app.py

import gradio as gr
import tempfile
import os
import requests
import json
import re
from bs4 import BeautifulSoup
from datetime import datetime
import urllib.parse
from pathlib import Path
from typing import List, Dict, Optional, Any, Tuple
import numpy as np
import time


# Configuration
SPACE_NAME = 'Language Learning Partner'
SPACE_DESCRIPTION = 'AI Italian conversation partner'

# Default configuration values
DEFAULT_CONFIG = {
    'name': SPACE_NAME,
    'description': SPACE_DESCRIPTION,
    'system_prompt': "You are Domenico from Sicily, a Juventus football fan, native Italian speaker serving as a conversational partner for university students in an Italian 101 class. Students will interact and converse with you in Italian, and you must respond EXCLUSIVELY IN ITALIAN without providing English translations, using vocabulary appropriate for beginner-level Italian 101 students. Focus your responses on topics suitable for beginners such as sports, daily life, routines, food, numbers, and hobbies. When students make errors, model the correct forms naturally in your response without explicitly pointing out mistakes, allowing them to learn through exposure to proper usage. Recognize when students demonstrate more advanced abilities and adjust your language complexity accordingly, while ensuring your Italian remains error-free. Keep all responses between 5-50 words, making sure sentences are grammatically complete. Limit all verb conjugations to the present tense only, avoiding all other verb forms and tenses. Address students using the informal second-person singular 'tu' form.",
    'temperature': 0.5,
    'max_tokens': 250,
    'model': 'google/gemma-3-27b-it',
    'api_key_var': 'API_KEY',
    'theme': 'Default',
    'grounding_urls': [],
    'enable_dynamic_urls': True,
    'enable_file_upload': True,
    'examples': ['Ciao! Come stai oggi?', 'Mi piace giocare a calcio. E tu?', 'Cosa mangi di solito a colazione?', 'A che ora ti svegli la mattina?', 'Qual è il tuo sport preferito?'],
    'language': 'Italian',
    'enable_tts': True,
    'tts_model': 'microsoft/speecht5_tts',
    'tts_voice': 'default',
    'locked': False
}

# Available themes with proper instantiation
AVAILABLE_THEMES = {
    "Default": gr.themes.Default(),
    "Soft": gr.themes.Soft(),
    "Glass": gr.themes.Glass(),
    "Monochrome": gr.themes.Monochrome(),
    "Base": gr.themes.Base()
}


class ConfigurationManager:
    """Manage configuration with validation and persistence"""
    
    def __init__(self):
        self.config_path = "config.json"
        self.backup_dir = "config_backups"
        self._config = None
    
    def load(self) -> Dict[str, Any]:
        """Load configuration from file with fallback to defaults"""
        try:
            with open(self.config_path, 'r') as f:
                self._config = json.load(f)
                print("✅ Loaded configuration from config.json")
                return self._config
        except FileNotFoundError:
            print("ℹ️ No config.json found, using default configuration")
            self._config = DEFAULT_CONFIG.copy()
            self.save(self._config)
            return self._config
        except Exception as e:
            print(f"⚠️ Error loading config.json: {e}, using defaults")
            self._config = DEFAULT_CONFIG.copy()
            return self._config
    
    def save(self, config: Dict[str, Any]) -> bool:
        """Save configuration with automatic backup"""
        try:
            # Create backup if config exists
            if os.path.exists(self.config_path):
                self._create_backup()
            
            # Save new configuration
            with open(self.config_path, 'w') as f:
                json.dump(config, f, indent=2)
            
            self._config = config
            return True
        except Exception as e:
            print(f"❌ Error saving configuration: {e}")
            return False
    
    def _create_backup(self):
        """Create timestamped backup"""
        try:
            os.makedirs(self.backup_dir, exist_ok=True)
            timestamp = datetime.now().strftime('%Y%m%d_%H%M%S')
            backup_path = os.path.join(self.backup_dir, f"config_{timestamp}.json")
            
            with open(self.config_path, 'r') as source:
                config_data = json.load(source)
                with open(backup_path, 'w') as backup:
                    json.dump(config_data, backup, indent=2)
            
            self._cleanup_old_backups()
        except Exception as e:
            print(f"⚠️ Error creating backup: {e}")
    
    def _cleanup_old_backups(self, keep=10):
        """Keep only the most recent backups"""
        try:
            backups = sorted([
                f for f in os.listdir(self.backup_dir)
                if f.startswith('config_') and f.endswith('.json')
            ])
            
            if len(backups) > keep:
                for old_backup in backups[:-keep]:
                    os.remove(os.path.join(self.backup_dir, old_backup))
        except Exception as e:
            print(f"⚠️ Error cleaning up backups: {e}")
    
    def get(self, key: str, default: Any = None) -> Any:
        """Get configuration value"""
        if self._config is None:
            self.load()
        return self._config.get(key, default)


