Update app.py

app.py

@@ -1,88 +1,125 @@
-# app.py
 import os
+import warnings
 from dotenv import load_dotenv
-from fastapi import FastAPI, HTTPException
-from fastapi.middleware.cors import CORSMiddleware
-from pydantic import BaseModel
+import gradio as gr
+
+
 from qdrant_search import QdrantSearch
 from langchain_groq import ChatGroq
 from nomic_embeddings import EmbeddingsModel
 
+
+
+
+# Load environment variables from .env file
 load_dotenv()
 
-import warnings
+# Suppress FutureWarnings
 warnings.filterwarnings("ignore", category=FutureWarning)
 
+# Disable tokenizers parallelism to avoid potential issues
 os.environ["TOKENIZERS_PARALLELISM"] = "FALSE"
 
-# Initialize FastAPI app
-app = FastAPI()
 
-# Allow CORS for frontend on Vercel
-app.add_middleware(
-    CORSMiddleware,
-    allow_origins=["*"],  # Replace "*" with your frontend URL for better security
-    allow_credentials=True,
-    allow_methods=["*"],
-    allow_headers=["*"],
-)
+
+
+
+
+
+
+
+
+
+
 
 # Initialize global variables
 collection_names = ["docs_v1_2", "docs_v2_2", "docs_v3_2"]
 limit = 5
-llm = ChatGroq(model="mixtral-8x7b-32768")
-embeddings = EmbeddingsModel()
-search = QdrantSearch(
-    qdrant_url=os.environ["QDRANT_CLOUD_URL"],
-    api_key=os.environ["QDRANT_API_KEY"],
     embeddings=embeddings
 )
 
-# Define request and response models
-class QueryRequest(BaseModel):
-    question: str
+def chat_endpoint(question: str):
+    """
+    Handles the chat functionality by processing the user's question,
+    retrieving relevant documents, generating an answer, and returning sources.
+
+    Args:
+        question (str): The user's question.
 
-class AnswerResponse(BaseModel):
-    answer: str
-    sources: list
 
-# API endpoint to handle user queries
-@app.post("/api/chat", response_model=AnswerResponse)
-async def chat_endpoint(request: QueryRequest):
-    query = request.question.strip()
+    Returns:
+        Tuple[str, str]: The generated answer and the sources used.
+    """
+    query = question.strip()
     if not query:
-        raise HTTPException(status_code=400, detail="Query cannot be empty.")
+        return "❌ **Error:** Query cannot be empty.", "No sources available."
 
     # Step 1: Retrieve relevant documents from Qdrant
     retrieved_docs = search.query_multiple_collections(query, collection_names, limit)
 
+    if not retrieved_docs:
+        return "⚠️ **No relevant documents found** for your query.", "No sources available."
+
+
+
+
     # Step 2: Prepare the context from retrieved documents
-    context = "\n".join([doc['text'] for doc in retrieved_docs])
-
-    # Step 3: Construct the prompt with context and question
-    prompt = (
-        "You are LangAssist, a knowledgeable assistant for the LangChain Python Library. "
-        "Given the following context from the documentation, provide a helpful answer to the user's question.\n\n"
-        "Context:\n{context}\n\n"
-        "Question: {question}\n\n"
-        "Answer:"
-    ).format(context=context, question=query)
-
-    # Step 4: Generate an answer using the language model
+    context = "\n\n".join([doc['text'] for doc in retrieved_docs])
     try:
         answer = llm.invoke(prompt)
     except Exception as e:
-        raise HTTPException(status_code=500, detail=str(e))
+        return f"⚠️ **Error generating answer:** {str(e)}", "No sources available."
 
     # Prepare sources
-    sources = [
-        {
-            "source": doc['source'],
-            "text": doc['text']
-        } for doc in retrieved_docs
-    ]
+    sources_md = "\n\n".join([
+        f"**Source:** {src['source']}\n**Excerpt:** {src['text']}"
+        for src in retrieved_docs
+    ])
+
+    return answer.content.strip(), sources_md
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+# Create Gradio Interface
+interface = gr.Interface(
+    fn=chat_endpoint,
+    inputs=gr.Textbox(
+        lines=2,
+        placeholder="Type your question here...",
+    description="Ask questions about the LangChain Python Library and get answers based on the latest documentation."
+)
+
+# If running locally, uncomment the following lines:
+
+
+
+
+
+
+
 
-    # Step 5: Return the answer and sources
-    # return AnswerResponse(answer=answer.strip(), sources=sources)
-    return AnswerResponse(answer=answer.content.strip(), sources=sources)
 
+# if __name__ == "__main__":
+#     interface.launch()