Update app.py

app.py

@@ -1,104 +1,200 @@
-import os
-import fitz  # PyMuPDF for PDF processing
-import faiss
-import numpy as np
 import streamlit as st
-from langchain.text_splitter import RecursiveCharacterTextSplitter
-from sentence_transformers import SentenceTransformer
-from groq import Groq
-from dotenv import load_dotenv
-
-# Load API key
-load_dotenv()
-GROQ_API_KEY = os.getenv("GROQ_API_KEY")
-
-# Initialize Groq client
-client = Groq(api_key=GROQ_API_KEY)
-
-# Load sentence transformer model for embedding
-embedding_model = SentenceTransformer('sentence-transformers/all-MiniLM-L6-v2')
-
-def extract_text_from_pdf(pdf_path):
-    """Extract text from a PDF file using PyMuPDF."""
-    doc = fitz.open(pdf_path)
-    text = ""
-    for page in doc:
-        text += page.get_text("text") + "\n"
-    return text.strip()
-
-def create_text_chunks(text, chunk_size=500, chunk_overlap=100):
-    """Split text into chunks of specified size with overlap."""
-    text_splitter = RecursiveCharacterTextSplitter(
-        chunk_size=chunk_size,
-        chunk_overlap=chunk_overlap
-    )
-    chunks = text_splitter.split_text(text)
-    return chunks
-
-def create_faiss_index(chunks):
-    """Generate embeddings for text chunks and store them in FAISS."""
-    embeddings = embedding_model.encode(chunks, convert_to_numpy=True)
-    dimension = embeddings.shape[1]
-
-    index = faiss.IndexFlatL2(dimension)  # L2 (Euclidean) distance
-    index.add(embeddings)  # Add embeddings to FAISS index
-
-    return index, embeddings, chunks
-
-def retrieve_similar_chunks(query, index, embeddings, chunks, top_k=3):
-    """Retrieve the most relevant text chunks using FAISS."""
-    query_embedding = embedding_model.encode([query], convert_to_numpy=True)
-    distances, indices = index.search(query_embedding, top_k)
-
-    results = [chunks[idx] for idx in indices[0]]
-    return results
-
-def query_groq_api(query, context):
-    """Send the query along with retrieved context to Groq API."""
-    prompt = f"Use the following context to answer the question:\n\n{context}\n\nQuestion: {query}\nAnswer:"
-
-    chat_completion = client.chat.completions.create(
-        messages=[{"role": "user", "content": prompt}],
-        model="llama-3.3-70b-versatile",
-    )
-
-    return chat_completion.choices[0].message.content
-
-# Streamlit UI
-st.title("📚 RAG-based PDF Query Application")
-st.write("Upload a PDF and ask questions!")
-
-# File Upload
-uploaded_file = st.file_uploader("Upload PDF", type="pdf")
-
-if uploaded_file is not None:
-    pdf_path = "uploaded_document.pdf"
-
-    # Save file temporarily
-    with open(pdf_path, "wb") as f:
-        f.write(uploaded_file.getbuffer())
-
-    # Process the PDF
-    st.write("Processing PDF...")
-    text = extract_text_from_pdf(pdf_path)
-    chunks = create_text_chunks(text)
-    index, embeddings, chunk_texts = create_faiss_index(chunks)
-
-    st.success("PDF processed! Now you can ask questions.")
-
-    # User Query
-    query = st.text_input("Ask a question about the PDF:")
-
-    if st.button("Get Answer"):
-        if query:
-            # Retrieve top chunks
-            relevant_chunks = retrieve_similar_chunks(query, index, embeddings, chunk_texts)
-            context = "\n\n".join(relevant_chunks)
-
-            # Query Groq API
-            response = query_groq_api(query, context)
-
-            st.subheader("Answer:")
-            st.write(response)
-        else:
-            st.warning("Please enter a question.")