# Initialize configuration manager
config_manager = ConfigurationManager()
config = config_manager.load()

# Load configuration values
SPACE_NAME = config.get('name', DEFAULT_CONFIG['name'])
SPACE_DESCRIPTION = config.get('description', DEFAULT_CONFIG['description'])
SYSTEM_PROMPT = config.get('system_prompt', DEFAULT_CONFIG['system_prompt'])
temperature = config.get('temperature', DEFAULT_CONFIG['temperature'])
max_tokens = config.get('max_tokens', DEFAULT_CONFIG['max_tokens'])
MODEL = config.get('model', DEFAULT_CONFIG['model'])
THEME = config.get('theme', DEFAULT_CONFIG['theme'])
GROUNDING_URLS = config.get('grounding_urls', DEFAULT_CONFIG['grounding_urls'])
ENABLE_DYNAMIC_URLS = config.get('enable_dynamic_urls', DEFAULT_CONFIG['enable_dynamic_urls'])
ENABLE_FILE_UPLOAD = config.get('enable_file_upload', DEFAULT_CONFIG.get('enable_file_upload', True))
LANGUAGE = config.get('language', DEFAULT_CONFIG.get('language', 'English'))
ENABLE_TTS = config.get('enable_tts', DEFAULT_CONFIG.get('enable_tts', False))
TTS_MODEL = config.get('tts_model', DEFAULT_CONFIG.get('tts_model', 'microsoft/speecht5_tts'))
TTS_VOICE = config.get('tts_voice', DEFAULT_CONFIG.get('tts_voice', 'default'))

# Environment variables
ACCESS_CODE = os.environ.get("ACCESS_CODE")
API_KEY_VAR = config.get('api_key_var', DEFAULT_CONFIG['api_key_var'])
API_KEY = os.environ.get(API_KEY_VAR, "").strip() or None
HF_TOKEN = os.environ.get('HF_TOKEN', '')
SPACE_ID = os.environ.get('SPACE_ID', '')


# Utility functions
def validate_api_key() -> bool:
    """Validate API key configuration"""
    if not API_KEY:
        print(f"⚠️  API KEY CONFIGURATION ERROR:")
        print(f"   Variable name: {API_KEY_VAR}")
        print(f"   Status: Not set or empty")
        print(f"   Action needed: Set '{API_KEY_VAR}' in HuggingFace Space secrets")
        return False
    elif not API_KEY.startswith('sk-or-'):
        print(f"⚠️  API KEY FORMAT WARNING:")
        print(f"   Variable name: {API_KEY_VAR}")
        print(f"   Note: OpenRouter keys should start with 'sk-or-'")
        return True
    else:
        print(f"✅ API Key configured successfully")
        return True


def validate_url_domain(url: str) -> bool:
    """Validate URL domain"""
    try:
        from urllib.parse import urlparse
        parsed = urlparse(url)
        return bool(parsed.netloc and parsed.scheme in ['http', 'https'])
    except:
        return False


def fetch_url_content(url: str, max_length: int = 3000) -> str:
    """Fetch and convert URL content to text"""
    try:
        if not validate_url_domain(url):
            return f"❌ Invalid URL format: {url}"
        
        headers = {
            'User-Agent': 'Mozilla/5.0 (compatible; HuggingFace-Space/1.0)'
        }
        
        response = requests.get(url, headers=headers, timeout=5)
        response.raise_for_status()
        
        content_type = response.headers.get('content-type', '').lower()
        
        if 'text/html' in content_type:
            soup = BeautifulSoup(response.text, 'html.parser')
            
            # Remove script and style elements
            for script in soup(["script", "style"]):
                script.extract()
            
            # Get text content
            text = soup.get_text(separator=' ', strip=True)
            
            # Clean up whitespace
            text = ' '.join(text.split())
            
            # Limit content length
            if len(text) > max_length:
                text = text[:max_length] + "... [truncated]"
            
            return f"📄 **Content from:** {url}\n\n{text}\n"
        
        elif any(ct in content_type for ct in ['text/plain', 'application/json']):
            text = response.text
            if len(text) > max_length:
                text = text[:max_length] + "... [truncated]"
            return f"📄 **Content from:** {url}\n\n{text}\n"
        
        else:
            return f"⚠️ Unsupported content type at {url}: {content_type}"
            
    except requests.exceptions.Timeout:
        return f"⏱️ Timeout accessing {url}"
    except requests.exceptions.RequestException as e:
        return f"❌ Error accessing {url}: {str(e)}"
    except Exception as e:
        return f"❌ Unexpected error with {url}: {str(e)}"


def extract_urls_from_text(text: str) -> List[str]:
    """Extract URLs from message text"""
    url_pattern = r'https?://[^\s<>"{}|\\^`\[\]]+(?:\.[^\s<>"{}|\\^`\[\]])*'
    urls = re.findall(url_pattern, text)
    return [url.rstrip('.,;:)?!') for url in urls]


def process_file_upload(file_path: str) -> str:
    """Process uploaded file with Gradio best practices"""
    if not file_path or not os.path.exists(file_path):
        return "❌ File not found"
    
    try:
        file_size = os.path.getsize(file_path)
        file_name = os.path.basename(file_path)
        _, ext = os.path.splitext(file_path.lower())
        
        # Text file extensions
        text_extensions = {
            '.txt', '.md', '.markdown', '.rst',
            '.py', '.js', '.jsx', '.ts', '.tsx', '.json', '.yaml', '.yml',
            '.html', '.htm', '.xml', '.css', '.scss',
            '.java', '.c', '.cpp', '.h', '.cs', '.go', '.rs',
            '.sh', '.bash', '.log', '.csv', '.sql'
        }
        
        max_chars = 5000  # Define max_chars limit for file reading
        
        if ext in text_extensions:
            with open(file_path, 'r', encoding='utf-8', errors='ignore') as f:
                content = f.read(max_chars)
                if len(content) == max_chars:
                    content += "\n... [truncated]"
                return f"📄 **{file_name}** ({file_size:,} bytes)\n```{ext[1:]}\n{content}\n```"
        
