Find corrupt files which were exported from snapchat memories

find-corrupt-snaps.py

import os
import subprocess
import json
import shutil
from datetime import datetime
from zoneinfo import ZoneInfo  # Python 3.9+

# === Konfiguration ===
input_dir = r"./downloads-with-captions"    # Ordner mit defekten Dateien
fixed_dir = r"./downloads-no-captions"      # Ordner mit funktionierenden Dateien (UTC)
output_dir = r"./downloads-with-captions"   # Zielordner für kopierte Dateien
json_path = os.path.join(input_dir, "corrupt-files.json")

corrupt_files = []

# === Schritt 1: Defekte Dateien erkennen ===
for root, _, files in os.walk(input_dir):
    for file in files:
        if file.lower().endswith(".mp4"):
            path = os.path.join(root, file)
            try:
                result = subprocess.run(
                    ['ffprobe', '-v', 'error',
                     '-select_streams', 'v',
                     '-show_entries', 'stream=codec_type',
                     '-of', 'csv=p=0', path],
                    stdout=subprocess.PIPE,
                    stderr=subprocess.PIPE,
                    text=True
                )
                if 'video' not in result.stdout:
                    corrupt_files.append(path)
                    print(f"Potential corrupt file: {path}")
            except Exception as e:
                print(f"Error while testing file {path}: {e}")

# === Schritt 2: Defekte Dateien in JSON speichern ===
with open(json_path, 'w', encoding='utf-8') as f:
    json.dump(corrupt_files, f, indent=4, ensure_ascii=False)
print(f"\n✅ Corrupt files saved to {json_path}")

# === Schritt 3: Funktionierende Dateien kopieren mit Zeitzonenanpassung ===
os.makedirs(output_dir, exist_ok=True)

for corrupt_path in corrupt_files:
    filename = os.path.basename(corrupt_path)
    name_part, ext = os.path.splitext(filename)

    try:
        # Datum aus Europe/Zurich-Dateiname parsen
        dt_zh = datetime.strptime(name_part, "%Y-%m-%d_%H-%M-%S").replace(tzinfo=ZoneInfo("Europe/Zurich"))

        # In UTC konvertieren (so heißt die Datei im fixed_dir)
        dt_utc = dt_zh.astimezone(ZoneInfo("UTC"))
        fixed_filename = dt_utc.strftime("%Y-%m-%d_%H-%M-%S") + ext
        fixed_file_path = os.path.join(fixed_dir, fixed_filename)

        if not os.path.exists(fixed_file_path):
            print(f"⚠️ Fixed file not found for {filename} → looked for {fixed_filename}")
            continue

        # Kopieren unter Originalnamen (Europe/Zurich)
        dst_path = os.path.join(output_dir, filename)
        shutil.copy2(fixed_file_path, dst_path)
        print(f"✅ Copied {fixed_file_path} → {dst_path}")

    except Exception as e:
        print(f"⚠️ Error processing {filename}: {e}")

Script made with ChatGPT