src/buildgraph.py

from src.graph import *
from langgraph.graph import END, StateGraph, START
import sys
from langgraph.checkpoint.memory import MemorySaver
from langgraph.errors import GraphRecursionError


memory = MemorySaver()

try:
    print("Initializing StateGraph...")
    workflow = StateGraph(GraphState)

    print("Adding nodes to the graph...")
    workflow.add_node("understand_intent", understand_intent)
    # workflow.add_node("intent_aware_response", intent_aware_response)
    workflow.add_node("greeting", greeting)
    workflow.add_node("off_topic", off_topic)
    workflow.add_node("route_question", route_question)
    workflow.add_node("retrieve", retrieve)
    workflow.add_node("web_search", web_search)
    workflow.add_node("grade_documents", grade_documents)
    workflow.add_node("generate", generate)
    workflow.add_node("transform_query", transform_query)
    workflow.add_node("grade_generation", grade_generation_v_documents_and_question)

    print("Nodes added successfully.")

    print("Building graph edges...")
    workflow.add_edge(START, "understand_intent")
    workflow.add_conditional_edges(
        "understand_intent",
        intent_aware_response,
        {
            "off_topic": "off_topic",
            "greeting": "greeting",
            "route_question": "route_question",
        }
)

    workflow.add_edge("greeting", END)
    workflow.add_edge("off_topic", END)
    
    workflow.add_conditional_edges(
        "route_question",
        lambda x: x["route_question"],
        {
            "web_search": "web_search",
            "vectorstore": "retrieve",
        }   
    )
    
    workflow.add_edge(
        "retrieve", "grade_documents",
    )
    
    workflow.add_edge(
        "generate", "grade_generation",
    )
    
    workflow.add_conditional_edges(
        "grade_generation",
        lambda x: x["grade_generation"],
        {
            "not supported": "generate",
            "useful": END,
            "not useful": "transform_query",
        }
    )
    
    workflow.add_edge("transform_query", "route_question")

    workflow.add_conditional_edges(
        "grade_documents",
        decide_to_generate,
        {
            "transform_query": "transform_query",
            "generate": "generate",
        },
    )
    
    workflow.add_edge("web_search", "generate")

    print("Graph edges built successfully.")


    print("Compiling the workflow...")
    app = workflow.compile(checkpointer=memory)
    print("Workflow compiled successfully.")
    
    try:
        from IPython import get_ipython
        from IPython.display import Image, display

        # Check if we're in an IPython environment
        if get_ipython() is not None:
            print("Attempting to display graph visualization...")
            graph_image = app.get_graph().draw_mermaid_png()
            display(Image(graph_image))
            print("Graph visualization displayed successfully.")
        else:
            print("Not running in IPython environment. Saving graph as JPG...")
            import os
            from PIL import Image
            import io

            graph_image = app.get_graph().draw_mermaid_png()
            img = Image.open(io.BytesIO(graph_image))
            img = img.convert('RGB')
            
            # Create a 'graphs' directory if it doesn't exist
            if not os.path.exists('graphs'):
                os.makedirs('graphs')
            
            img.save('graphs/workflow_graph.jpg', 'JPEG')
            print("Graph saved as 'graphs/workflow_graph.jpg'")
    except ImportError as e:
        print(f"Required libraries not available. Graph visualization skipped. Error: {e}")
    except Exception as e:
        print(f"Error handling graph visualization: {e}")
        print("Graph visualization skipped.")
    
except GraphRecursionError:
    print("Graph recursion limit reached during compilation.")
    # Handle the error as needed
except Exception as e:
    print(f"Error building the graph: {e}")
    sys.exit(1)

def run_workflow(user_input, config):
    try:
        print(f"Running workflow for question: {user_input}")

        # Ensure user_input is a string, not a dict
        if isinstance(user_input, dict):
            print("user_input is a dict, extracting content")
            user_input = user_input.get('content', str(user_input))

        print(f"Processed user_input: {user_input}")

        # Initialize input_state with required fields
        input_state = {
            "messages": [{"role": "user", "content": user_input}]
        }
        
        print(f"Initial input state: {input_state}")

        use_web_search = False
        final_output = None
        
        try:
            print("Starting graph execution")
            for output in app.stream(input_state, config):
                # print(f"Graph output: {output}")
                for key, value in output.items():
                    print(f"Node '{key}'")
                    if key in ["grade_generation", "off_topic", "greeting", "web_search"]:
                        print(f"Setting final_output from node '{key}'")
                        final_output = value
            print("Graph execution completed")
        except GraphRecursionError:
            print("Graph recursion limit reached, switching to web search")
            use_web_search = True
        
        if use_web_search:
            print("Executing web search fallback")
            web_search_result = web_search(input_state)
            print(f"Web search result: {web_search_result}")
            generate_result = generate(web_search_result)
            print(f"Generate result: {generate_result}")
            final_output = generate_result
        
        print(f"Final output before processing: {final_output}")
        
        if final_output is None:
            print("No final output generated")
            return {"generation": "I'm sorry, I couldn't generate a response. Could you please rephrase your question?"}
        elif isinstance(final_output, dict) and "generation" in final_output:
            print("Final output is a dict with 'generation' key")
            return {"generation": str(final_output["generation"])}
        elif isinstance(final_output, str):
            print("Final output is a string")
            return {"generation": final_output}
        else:
            print(f"Unexpected final output type: {type(final_output)}")
            return {"generation": str(final_output)}

    except Exception as e:
        print(f"Error running the workflow: {e}")
        print("Full traceback:")
        import traceback
        traceback.print_exc()
        return {"generation": "I encountered an error while processing your question. Please try again."}

if __name__ == "__main__":
    config = {"configurable": {"thread_id": "1"}}
    while True:
        question = input("Enter your question (or 'quit' to exit): ")
        if question.lower() == 'quit':
            break
        result = run_workflow(question, config)
        print("Chatbot:", result["generation"])