        # Special file types
        elif ext == '.pdf':
            return f"📑 **{file_name}** (PDF, {file_size:,} bytes)\n⚠️ PDF support requires PyPDF2"
        elif ext in {'.jpg', '.jpeg', '.png', '.gif', '.webp'}:
            return f"🖼️ **{file_name}** (Image, {file_size:,} bytes)"
        elif ext in {'.xlsx', '.xls'}:
            return f"📊 **{file_name}** (Spreadsheet, {file_size:,} bytes)"
        elif ext in {'.zip', '.tar', '.gz', '.rar'}:
            return f"🗜️ **{file_name}** (Archive, {file_size:,} bytes)"
        else:
            return f"📎 **{file_name}** ({ext or 'no extension'}, {file_size:,} bytes)"
            
    except Exception as e:
        return f"❌ Error processing file: {str(e)}"


# URL content cache
_url_content_cache = {}


def get_grounding_context() -> str:
    """Get grounding context from configured URLs with caching"""
    urls = GROUNDING_URLS
    if isinstance(urls, str):
        try:
            urls = json.loads(urls)
        except:
            return ""
    
    if not urls:
        return ""
    
    context_parts = []
    
    # Process primary sources (first 2 URLs with 8000 char limit)
    primary_urls = urls[:2]
    if primary_urls:
        context_parts.append("📚 **PRIMARY SOURCES:**\n")
        for i, url in enumerate(primary_urls, 1):
            if url in _url_content_cache:
                content = _url_content_cache[url]
            else:
                content = fetch_url_content(url, max_length=8000)
                _url_content_cache[url] = content
            
            if not content.startswith("❌") and not content.startswith("⏱️"):
                context_parts.append(f"\n**Primary Source {i} - {url}:**\n{content}")
    
    # Process secondary sources (URLs 3+ with 2500 char limit)
    secondary_urls = urls[2:]
    if secondary_urls:
        context_parts.append("\n\n📎 **SECONDARY SOURCES:**\n")
        for i, url in enumerate(secondary_urls, 1):
            if url in _url_content_cache:
                content = _url_content_cache[url]
            else:
                content = fetch_url_content(url, max_length=2500)
                _url_content_cache[url] = content
            
            if not content.startswith("❌") and not content.startswith("⏱️"):
                context_parts.append(f"\n**Secondary Source {i} - {url}:**\n{content}")
    
    if len(context_parts) > 0:
        return "\n".join(context_parts)
    return ""


def export_conversation_to_markdown(history: List[Dict[str, str]]) -> str:
    """Export conversation history to markdown"""
    if not history:
        return "No conversation to export."
    
    markdown_content = f"""# Conversation Export
Generated on: {datetime.now().strftime('%Y-%m-%d %H:%M:%S')}
Space: {SPACE_NAME}
Model: {MODEL}

---

"""
    
    message_count = 0
    for message in history:
        if isinstance(message, dict):
            role = message.get('role', 'unknown')
            content = message.get('content', '')
            
            if role == 'user':
                message_count += 1
                markdown_content += f"## User Message {message_count}\n\n{content}\n\n"
            elif role == 'assistant':
                markdown_content += f"## Assistant Response {message_count}\n\n{content}\n\n---\n\n"
    
    return markdown_content


def generate_response(message: str, history: List[Dict[str, str]], files: Optional[List] = None) -> str:
    """Generate response using OpenRouter API with file support"""
    
    # API key validation
    if not API_KEY:
        return f"""🔑 **API Key Required**

Please configure your OpenRouter API key:
1. Go to Settings (⚙️) in your HuggingFace Space
2. Click 'Variables and secrets'
3. Add secret: **{API_KEY_VAR}**
4. Value: Your OpenRouter API key (starts with `sk-or-`)

Get your API key at: https://openrouter.ai/keys"""
    
    # Process files if provided
    file_context = ""
    file_notification = ""
    
    if files:
        file_contents = []
        file_names = []
        
        for file_info in files:
            if isinstance(file_info, dict):
                file_path = file_info.get('path', file_info.get('name', ''))
            else:
                file_path = str(file_info)
            
            if file_path and os.path.exists(file_path):
                try:
                    content = process_file_upload(file_path)
                    file_contents.append(content)
                    file_names.append(os.path.basename(file_path))
                    print(f"📄 Processed file: {os.path.basename(file_path)}")
                except Exception as e:
                    print(f"❌ Error processing file: {e}")
        
        if file_contents:
            file_context = "\n\n[UPLOADED FILES]\n" + "\n\n".join(file_contents) + "\n"
            file_notification = f"\n\n[Note: Uploaded files: {', '.join(file_names)}]"
    
    # Get grounding context
    grounding_context = get_grounding_context()
    
    # Check for dynamic URLs in message
    if ENABLE_DYNAMIC_URLS:
        urls_in_message = extract_urls_from_text(message)
        if urls_in_message:
            print(f"🔗 Found {len(urls_in_message)} URLs in message")
            dynamic_context = "\n📎 **Dynamic Context:**\n"
            for url in urls_in_message[:3]:  # Limit to 3 URLs
                content = fetch_url_content(url)
                if not content.startswith("❌"):
                    dynamic_context += f"\n{content}"
            grounding_context += dynamic_context
    
