#!/usr/bin/env python3
# -*- coding: utf-8 -*-

import os
import shutil
import pymysql
import re
from pathlib import Path
from datetime import datetime
from collections import defaultdict
import time
import sys

# Force UTF-8 output even under Windows Task Scheduler
import sys
try:
    sys.stdout.reconfigure(encoding='utf-8')
    sys.stderr.reconfigure(encoding='utf-8')
except AttributeError:
    # Python < 3.7 fallback (not needed for you, but safe)
    import io
    sys.stdout = io.TextIOWrapper(sys.stdout.buffer, encoding='utf-8')
    sys.stderr = io.TextIOWrapper(sys.stderr.buffer, encoding='utf-8')

# ==============================
# 🛡 SAFE PRINT FOR CP1250 / EMOJI
# ==============================
def safe_print(text: str = ""):
    enc = sys.stdout.encoding or ""
    if not enc.lower().startswith("utf"):
        # Strip emoji and characters outside BMP for Task Scheduler
        text = ''.join(ch for ch in text if ord(ch) < 65536)
    try:
        print(text)
    except UnicodeEncodeError:
        # ASCII fallback
        text = ''.join(ch for ch in text if ord(ch) < 128)
        print(text)


# ==============================
# ⚙️ CONFIGURATION
# ==============================
DB_CONFIG = {
    "host": "192.168.1.76",
    "port": 3306,
    "user": "root",
    "password": "Vlado9674+",
    "database": "medevio",
    "charset": "utf8mb4",
}

BASE_DIR = Path(r"u:\Dropbox\Ordinace\Dokumentace_ke_zpracování\MP")
BASE_DIR.mkdir(parents=True, exist_ok=True)


def sanitize_name(name: str) -> str:
    """Replace invalid filename characters with underscore."""
    return re.sub(r'[<>:"/\\|?*\x00-\x1F]', "_", name).strip()


def make_abbrev(title: str) -> str:
    if not title:
        return ""
    words = re.findall(r"[A-Za-zÁ-Žá-ž0-9]+", title)
    abbr = ""
    for w in words:
        if w.isdigit():
            abbr += w
        else:
            abbr += w[0]
    return abbr.upper()


# ==============================
# 🧹 DELETE UNEXPECTED FILES
# ==============================
def clean_folder(folder: Path, valid_files: set):
    if not folder.exists():
        return

    for f in folder.iterdir():
        if f.is_file():
            if f.name.startswith("▲"):
                continue
            sanitized = sanitize_name(f.name)
            if sanitized not in valid_files:
                safe_print(f"🗑️ Removing unexpected file: {f.name}")
                try:
                    f.unlink()
                except Exception as e:
                    safe_print(f"⚠️ Could not delete {f}: {e}")


# ==============================
# 📦 DB CONNECTION
# ==============================
conn = pymysql.connect(**DB_CONFIG)

cur_meta = conn.cursor(pymysql.cursors.DictCursor)
cur_blob = conn.cursor()

safe_print("🔍 Loading metadata from DB (FAST)…")

cur_meta.execute("""
    SELECT d.id AS download_id,
           d.request_id,
           d.filename,
           d.created_at,
           p.updatedAt AS req_updated_at,
           p.pacient_jmeno AS jmeno,
           p.pacient_prijmeni AS prijmeni,
           p.displayTitle
    FROM medevio_downloads d
    JOIN pozadavky p ON d.request_id = p.id
    WHERE p.updatedAt >= DATE_SUB(NOW(), INTERVAL 14 DAY)
    ORDER BY p.updatedAt DESC
""")

rows = cur_meta.fetchall()
safe_print(f"📋 Found {len(rows)} attachment records.\n")

# ==============================
# 🧠 MAIN LOOP WITH PROGRESS
# ==============================

# Group rows by request_id in Python — avoids N extra SELECT filename queries
rows_by_request = defaultdict(list)
for r in rows:
    rows_by_request[r["request_id"]].append(r)

total_requests = len(rows_by_request)
safe_print(f"🔄 Processing {total_requests} unique requests...\n")

