Solr Vector Search

🚀 Solr MCP Introduction

A Python package for accessing Apache Solr indexes via the Model Context Protocol (MCP). This integration enables AI assistants like Claude to perform powerful search queries on your Solr indexes, combining keyword and vector search capabilities.

🚀 Quick Start

Prerequisites

• Python 3.10 or higher
• Docker and Docker Compose
• SolrCloud 9.x
• Ollama (for embedding generation)

Steps

1. Clone this repository.
2. Start SolrCloud using Docker:

   docker-compose up -d
   

3. Install dependencies:

   python -m venv venv
   source venv/bin/activate  # On Windows: venv\Scripts\activate
   pip install poetry
   poetry install
   

4. Process and index sample documents:

   python scripts/process_markdown.py data/bitcoin-whitepaper.md --output data/processed/bitcoin_sections.json
   python scripts/create_unified_collection.py unified
   python scripts/unified_index.py data/processed/bitcoin_sections.json --collection unified
   

5. Run the MCP server:

   poetry run python -m solr_mcp.server
   

For more detailed setup and usage instructions, please refer to the QUICKSTART.md guide.

✨ Features

• MCP Server: Implements the Model Context Protocol for integration with AI assistants.
• Hybrid Search: Combines the precision of keyword search with the semantic understanding of vector search.
• Vector Embeddings: Generates document embeddings using Ollama and nomic-embed-text.
• Unified Collection: Stores document content and vector embeddings in the same collection.
• Docker Integration: Enables easy installation via Docker and docker-compose.
• Optimized Vector Search: Pushes SQL filtering conditions to the vector search phase, efficiently handling combinations of vector and SQL queries, ensuring optimal performance even with large result sets and pagination.

🔧 Technical Details

Vector Search Optimization

The system employs an important optimization method to handle queries containing both vector similarity and SQL filtering conditions:

1. Pushes SQL filtering conditions (WHERE clause) to the vector search phase.
2. Ensures that vector similarity calculations are only performed on documents that ultimately meet the SQL criteria.
3. Significantly improves the performance of the following queries:
   • Selective WHERE clauses
   • Pagination (LIMIT/OFFSET)
   • Large result sets

This optimization minimizes the number of required vector similarity calculations by reducing computational overhead and network transmission.

📄 License

This project is licensed under the MIT License. Please see the LICENSE file for details.

💪 Contribution

Contributions are welcome! Please refer to the CONTRIBUTING.md guide for details.