bot.py

import logging
import os
import io
import html
import aiohttp
import socket
from urllib.parse import urlparse
from dotenv import load_dotenv  # <--- NEW IMPORT
from telegram import Update, constants
from telegram.ext import ApplicationBuilder, ContextTypes, MessageHandler, filters

# ==========================================
# ⚙️ CONFIGURATION
# ==========================================

# 1. Load environment variables from the .env file
load_dotenv() 

# 2. Retrieve values
BOT_TOKEN = os.getenv("BOT_TOKEN", "").strip().replace('"', '').replace("'", "")
BACKEND_API_URL = os.getenv("EXTERNAL_ANALYSIS_API_URL", "").strip().replace('"', '').replace("'", "")
API_KEY = os.getenv("API_KEY")

# Check if critical vars are missing
if not BOT_TOKEN or not BACKEND_API_URL:
    raise ValueError("❌ Error: BOT_TOKEN or BACKEND_API_URL is missing from .env file")

# ==========================================
# 📝 LOGGING SETUP
# ==========================================
logging.basicConfig(
    format='%(asctime)s - %(name)s - %(levelname)s - %(message)s',
    level=logging.INFO
)
logger = logging.getLogger(__name__)

# ==========================================
# 🧠 REPORT FORMATTER (JSON -> HTML)
# ==========================================
def format_analysis_report(data):
    """
    Converts the complex Backend JSON into a readable HTML Telegram message.
    """
    try:
        # --- HEADER ---
        tag = data.get("tag", "Analysis")
        overall_summary = data.get("overall_summary", "No summary provided.")
        source_cred_list = data.get("source_credibility_summary", [])
        
        # Determine icon based on tag content
        tag_lower = tag.lower()
        if "true" in tag_lower or "verified" in tag_lower:
            icon = "🟢"
        elif "false" in tag_lower or "misinfo" in tag_lower or "fake" in tag_lower:
            icon = "🔴"
        else:
            icon = "⚠️"
        
        # Start building the message
        message = f"<b>🚨 VERIFACT ANALYSIS REPORT</b>\n"
        message += "━━━━━━━━━━━━━━━━━━━\n"
        message += f"<b>Result:</b> {icon} <b>{html.escape(tag.upper())}</b>\n"
        
        # Source Credibility Summary (Average)
        if source_cred_list:
            total_score = 0
            count = 0
            for item in source_cred_list:
                if isinstance(item, dict) and 'credibility_score' in item:
                    try:
                        total_score += int(item['credibility_score'])
                        count += 1
                    except (ValueError, TypeError):
                        pass
            
            if count > 0:
                avg_score = int(total_score / count)
                # Determine label based on average
                if avg_score >= 80:
                    cred_label = "High"
                elif avg_score >= 60:
                    cred_label = "Moderate"
                else:
                    cred_label = "Low"
                message += f"<b>Source Credibility:</b> {cred_label} ({avg_score}%)\n"

        message += "\n<b>📝 Summary:</b>\n"
        message += f"<i>{html.escape(overall_summary)}</i>\n\n"

        # --- CLAIMS ANALYSIS ---
        claims = data.get("analyzed_claims", [])
        if claims:
            message += "<b>🔍 CLAIMS ANALYSIS</b>\n"
            message += "━━━━━━━━━━━━━━━━━━━\n"
            
            for i, claim in enumerate(claims, 1):
                claim_text = claim.get("claim_text", "N/A")
                conclusion = claim.get("conclusion", "N/A")
                
                message += f"<b>{i}️⃣ Claim:</b> \"{html.escape(claim_text)}\"\n"
                message += f"<b>💡 Conclusion:</b> {html.escape(conclusion)}\n"
                
                # Evidence
                supporting = claim.get("supporting_evidence", [])
                opposing = claim.get("opposing_evidence", [])
                
                if supporting:
                    message += "<b>✅ Supporting Evidence:</b>\n"
                    for ev in supporting:
                        src = html.escape(ev.get('source', 'Unknown'))
                        summ = html.escape(ev.get('summary', ''))
                        # Try to shorten source URL for display if it's a URL
                        if src.startswith('http'):
                            from urllib.parse import urlparse
                            try:
                                domain = urlparse(src).netloc
                                src_display = domain
                            except:
                                src_display = "Link"
                        else:
                            src_display = src
                            
