A basic document parsing and loading utility - Database Interfaces

Overview

The docp-* project suite is designed as a comprehensive (doc)ument (p)arsing library. Built in CPython, it consolidates the capabilities of various lower-level libraries, offering a unified solution for parsing binary document structures.

The suite is extended by several sister projects, each providing unique functionality:

Project	Description
docp-core	Centralized core objects, functionality and settings.
docp-parsers	Parse binary documents (e.g. PDF, PPTX, etc.) into Python objects.
docp-loaders	Load a parsed document's embeddings into a Chroma vector database, for RAG-enabled LLM use.
docp-docling	Convert a PDF into Markdown format via wrappers to the docling libraries.
docp-dbi	Interfaces to document databases such as ChromaDB, and Neo4j (coming soon).

This library (docp-dbi) extends the document parsing capability by adding access to a ChromaDB vector database for storing text embeddings. This is particularly useful for implementing RAG-enabled pipelines.

The Toolset (Interfaces)

As of this release, the following database interfaces are supported:

• ChromaDB (via langchain_chroma)
• Neo4j (coming soon)

Quickstart

Installation

To install docp-dbi, first activate your target virtual environment, then use pip:

pip install docp-dbi

For older releases, visit PyPI or the GitHub Releases page.

Example Usage

For convenience, here are a couple examples for how to create and interact with a database interface for your project.

Create an interface to ChromaDB:

    >>> from docp_dbi import ChromaDB

    # Create a database interface.
    >>> db = ChromaDB(path='/path/to/chromadb', collection='test-collection')

    # Display a list of all collections in the database.
    >>> db.client.list_collections()

    # Debug: Retrieve records from the database.
    >>> records = db.show_all()

Load a new PDF document into the database, and query against it:

    >>> from docp_dbi import ChromaDB
    >>> from docp_parsers import PDFParser
    >>> from langchain_text_splitters import RecursiveCharacterTextSplitter

    # Parse the PDF document.
    >>> pdf = PDFParser(path='/path/to/documents/rag-pipelines-how-to.pdf')
    >>> pdf.extract_text()

    # Setup a text splitter (for chunking the document).
    >>> splitter = RecursiveCharacterTextSplitter(
    ...     separators=['\n\n\n', '\n\n', '\n', '.'],
    ...     chunk_size=512,
    ...     chunk_overlap=128
    ... )
    # Split the document for storage.
    >>> docs = splitter.split_documents(pdf.doc.documents)

    # Create a database interface, using an offline, local user-defined embedding model.
    >>> db = ChromaDB(path='/path/to/databases/chroma/', 
    ...               collection='test', 
    ...               embedding_model_path='/path/to/models/sentence-transformers/all-MiniLM-L6-v2', 
    ...               offline=True)
    # Embed and store the document chunks.
    >>> db.add_documents(documents=docs)

    # Run your first query.
    >>> result = db.collection.query(query_texts=['How do I implement a RAG pipeline?'])

Using the Library

The documentation suite provides detailed explanations and usage examples for each importable module. For in-depth documentation, code examples, and source links, refer to the Library API page.

A search field is available in the left navigation bar to help you quickly locate specific modules or methods.

Troubleshooting

No troubleshooting guidance is available at this time.

For questions not covered here, or to report bugs, issues, or suggestions, please open an issue on GitHub.