# Pre-index BASE_DIR once — avoids iterdir() called twice per request
folder_list = [(f, f.name) for f in BASE_DIR.iterdir() if f.is_dir()]

for current_index, (req_id, req_rows) in enumerate(rows_by_request.items(), 1):
    percent = (current_index / total_requests) * 100
    safe_print(f"\n[ {percent:5.1f}% ] Processing request {current_index} / {total_requests} → {req_id}")

    # ========== VALID FILENAMES from already-loaded rows ==========
    # original filename → sanitized name (needed for DB query later)
    file_map = {sanitize_name(r["filename"]): r["filename"] for r in req_rows}
    valid_files = set(file_map.keys())

    # ========== BUILD FOLDER NAME ==========
    r = req_rows[0]
    updated_at = r["req_updated_at"] or datetime.now()
    date_str = updated_at.strftime("%Y-%m-%d")

    prijmeni = sanitize_name(r["prijmeni"] or "Unknown")
    jmeno = sanitize_name(r["jmeno"] or "")
    title = r.get("displayTitle") or ""
    abbr = make_abbrev(title)

    clean_folder_name = sanitize_name(
        f"{date_str} {prijmeni}, {jmeno} [{abbr}] {req_id}"
    )

    # ========== DETECT EXISTING FOLDER from pre-built index ==========
    req_id_str = str(req_id)
    matching = [f for f, name in folder_list if req_id_str in name]
    existing_folder = matching[0] if matching else None

    main_folder = existing_folder if existing_folder else BASE_DIR / clean_folder_name

    # ========== MERGE DUPLICATES ==========
    possible_dups = [f for f, name in folder_list if req_id_str in name and f != main_folder]

    for dup in possible_dups:
        safe_print(f"♻️ Merging duplicate folder: {dup.name}")

        clean_folder(dup, valid_files)
        main_folder.mkdir(parents=True, exist_ok=True)

        for f in dup.iterdir():
            if f.is_file():
                target = main_folder / f.name
                if not target.exists():
                    f.rename(target)

        shutil.rmtree(dup, ignore_errors=True)

    # ========== CLEAN MAIN FOLDER ==========
    clean_folder(main_folder, valid_files)

    # ========== DOWNLOAD MISSING FILES (batch blob fetch per request) ==========
    main_folder.mkdir(parents=True, exist_ok=True)
    added_new_file = False

    missing_san = [
        fn for fn in valid_files
        if not (main_folder / fn).exists() and not (main_folder / ("▲" + fn)).exists()
    ]

    if missing_san:
        # Fetch all missing blobs in a single query instead of one per file
        missing_orig = [file_map[fn] for fn in missing_san]
        placeholders = ",".join(["%s"] * len(missing_orig))
        cur_blob.execute(
            f"SELECT filename, file_content FROM medevio_downloads "
            f"WHERE request_id=%s AND filename IN ({placeholders})",
            [req_id] + missing_orig,
        )
        for blob_filename, content in cur_blob.fetchall():
            if not content:
                continue
            dest_plain = main_folder / sanitize_name(blob_filename)
            with open(dest_plain, "wb") as fh:
                fh.write(content)
            safe_print(f"💾 Wrote: {dest_plain.relative_to(BASE_DIR)}")
            added_new_file = True

    # ========== REMOVE ▲ FLAG IF NEW FILES ADDED ==========
    if added_new_file and "▲" in main_folder.name:
        new_name = main_folder.name.replace("▲", "").strip()
        new_path = main_folder.parent / new_name

        if new_path != main_folder:
            try:
                main_folder.rename(new_path)
                safe_print(f"🔄 Folder flag ▲ removed → {new_name}")
                main_folder = new_path
            except Exception as e:
                safe_print(f"⚠️ Could not rename folder: {e}")

safe_print("\n🎯 Export complete.\n")

cur_blob.close()
cur_meta.close()
conn.close()