
    table_output_format: str = 'html'
) -> list[Path]

Convert documents into a Deep Search document format using EasyOCR with CPU only, and export the document and its tables to the specified output directory.
Supported formats:
PDF, IMAGE, DOCX, HTML, PPTX, ASCIIDOC, MD,

Parameters:

Name	Description
input_file_path	The path to the input file. Type: Annotated[pathlib.Path | str, ‘Path to the input file or directory’]
output_dir_path	The path to the output directory. Type: Annotated[pathlib.Path | str | None, ‘Path to the output directory’] Default: None
output_formats	The output formats. Defaults to [“markdown”]. Type: Annotated[list[str] | None, ‘List of output formats (markdown, json)‘] Default: None
table_output_format	The output format for tables. Defaults to “html”. Type: str Default: ‘html’

Returns:

Type	Description
list[pathlib.Path]	list[ConversionResult]: The result of the conversion.