    # Build messages with grounding context and file context in system prompt
    system_content = SYSTEM_PROMPT
    
    # Add language instruction if not English
    if LANGUAGE != 'English':
        system_content += f"\n\nIMPORTANT: You must respond EXCLUSIVELY in {LANGUAGE}. All your responses should be written entirely in {LANGUAGE}, even when user input is in a different language, particularly English."
    
    if grounding_context:
        system_content += "\n\nIMPORTANT: When providing information from the reference sources below, please cite the specific URL(s) where the information can be found. Format citations as plain URLs without brackets, or use markdown link format like [text](url). Never use hard brackets like 【url】 as they break the links."
        system_content = f"{system_content}\n\n{grounding_context}"
    if file_context:
        system_content = f"{system_content}\n\n{file_context}"
    
    messages = [{"role": "system", "content": system_content}]
    
    # Add conversation history
    for msg in history:
        if isinstance(msg, dict) and 'role' in msg and 'content' in msg:
            messages.append({
                "role": msg['role'],
                "content": msg['content']
            })
    
    # Add current message
    messages.append({
        "role": "user",
        "content": message
    })
    
    # Make API request
    try:
        # Make API request
        headers = {
            "Authorization": f"Bearer {API_KEY}",
            "Content-Type": "application/json",
            "HTTP-Referer": f"https://huggingface.co/spaces/{SPACE_ID}" if SPACE_ID else "https://huggingface.co",
            "X-Title": SPACE_NAME
        }
        
        data = {
            "model": MODEL,
            "messages": messages,
            "temperature": temperature,
            "max_tokens": max_tokens,
            "stream": False
        }
        
        response = requests.post(
            "https://openrouter.ai/api/v1/chat/completions",
            headers=headers,
            json=data,
            timeout=30
        )
        
        if response.status_code == 200:
            try:
                result = response.json()
                ai_response = result['choices'][0]['message']['content']
            except (json.JSONDecodeError, KeyError) as e:
                return f"❌ Error parsing API response: {str(e)}"
            
            # Add file notification if files were uploaded
            if file_notification:
                ai_response += file_notification
            
            return ai_response
        else:
            try:
                error_data = response.json()
                error_message = error_data.get('error', {}).get('message', 'Unknown error')
            except:
                error_message = response.text if response.text else 'Unknown error'
            return f"❌ API Error ({response.status_code}): {error_message}"
            
    except requests.exceptions.Timeout:
        return "⏰ Request timeout (30s limit). Try a shorter message or different model."
    except requests.exceptions.ConnectionError:
        return "🌐 Connection error. Check your internet connection and try again."
    except Exception as e:
        return f"❌ Error: {str(e)}"


# Chat history for export
chat_history_store = []


def verify_hf_token_access() -> Tuple[bool, str]:
    """Verify HuggingFace token and access"""
    if not HF_TOKEN:
        return False, "No HF_TOKEN found"
    
    if not SPACE_ID:
        return False, "No SPACE_ID found - running locally?"
    
    try:
        headers = {"Authorization": f"Bearer {HF_TOKEN}"}
        response = requests.get(
            f"https://huggingface.co/api/spaces/{SPACE_ID}",
            headers=headers,
            timeout=5
        )
        if response.status_code == 200:
            return True, f"HF Token valid for {SPACE_ID}"
        else:
            return False, f"HF Token invalid or no access to {SPACE_ID}"
    except Exception as e:
        return False, f"Error verifying HF token: {str(e)}"


def generate_tts(text: str, max_retries: int = 2) -> Tuple[Optional[Tuple[int, np.ndarray]], str]:
    """Generate TTS audio using HuggingFace Inference API"""
    if not ENABLE_TTS or not text:
        return None, "TTS disabled or no text provided"
    
    hf_token = os.getenv("HF_TOKEN")
    if not hf_token:
        return None, "⚠️ HF_TOKEN not configured for TTS"
    
    # Limit text length for TTS
    text = text[:500]
    
    # Select speaker embedding based on voice preference
    speaker_embeddings = None
    if TTS_MODEL == "microsoft/speecht5_tts":
        # For SpeechT5, we need speaker embeddings
        # Using a predefined speaker ID (7306 is a clear female voice)
        speaker_id = {
            "default": 7306,
            "female": 7306,
            "male": 5105,
            "neutral": 6678
        }.get(TTS_VOICE, 7306)
        
        # Note: In production, you'd load actual embeddings from the dataset
        # For now, we'll let the API handle default voice
        speaker_embeddings = {"speaker_id": speaker_id}
    
    for attempt in range(max_retries):
        try:
            headers = {"Authorization": f"Bearer {hf_token}"}
            api_url = f"https://api-inference.huggingface.co/models/{TTS_MODEL}"
            
