app.py

import streamlit as st
import openai
import os
from langchain_community.document_loaders import PyPDFLoader
from langchain.text_splitter import RecursiveCharacterTextSplitter
from langchain.embeddings import OpenAIEmbeddings
from langchain.vectorstores import FAISS
from openai import OpenAI
import base64
import requests

# Initialize clients
client = OpenAI(api_key=os.getenv("OPENAI_API_KEY"))
elevenlabs_key = os.getenv("ELEVENLABS_API_KEY")
tavily_key = os.getenv("TAVILY_API_KEY")

# Initialize session state
if "messages" not in st.session_state:
    st.session_state.messages = []
if "document_vectors" not in st.session_state:
    st.session_state.document_vectors = None

# App title
st.title("🤖 SuperBot Pro")
st.caption("An AI Assistant with Superpowers")

# ===== Sidebar Settings =====
with st.sidebar:
    st.header("⚙️ Settings")
    
    # Personality and Mode
    tone = st.selectbox("Personality:", ["Assistant", "Sarcastic", "Academic", "Shakespeare"])
    mode = st.radio("Mode:", ["Chat", "Document Q&A", "Web Researcher"])
    
    # Advanced Features
    web_access = st.checkbox("Enable Web Search", value=False)
    voice_enabled = st.checkbox("Enable Voice Response", value=False)
    
    # File Uploaders
    uploaded_file = st.file_uploader("Upload Document (PDF)", type=["pdf"])
    uploaded_image = st.file_uploader("Upload Image", type=["jpg", "png"])

# ===== Document Processing =====
if uploaded_file and mode == "Document Q&A":
    with st.spinner("Processing document..."):
        # Save and load PDF
        with open(uploaded_file.name, "wb") as f:
            f.write(uploaded_file.getbuffer())
        loader = PyPDFLoader(uploaded_file.name)
        pages = loader.load()
        
        # Split and embed
        text_splitter = RecursiveCharacterTextSplitter(chunk_size=1000, chunk_overlap=200)
        texts = text_splitter.split_documents(pages)
        embeddings = OpenAIEmbeddings()
        st.session_state.document_vectors = FAISS.from_documents(texts, embeddings)

# ===== Chat Interface =====
for message in st.session_state.messages:
    with st.chat_message(message["role"]):
        st.markdown(message["content"])
        if "audio" in message:
            st.audio(message["audio"], format="audio/mp3")

# Input area
if prompt := st.chat_input("Ask me anything..."):
    # Add user message to history
    st.session_state.messages.append({"role": "user", "content": prompt})
    
    with st.chat_message("user"):
        st.markdown(prompt)
    
    # ===== Generate Response =====
    with st.chat_message("assistant"):
        response_placeholder = st.empty()
        full_response = ""
        
        try:
            # ===== Content Moderation =====
            moderation = client.moderations.create(input=prompt)
            if moderation.results[0].flagged:
                st.error("Content violates policies")
                st.stop()
            
            # ===== System Prompt Engineering =====
            system_prompt = f"You are a {tone} assistant. Respond concisely."
            
            if mode == "Document Q&A" and st.session_state.document_vectors:
                docs = st.session_state.document_vectors.similarity_search(prompt, k=3)
                system_prompt += f"\nDocument context: {[doc.page_content for doc in docs]}"
            
            if web_access and tavily_key:
                search_response = requests.post(
                    "https://api.tavily.com/search",
                    json={"query": prompt, "api_key": tavily_key}
                )
                web_results = search_response.json()["results"]
                system_prompt += f"\nWeb results: {web_results[:2]}"
            
            # ===== Generate with OpenAI =====
            response = client.chat.completions.create(
                model="gpt-4",
                messages=[
                    {"role": "system", "content": system_prompt},
                    *st.session_state.messages
                ],
                stream=True
            )
            
            # Stream response
            for chunk in response:
                if chunk.choices[0].delta.content:
                    full_response += chunk.choices[0].delta.content
                    response_placeholder.markdown(full_response + "▌")
            
            response_placeholder.markdown(full_response)
            
            # ===== Voice Output =====
            if voice_enabled and elevenlabs_key:
                audio_response = requests.post(
                    "https://api.elevenlabs.io/v1/text-to-speech/21m00Tcm4TlvDq8ikWAM",
                    headers={"xi-api-key": elevenlabs_key},
                    json={"text": full_response}
                )
                st.audio(audio_response.content, format="audio/mp3")
            
            # Add to history
            st.session_state.messages.append({
                "role": "assistant", 
                "content": full_response,
                "audio": audio_response.content if voice_enabled else None
            })
            
        except Exception as e:
            st.error(f"Error: {str(e)}")

# ===== Feedback Buttons =====
col1, col2 = st.columns(2)
with col1:
    if st.button("👍 Good Response"):
        st.toast("Thanks for your feedback!")
with col2:
    if st.button("👎 Needs Improvement"):
        st.toast("We'll do better next time!")

# ===== Image Processing =====
if uploaded_image:
    st.image(uploaded_image, caption="Uploaded Image")
    with st.spinner("Analyzing image..."):
        response = client.chat.completions.create(
            model="gpt-4-vision-preview",
            messages=[{
                "role": "user",
                "content": [
                    {"type": "text", "text": "Describe this image"},
                    {"type": "image_url", "image_url": {"url": f"data:image/jpeg;base64,{base64.b64encode(uploaded_image.getvalue()).decode('utf-8')}"}}
                ]
            }]
        )
        st.write(response.choices[0].message.content)