app.py (Streamlit interface)

import streamlit as st
from pipeline import process_pdf, enrich_metadata, semantic_split, build_vector_db, load_vector_db, get_qa_chain
import tempfile, os

st.set_page_config(page_title="Bangladesh Law RAG", layout="wide")

st.title("📜 Bangladesh Law Assistant")
st.markdown("Ask questions about Bangladesh ICT, Labour, Penal Code, and Constitution documents.")

# Upload and process PDF
with st.sidebar:
    st.header("📂 Upload Legal PDF")
    uploaded_file = st.file_uploader("Choose a PDF", type="pdf")
    if uploaded_file:
        with tempfile.NamedTemporaryFile(delete=False, suffix=".pdf") as tmp:
            tmp.write(uploaded_file.read())
            path = tmp.name
        st.success("PDF uploaded. Processing...")
        pages = process_pdf(path)
        docs = enrich_metadata(pages)
        chunks = semantic_split(docs)
        build_vector_db(chunks)
        st.success("Vector DB updated successfully.")

# Load vector DB
vectorstore = load_vector_db()
qa_chain = get_qa_chain(vectorstore)

# Query box
query = st.text_input("🔎 Enter your legal query:")
if query:
    result = qa_chain({"query": query})
    answer = result["result"]
    sources = result["source_documents"]

    st.markdown("### 🧠 Answer")
    st.write(answer)

    st.markdown("### 📚 Sources")
    for doc in sources:
        meta = doc.metadata
        st.markdown(f"- `{meta.get('law_name', 'Unknown')} - {meta.get('section_heading', 'Unknown')}` ({meta.get('source')})")