            # Prepare payload
            payload = {"inputs": text}
            if speaker_embeddings and TTS_MODEL == "microsoft/speecht5_tts":
                # For models that support speaker embeddings
                payload["parameters"] = speaker_embeddings
            
            response = requests.post(
                api_url,
                headers=headers,
                json=payload,
                timeout=20
            )
            
            if response.status_code == 200:
                # Convert audio bytes to numpy array
                audio_array = np.frombuffer(response.content, dtype=np.int16)
                # Most TTS models output at 16kHz
                sample_rate = 16000
                return (sample_rate, audio_array), "✅ Audio generated successfully"
            
            elif response.status_code == 503:
                # Model is loading
                if attempt < max_retries - 1:
                    time.sleep(20)  # Wait for model to load
                    continue
                else:
                    return None, "⏳ Model is loading, please try again in a moment"
            
            else:
                try:
                    error_msg = response.json().get('error', 'Unknown error')
                except:
                    error_msg = response.text if response.text else 'Unknown error'
                return None, f"❌ API Error ({response.status_code}): {error_msg}"
                
        except requests.exceptions.Timeout:
            return None, "⏰ TTS request timeout"
        except Exception as e:
            if attempt == max_retries - 1:
                return None, f"❌ TTS Error: {str(e)}"
            time.sleep(2)
    
    return None, "❌ Max retries exceeded"


# Create main interface with clean tab structure
def create_interface():
    """Create the Gradio interface with clean tab structure"""
    
    # Get theme
    theme = AVAILABLE_THEMES.get(THEME, gr.themes.Default())
    
    # Validate API key on startup
    API_KEY_VALID = validate_api_key()
    
    # Check HuggingFace access
    HF_ACCESS_VALID, HF_ACCESS_MESSAGE = verify_hf_token_access()
    
    # Access control check
    has_access = ACCESS_CODE is None  # No access code required
    
    with gr.Blocks(title=SPACE_NAME, theme=theme) as demo:
        # State for access control
        access_granted = gr.State(has_access)
        
        # Header - always visible
        gr.Markdown(f"# {SPACE_NAME}")
        gr.Markdown(SPACE_DESCRIPTION)
        
        # Access control panel (visible when access not granted)
        with gr.Column(visible=(not has_access)) as access_panel:
            gr.Markdown("### 🔐 Access Required")
            gr.Markdown("Please enter the access code:")
            
            with gr.Row():
                access_input = gr.Textbox(
                    label="Access Code",
                    placeholder="Enter access code...",
                    type="password",
                    scale=3
                )
                access_btn = gr.Button("Submit", variant="primary", scale=1)
            
            access_status = gr.Markdown()
        
        # Main interface (visible when access granted)
        with gr.Column(visible=has_access) as main_panel:
            with gr.Tabs() as tabs:
                # Chat Tab
                with gr.Tab("💬 Chat"):
                    # Get examples
                    examples = config.get('examples', [])
                    if isinstance(examples, str):
                        try:
                            examples = json.loads(examples)
                        except:
                            examples = []
                    
                    # State to hold uploaded files
                    uploaded_files = gr.State([])
                    
                    # Create chat interface
                    chatbot = gr.Chatbot(type="messages", height=400)
                    msg = gr.Textbox(label="Message", placeholder="Type your message here...", lines=2)
                    
                    with gr.Row():
                        submit_btn = gr.Button("Send", variant="primary")
                        clear_btn = gr.Button("Clear")
                    
                    # Export functionality
                    with gr.Row():
                        export_btn = gr.DownloadButton(
                            "📥 Export Conversation",
                            variant="secondary",
                            size="sm"
                        )
                    
                    # Export handler
                    def prepare_export():
                        if not chat_history_store:
                            return None
                        
                        content = export_conversation_to_markdown(chat_history_store)
                        
                        # Create filename
                        space_name_safe = re.sub(r'[^a-zA-Z0-9]+', '_', SPACE_NAME).lower()
                        timestamp = datetime.now().strftime('%Y%m%d_%H%M%S')
                        filename = f"{space_name_safe}_conversation_{timestamp}.md"
                        
                        # Save to temp file
                        temp_path = Path(tempfile.gettempdir()) / filename
                        temp_path.write_text(content, encoding='utf-8')
                        
                        return str(temp_path)
                    
                    export_btn.click(
                        prepare_export,
                        outputs=[export_btn]
                    )
                    
                    # TTS functionality
                    if ENABLE_TTS:
                        with gr.Row():
                            tts_btn = gr.Button("🔊 Read Last Response", variant="secondary", size="sm")
                            audio_output = gr.Audio(label="TTS Output", visible=False, autoplay=True)
                        
                        tts_status = gr.Markdown("", visible=False)
                        last_assistant_message = gr.State("")
                        
                        def update_last_message(chat_history):
                            """Extract the last assistant message from chat history"""
                            if not chat_history:
                                return ""
                            
                            # Find the last assistant message
                            for message in reversed(chat_history):
                                if isinstance(message, dict) and message.get('role') == 'assistant':
                                    return message.get('content', '')
                            return ""
                        
                        def handle_tts_click(last_message):
                            """Handle TTS button click"""
                            if not last_message:
                                return None, gr.update(visible=False), gr.update(value="⚠️ No message to read", visible=True)
                            
                            audio_data, status_msg = generate_tts(last_message)
                            
                            if audio_data:
                                return (
                                    audio_data,
                                    gr.update(visible=True),
                                    gr.update(value=status_msg, visible=True)
                                )
                            else:
                                return (
                                    None,
                                    gr.update(visible=False),
                                    gr.update(value=status_msg, visible=True)
                                )
                        
                        # Update last message whenever chat updates
                        chatbot.change(
                            update_last_message,
                            inputs=[chatbot],
                            outputs=[last_assistant_message]
                        )
                        
