app.py

import os
import streamlit as st
from langchain_community.document_loaders import PyPDFLoader
from langchain.text_splitter import CharacterTextSplitter
from langchain_community.vectorstores import FAISS
from langchain.embeddings import HuggingFaceEmbeddings
from langchain.chains import RetrievalQA
from langchain_groq import ChatGroq  # NEW import

# -------------------------------
# Sidebar for API key input
# -------------------------------
st.set_page_config(page_title="Groq PDF Chatbot")
st.title("📄 Chat with your PDF using Groq + LLaMA3")

api_key = st.sidebar.text_input("🔑 Enter your Groq API Key", type="password")
if not api_key:
    st.warning("Please enter your Groq API key in the sidebar.")
    st.stop()

os.environ["GROQ_API_KEY"] = api_key

# -------------------------------
# Load and split the PDF
# -------------------------------
pdf_path = "docs/acca.pdf"  # Make sure this file is in your Space
loader = PyPDFLoader(pdf_path)
pages = loader.load()

text_splitter = CharacterTextSplitter(chunk_size=1000, chunk_overlap=200)
docs = text_splitter.split_documents(pages)

# -------------------------------
# Vector store using FAISS
# -------------------------------
embeddings = HuggingFaceEmbeddings(model_name="sentence-transformers/all-MiniLM-L6-v2")
vectorstore = FAISS.from_documents(docs, embeddings)

# -------------------------------
# Groq LLM setup
# -------------------------------
llm = ChatGroq(
    temperature=0.7,
    model_name="llama3-8b-8192",  # Make sure to use correct lowercase name
    groq_api_key=api_key
)

qa_chain = RetrievalQA.from_chain_type(
    llm=llm,
    retriever=vectorstore.as_retriever(),
    return_source_documents=True
)

# -------------------------------
# User input and response
# -------------------------------
query = st.text_input("Ask a question based on the PDF:")
if query:
    with st.spinner("Generating answer..."):
        result = qa_chain.invoke(query)
        st.subheader("📌 Answer")
        st.write(result["result"])