langchain-opendataloader-pdf

This package integrates the OpenDataLoader PDF engine with LangChain by providing a document loader which parses PDFs into structured Document objects.

Requirements

• Python >= 3.9
• Java 11 or newer available on the system PATH
• opendataloader-pdf >= 1.1.0

Installation

pip install -U langchain-opendataloader-pdf

Quick start

from langchain_opendataloader_pdf import OpenDataLoaderPDFLoader

loader = OpenDataLoaderPDFLoader(
    file_path=["path/to/document.pdf", "path/to/folder"], 
    format="text"
)
documents = loader.load()

for doc in documents:
    print(doc.metadata, doc.page_content[:80])

Parameters

Parameter	Type	Required	Default	Description
file_path	List[str]	✅ Yes	—	One or more PDF file paths or directories to process.
format	str	No	None	Output formats (e.g. "json", "html", "markdown", "text").
quiet	bool	No	False	Suppresses CLI logging output when True.
content_safety_off	Optional[List[str]]	No	None	List of content safety filters to disable (e.g. "all", "hidden-text", "off-page", "tiny", "hidden-ocg").

Development workflow

This repository uses Poetry for dependency management. If you don't have Poetry installed, please follow the official installation guide.

Once Poetry is installed, you can install the project dependencies:

poetry install --with dev

Common tasks are mirrored in the Makefile so you can run them with or without Poetry.

Quality checks

make lint      # ruff + mypy
make test      # unit test suite (network disabled)
make integration_tests  # runs tests that may touch the network

You can also call the underlying Poetry commands directly (e.g., poetry run pytest).

Note for Windows Users:

If the make command is not available on your system, you can run the quality checks using the following commands directly:

• Linting:

  poetry run ruff check .
  poetry run mypy .

• Unit Tests:

  poetry run pytest --disable-socket --allow-unix-socket

• Integration Tests:

  poetry run pytest

Publishing notes

Run poetry check and poetry build to verify the package metadata before uploading to PyPI. Confirm that langchain_opendataloader_pdf/py.typed is present in the wheel so consumers benefit from typing information.

License

Distributed under the MIT License. See LICENSE for full text.