                        # Handle TTS button click
                        tts_btn.click(
                            handle_tts_click,
                            inputs=[last_assistant_message],
                            outputs=[audio_output, audio_output, tts_status]
                        )
                    
                    # Examples section
                    if examples:
                        gr.Examples(examples=examples, inputs=msg)
                    
                    # Chat functionality
                    def respond(message, chat_history, files_state, is_granted):
                        if not is_granted:
                            return chat_history, "", is_granted
                        
                        if not message:
                            return chat_history, "", is_granted
                        
                        # Format history for the generate_response function
                        formatted_history = []
                        for h in chat_history:
                            if isinstance(h, dict):
                                formatted_history.append(h)
                        
                        # Get response
                        response = generate_response(message, formatted_history, files_state)
                        
                        # Update chat history
                        chat_history = chat_history + [
                            {"role": "user", "content": message},
                            {"role": "assistant", "content": response}
                        ]
                        
                        # Update stored history for export
                        global chat_history_store
                        chat_history_store = chat_history
                        
                        return chat_history, "", is_granted
                    
                    # Wire up the interface
                    msg.submit(respond, [msg, chatbot, uploaded_files, access_granted], [chatbot, msg, access_granted])
                    submit_btn.click(respond, [msg, chatbot, uploaded_files, access_granted], [chatbot, msg, access_granted])
                    clear_btn.click(lambda: ([], ""), outputs=[chatbot, msg])
                    
                    # File upload accordion
                    if ENABLE_FILE_UPLOAD:
                        with gr.Accordion("📎 Upload Files", open=False):
                            file_upload = gr.File(
                                label="Upload Files",
                                file_types=None,
                                file_count="multiple",
                                visible=True,
                                interactive=True
                            )
                            clear_files_btn = gr.Button("Clear Files", size="sm", variant="secondary")
                            uploaded_files_display = gr.Markdown("", visible=False)
                            
                            def handle_file_upload(files):
                                if not files:
                                    return [], "", gr.update(visible=False)
                                
                                file_names = []
                                for file_info in files:
                                    if isinstance(file_info, dict):
                                        file_path = file_info.get('path', file_info.get('name', ''))
                                    else:
                                        file_path = str(file_info)
                                    
                                    if file_path and os.path.exists(file_path):
                                        file_names.append(os.path.basename(file_path))
                                
                                if file_names:
                                    display_text = f"📎 **Uploaded files:** {', '.join(file_names)}"
                                    return files, display_text, gr.update(visible=True)
                                return [], "", gr.update(visible=False)
                            
                            def clear_files():
                                return None, [], "", gr.update(visible=False)
                            
                            file_upload.change(
                                handle_file_upload,
                                inputs=[file_upload],
                                outputs=[uploaded_files, uploaded_files_display, uploaded_files_display]
                            )
                            
                            clear_files_btn.click(
                                clear_files,
                                outputs=[file_upload, uploaded_files, uploaded_files_display, uploaded_files_display]
                            )
                    
                    # Configuration accordion
                    with gr.Accordion("ℹ️ Configuration", open=False):
                        gr.JSON(
                            value=config,
                            label="config.json",
                            show_label=True
                        )
                
                # Configuration Tab
                with gr.Tab("⚙️ Configuration"):
                    gr.Markdown("## Configuration Management")
                    
                    # State for config tab authentication
                    config_authenticated = gr.State(False)
                    
                    # Authentication panel
                    with gr.Column(visible=True) as config_auth_panel:
                        gr.Markdown("### 🔐 Authentication Required")
                        gr.Markdown("Enter your HF_TOKEN to access configuration settings:")
                        
                        with gr.Row():
                            config_password = gr.Textbox(
                                label="HF Token",
                                placeholder="Enter your HF_TOKEN...",
                                type="password",
                                scale=3
                            )
                            config_auth_btn = gr.Button("Authenticate", variant="primary", scale=1)
                        
                        config_auth_status = gr.Markdown()
                    
                    # Configuration panel (hidden until authenticated)
                    with gr.Column(visible=False) as config_panel:
                        # Show authentication status
                        if HF_ACCESS_VALID:
                            gr.Markdown(f"✅ {HF_ACCESS_MESSAGE}")
                            gr.Markdown("Configuration changes will be saved to the HuggingFace repository.")
                        else:
                            gr.Markdown(f"ℹ️ {HF_ACCESS_MESSAGE}")
                            gr.Markdown("Set HF_TOKEN in Space secrets to enable auto-save.")
                        
                        # Configuration editor
                        gr.Markdown("### ⚙️ Configuration Editor")
                        
                        # Show lock status if locked
                        if config.get('locked', False):
                            gr.Markdown("⚠️ **Note:** Configuration is locked.")
                        
