#!/usr/bin/env python3
"""
Consolidates all individual email PDFs into one giant PDF for NotebookLM.
Sorted by date (prefix in filename). Supports auto-splitting into parts if size > 180MB.
"""
import os
import sys
import fitz  # PyMuPDF
from pathlib import Path
import logging
import requests
from datetime import datetime

# === CONFIG ===
BASE_DIR = Path(__file__).parent
SOURCE_DIR = BASE_DIR / "exported_emails" / "all"
OUTPUT_FILE = BASE_DIR / "ALL_EMAILS_CONSOLIDATED.pdf"
BOT_TOKEN = "8725618164:AAH1tGalq-pw1l0t4P5c0sdLkCVJdh7IE7M"
CHAT_ID = "197957361"
PROXY = "socks5h://127.0.0.1:10808"

logging.basicConfig(level=logging.INFO, format="%(asctime)s [%(levelname)s] %(message)s")
log = logging.getLogger("Consolidator")

def send_tg_alert(count, size_mb, parts_count):
    url = f"https://api.telegram.org/bot{BOT_TOKEN}/sendMessage"
    parts_text = f"📦 <b>Частей архива:</b> {parts_count}\n" if parts_count > 1 else ""
    text = f"""
🔥 <b>СВОДНЫЙ АРХИВ ПОЧТЫ ОБНОВЛЕН!</b> 🔥

📑 <b>Файлы:</b> <code>ALL_EMAILS_CONSOLIDATED.pdf</code> (или части _partX.pdf)
📧 <b>Всего писем в архиве:</b> {count}
{parts_text}🐘 <b>Общий размер:</b> {size_mb:.2f} МБ
📅 <b>Обновлено:</b> {datetime.now().strftime('%Y-%m-%d %H:%M')}

🚀 Данные успешно отправлены в Google Drive (Domovoy_AI).
"""
    try:
        requests.post(url, json={"chat_id": CHAT_ID, "text": text, "parse_mode": "HTML"}, 
                      proxies={"https": PROXY}, timeout=20)
        log.info("✅ Telegram alert sent.")
    except Exception as e:
        log.error(f"❌ TG Alert failed: {e}")

def consolidate():
    log.info("Starting consolidation...")
    
    if not SOURCE_DIR.exists():
        log.error(f"Source directory {SOURCE_DIR} not found!")
        return

    pdf_files = list(SOURCE_DIR.glob("*.pdf"))
    if not pdf_files:
        log.warning("No PDF files found to consolidate.")
        return

    # Sort by filename (which starts with date YYYY-MM-DD or contains it)
    pdf_files.sort(key=lambda x: x.name)

    log.info(f"Found {len(pdf_files)} PDFs. Merging with part splitting (max 180MB)...")

    # Clean old parts
    for p in BASE_DIR.glob("ALL_EMAILS_CONSOLIDATED_part*.pdf"):
        try: os.unlink(p)
        except Exception as e: log.warning(f"Failed to delete old part {p}: {e}")
    if OUTPUT_FILE.exists():
        try: os.unlink(OUTPUT_FILE)
        except Exception as e: log.warning(f"Failed to delete old output: {e}")

    parts = []
    current_merger = fitz.open()
    current_size = 0
    part_num = 1
    
    for pdf_path in pdf_files:
        file_size = pdf_path.stat().st_size
        # Limit to 195 MB to guarantee fitting in NotebookLM 195MB
        if current_size + file_size > 195 * 1024 * 1024 and len(current_merger) > 0:
            part_file = BASE_DIR / f"ALL_EMAILS_CONSOLIDATED_part{part_num}.pdf"
            current_merger.save(part_file, garbage=3, deflate=True)
            log.info(f"Saved part {part_num}: {part_file} ({part_file.stat().st_size / (1024*1024):.2f} MB)")
            parts.append(part_file)
            current_merger.close()
            
            current_merger = fitz.open()
            current_size = 0
            part_num += 1

        try:
            doc = fitz.open(pdf_path)
            current_merger.insert_pdf(doc)
            doc.close()
            current_size += file_size
        except Exception as e:
            log.warning(f"Failed to add {pdf_path.name}: {e}")

    # Save last part
    if len(current_merger) > 0:
        if part_num == 1:
            current_merger.save(OUTPUT_FILE, garbage=3, deflate=True)
            log.info(f"✅ Consolidated single PDF saved: {OUTPUT_FILE} ({OUTPUT_FILE.stat().st_size / (1024*1024):.2f} MB)")
            parts.append(OUTPUT_FILE)
        else:
            part_file = BASE_DIR / f"ALL_EMAILS_CONSOLIDATED_part{part_num}.pdf"
            current_merger.save(part_file, garbage=3, deflate=True)
            log.info(f"Saved part {part_num}: {part_file} ({part_file.stat().st_size / (1024*1024):.2f} MB)")
            parts.append(part_file)
        current_merger.close()

    total_size_mb = sum(p.stat().st_size for p in parts) / (1024 * 1024)
    send_tg_alert(len(pdf_files), total_size_mb, len(parts))

if __name__ == "__main__":
    consolidate()