#!/usr/bin/env python3
import fitz  # PyMuPDF
import re
from pathlib import Path
import logging

# === CONFIG ===
PDF_PATH = "/home/matrixhasyou/mutt/ALL_EMAILS_CONSOLIDATED.pdf"
OUTPUT_MD = "/home/matrixhasyou/mutt/LEGAL_REFERENCES.md"

logging.basicConfig(level=logging.INFO, format="%(message)s")
log = logging.getLogger("LegalScanner")

# Паттерны для поиска законов
PATTERNS = [
    r'(?i)ЖК\s?РФ',
    r'(?i)ГПК\s?РФ',
    r'(?i)ГК\s?РФ',
    r'(?i)КоАП\s?РФ',
    r'(?i)УК\s?РФ',
    r'(?i)НК\s?РФ',
    r'(?i)Федеральн[а-яё]+\sзакон[а-яё]*\s(?:от\s)?\d{1,2}\.\d{2}\.\d{2,4}\s?г?\s?№\s?\d+[\-ФЗ]*',
    r'(?i)Постановлен[а-яё]+\sПравительства\s(?:РФ\s)?(?:от\s)?\d{1,2}\.\d{2}\.\d{2,4}\s?г?\s?№\s?\d+',
    r'(?i)Приказ[а-яё]*\s[А-Яа-яЁё\s]+(?:от\s)?\d{1,2}\.\d{2}\.\d{2,4}\s?г?\s?№\s?\d+',
    r'(?i)СанПиН\s\d+\.\d+\.\d+\-\d+',
    r'(?i)ГОСТ\s[Р\s]*\d+\-\d+',
    r'(?i)ст\.\s?\d+(?:\.\d+)?\s?[А-Яа-яЁё\s]{2,20}РФ', # ст. 36 ЖК РФ
]

def extract_legal():
    if not Path(PDF_PATH).exists():
        log.error(f"File {PDF_PATH} not found!")
        return

    log.info(f"Opening {PDF_PATH}...")
    doc = fitz.open(PDF_PATH)
    found_refs = set()

    log.info(f"Scanning {len(doc)} pages...")
    for page_num in range(len(doc)):
        text = doc[page_num].get_text()
        for pattern in PATTERNS:
            matches = re.findall(pattern, text)
            for m in matches:
                # Очистка: убираем лишние пробелы и переносы
                clean_m = " ".join(m.split()).strip()
                found_refs.add(clean_m)
        
        if (page_num + 1) % 50 == 0:
            log.info(f"  Processed {page_num + 1} pages...")

    doc.close()

    if found_refs:
        sorted_refs = sorted(list(found_refs))
        with open(OUTPUT_MD, "w", encoding="utf-8") as f:
            f.write(f"# ⚖️ Список юридических ссылок из переписки\n\n")
            f.write(f"*Данные извлечены автоматически из ALL_EMAILS_CONSOLIDATED.pdf*\n")
            f.write(f"*Дата анализа: {Path(PDF_PATH).stat().st_mtime}*\n\n")
            
            # Группировка
            f.write("## 📚 Кодексы и Общие нормы\n")
            for ref in sorted_refs:
                if any(k in ref.upper() for k in ["РФ", "КОДЕКС"]):
                    f.write(f"- [ ] {ref}\n")
            
            f.write("\n## 📜 Постановления и Законы\n")
            for ref in sorted_refs:
                if any(k in ref.upper() for k in ["ПОСТАНОВЛЕН", "ФЕДЕРАЛЬН", "ЗАКОН", "№"]):
                    if not any(k in ref.upper() for k in ["РФ"]): # Чтобы не дублировать кодексы
                         f.write(f"- [ ] {ref}\n")

            f.write("\n## 🛠 Технические нормы (СанПиН, ГОСТ)\n")
            for ref in sorted_refs:
                if any(k in ref.upper() for k in ["САНПИН", "ГОСТ"]):
                    f.write(f"- [ ] {ref}\n")

        log.info(f"✅ Found {len(found_refs)} unique references. Saved to {OUTPUT_MD}")
    else:
        log.warning("No legal references found.")

if __name__ == "__main__":
    extract_legal()