app.py

"""
🤖 AI Agent PDF Extractor with Google Gemini
Ultra-Modern 2025 UI Design with Glassmorphism & Bento Grid
"""

import streamlit as st
import google.generativeai as genai
from pathlib import Path
import PyPDF2
import tempfile
import time
import json
from datetime import datetime
from config import Config

# ============================================================================
# PAGE CONFIG
# ============================================================================

st.set_page_config(
    page_title="AI PDF Extractor | Gemini",
    page_icon="🤖",
    layout="wide",
    initial_sidebar_state="collapsed"
)

# Validate config
try:
    Config.validate()
    genai.configure(api_key=Config.GOOGLE_API_KEY)
except Exception as e:
    st.error(f"⚠️ Configuration Error: {e}")
    st.info("💡 Get your free API key: https://aistudio.google.com/app/apikey")
    st.stop()

# ============================================================================
# ULTRA-MODERN UI STYLING (2025 TRENDS)
# ============================================================================

st.markdown("""
<style>
    @import url('https://fonts.googleapis.com/css2?family=Inter:wght@300;400;500;600;700;800;900&display=swap');
    
    /* ========== GLOBAL RESET ========== */
    * {
        font-family: 'Inter', -apple-system, BlinkMacSystemFont, sans-serif;
        margin: 0;
        padding: 0;
    }
    
    /* Hide Streamlit Branding */
    #MainMenu, footer, header {visibility: hidden;}
    
    /* ========== ANIMATED GRADIENT BACKGROUND ========== */
    .main {
        background: linear-gradient(-45deg, #667eea, #764ba2, #f093fb, #4facfe);
        background-size: 400% 400%;
        animation: gradientShift 15s ease infinite;
        padding: 0;
        min-height: 100vh;
    }
    
    @keyframes gradientShift {
        0% {background-position: 0% 50%;}
        50% {background-position: 100% 50%;}
        100% {background-position: 0% 50%;}
    }
    
            /* ========== GLASSMORPHISM HEADER (FIXED VISIBILITY) ========== */
    .glass-header {
        background: linear-gradient(135deg, rgba(102, 126, 234, 0.95) 0%, rgba(118, 75, 162, 0.95) 100%);
        backdrop-filter: blur(20px) saturate(180%);
        -webkit-backdrop-filter: blur(20px) saturate(180%);
        border: 1px solid rgba(255, 255, 255, 0.3);
        padding: 3rem 3rem;
        border-radius: 24px;
        margin: 2rem auto 1rem auto;
        max-width: 1400px;
        box-shadow: 0 8px 32px 0 rgba(31, 38, 135, 0.5), 0 0 60px rgba(102, 126, 234, 0.3);
        text-align: center;
        position: relative;
        overflow: hidden;
    }
    
    .glass-header::before {
        content: '';
        position: absolute;
        top: 0;
        left: -100%;
        width: 100%;
        height: 100%;
        background: linear-gradient(90deg, transparent, rgba(255,255,255,0.15), transparent);
        animation: shimmer 3s infinite;
        z-index: 1;
    }
    
    @keyframes shimmer {
        0% {left: -100%;}
        100% {left: 100%;}
    }
    
    .glass-header h1 {
        font-size: 3.5rem !important;
        font-weight: 900 !important;
        color: #ffffff !important;
        letter-spacing: -2px !important;
        margin-bottom: 0.5rem !important;
        position: relative !important;
        z-index: 10 !important;
        text-shadow: 3px 3px 6px rgba(0,0,0,0.4) !important;
        line-height: 1.2 !important;
    }
    
    .glass-header p {
        color: rgba(255, 255, 255, 0.95) !important;
        font-size: 1.2rem !important;
        font-weight: 500 !important;
        letter-spacing: 0.5px !important;
        text-shadow: 2px 2px 4px rgba(0,0,0,0.3) !important;
        position: relative !important;
        z-index: 10 !important;
    }
    
    
        /* ========== BENTO GRID CONTAINER ========== */
    .bento-container {
        background: rgba(255, 255, 255, 0.95);
        backdrop-filter: blur(10px);
        padding: 0;
        border-radius: 24px;
        margin: 0rem auto;
        max-width: 1400px;
        box-shadow: 0 20px 60px rgba(0,0,0,0.3);
        overflow: hidden;
    }

    }
    
    /* ========== STAT CARDS (BENTO STYLE) ========== */
    .stat-grid {
        display: grid;
        grid-template-columns: repeat(auto-fit, minmax(240px, 1fr));
        gap: 1.25rem;
        margin: 2rem 0;
    }
    