+import random
+import time
+from streamlit.components.v1 import html
+
+# Set page config with light purple theme
+st.set_page_config(
+    page_title="Emotion Mirror Chatbot",
+    page_icon="😊",
+    layout="centered",
+    initial_sidebar_state="collapsed"
+)
+
+# Custom CSS for light purple theme
+st.markdown("""
+<style>
+:root {
+    --primary: #b19cd9;
+    --background: #f5f0ff;
+    --secondary-background: #e6e0fa;
+    --text: #4a4a4a;
+    --font: "Arial", sans-serif;
+}
+
+body {
+    background-color: var(--background);
+    color: var(--text);
+    font-family: var(--font);
+}
+
+.stTextInput>div>div>input {
+    background-color: var(--secondary-background) !important;
+    color: var(--text) !important;
+}
+
+.stButton>button {
+    background-color: var(--primary) !important;
+    color: white !important;
+    border: none;
+    border-radius: 8px;
+    padding: 8px 16px;
+}
+
+.stMarkdown {
+    font-family: monospace !important;
+    font-size: 16px !important;
+}
+
+.chat-message {
+    padding: 12px;
+    border-radius: 12px;
+    margin: 8px 0;
+    max-width: 80%;
+}
+
+.user-message {
+    background-color: var(--secondary-background);
+    margin-left: auto;
+    text-align: right;
+}
+
+.bot-message {
+    background-color: var(--primary);
+    color: white;
+    margin-right: auto;
+}
+
+.face-container {
+    text-align: center;
+    padding: 20px;
+    background: white;
+    border-radius: 16px;
+    box-shadow: 0 4px 12px rgba(0,0,0,0.1);
+    margin: 20px 0;
+}
+</style>
+""", unsafe_allow_html=True)
+
+# Emotion databases
+POSITIVE_WORDS = {"happy", "awesome", "great", "joy", "excited", "good", "wonderful", "fantastic", "amazing"}
+NEGATIVE_WORDS = {"sad", "depressed", "angry", "cry", "lonely", "bad", "terrible", "awful", "miserable"}
+HELP_RESPONSES = [
+    "Would you like to talk about it? 💬",
+    "I'm here to listen 💙",
+    "Want some uplifting quotes? 📜",
+    "Would a virtual hug help? 🤗",
+    "Let's focus on something positive 🌈"
+]
+
+# ASCII Art Library
+FACES = {
+    "happy": r"""
+  ╔════════╗
+  😄 AWESOME!  
+  ╚════════╝
+    """,
+    "sad": r"""
+  ╔════════╗
+  😢 SAD DAY?  
+  ╚════════╝
+    """,
+    "neutral": r"""
+  ╔════════╗
+  😐 HELLO   
+  ╚════════╝
+    """,
+    "love": r"""
+  ╔════════╗
+  😍 LOVELY!  
+  ╚════════╝
+    """
+}
+
+# Confetti effect using JavaScript
+def confetti_effect():
+    confetti_js = """
+    <script src="https://cdn.jsdelivr.net/npm/[email protected]/dist/confetti.browser.min.js"></script>
+    <script>
+    const count = 200;
+    const defaults = {
+        origin: { y: 0.7 }
+    };
+
+    function fire(particleRatio, opts) {
+        confetti(Object.assign({}, defaults, opts, {
+            particleCount: Math.floor(count * particleRatio)
+        }));
+    }
+
+    fire(0.25, { spread: 26, startVelocity: 55 });
+    fire(0.2, { spread: 60 });
+    fire(0.35, { spread: 100, decay: 0.91, scalar: 0.8 });
+    fire(0.1, { spread: 120, startVelocity: 25, decay: 0.92, scalar: 1.2 });
+    fire(0.1, { spread: 120, startVelocity: 45 });
+    </script>
+    """
+    html(confetti_js)
+
+# Emotion detection function
+def detect_emotion(text):
+    text = text.lower()
+    if any(word in text for word in POSITIVE_WORDS):
+        return "happy"
+    elif any(word in text for word in NEGATIVE_WORDS):
+        return "sad"
+    elif "love" in text or "heart" in text:
+        return "love"
+    return "neutral"
+
+# Initialize chat history
+if "messages" not in st.session_state:
+    st.session_state.messages = []
+    st.session_state.current_emotion = "neutral"
+
+# Title and description
+st.title("✨ Emotion Mirror Chatbot")
+st.markdown("I'm a reactive AI agent that mirrors your emotions! Try words like *happy*, *sad*, or *awesome*")
+
+# Display current face
+with st.container():
+    st.markdown(f"<div class='face-container'>\n{FACES[st.session_state.current_emotion]}\n</div>", 
+                unsafe_allow_html=True)
+
+# Display chat messages
+for message in st.session_state.messages:
+    with st.chat_message(message["role"]):
+        st.markdown(f"<div class='chat-message {message['role']}-message'>{message['content']}</div>", 
+                    unsafe_allow_html=True)
+
+# User input
+if prompt := st.chat_input("How are you feeling today?"):
+    # Add user message to chat history
+    st.session_state.messages.append({"role": "user", "content": prompt})
+    
+    # Detect emotion
+    emotion = detect_emotion(prompt)
+    st.session_state.current_emotion = emotion
+    
+    # Generate bot response
+    if emotion == "happy":
+        response = FACES["happy"] + "\n\n🌟 That's wonderful to hear!"
+        confetti_effect()
+    elif emotion == "sad":
+        response = FACES["sad"] + "\n\n" + random.choice(HELP_RESPONSES)
+    elif emotion == "love":
+        response = FACES["love"] + "\n\n💖 Love is in the air!"
+    else:
+        response = FACES["neutral"] + "\n\nTell me more about your feelings..."
+    
+    # Add bot response to chat history
+    st.session_state.messages.append({"role": "bot", "content": response})
+    
+    # Rerun to update the display
+    st.experimental_rerun()
+
+# Add reset button
+if st.button("Reset Conversation"):
+    st.session_state.messages = []
+    st.session_state.current_emotion = "neutral"
+    st.experimental_rerun()