                        # Basic settings
                        with gr.Column():
                            edit_name = gr.Textbox(
                            label="Space Name",
                            value=config.get('name', ''),
                            max_lines=1
                        )
                        edit_model = gr.Dropdown(
                            label="Model",
                            choices=[
                                # Google models
                                "google/gemini-2.0-flash-001",
                                "google/gemma-3-27b-it",
                                # Anthropic models
                                "anthropic/claude-3.5-sonnet",
                                "anthropic/claude-3.5-haiku",
                                # OpenAI models
                                "openai/gpt-4o-mini",
                                "openai/gpt-4o-mini-search-preview",
                                "openai/gpt-oss-120b",
                                # MistralAI models
                                "mistralai/mistral-medium-3",
                                # DeepSeek models
                                "deepseek/deepseek-r1-distill-qwen-32b",
                                # NVIDIA models
                                "nvidia/llama-3.1-nemotron-70b-instruct",
                                # Qwen models
                                "qwen/qwen3-30b-a3b-instruct-2507"
                            ],
                            value=config.get('model', ''),
                            allow_custom_value=True
                        )
                        
                        edit_language = gr.Dropdown(
                            label="Language",
                            choices=[
                                "Arabic",
                                "Bengali",
                                "English",
                                "French",
                                "German",
                                "Hindi",
                                "Italian",
                                "Japanese",
                                "Korean",
                                "Mandarin",
                                "Portuguese",
                                "Russian",
                                "Spanish",
                                "Turkish"
                            ],
                            value=config.get('language', 'English')
                        )
                        
                        edit_description = gr.Textbox(
                            label="Description",
                            value=config.get('description', ''),
                            max_lines=2
                        )
                        
                        edit_system_prompt = gr.Textbox(
                            label="System Prompt",
                            value=config.get('system_prompt', ''),
                            lines=5
                        )
                        
                        with gr.Row():
                            edit_temperature = gr.Slider(
                                label="Temperature",
                                minimum=0,
                                maximum=2,
                                value=config.get('temperature', 0.7),
                                step=0.1
                            )
                            edit_max_tokens = gr.Slider(
                                label="Max Tokens",
                                minimum=50,
                                maximum=4096,
                                value=config.get('max_tokens', 750),
                                step=50
                            )
                        
                        edit_examples = gr.Textbox(
                            label="Example Prompts (one per line)",
                            value='\n'.join(config.get('examples', [])),
                            lines=3
                        )
                        
                        # URL Grounding
                        gr.Markdown("### URL Grounding")
                        edit_grounding_urls = gr.Textbox(
                            label="Grounding URLs (one per line)",
                            placeholder="https://example.com/docs\nhttps://example.com/api",
                            value='\n'.join(config.get('grounding_urls', [])),
                            lines=5,
                            info="First 2 URLs: Primary sources (8000 chars). URLs 3+: Secondary sources (2500 chars)."
                        )
                        
                        with gr.Row():
                            edit_enable_dynamic_urls = gr.Checkbox(
                                label="Enable Dynamic URL Extraction",
                                value=config.get('enable_dynamic_urls', True),
                                info="Extract and fetch URLs from user messages"
                            )
                            edit_enable_file_upload = gr.Checkbox(
                                label="Enable File Upload",
                                value=config.get('enable_file_upload', True),
                                info="Allow users to upload files for context"
                            )
                        
                        # TTS Configuration
                        gr.Markdown("### 🔊 Text-to-Speech")
                        with gr.Row():
                            edit_enable_tts = gr.Checkbox(
                                label="Enable TTS",
                                value=config.get('enable_tts', False),
                                info="Enable text-to-speech for assistant responses"
                            )
                            edit_tts_model = gr.Dropdown(
                                label="TTS Model",
                                choices=[
                                    "microsoft/speecht5_tts",
                                    "facebook/mms-tts-eng",
                                    "suno/bark",
                                    "parler-tts/parler-tts-mini-v1"
                                ],
                                value=config.get('tts_model', 'microsoft/speecht5_tts'),
                                allow_custom_value=True
                            )
                            edit_tts_voice = gr.Dropdown(
                                label="Voice",
                                choices=["default", "female", "male", "neutral"],
                                value=config.get('tts_voice', 'default')
                            )
                        
                        # Configuration actions
                        with gr.Row():
                            save_btn = gr.Button("💾 Save Configuration", variant="primary")
                            reset_btn = gr.Button("↩️ Reset to Defaults", variant="secondary")
                        
                        config_status = gr.Markdown()
                        
                        def save_configuration(name, description, system_prompt, model, language, temp, tokens, examples, grounding_urls, enable_dynamic_urls, enable_file_upload, enable_tts, tts_model, tts_voice):
                            """Save updated configuration"""
                            try:
                                updated_config = config.copy()
                                updated_config.update({
                                    'name': name,
                                    'description': description,
                                    'system_prompt': system_prompt,
                                    'model': model,
                                    'language': language,
                                    'temperature': temp,
                                    'max_tokens': int(tokens),
                                    'examples': [ex.strip() for ex in examples.split('\n') if ex.strip()],
                                    'grounding_urls': [url.strip() for url in grounding_urls.split('\n') if url.strip()],
                                    'enable_dynamic_urls': enable_dynamic_urls,
                                    'enable_file_upload': enable_file_upload,
                                    'enable_tts': enable_tts,
                                    'tts_model': tts_model,
                                    'tts_voice': tts_voice,
                                    'locked': config.get('locked', False)
                                })
                                
                                if config_manager.save(updated_config):
                                    # Auto-commit if HF token is available
                                    if HF_TOKEN and SPACE_ID:
                                        try:
                                            from huggingface_hub import HfApi, CommitOperationAdd
                                            api = HfApi(token=HF_TOKEN)
                                            
                                            operations = [
                                                CommitOperationAdd(
                                                    path_or_fileobj=config_manager.config_path,
                                                    path_in_repo="config.json"
                                                )
                                            ]
                                            