    .stat-card {
        background: linear-gradient(135deg, #667eea 0%, #764ba2 100%);
        padding: 2rem;
        border-radius: 20px;
        text-align: center;
        color: white;
        box-shadow: 0 10px 30px rgba(102, 126, 234, 0.4);
        transition: all 0.4s cubic-bezier(0.175, 0.885, 0.32, 1.275);
        position: relative;
        overflow: hidden;
    }
    
    .stat-card::before {
        content: '';
        position: absolute;
        top: -50%;
        left: -50%;
        width: 200%;
        height: 200%;
        background: radial-gradient(circle, rgba(255,255,255,0.1) 0%, transparent 70%);
        opacity: 0;
        transition: opacity 0.3s;
    }
    
    .stat-card:hover {
        transform: translateY(-8px) scale(1.02);
        box-shadow: 0 15px 40px rgba(102, 126, 234, 0.6);
    }
    
    .stat-card:hover::before {
        opacity: 1;
    }
    
    .stat-card h3 {
        font-size: 3rem;
        margin: 0;
        font-weight: 800;
        line-height: 1;
    }
    
    .stat-card p {
        margin: 0.75rem 0 0 0;
        opacity: 0.95;
        font-size: 0.95rem;
        text-transform: uppercase;
        letter-spacing: 2px;
        font-weight: 600;
    }
    
    /* ========== MODERN BUTTONS ========== */
    .stButton>button {
        background: linear-gradient(135deg, #667eea 0%, #764ba2 100%);
        color: white;
        border: none;
        padding: 1rem 2.5rem;
        font-size: 1.05rem;
        font-weight: 700;
        border-radius: 16px;
        cursor: pointer;
        transition: all 0.3s ease;
        width: 100%;
        box-shadow: 0 8px 25px rgba(102, 126, 234, 0.4);
        letter-spacing: 0.5px;
        text-transform: uppercase;
        position: relative;
        overflow: hidden;
    }
    
    .stButton>button::before {
        content: '';
        position: absolute;
        top: 50%;
        left: 50%;
        width: 0;
        height: 0;
        border-radius: 50%;
        background: rgba(255, 255, 255, 0.3);
        transform: translate(-50%, -50%);
        transition: width 0.6s, height 0.6s;
    }
    
    .stButton>button:hover::before {
        width: 300px;
        height: 300px;
    }
    
    .stButton>button:hover {
        transform: translateY(-3px);
        box-shadow: 0 12px 35px rgba(102, 126, 234, 0.7);
    }
    
    .stButton>button:active {
        transform: translateY(-1px);
    }

        /* ========== RESULT BOX (UNIFORM STYLE) ========== */
        /* ========== JUSTIFIED TEXT FOR SUMMARIES ========== */
    .justified-text {
        text-align: justify;
        line-height: 1.8;
        color: #333;
        font-size: 1.05rem;
    }
    
        /* ========== INFO BLOCKS (COMPACT VERSION) ========== */
    .info-block {
        background: linear-gradient(135deg, rgba(102, 126, 234, 0.06) 0%, rgba(118, 75, 162, 0.06) 100%);
        backdrop-filter: blur(10px);
        padding: 1.25rem;
        border-radius: 16px;
        border: 1px solid rgba(102, 126, 234, 0.15);
        margin: 0.75rem 0;
        box-shadow: 0 4px 12px rgba(0,0,0,0.04);
        transition: all 0.3s ease;
    }
    
    .info-block:hover {
        transform: translateY(-2px);
        box-shadow: 0 6px 20px rgba(102, 126, 234, 0.12);
        border-color: rgba(102, 126, 234, 0.25);
    }
    
    .info-block h4 {
        color: #667eea;
        margin-top: 0;
        margin-bottom: 1rem;
        font-size: 1.2rem;
        font-weight: 700;
        text-align: center;
    }

    
    /* ========== FEATURE ITEM ========== */
    .feature-item {
        padding: 0.75rem 0;
        border-bottom: 1px solid rgba(102, 126, 234, 0.1);
    }
    
    .feature-item:last-child {
        border-bottom: none;
    }
    
    .feature-item strong {
        color: #667eea;
        font-size: 1.05rem;
    }
    
    .feature-item p {
        color: #666;
        margin: 0.25rem 0 0 0;
        font-size: 0.95rem;
        line-height: 1.6;
    }

    /* ========== MODERN BADGES ========== */
    .badge {
        display: inline-block;
        padding: 0.5rem 1rem;
        background: linear-gradient(135deg, #667eea 0%, #764ba2 100%);
        color: white;
        border-radius: 25px;
        font-size: 0.9rem;
        font-weight: 600;
        margin: 0.35rem;
        box-shadow: 0 4px 12px rgba(102, 126, 234, 0.3);
        transition: all 0.3s ease;
    }
    
    .badge:hover {
        transform: scale(1.05);
        box-shadow: 0 6px 18px rgba(102, 126, 234, 0.5);
    }
    
