Releases: TigerGraph-DevLabs/tigergraph-mcp-utils
Releases · TigerGraph-DevLabs/tigergraph-mcp-utils
Release v0.2.0
- docs: add copywrie to all Python files; add document LICENSE
- perf: importing fewer classes in
tigergraphx/__init__.py
- feat: add support for TigerGraph APIs (ping, gsql, get_schema)
- feat: add support for TigerGraph APIs (run_interpreted_query)
- test: add unit test cases to class TigerGraphConnectionConfig
- fix: get_schema_from_db should consider vector attributes
- fix: add multi-edge support in
get_edge_data
- feat: add method
bfs
- feat: integrate Ragas for GraphRAG evaluation
- feat: add Ragas-based evaluation for LightRAG
- feat: add Ragas-based evaluation for MSFT GraphRAG
- perf: replace mkdocs-jupyter with jupyter nbconvert for faster ipynb-to-md conversion
- docs: add more examples in "TigerGraphX Quick Start: Using TigerGraph as Graph Database"
- docs: add more examples in "TigerGraphX Quick Start: Using TigerGraph as Vector Database"
- docs: add more examples in "TigerGraphX Quick Start: Using TigerGraph for Graph and Vector Database"
- docs: add evaluation section to LightRAG
- docs: add evaluation content to MSFT GraphRAG
- docs: add examples for query operations APIs
- docs: add BFS example by using method get_neighbors
- docs: add examples for vector operations APIs
Release v0.1.12
- docs: add CHANGELOG.md
- fix: improve error messages and logging for schema creation
- feat: add aliases to TigerGraphConnectionConfig for environment variables
- fix: correct the order of attributes when getting schema from TigerGraph
- fix: consider the discriminators in multi-edges when getting schema from TigerGraph
- fix: improve error messages and logging for data loading
- fix: support for integer node IDs
- fix: add alias to
get_nodes
andget_neighbors
methods - docs: add filtering on multiple attributes in the
get_nodes
example - fix: return empty DataFrame when the result is empty for
get_nodes
andget_neighbors
methods - fix: check the existence of the edge types for the method
degree
,get_node_edges
andget_neighbors
- fix: convert the types of
edge_types
andtarget_node_types
toSet
in theget_neighbors
method - fix: ensure undirected edges are counted once in
number_of_edges
method - docs: rewrite API Reference Introduction
- docs: add examples for schema operation APIs
- docs: add examples for data loading operation APIs
- docs: add examples for node operations APIs
- docs: add examples for edge operations APIs
- docs: add examples for statistics operations APIs
Release v0.1.11
TigerGraphX v0.1.11 – Initial Release
We are excited to announce the initial release of TigerGraphX version 0.1.11! This release marks the first step in delivering a robust connector designed to integrate seamlessly with TigerGraph databases for Graph-Powered RAG workflows.
Key Features
1. Schema Management
- Easily create and modify schemas using YAML, JSON, or Python dictionaries.
- No GSQL knowledge is required.
- Pythonic tools for designing database structures effortlessly.
2. Data Loading
- Automated loading jobs for streamlined data imports.
- High-efficiency workflows with support for Parquet files.
- Simplified data ingestion processes for faster setup.
3. Graph Library Interface
- Python-native APIs for CRUD operations.
- Comprehensive tools for graph reporting and visualization.
- Built-in graph algorithms including centrality, community detection, and path analysis algorithms
4. Graph Query Interface
- Simplified advanced querying with intuitive APIs.
- Seamless integration into analytics workflows via DataFrame outputs.
- Support for advanced multi-hop query traversal and manipulation
5. Vector Search Capabilities
- AI-driven applications with integrated vector embeddings.
- Efficient top-K entity retrieval for enhanced intelligence.
- Ideal for recommendation systems and contextual analysis.
6. LLM Integration and GraphRAG support
- Full support for GraphRAG workflows.
- Flexible, token-aware context builders for advanced applications.
- Tools for token optimization and seamless LLM integration.
7. Machine Learning Ready [Planned Feature]
- Seamless integration with popular ML libraries
- Graph feature extraction
- Native support for graph neural networks (GNNs)
Getting Started
To begin using TigerGraphX, please refer to our documentation for installation instructions and usage examples. We encourage you to try out the connector and provide feedback, report issues, or suggest new features via our GitHub repository.
Thank you for choosing TigerGraphX — we look forward to evolving this project with your insights and contributions.
Start unlocking the power of graphs with TigerGraphX today!