                                            api.create_commit(
                                                repo_id=SPACE_ID,
                                                operations=operations,
                                                commit_message="Update configuration via web UI",
                                                commit_description=f"Configuration update at {datetime.now().strftime('%Y-%m-%d %H:%M:%S')}",
                                                repo_type="space",
                                                token=HF_TOKEN
                                            )
                                            return "✅ Configuration saved and committed to repository!"
                                        except Exception as e:
                                            return f"✅ Configuration saved locally. ⚠️ Auto-commit failed: {str(e)}"
                                    else:
                                        return "✅ Configuration saved locally (no HF token for auto-commit)"
                                else:
                                    return "❌ Failed to save configuration"
                                    
                            except Exception as e:
                                return f"❌ Error: {str(e)}"
                        
                        save_btn.click(
                            save_configuration,
                            inputs=[edit_name, edit_description, edit_system_prompt, edit_model, edit_language,
                                   edit_temperature, edit_max_tokens, edit_examples, edit_grounding_urls,
                                   edit_enable_dynamic_urls, edit_enable_file_upload, edit_enable_tts, edit_tts_model, edit_tts_voice],
                            outputs=[config_status]
                        )
                        
                        def reset_configuration():
                            """Reset to default configuration"""
                            try:
                                if config_manager.save(DEFAULT_CONFIG):
                                    return (
                                        DEFAULT_CONFIG['name'],
                                        DEFAULT_CONFIG['description'],
                                        DEFAULT_CONFIG['system_prompt'],
                                        DEFAULT_CONFIG['model'],
                                        DEFAULT_CONFIG.get('language', 'English'),
                                        DEFAULT_CONFIG['temperature'],
                                        DEFAULT_CONFIG['max_tokens'],
                                        '\n'.join(DEFAULT_CONFIG['examples']),
                                        '\n'.join(DEFAULT_CONFIG['grounding_urls']),
                                        DEFAULT_CONFIG['enable_dynamic_urls'],
                                        DEFAULT_CONFIG['enable_file_upload'],
                                        DEFAULT_CONFIG.get('enable_tts', False),
                                        DEFAULT_CONFIG.get('tts_model', 'microsoft/speecht5_tts'),
                                        DEFAULT_CONFIG.get('tts_voice', 'default'),
                                        "✅ Reset to default configuration"
                                    )
                                else:
                                    return (*[gr.update() for _ in range(14)], "❌ Failed to reset")
                            except Exception as e:
                                return (*[gr.update() for _ in range(14)], f"❌ Error: {str(e)}")
                        
                        reset_btn.click(
                            reset_configuration,
                            outputs=[edit_name, edit_description, edit_system_prompt, edit_model, edit_language,
                                    edit_temperature, edit_max_tokens, edit_examples, edit_grounding_urls,
                                    edit_enable_dynamic_urls, edit_enable_file_upload, edit_enable_tts, edit_tts_model, edit_tts_voice, config_status]
                        )
                        
                        # Configuration tab authentication handler
                        def handle_config_auth(password):
                            """Handle configuration tab authentication"""
                            if not HF_TOKEN:
                                return (
                                    gr.update(visible=True),   # Keep auth panel visible
                                    gr.update(visible=False),  # Keep config panel hidden
                                    gr.update(value="❌ No HF_TOKEN is set in Space secrets. Configuration cannot be enabled."),
                                    False
                                )
                            
                            if password == HF_TOKEN:
                                return (
                                    gr.update(visible=False),  # Hide auth panel
                                    gr.update(visible=True),   # Show config panel
                                    gr.update(value="✅ Authentication successful!"),
                                    True
                                )
                            else:
                                return (
                                    gr.update(visible=True),   # Keep auth panel visible
                                    gr.update(visible=False),  # Keep config panel hidden
                                    gr.update(value="❌ Invalid HF_TOKEN. Please try again."),
                                    False
                                )
                        
                        config_auth_btn.click(
                            handle_config_auth,
                            inputs=[config_password],
                            outputs=[config_auth_panel, config_panel, config_auth_status, config_authenticated]
                        )
                        
                        config_password.submit(
                            handle_config_auth,
                            inputs=[config_password],
                            outputs=[config_auth_panel, config_panel, config_auth_status, config_authenticated]
                        )
        
        # Access control handler
        if ACCESS_CODE:
            def handle_access(code, current_state):
                if code == ACCESS_CODE:
                    return (
                        gr.update(visible=False),  # Hide access panel
                        gr.update(visible=True),   # Show main panel
                        gr.update(value="✅ Access granted!"),  # Status message
                        True  # Update state
                    )
                else:
                    return (
                        gr.update(visible=True),   # Keep access panel visible
                        gr.update(visible=False),  # Keep main panel hidden
                        gr.update(value="❌ Invalid access code. Please try again."),  # Status message
                        False  # State remains false
                    )
            
            access_btn.click(
                handle_access,
                inputs=[access_input, access_granted],
                outputs=[access_panel, main_panel, access_status, access_granted]
            )
            
            access_input.submit(
                handle_access,
                inputs=[access_input, access_granted],
                outputs=[access_panel, main_panel, access_status, access_granted]
            )
    
    return demo


# Create and launch the interface
if __name__ == "__main__":
    demo = create_interface()
    demo.launch()