        /* ========== MODERN TABS (CENTERED) ========== */
    .stTabs [data-baseweb="tab-list"] {
        gap: 1rem;
        background: linear-gradient(135deg, rgba(102, 126, 234, 0.1) 0%, rgba(118, 75, 162, 0.1) 100%);
        border-bottom: none;
        padding: 1.5rem 2rem;
        display: flex;
        justify-content: center;
        align-items: center;
    }
    
    .stTabs [data-baseweb="tab"] {
        background: rgba(255, 255, 255, 0.8);
        backdrop-filter: blur(5px);
        border-radius: 16px;
        padding: 1rem 2rem;
        font-weight: 700;
        font-size: 1.05rem;
        border: 2px solid rgba(102, 126, 234, 0.2);
        color: #667eea;
        transition: all 0.3s ease;
        min-width: 180px;
        text-align: center;
    }
    
    .stTabs [data-baseweb="tab"]:hover {
        background: rgba(255, 255, 255, 1);
        transform: translateY(-3px);
        box-shadow: 0 8px 20px rgba(102, 126, 234, 0.3);
        border-color: #667eea;
    }
    
    .stTabs [aria-selected="true"] {
        background: linear-gradient(135deg, #667eea 0%, #764ba2 100%);
        color: white;
        box-shadow: 0 8px 25px rgba(102, 126, 234, 0.5);
        border-color: transparent;
        transform: translateY(-3px);
    }
    
        .stTabs [data-baseweb="tab-panel"] {
        padding: 2.5rem;
        text-align: center;
        max-width: 1200px;
        margin: 0 auto;
    }
    
    /* Center all tab content */
    .stTabs h3, .stTabs h4 {
        text-align: center !important;
    }
    
    .stTabs .stButton {
        display: flex;
        justify-content: center;
    }


    
    /* ========== UPLOAD AREA ========== */
    .uploadedFile {
        background: rgba(102, 126, 234, 0.05);
        border-radius: 16px;
        padding: 1.5rem;
        border: 3px dashed #667eea;
        transition: all 0.3s ease;
    }
    
    .uploadedFile:hover {
        background: rgba(102, 126, 234, 0.1);
        border-color: #764ba2;
    }
    
    /* ========== INFO BOXES ========== */
    .stInfo, .stSuccess, .stWarning, .stError {
        border-radius: 16px;
        padding: 1.25rem;
        margin: 1rem 0;
        font-weight: 500;
        box-shadow: 0 4px 15px rgba(0,0,0,0.1);
    }
    
    /* ========== TEXT AREA ========== */
    .stTextArea textarea {
        border-radius: 16px;
        border: 2px solid rgba(102, 126, 234, 0.3);
        font-family: 'Inter', monospace;
        transition: all 0.3s ease;
    }
    
    .stTextArea textarea:focus {
        border-color: #667eea;
        box-shadow: 0 0 0 3px rgba(102, 126, 234, 0.2);
    }
    
    /* ========== LOADING ANIMATION ========== */
    @keyframes pulse {
        0%, 100% { opacity: 1; }
        50% { opacity: 0.6; }
    }
    
    .loading {
        animation: pulse 2s ease-in-out infinite;
    }
    
    /* ========== FILE UPLOADER ========== */
    [data-testid="stFileUploader"] {
        background: rgba(255, 255, 255, 0.7);
        backdrop-filter: blur(5px);
        border-radius: 16px;
        padding: 1.5rem;
        border: 2px dashed rgba(102, 126, 234, 0.4);
        transition: all 0.3s ease;
    }
    
    [data-testid="stFileUploader"]:hover {
        border-color: #667eea;
        background: rgba(255, 255, 255, 0.9);
    }
    
    /* ========== RESPONSIVE DESIGN ========== */
    @media (max-width: 768px) {
        .glass-header h1 {
            font-size: 2.5rem;
        }
        
        .glass-header p {
            font-size: 1rem;
        }
        
        .stat-card h3 {
            font-size: 2.5rem;
        }
        
        .bento-container {
            padding: 1.5rem;
        }
    }
</style>
""", unsafe_allow_html=True)

# ============================================================================
# SESSION STATE
# ============================================================================

if 'pdf_data' not in st.session_state:
    st.session_state.pdf_data = None
if 'summary' not in st.session_state:
    st.session_state.summary = None
if 'key_info' not in st.session_state:
    st.session_state.key_info = None
if 'processing_time' not in st.session_state:
    st.session_state.processing_time = 0

# ============================================================================
# CORE FUNCTIONS
# ============================================================================

def extract_pdf_text(pdf_path: Path):
    """Extract text from PDF with metadata."""
    try:
        with open(pdf_path, 'rb') as file:
            reader = PyPDF2.PdfReader(file)
            
            text_data = {
                'full_text': '',
                'total_pages': len(reader.pages),
                'metadata': {}
            }
            
