JanusGraph Integration

[zhaoJoseph]

Checked

• I searched existing ideas and did not find a similar one
• I added a very descriptive title
• I've clearly described the feature request and motivation for it

Feature request

We are a team of CS students at UofT and propose to create an integration of the graph database system JanusGraph into Langchain, we will submit a pull request some time around mid November.

Motivation

We have found partial solutions for integration and requests for integration support here;
#16082, #23957

Proposal (If applicable)

Our implementation will provide similar support as done for the graph database system Neo4J such as loading a graph from GraphDocument objects. Some key integrations will be querying the graph using an LLM to return a result, generating Gremlin statements for specific prompts such as adding new nodes to a graph, ignoring/including nodes in a graph while querying and passing tools to function outputs.

Pseudocode:

from langchain_community.graphs.graph_document import GraphDocument, Node, Relationship
from langchain_community.graphs.graph_store import GraphStore

class JanusGraph(GraphStore):

	Parameters:
	url (Optional[str]): The URL of the JanusGraph database server.
	username (Optional[str]): The username for database authentication.
	password (Optional[str]): The password for database authentication.

	def __init__(
    	  self,
    	  url: Optional[str] = None,
    	  username: Optional[str] = None,
    	  password: Optional[str] = None,
	) -> None:
    	"""Create a new JanusGraph wrapper instance."""
        # import gremlinpython
        # connect to database
    	self,
    	url: Optional[str] = None,
    	username: Optional[str] = None,
    	password: Optional[str] = None,
        # set schema

	def query(
    	  self,
    	  query: str,
    	  params: dict = {},
	) -> List[Dict[str, Any]]:
    	"""Query JanusGraph database.

    	Args:
        	query (str): The Gremlin query to execute.
        	params (dict): The parameters to pass to the query.

    	Returns:
        	List[Dict[str, Any]]: The list of dictionaries containing the query results.
    	"""
	pass

	@property
	def get_structured_schema(self) -> Dict[str, Any]:
    	"""Returns the structured schema of the Graph"""
    		return self.structured_schema
	
	def refresh_schema(self) -> None:
    	"""
    	Refreshes the JanusGraph graph schema information.
    	"""
	pass

	def add_graph_documents(
    	self,
    	graph_documents: List[GraphDocument],
    	include_source: bool = False,
    	baseEntityLabel: bool = False,
	) -> None:
    	"""
    	This method constructs nodes and relationships in the graph based on the
    	provided GraphDocument objects.

    	Parameters:
    	- graph_documents (List[GraphDocument]): A list of GraphDocument objects
    	that contain the nodes and relationships to be added to the graph. Each
    	GraphDocument should encapsulate the structure of part of the graph,
    	including nodes, relationships, and the source document information..
    	"""
	pass 

	
        def add_edge(self, relationship: Relationship) -> Any:
	        “”” Add edge to graph“””
	        pass
	        
        def add_node(self, node: Node, node_cache: dict = {}) -> Node:
	        “””Add node to graph“””
	        pass

Files Added/Modified:

langchain-ai/langchain/blob/master/libs/community/langchain_community/graphs/janus_graph.py
langchain-ai/langchain/blob/master/libs/community/tests/unit_tests/graphs/test_janus_graph.py
langchain-ai/langchain/blob/master/docs/docs/integrations/graphs/janusgraph_gremlin.ipynb

Steps:

1. Add implementation for JanusGraph integration
2. Add unit testing for each features added and loading JanusGraph database
3. Add documentation for all features in python notebook and how to integrate JanusGraph and documentation for each method in file