Files
medevio/10ReadPozadavky/PRAVIDELNE_5_SaveToFileSystem single step.py
2025-11-17 06:59:45 +01:00

147 lines
3.9 KiB
Python
Raw Permalink Blame History

This file contains ambiguous Unicode characters
This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.
#!/usr/bin/env python3
# -*- coding: utf-8 -*-
import os
import shutil
import pymysql
import re
from pathlib import Path
from datetime import datetime
# ==============================
# ⚙️ CONFIGURATION
# ==============================
DB_CONFIG = {
"host": "192.168.1.76",
"port": 3307,
"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:
return re.sub(r'[<>:"/\\|?*\x00-\x1F]', "_", name).strip()
def clean_folder(folder: Path, valid_files: set):
"""Remove files that do NOT exist in MySQL for this request."""
if not folder.exists():
return
for f in folder.iterdir():
if f.is_file() and sanitize_name(f.name) not in valid_files:
print(f"🗑️ Removing unexpected file: {f.name}")
try:
f.unlink()
except Exception as e:
print(f"⚠️ Cannot delete {f}: {e}")
# ==============================
# 📥 LOAD EVERYTHING IN ONE QUERY
# ==============================
conn = pymysql.connect(**DB_CONFIG)
cur = conn.cursor(pymysql.cursors.DictCursor)
print("📥 Loading ALL metadata + BLOBs with ONE MySQL query…")
cur.execute("""
SELECT
d.id AS download_id,
d.request_id,
d.filename,
d.file_content,
p.updatedAt AS req_updated_at,
p.pacient_jmeno AS jmeno,
p.pacient_prijmeni AS prijmeni
FROM medevio_downloads d
JOIN pozadavky p ON d.request_id = p.id
ORDER BY p.updatedAt DESC, d.created_at ASC
""")
rows = cur.fetchall()
print(f"📦 Loaded {len(rows)} total file rows.\n")
conn.close()
# ==============================
# 🔄 ORGANIZE ROWS PER REQUEST
# ==============================
requests = {} # req_id → list of file dicts
for r in rows:
req_id = r["request_id"]
if req_id not in requests:
requests[req_id] = []
requests[req_id].append(r)
print(f"📌 Unique requests: {len(requests)}\n")
# ==============================
# 🧠 MAIN LOOP SAME LOGIC AS BEFORE
# ==============================
for req_id, filelist in requests.items():
# ========== GET UPDATEDAT (same logic) ==========
any_row = filelist[0]
updated_at = any_row["req_updated_at"] or datetime.now()
date_str = updated_at.strftime("%Y-%m-%d")
prijmeni = sanitize_name(any_row["prijmeni"] or "Unknown")
jmeno = sanitize_name(any_row["jmeno"] or "")
folder_name = sanitize_name(f"{date_str} {prijmeni}, {jmeno} {req_id}")
main_folder = BASE_DIR / folder_name
# ========== VALID FILES ==========
valid_files = {sanitize_name(r["filename"]) for r in filelist}
# ========== FIND OLD FOLDERS ==========
possible_dups = [
f for f in BASE_DIR.iterdir()
if f.is_dir() and req_id in f.name and f != main_folder
]
# ========== MERGE OLD FOLDERS ==========
for dup in possible_dups:
print(f"♻️ Merging 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 ==========
main_folder.mkdir(parents=True, exist_ok=True)
clean_folder(main_folder, valid_files)
# ========== SAVE FILES (fast now) ==========
for r in filelist:
filename = sanitize_name(r["filename"])
dest = main_folder / filename
if dest.exists():
continue
content = r["file_content"]
if not content:
continue
with open(dest, "wb") as f:
f.write(content)
print(f"💾 Saved: {dest.relative_to(BASE_DIR)}")
print("\n🎯 Export complete.\n")