            # Extract metadata
            if reader.metadata:
                text_data['metadata'] = {
                    'title': reader.metadata.get('/Title', 'N/A'),
                    'author': reader.metadata.get('/Author', 'N/A'),
                    'creator': reader.metadata.get('/Creator', 'N/A')
                }
            
            # Extract text from all pages
            for page in reader.pages:
                try:
                    text = page.extract_text()
                    if text.strip():
                        text_data['full_text'] += text + "\n\n"
                except:
                    continue
            
            text_data['word_count'] = len(text_data['full_text'].split())
            text_data['char_count'] = len(text_data['full_text'])
            
            return text_data
    except Exception as e:
        raise Exception(f"PDF extraction failed: {str(e)}")

def generate_ai_summary(text: str):
    """Generate AI summary using Gemini."""
    try:
        model = genai.GenerativeModel(Config.MODEL_NAME)
        
        prompt = f"""Analyze this document and provide a concise, well-structured summary in approximately 150-200 words.
Focus on key points, main themes, and actionable insights.

DOCUMENT:
{text[:15000]}

Provide a clear, professional summary:"""
        
        response = model.generate_content(prompt)
        return response.text.strip()
    except Exception as e:
        raise Exception(f"AI summarization failed: {str(e)}")

def extract_key_information(text: str):
    """Extract structured information using Gemini."""
    try:
        model = genai.GenerativeModel(Config.MODEL_NAME)
        
        prompt = f"""Analyze this document and extract key information in a structured format:

DOCUMENT:
{text[:15000]}

Provide the following information:
1. MAIN TOPICS: (List 3-5 key topics, comma-separated)
2. KEY ENTITIES: (People, organizations, locations - comma-separated)
3. IMPORTANT DATES/NUMBERS: (Any significant data points - comma-separated)
4. DOCUMENT TYPE: (Single category/type)
5. KEY TAKEAWAYS: (3-5 bullet points starting with -)

Format your response exactly as shown above."""
        
        response = model.generate_content(prompt)
        return parse_response(response.text)
    except Exception as e:
        raise Exception(f"Information extraction failed: {str(e)}")

def parse_response(text: str):
    """Parse AI response into structured data."""
    result = {
        'topics': [],
        'entities': [],
        'dates_numbers': [],
        'doc_type': 'Unknown',
        'takeaways': []
    }
    
    lines = text.split('\n')
    current_section = None
    
    for line in lines:
        line = line.strip()
        if 'MAIN TOPICS:' in line.upper() or line.startswith('1.'):
            current_section = 'topics'
            if ':' in line:
                content = line.split(':', 1)[1].strip()
                if content:
                    result['topics'] = [t.strip() for t in content.split(',') if t.strip()]
        elif 'KEY ENTITIES:' in line.upper() or line.startswith('2.'):
            current_section = 'entities'
            if ':' in line:
                content = line.split(':', 1)[1].strip()
                if content:
                    result['entities'] = [e.strip() for e in content.split(',') if e.strip()]
        elif 'IMPORTANT DATES' in line.upper() or 'NUMBERS:' in line.upper() or line.startswith('3.'):
            current_section = 'dates'
            if ':' in line:
                content = line.split(':', 1)[1].strip()
                if content:
                    result['dates_numbers'] = [d.strip() for d in content.split(',') if d.strip()]
        elif 'DOCUMENT TYPE:' in line.upper() or line.startswith('4.'):
            current_section = 'type'
            if ':' in line:
                result['doc_type'] = line.split(':', 1)[1].strip()
        elif 'KEY TAKEAWAYS:' in line.upper() or 'TAKEAWAYS:' in line.upper() or line.startswith('5.'):
            current_section = 'takeaways'
        elif line.startswith(('- ', '• ', '* ', '·')) and current_section == 'takeaways':
            takeaway = line.lstrip('- •*· ').strip()
            if takeaway:
                result['takeaways'].append(takeaway)
    
    return result

# ============================================================================
# UI - HEADER
# ============================================================================

st.markdown("""
<div class="glass-header">
    <h1 style="color: #ffffff !important; text-shadow: 3px 3px 10px rgba(0,0,0,0.7), 0 0 30px rgba(255,255,255,0.3) !important;">🤖 AI PDF EXTRACTOR</h1>
    <p style="color: #ffffff !important; text-shadow: 2px 2px 8px rgba(0,0,0,0.6) !important;">Powered by Google Gemini 2.5 • Extract • Analyze • Summarize</p>
</div>
""", unsafe_allow_html=True)

# ============================================================================
# UI - MAIN CONTAINER
# ============================================================================

st.markdown('<div class="bento-container">', unsafe_allow_html=True)

