#!/usr/bin/env python3
"""Scan 4TB Mylio folder and find files not in Immich."""

import subprocess
import hashlib
import sqlite3
import os

REMOTE = "macmini"
REMOTE_PATH = "/Volumes/4TB/Mylio"
LOCAL_DB = "/home/johan/immich-compare/hash_index.db"
EXTENSIONS = {'.jpg', '.jpeg', '.png', '.heic', '.gif', '.mp4', '.mov', '.avi', '.m4v', '.3gp'}

def get_immich_hashes():
    """Load all Immich hashes from existing database."""
    conn = sqlite3.connect(LOCAL_DB)
    c = conn.cursor()
    hashes = set()
    for row in c.execute("SELECT hash64k FROM files WHERE source='immich'"):
        hashes.add(row[0])
    conn.close()
    print(f"Loaded {len(hashes)} Immich hashes")
    return hashes

def scan_remote():
    """Find all media files on remote 4TB Mylio."""
    print(f"\nFinding files on {REMOTE}:{REMOTE_PATH}...")
    cmd = f"ssh {REMOTE} \"find '{REMOTE_PATH}' -type f 2>/dev/null\""
    result = subprocess.run(cmd, shell=True, capture_output=True, text=True)
    
    files = []
    for line in result.stdout.strip().split('\n'):
        if not line:
            continue
        ext = os.path.splitext(line)[1].lower()
        if ext in EXTENSIONS:
            files.append(line)
    
    print(f"Found {len(files)} media files")
    return files

def hash_remote_file(filepath):
    """Get 64KB hash of remote file."""
    cmd = f"ssh {REMOTE} \"head -c 65536 '{filepath}' 2>/dev/null\" | md5sum"
    result = subprocess.run(cmd, shell=True, capture_output=True, text=True)
    if result.returncode == 0:
        return result.stdout.split()[0]
    return None

def main():
    immich_hashes = get_immich_hashes()
    files = scan_remote()
    
    missing = []
    matched = 0
    
    print(f"\nHashing files and comparing...")
    for i, filepath in enumerate(files):
        if (i + 1) % 100 == 0:
            print(f"  {i+1}/{len(files)} - {matched} matched, {len(missing)} missing")
        
        h = hash_remote_file(filepath)
        if h is None:
            continue
        
        if h in immich_hashes:
            matched += 1
        else:
            missing.append(filepath)
    
    print(f"\n{'='*50}")
    print(f"Results:")
    print(f"  Total scanned: {len(files)}")
    print(f"  Already in Immich: {matched}")
    print(f"  NOT in Immich: {len(missing)}")
    
    # Save missing list
    with open('/tmp/4tb_mylio_missing.txt', 'w') as f:
        for p in missing:
            f.write(p + '\n')
    
    print(f"\nMissing files saved to /tmp/4tb_mylio_missing.txt")
    return missing

if __name__ == "__main__":
    main()