                        message += f"• {summ} <i>({src_display})</i>\n"
                
                if opposing:
                    message += "<b>❌ Opposing Evidence:</b>\n"
                    for ev in opposing:
                        src = html.escape(ev.get('source', 'Unknown'))
                        summ = html.escape(ev.get('summary', ''))
                        # Try to shorten source URL for display
                        if src.startswith('http'):
                            from urllib.parse import urlparse
                            try:
                                domain = urlparse(src).netloc
                                src_display = domain
                            except:
                                src_display = "Link"
                        else:
                            src_display = src

                        message += f"• {summ} <i>({src_display})</i>\n"
                
                message += "\n"

        # --- FACT CHECKS ---
        all_fact_checks = []
        for claim in claims:
            all_fact_checks.extend(claim.get("fact_checking_results", []))
            
        # Filter out "None" URLs or empty results
        valid_fact_checks = [fc for fc in all_fact_checks if fc.get('url') and fc.get('url') != "None"]

        if valid_fact_checks:
            message += "<b>🔗 FACT CHECKS</b>\n"
            seen_urls = set()
            for fc in valid_fact_checks:
                url = fc.get('url', '#')
                if url not in seen_urls:
                    # Use inference or source name if available, else domain
                    source = fc.get('source', 'Fact Check')
                    if source == 'Fact Check' and url != '#':
                         from urllib.parse import urlparse
                         try:
                             source = urlparse(url).netloc
                         except:
                             pass
                    
                    source = html.escape(source)
                    message += f"• <a href='{url}'>{source}</a>\n"
                    seen_urls.add(url)
            message += "\n"

        # --- SOURCE CREDIBILITY DETAILS ---
        if source_cred_list:
             message += "<b>🛡️ SOURCE CREDIBILITY</b>\n"
             for item in source_cred_list[:5]: # Limit to top 5 to avoid spam
                 url = item.get('url', '')
                 score = item.get('credibility_score', 'N/A')
                 category = item.get('category', 'Unknown')
                 
                 # Extract domain
                 domain = "Unknown Source"
                 if url:
                     from urllib.parse import urlparse
                     try:
                         domain = urlparse(url).netloc
                     except:
                         domain = url
                 
                 message += f"• <b>{domain}</b>: {category} ({score})\n"
             message += "\n"

        # --- REVERSE IMAGE SEARCH (Optional) ---
        ris = data.get("reverse_image_search_data")
        if ris:
            ris_summary = ris.get("summary", "")
            matched = ris.get("matched_links", [])
            
            if ris_summary or matched:
                message += "<b>🖼️ IMAGE ANALYSIS</b>\n"
                if ris_summary:
                    message += f"{html.escape(ris_summary)}\n"
                
                if matched:
                    for match in matched[:3]: 
                        domain = html.escape(match.get('domain', 'Link'))
                        url = match.get('url', '#')
                        date = html.escape(match.get('date', ''))
                        message += f"• <a href='{url}'>{domain}</a> ({date})\n"

        message += "\n<i>🤖 Analysis generated by Verifact</i>"
        return message

    except Exception as e:
        logger.error(f"Formatting Error: {e}")
        return "⚠️ <b>Format Error:</b> Data received, but could not be displayed properly."

# ==========================================
# 📡 BACKEND CONNECTOR
# ==========================================
async def query_backend_pipeline(form_data):
    """
    Sends Multipart Form Data (Text + Files) to Cloud Run.
    """
    headers = {}
    if API_KEY:
        headers["Authorization"] = f"Bearer {API_KEY}"
    
    timeout = aiohttp.ClientTimeout(total=60)

    async with aiohttp.ClientSession(timeout=timeout) as session:
        try:
            async with session.post(BACKEND_API_URL, data=form_data, headers=headers) as response:
                if response.status == 200:
                    return await response.json()
                else:
                    error_text = await response.text()
                    logger.error(f"Backend Error {response.status}: {error_text}")
                    return None
        except Exception as e:
            logger.error(f"Connection Error: {e}")
            return None

# ==========================================
# 🎮 BOT HANDLERS
# ==========================================

async def start(update: Update, context: ContextTypes.DEFAULT_TYPE):
    welcome_text = (
        "👋 <b>Verifact Forwarding Bot</b>\n\n"
        "I am connected to the misinformation analysis pipeline.\n"
        "Forward me any <b>Text</b> or <b>Image</b> to verify it."
    )
    await context.bot.send_message(chat_id=update.effective_chat.id, text=welcome_text, parse_mode='HTML')

async def handle_text(update: Update, context: ContextTypes.DEFAULT_TYPE):
    user_text = update.message.text
    
    await context.bot.send_chat_action(chat_id=update.effective_chat.id, action=constants.ChatAction.TYPING)
    