# Create tabs with modern icons
tab1, tab2, tab3, tab4 = st.tabs([
    "📤 Upload & Extract",
    "🤖 AI Analysis",
    "📊 Complete Results",
    "ℹ️ About"
])

# ============================================================================
# TAB 1: UPLOAD & EXTRACT
# ============================================================================

with tab1:
    st.markdown("<h3 style='text-align: center;'>📄 Upload Your PDF Document</h3>", unsafe_allow_html=True)
    st.markdown("<p style='text-align: center; color: #666; margin-bottom: 2rem;'>Drag and drop or browse to upload your PDF file (max 10MB)</p>", unsafe_allow_html=True)
    
    # Center the upload area
    col_left, col_center, col_right = st.columns([1, 2, 1])
    
    with col_center:
        uploaded_file = st.file_uploader(
            "Choose PDF",
            type=['pdf'],
            help="Maximum file size: 10MB",
            label_visibility="collapsed"
        )
    
    if uploaded_file:
        col1, col2 = st.columns(2)
        with col1:
            st.info(f"**📁 {uploaded_file.name}**")
        with col2:
            file_size = len(uploaded_file.getvalue()) / (1024 * 1024)
            st.success(f"**📊 {file_size:.2f} MB**")
        
        st.markdown("---")
        
        # Center the button
        col_left, col_center, col_right = st.columns([1, 2, 1])
        with col_center:
            extract_btn = st.button("🚀 EXTRACT TEXT", use_container_width=True)
        
        if extract_btn:
            start_time = time.time()
            
            with st.spinner("🔄 Processing PDF..."):
                try:
                    # Save temp file
                    with tempfile.NamedTemporaryFile(delete=False, suffix='.pdf') as tmp:
                        tmp.write(uploaded_file.getvalue())
                        tmp_path = Path(tmp.name)
                    
                    # Extract
                    pdf_data = extract_pdf_text(tmp_path)
                    st.session_state.pdf_data = pdf_data
                    st.session_state.processing_time = time.time() - start_time
                    
                    # Cleanup
                    tmp_path.unlink()
                    
                    st.success("✅ Text extracted successfully!")
                    
                    # Display stats in Bento grid
                    st.markdown('<div class="stat-grid">', unsafe_allow_html=True)
                    col1, col2, col3, col4 = st.columns(4)
                    
                    with col1:
                        st.markdown(f"""
                        <div class="stat-card">
                            <h3>{pdf_data['total_pages']}</h3>
                            <p>Pages</p>
                        </div>
                        """, unsafe_allow_html=True)
                    
                    with col2:
                        st.markdown(f"""
                        <div class="stat-card">
                            <h3>{pdf_data['word_count']:,}</h3>
                            <p>Words</p>
                        </div>
                        """, unsafe_allow_html=True)
                    
                    with col3:
                        st.markdown(f"""
                        <div class="stat-card">
                            <h3>{pdf_data['char_count']:,}</h3>
                            <p>Characters</p>
                        </div>
                        """, unsafe_allow_html=True)
                    
                    with col4:
                        st.markdown(f"""
                        <div class="stat-card">
                            <h3>{st.session_state.processing_time:.2f}s</h3>
                            <p>Time</p>
                        </div>
                        """, unsafe_allow_html=True)
                    
                    st.markdown('</div>', unsafe_allow_html=True)
                    
                except Exception as e:
                    st.error(f"❌ Error: {str(e)}")
        
        # Preview extracted text
        if st.session_state.pdf_data:
            st.markdown("---")
            with st.expander("👁️ Preview Extracted Text", expanded=False):
                preview_text = st.session_state.pdf_data['full_text'][:3000]
                st.text_area(
                    "Document Content",
                    preview_text + "..." if len(st.session_state.pdf_data['full_text']) > 3000 else preview_text,
                    height=300,
                    disabled=True,
                    label_visibility="collapsed"
                )
    else:
        st.markdown("<br><br>", unsafe_allow_html=True)
        st.info("👆 Please upload a PDF file to get started")

# ============================================================================
# TAB 2: AI ANALYSIS
# ============================================================================

with tab2:
    st.markdown("<h3 style='text-align: center;'>🤖 AI-Powered Document Analysis</h3>", unsafe_allow_html=True)
    st.markdown("<br>", unsafe_allow_html=True)
    
    if not st.session_state.pdf_data:
        st.warning("⚠️ Please upload and extract a PDF first in the Upload tab.")
    else:
        st.info(f"🎯 Analyzing document: **{uploaded_file.name if uploaded_file else 'Document'}**")
        st.markdown("<br>", unsafe_allow_html=True)
        
        col1, col2 = st.columns(2)
        
        with col1:
            if st.button("📝 GENERATE AI SUMMARY", use_container_width=True):
                with st.spinner("🤖 Generating intelligent summary..."):
                    try:
                        summary = generate_ai_summary(st.session_state.pdf_data['full_text'])
                        st.session_state.summary = summary
                        st.success("✅ Summary generated successfully!")
                        st.balloons()
                    except Exception as e:
                        st.error(f"❌ Error: {str(e)}")
        
        with col2:
            if st.button("🔍 EXTRACT KEY INFORMATION", use_container_width=True):
                with st.spinner("🤖 Extracting key insights..."):
                    try:
                        key_info = extract_key_information(st.session_state.pdf_data['full_text'])
                        st.session_state.key_info = key_info
                        st.success("✅ Information extracted successfully!")
                        st.balloons()
                    except Exception as e:
                        st.error(f"❌ Error: {str(e)}")
        
