Multi-Format Conversion

Convert mixed document types with format-specific settings

Process multiple file formats (PDF, DOCX, PPTX, HTML, images, etc.) with customized handling per format.

Overview

This example shows how to:

• Convert a mixed list of file formats
• Restrict allowed formats with an explicit whitelist
• Override pipeline and backend settings per format
• Export results to Markdown, JSON, and YAML

Basic Multi-Format Conversion

run_with_formats.py

from pathlib import Path
import json
import yaml
from docling.datamodel.base_models import InputFormat
from docling.backend.pypdfium2_backend import PyPdfiumDocumentBackend
from docling.document_converter import (
    DocumentConverter,
    PdfFormatOption,
    WordFormatOption,
)
from docling.pipeline.simple_pipeline import SimplePipeline
from docling.pipeline.standard_pdf_pipeline import StandardPdfPipeline

input_paths = [
    Path("README.md"),
    Path("tests/data/html/wiki_duck.html"),
    Path("tests/data/docx/word_sample.docx"),
    Path("tests/data/pptx/powerpoint_sample.pptx"),
    Path("tests/data/2305.03393v1-pg9-img.png"),
    Path("tests/data/pdf/2206.01062.pdf"),
]

Configure Format-Specific Options

1. Whitelist Formats

   Specify which formats to process. Non-matching files are ignored.

2. Override Per-Format Settings

   Customize pipeline and backend for specific formats.

3. Convert All Documents

   Process the mixed document list.

doc_converter = DocumentConverter(
    allowed_formats=[
        InputFormat.PDF,
        InputFormat.IMAGE,
        InputFormat.DOCX,
        InputFormat.HTML,
        InputFormat.PPTX,
        InputFormat.ASCIIDOC,
        InputFormat.CSV,
        InputFormat.MD,
    ],
    format_options={
        InputFormat.PDF: PdfFormatOption(
            pipeline_cls=StandardPdfPipeline,
            backend=PyPdfiumDocumentBackend
        ),
        InputFormat.DOCX: WordFormatOption(
            pipeline_cls=SimplePipeline
        ),
    },
)

conv_results = doc_converter.convert_all(input_paths)

Files not in allowed_formats are silently skipped during conversion.

Export Results

output_dir = Path("scratch")
output_dir.mkdir(parents=True, exist_ok=True)

for res in conv_results:
    doc_filename = res.input.file.stem
    
    # Export to Markdown
    with (output_dir / f"{doc_filename}.md").open("w") as fp:
        fp.write(res.document.export_to_markdown())
    
    # Export to JSON
    with (output_dir / f"{doc_filename}.json").open("w") as fp:
        fp.write(json.dumps(res.document.export_to_dict()))
    
    # Export to YAML
    with (output_dir / f"{doc_filename}.yaml").open("w") as fp:
        fp.write(yaml.safe_dump(res.document.export_to_dict()))
    
    print(f"Converted {res.input.file.name}")

Format Options

PDF with PyPdfium

from docling.backend.pypdfium2_backend import PyPdfiumDocumentBackend
from docling.pipeline.standard_pdf_pipeline import StandardPdfPipeline

InputFormat.PDF: PdfFormatOption(
    pipeline_cls=StandardPdfPipeline,
    backend=PyPdfiumDocumentBackend
)

Supported Formats

Docling supports:

• PDF: Native and scanned PDFs
• DOCX: Microsoft Word documents
• PPTX: PowerPoint presentations
• HTML: Web pages
• Images: PNG, JPG, TIFF
• Markdown: MD files
• AsciiDoc: ASCIIDOC files
• CSV: Comma-separated values

Default vs Custom Configuration

# Default: No explicit configuration needed
doc_converter = DocumentConverter()

# Custom: Override specific formats
doc_converter = DocumentConverter(
    allowed_formats=[InputFormat.PDF, InputFormat.DOCX],
    format_options={
        InputFormat.PDF: PdfFormatOption(...),
    },
)