# -*- coding: utf-8 -*-
# =============================================================================
# Nazev:   store_cda_batch_v1.2.py
# Verze:   1.2
# Datum:   2026-06-11
# Popis:   Davkove ulozi binarky CDA (PDF) do Mongo k investigatorum
#          (feasibility.investigators -> cda.data_*). Zdroj = .msg soubory na
#          Toweru (/mnt/user/JNJEMAILS), stazene pres SFTP, priloha vytazena
#          extract_msg. Mapovani investigator -> (.msg, attachment) je
#          explicitni (zadne hadani za behu). Drzi se domluvy: fyzicky
#          dokument z e-mailu -> do Mongo (CDA fyzicky ulozeno k lekarum).
#          Zapise: cda.data_base64, cda.data_sha256, cda.data_filename,
#                  cda.data_mime, cda.data_size, cda.data_stored_at,
#                  cda.data_source_msg; doplni cda.soubor pokud chybi.
#          Existujici cda.* (stav, datum_*, zdroj, poznamka) NEMENI.
# Pouziti: python store_cda_batch_v1.2.py            (dry-run / nahled)
#          python store_cda_batch_v1.2.py --apply     (zapise do Mongo)
# Zmeny v1.2: DAVKA 5 (11JUN2026) - Mudr Robert (krok 4 -> 5).
# =============================================================================

import os
import sys
import base64
import hashlib
import unicodedata
import paramiko
import extract_msg
from pymongo import MongoClient
from bson import ObjectId

MONGO_URI = os.environ.get("MONGO_URI", "mongodb://192.168.1.76:27017")
TOWER_HOST = "192.168.1.76"
TOWER_USER = "root"
TOWER_PASS = "7309208104"
REMOTE_DIR = "/mnt/user/JNJEMAILS"
TMPDIR = r"u:\Dropbox\!!!Days\Downloads Z230\_cda_tmp"
STORED_AT = "2026-06-11"

# investigator_id -> (msg_filename, attachment_filename, label)
# DAVKA 5 (11JUN2026): nove CDA z 11.6. (krok 4 -> 5)
MAPPING = [
    ("6a19832b5fc2213518257978", "FC130007DE92C2320000.msg",
     "CZ_ CDA PI_MUDr. Robert Mudr_11Jun2026.pdf",
     "Mudr Robert (Nemocnice Milosrdnych sester sv. Karla Boromejskeho)"),
]

# HISTORIE drivejsich davek (jiz ulozeno):
# DAVKA 4 (11JUN2026): Konecny Michal FC130007DE92C231, Balaz Jozef FC130007DE92C20F.
# DAVKA 3 (10JUN2026): Gregar FC130007DE92C204, Durina FC130007DE92C203,
#   Horvath/Accout Center FC130007DE92C1FE.
# DAVKA 1+2 (09JUN2026): Hlavaty/Cliniq FC1300053049739C, Fedurco/ENDOMED
#   FC1300053049739B, Tichy FC13000530495B95, Falc FC130007D8A1F0E6, Pesta
#   FC130007D8A1F0E1, Jungwirthova FC130007D8A1F0E2, Lukac FC130007C9E971FF
#   (store_cda_to_mongo_v1.0), Matous/Axon FC130007D8A1F0E3, Mihalkanin/GastroLM
#   FC130007D8A1F0E6, Krizova/Motol FC130007C1643CA1.


def norm(s):
    """lowercase, bez diakritiky, sjednocene mezery"""
    s = s or ""
    s = unicodedata.normalize("NFKD", s)
    s = "".join(c for c in s if not unicodedata.combining(c))
    return " ".join(s.lower().split())


def main():
    apply = "--apply" in sys.argv
    os.makedirs(TMPDIR, exist_ok=True)

    # SSH/SFTP
    ssh = paramiko.SSHClient()
    ssh.set_missing_host_key_policy(paramiko.AutoAddPolicy())
    ssh.connect(TOWER_HOST, username=TOWER_USER, password=TOWER_PASS, timeout=30)
    sftp = ssh.open_sftp()

    client = MongoClient(MONGO_URI)
    col = client["feasibility"]["investigators"]

    plan = []
    for inv_id, msg_name, att_name, label in MAPPING:
        local_msg = os.path.join(TMPDIR, msg_name)
        if not os.path.exists(local_msg):
            sftp.get(f"{REMOTE_DIR}/{msg_name}", local_msg)
        m = extract_msg.Message(local_msg)
        target = norm(att_name)
        chosen = None
        for att in m.attachments:
            name = att.longFilename or att.shortFilename or ""
            if norm(name) == target or (target in norm(name)) or (norm(name) in target and name.lower().endswith(".pdf")):
                chosen = (name, att.data)
                break
        m.close()
        if not chosen:
            plan.append((inv_id, label, msg_name, att_name, None, "!!! PRILOHA NENALEZENA"))
            continue
        raw = chosen[1]
        sha = hashlib.sha256(raw).hexdigest()
        plan.append((inv_id, label, msg_name, chosen[0], (len(raw), sha, raw), "OK"))

    sftp.close(); ssh.close()

    # Nahled
    print("=== NAHLED DAVKY (CDA -> Mongo cda.data) ===\n")
    for inv_id, label, msg_name, att_name, info, status in plan:
        doc = col.find_one({"_id": ObjectId(inv_id)}, {"prijmeni": 1, "jmeno": 1, "cda.data_base64": 1})
        has = bool(doc and doc.get("cda", {}).get("data_base64"))
        print(f"[{status}] {label}  (_id {inv_id})")
        print(f"    .msg: {msg_name}")
        print(f"    priloha: {att_name}")
        if info:
            print(f"    velikost: {info[0]} B   sha256: {info[1]}")
        print(f"    data_base64 jiz existuje: {has}")
        print()

    if not apply:
        print(">>> DRY-RUN. Pro zapis spust s --apply")
        return

    n = 0
    for inv_id, label, msg_name, att_name, info, status in plan:
        if status != "OK" or not info:
            print(f"PRESKAKUJI {label}: {status}")
            continue
        size, sha, raw = info
        b64 = base64.b64encode(raw).decode("ascii")
        res = col.update_one(
            {"_id": ObjectId(inv_id)},
            {"$set": {
                "cda.data_base64": b64,
                "cda.data_sha256": sha,
                "cda.data_filename": att_name,
                "cda.data_mime": "application/pdf",
                "cda.data_size": size,
                "cda.data_stored_at": STORED_AT,
                "cda.data_source_msg": msg_name,
                "cda.soubor": att_name,
            }},
        )
        n += res.modified_count
        print(f"ZAPSANO: {label}  (modified={res.modified_count})")
    print(f"\n>>> CELKEM ZAPSANO: {n}")


if __name__ == "__main__":
    main()