                # Display Summary
        if st.session_state.summary:
            st.markdown("---")
            st.markdown('<div class="result-box">', unsafe_allow_html=True)
            st.markdown("#### 📝 AI-Generated Summary")
            st.markdown(f'<div class="justified-text">{st.session_state.summary}</div>', unsafe_allow_html=True)
            st.caption(f"✨ Generated by {Config.MODEL_NAME} • {len(st.session_state.summary.split())} words")
            st.markdown('</div>', unsafe_allow_html=True)

        
                # Display Key Information
        if st.session_state.key_info:
            st.markdown("---")
            st.markdown('<div class="result-box">', unsafe_allow_html=True)
            st.markdown("#### 🔍 Key Information Extracted")
            st.markdown("<br>", unsafe_allow_html=True)
            
            col1, col2 = st.columns(2)
            
            with col1:
                st.markdown("**📌 Main Topics**")
                if st.session_state.key_info['topics']:
                    for topic in st.session_state.key_info['topics']:
                        st.markdown(f'<span class="badge">{topic}</span>', unsafe_allow_html=True)
                else:
                    st.caption("No topics identified")
                
                st.markdown("<br>**👥 Key Entities**", unsafe_allow_html=True)
                if st.session_state.key_info['entities']:
                    entities_text = "<div style='text-align: left;'>"
                    for entity in st.session_state.key_info['entities'][:6]:
                        entities_text += f"<p style='margin: 0.5rem 0; color: #555;'>• {entity}</p>"
                    entities_text += "</div>"
                    st.markdown(entities_text, unsafe_allow_html=True)
                else:
                    st.caption("No entities identified")
            
            with col2:
                st.markdown(f"**📄 Document Type**")
                st.code(st.session_state.key_info['doc_type'], language=None)
                
                st.markdown("**📅 Important Data**")
                if st.session_state.key_info['dates_numbers']:
                    dates_text = "<div style='text-align: left;'>"
                    for data in st.session_state.key_info['dates_numbers'][:6]:
                        dates_text += f"<p style='margin: 0.5rem 0; color: #555;'>• {data}</p>"
                    dates_text += "</div>"
                    st.markdown(dates_text, unsafe_allow_html=True)
                else:
                    st.caption("No significant dates/numbers found")
            
            if st.session_state.key_info['takeaways']:
                st.markdown("<br>**💡 Key Takeaways**", unsafe_allow_html=True)
                takeaways_text = "<div class='justified-text'>"
                for i, takeaway in enumerate(st.session_state.key_info['takeaways'], 1):
                    takeaways_text += f"<p><strong>{i}.</strong> {takeaway}</p>"
                takeaways_text += "</div>"
                st.markdown(takeaways_text, unsafe_allow_html=True)
            
            st.markdown('</div>', unsafe_allow_html=True)


# ============================================================================
# TAB 3: COMPLETE RESULTS
# ============================================================================

with tab3:
    st.markdown("<h3 style='text-align: center;'>📊 Complete Analysis Results</h3>", unsafe_allow_html=True)
    st.markdown("<br>", unsafe_allow_html=True)
    
    if not st.session_state.pdf_data:
        st.warning("⚠️ No results available. Please extract a PDF first.")
    else:
        # Document Information Block
        st.markdown('<div class="result-box">', unsafe_allow_html=True)
        st.markdown("#### 📄 Document Information")
        
        col1, col2, col3 = st.columns(3)
        with col1:
            st.metric("📄 Filename", uploaded_file.name if uploaded_file else "N/A")
        with col2:
            st.metric("📊 Total Pages", st.session_state.pdf_data['total_pages'])
        with col3:
            st.metric("📝 Word Count", f"{st.session_state.pdf_data['word_count']:,}")
        
        col1, col2, col3 = st.columns(3)
        with col1:
            st.metric("🔤 Characters", f"{st.session_state.pdf_data['char_count']:,}")
        with col2:
            st.metric("⏱️ Processing Time", f"{st.session_state.processing_time:.2f}s")
        with col3:
            st.metric("🤖 AI Model", Config.MODEL_NAME.split('-')[1].upper())
        
        st.markdown('</div>', unsafe_allow_html=True)
        
