LangChain OpenDMA

LangChain document loaders and retrievers for OpenDMA.

OpenDMA is a vendor-neutral abstraction layer for enterprise content management systems. It provides a common API for repositories such as Alfresco, CMOD, Documentum, FileNet P8, OnBase, SharePoint, and other ECM or document management platforms. This package connects that API to LangChain by loading and retrieving OpenDMA documents as langchain_core.documents.Document objects.

Use this package when you want to build LangChain applications, RAG pipelines, or content analysis workflows on top of documents stored in ECM systems.

Features

• Load documents from an OpenDMA REST service by document ID, folder ID, or query.
• Retrieve documents from OpenDMA search results through LangChain's retriever API.
• Use specialized retrievers for Alfresco, Documentum, FileNet P8, and OnBase.
• Preserve OpenDMA and repository metadata on every LangChain Document.
• Process plain text content out of the box.
• Process richer document formats with optional Unstructured or Docling handlers.
• Use LangChain's sync and async document loader APIs.

Installation

Install OpenDMA and this integration from PyPI:

pip install langchain-opendma

Install optional parser integrations when you need Office, PDF, HTML, images, or other rich formats:

pip install "langchain-opendma[unstructured]"
pip install "langchain-opendma[docling]"
pip install "langchain-opendma[all]"

Quickstart

from langchain_opendma import OpenDMALoader

loader = OpenDMALoader(
    endpoint="http://localhost:8080/opendma",
    username="admin",
    password="admin",
    repository_id="my-repository",
    document_ids=["some-document-id"],
)

documents = loader.load()

for document in documents:
    print(document.metadata["source"])
    print(document.metadata.get("opendma:Title"))
    print(document.page_content)

By default, OpenDMALoader handles text/plain content. For PDF, Office, HTML, image, and other rich formats, configure an Unstructured or Docling content handler. See the documentation for details.

Use OpenDMARetriever when you want LangChain to call an OpenDMA search as part of a retrieval pipeline:

from langchain_opendma import OpenDMARetriever

retriever = OpenDMARetriever(
    endpoint="http://localhost:8080/opendma",
    username="admin",
    password="admin",
    repository_id="my-repository",
    query_language="opendma:sfts",
)

documents = retriever.invoke("needle keyword")

Documentation

• Tutorials: guided LangChain application tutorials
• Documentation: usage, loader options, and content handlers
• Examples: runnable examples using the tutorial repository

Development

This project uses uv for dependency management.

uv sync --all-extras
uv run pytest
uv run ruff check src tests
uv run mypy src tests

Related Projects

• LangChain
• OpenDMA
• opendma-api
• opendma-remote
• Unstructured
• Docling