    status_msg = await context.bot.send_message(
        chat_id=update.effective_chat.id, 
        text="📡 <i>Verifact is analyzing text...</i>",
        parse_mode='HTML'
    )

    data = aiohttp.FormData()
    data.add_field('text', user_text)
    data.add_field('source', 'Telegram')

    json_response = await query_backend_pipeline(data)

    if json_response:
        report = format_analysis_report(json_response)
        await context.bot.edit_message_text(
            chat_id=update.effective_chat.id,
            message_id=status_msg.message_id,
            text=report,
            parse_mode='HTML',
            disable_web_page_preview=True
        )
    else:
        await context.bot.edit_message_text(
            chat_id=update.effective_chat.id,
            message_id=status_msg.message_id,
            text="⚠️ <b>System Error:</b> The pipeline is currently unreachable or timed out.",
            parse_mode='HTML'
        )

async def handle_photo(update: Update, context: ContextTypes.DEFAULT_TYPE):
    await context.bot.send_chat_action(chat_id=update.effective_chat.id, action=constants.ChatAction.UPLOAD_PHOTO)
    
    status_msg = await context.bot.send_message(
        chat_id=update.effective_chat.id, 
        text="📡 <i>Downloading media & analyzing...</i>",
        parse_mode='HTML'
    )

    try:
        photo = update.message.photo[-1]
        file_obj = await context.bot.get_file(photo.file_id)
        
        f_memory = io.BytesIO()
        await file_obj.download_to_memory(out=f_memory)
        f_memory.seek(0) 

        data = aiohttp.FormData()
        caption_text = update.message.caption if update.message.caption else "Image analysis request"
        
        data.add_field('text', caption_text)
        data.add_field('source', 'Telegram')
        data.add_field('file', f_memory, filename='telegram_image.jpg', content_type='image/jpeg')

        json_response = await query_backend_pipeline(data)

        if json_response:
            report = format_analysis_report(json_response)
            await context.bot.edit_message_text(
                chat_id=update.effective_chat.id,
                message_id=status_msg.message_id,
                text=report,
                parse_mode='HTML',
                disable_web_page_preview=True
            )
        else:
            await context.bot.edit_message_text(
                chat_id=update.effective_chat.id,
                message_id=status_msg.message_id,
                text="⚠️ <b>Error:</b> Analysis failed or timed out.",
                parse_mode='HTML'
            )

    except Exception as e:
        logger.error(f"Image Handler Error: {e}")
        await context.bot.edit_message_text(
            chat_id=update.effective_chat.id,
            message_id=status_msg.message_id,
            text="❌ <b>Error:</b> Could not process the image file.",
            parse_mode='HTML'
        )

# ==========================================
# 🛠️ DIAGNOSTICS
# ==========================================
def check_network():
    """Checks DNS resolution for critical services."""
    logger.info("--- NETWORK DIAGNOSTICS ---")
    
    targets = [
        ("Telegram API", "api.telegram.org"),
        ("Google", "google.com")
    ]
    
    # Add Backend Host if parseable
    try:
        if BACKEND_API_URL:
            backend_host = urlparse(BACKEND_API_URL).netloc
            targets.append(("Backend API", backend_host))
    except:
        pass

    for name, host in targets:
        try:
            ip = socket.gethostbyname(host)
            logger.info(f"✅ {name} ({host}) resolved to {ip}")
        except socket.gaierror as e:
            logger.error(f"❌ {name} ({host}) DNS FAILURE: {e}")
        except Exception as e:
            logger.error(f"❌ {name} ({host}) Unexpected Error: {e}")
            
    logger.info("---------------------------")

# ==========================================
# 🚀 MAIN RUNNER
# ==========================================
if __name__ == '__main__':
    check_network()
    
    from telegram.request import HTTPXRequest
    
    # Use a robust request object with longer timeouts
    trequest = HTTPXRequest(connection_pool_size=8, read_timeout=20.0, write_timeout=20.0, connect_timeout=20.0)
    
    application = ApplicationBuilder().token(BOT_TOKEN).request(trequest).build()
    
    application.add_handler(MessageHandler(filters.COMMAND & filters.Regex(r'^/start$'), start))
    application.add_handler(MessageHandler(filters.TEXT & (~filters.COMMAND), handle_text))
    application.add_handler(MessageHandler(filters.PHOTO, handle_photo))

    print(f"✅ Bot is running.")
    print(f"🔗 Connected to Backend: {BACKEND_API_URL}")
    
    application.run_polling()