        # AI Analysis Block
        if st.session_state.summary or st.session_state.key_info:
            st.markdown('<div class="result-box">', unsafe_allow_html=True)
            st.markdown("#### 🤖 AI Analysis Summary")
            
            if st.session_state.summary:
                st.markdown("**📝 Document Summary:**")
                st.markdown(f'<div class="justified-text">{st.session_state.summary}</div>', unsafe_allow_html=True)
                st.markdown("<br>", unsafe_allow_html=True)
            
            if st.session_state.key_info:
                col1, col2 = st.columns(2)
                
                with col1:
                    st.markdown("**📄 Document Type:**")
                    st.info(st.session_state.key_info['doc_type'])
                    
                    if st.session_state.key_info['topics']:
                        st.markdown("**📌 Topics:**")
                        for topic in st.session_state.key_info['topics']:
                            st.markdown(f"- {topic}")
                
                with col2:
                    if st.session_state.key_info['entities']:
                        st.markdown("**👥 Entities:**")
                        for entity in st.session_state.key_info['entities'][:5]:
                            st.markdown(f"- {entity}")
                    
                    if st.session_state.key_info['dates_numbers']:
                        st.markdown("**📅 Key Data:**")
                        for data in st.session_state.key_info['dates_numbers'][:5]:
                            st.markdown(f"- {data}")
            
            st.markdown('</div>', unsafe_allow_html=True)
        
        st.markdown("<br>", unsafe_allow_html=True)
        
        # Download Options
        st.markdown('<div class="result-box">', unsafe_allow_html=True)
        st.markdown("#### ⬇️ Download Options")
        
        # Compile results
        results = {
            'document_info': {
                'filename': uploaded_file.name if uploaded_file else 'N/A',
                'total_pages': st.session_state.pdf_data['total_pages'],
                'word_count': st.session_state.pdf_data['word_count'],
                'character_count': st.session_state.pdf_data['char_count'],
                'processing_time_seconds': round(st.session_state.processing_time, 2),
                'metadata': st.session_state.pdf_data['metadata']
            },
            'ai_analysis': {
                'summary': st.session_state.summary,
                'key_information': st.session_state.key_info,
                'model_used': Config.MODEL_NAME
            },
            'timestamp': datetime.now().strftime("%Y-%m-%d %H:%M:%S IST")
        }
        
        col1, col2 = st.columns(2)
        
        with col1:
            json_data = json.dumps(results, indent=2)
            st.download_button(
                label="📥 DOWNLOAD JSON",
                data=json_data,
                file_name=f"analysis_{datetime.now().strftime('%Y%m%d_%H%M%S')}.json",
                mime="application/json",
                use_container_width=True
            )
        
        with col2:
            # Create text report
            text_report = f"""
AI PDF EXTRACTION REPORT
========================

Document: {results['document_info']['filename']}
Processed: {results['timestamp']}
Model: {Config.MODEL_NAME}

STATISTICS:
- Pages: {results['document_info']['total_pages']}
- Words: {results['document_info']['word_count']}
- Characters: {results['document_info']['character_count']}
- Processing Time: {results['document_info']['processing_time_seconds']}s

AI SUMMARY:
{st.session_state.summary if st.session_state.summary else 'Not generated'}

KEY INFORMATION:
- Document Type: {st.session_state.key_info['doc_type'] if st.session_state.key_info else 'N/A'}
- Topics: {', '.join(st.session_state.key_info['topics']) if st.session_state.key_info and st.session_state.key_info['topics'] else 'N/A'}

Generated by AI PDF Extractor
"""
            
            st.download_button(
                label="📥 DOWNLOAD TXT",
                data=text_report,
                file_name=f"report_{datetime.now().strftime('%Y%m%d_%H%M%S')}.txt",
                mime="text/plain",
                use_container_width=True
            )
        
        st.markdown('</div>', unsafe_allow_html=True)
# ============================================================================
# TAB 4: ABOUT & INFO
# ============================================================================

with tab4:
    st.markdown("<h3 style='text-align: center; margin-bottom: 0.5rem;'>📖 About AI PDF Extractor</h3>", unsafe_allow_html=True)
    st.markdown("<p style='text-align: center; color: #666; margin-bottom: 2rem;'>Intelligent document processing powered by Google Gemini AI</p>", unsafe_allow_html=True)
    
