Skip to content
This repository was archived by the owner on Jan 21, 2026. It is now read-only.

Releases: MegaGrindStone/go-light-rag

v0.1.2

05 Apr 23:57

Choose a tag to compare

This update expands database support with two new integrations: Redis for high-performance key-value operations and Milvus for vector similarity searches, enhancing the framework's flexibility for different data storage and retrieval requirements.

Added

  • Add Redis struct enabling integration with Redis key-value database for fast, in-memory data storage and retrieval operations.
  • Add Milvus struct for Milvus vector database integration, supporting efficient similarity searches and vector operations.

v0.1.1

31 Mar 23:07

Choose a tag to compare

This update introduces semantic text chunking capabilities through a new handler and improves code quality with comprehensive tests, while also addressing a concurrency setting issue that could impact processing performance.

Added

  • Add Semantic handler that intelligently chunks text documents based on content meaning rather than arbitrary size limits.
  • Add comprehensive unit tests for the handler package to ensure reliability and correctness.

Fixed

  • Fix incorrect default value in Default handler's concurrency count setting that could affect processing performance.

v0.1.0

24 Mar 21:49

Choose a tag to compare

This initial release introduces a comprehensive data access framework with support for multiple database types (graph, vector, and key-value) and AI service integrations. New functionality includes query and insertion operations, document processing capabilities for both standard text and Go source code, and standardized interfaces for future extensibility.

Added

  • Add Query function for database querying operations.
  • Add Insert function for database insertion operations.
  • Implement core interfaces required by Query and Insert functions.
  • Add Neo4J struct for Neo4j graph database integration.
  • Add Chromem struct for chromem-go vector database integration.
  • Add Bolt struct for bbolt key-value database integration.
  • Implement AI service integrations (Anthropic, Ollama, OpenAI, and OpenRouter) through the LLM interface.
  • Add Default struct for processing standard text documents.
  • Add Go struct for parsing and analyzing Go source code.