    # Row 1: Key Features and How to Use
    col1, col2 = st.columns(2, gap="large")
    
    with col1:
        with st.container():
            st.markdown("#### ✨ Key Features")
            st.markdown("")
            
            st.markdown("**📄 PDF Text Extraction**")
            st.caption("Extract text from any PDF document with complete metadata and structural information")
            st.markdown("")
            
            st.markdown("**🤖 AI Summarization**")
            st.caption("Generate intelligent, context-aware summaries using Google Gemini 2.5 Flash model")
            st.markdown("")
            
            st.markdown("**🔍 Information Extraction**")
            st.caption("Automatically identify topics, entities, key insights, and important data points")
            st.markdown("")
            
            st.markdown("**📊 Export Results**")
            st.caption("Download complete analysis in JSON or TXT format for easy integration")
            st.markdown("")
            
            st.markdown("**💯 100% Free**")
            st.caption("Powered by Google Gemini's generous free tier with no credit card required")
    
    with col2:
        with st.container():
            st.markdown("#### 🚀 How to Use")
            st.markdown("")
            
            st.markdown("**1. Upload Tab**")
            st.caption("Upload your PDF document (max 10MB) and click extract to process the content")
            st.markdown("")
            
            st.markdown("**2. AI Analysis Tab**")
            st.caption("Generate AI-powered summary and extract structured key information from your document")
            st.markdown("")
            
            st.markdown("**3. Results Tab**")
            st.caption("View complete analysis with metrics, download results in your preferred format")
            st.markdown("")
            
            st.markdown("**4. Enjoy!**")
            st.caption("Use the insights for research, work, study, or any document analysis needs")
    
    st.markdown("<br>", unsafe_allow_html=True)
    st.markdown("---")
    
    # Row 2: Tech Stack and Gemini Limits
    col1, col2 = st.columns(2, gap="large")
    
    with col1:
        with st.container():
            st.markdown("#### 🛠️ Tech Stack")
            st.markdown("")
            
            tech_stack = [
                ("🐍 Python 3.8+", "Core programming language"),
                ("🤖 Google Gemini 2.5", "AI model for analysis"),
                ("📄 PyPDF2", "PDF parsing library"),
                ("🎨 Streamlit", "Web framework"),
                ("🔒 python-dotenv", "Secure configuration")
            ]
            
            for tech, desc in tech_stack:
                col_a, col_b = st.columns([2, 3])
                with col_a:
                    st.markdown(f"**{tech}**")
                with col_b:
                    st.caption(desc)
    
    with col2:
        with st.container():
            st.markdown("#### 📊 Gemini Free Tier")
            st.markdown("")
            
            limits = [
                ("✅ Requests/Minute", "15"),
                ("✅ Tokens/Day", "1 Million"),
                ("✅ Requests/Day", "1,500"),
                ("✅ Credit Card", "Not Required")
            ]
            
            for feature, value in limits:
                col_a, col_b = st.columns([3, 2])
                with col_a:
                    st.markdown(feature)
                with col_b:
                    st.markdown(f"**{value}**")
            
            st.markdown("")
            st.success("✨ Ideal for production-ready AI applications")
    
    st.markdown("<br>", unsafe_allow_html=True)
    st.markdown("---")
    
    # Row 3: Coming Soon Features
    st.markdown("#### 🚀 Upcoming Features & Roadmap")
    st.markdown("")
    
    col1, col2, col3 = st.columns(3)
    
    with col1:
        st.markdown("**📊 Advanced Analytics**")
        st.caption("Visual charts and statistics for deeper insights")
        st.markdown("")
        
        st.markdown("**📁 Batch Processing**")
        st.caption("Analyze multiple PDFs simultaneously")
    
    with col2:
        st.markdown("**🔍 Smart Search**")
        st.caption("Semantic search across documents")
        st.markdown("")
        
        st.markdown("**💬 Chat with PDF**")
        st.caption("Interactive Q&A using RAG")
    
    with col3:
        st.markdown("**🌍 Multi-language**")
        st.caption("Support for 50+ languages")
        st.markdown("")
        
        st.markdown("**🔗 RESTful API**")
        st.caption("Integration with applications")
    
    st.markdown("<br>", unsafe_allow_html=True)
    st.markdown("---")
    
    # System Info Metrics
    st.markdown("#### ⚙️ Current System Configuration")
    st.markdown("")
    
    col1, col2, col3 = st.columns(3)
    
    with col1:
        st.metric(
            label="🤖 AI Model",
            value="Gemini 2.5 Flash"
        )
    
    with col2:
        st.metric(
            label="📁 Max File Size",
            value=f"{Config.MAX_FILE_SIZE_MB} MB"
        )
    
    with col3:
        st.metric(
            label="🌡️ Temperature",
            value=Config.TEMPERATURE
        )

# ============================================================================
# FOOTER
# ============================================================================

st.markdown("""
<div style="text-align: center; padding: 2rem; margin-top: 2rem;">
    <p style="color: white; font-size: 0.95rem; font-weight: 500;">
        Built for AI Agent Applications • Powered by Google Gemini • 2025
    </p>
</div>
""", unsafe_allow_html=True)