Google Drive Duplicates

google_drive_duplicates.py

import os
import re
import sys
import shutil


def get_size(start_path):
    total_size = 0
    for dirpath, dirnames, filenames in os.walk(start_path):
        for f in filenames:
            fp = os.path.join(dirpath, f)
            # skip if it is symbolic link
            if not os.path.islink(fp):
                total_size += os.path.getsize(fp)

    if os.path.isfile(start_path) and not os.path.islink(start_path):
        total_size += os.path.getsize(start_path)

    return total_size


def sizeof_fmt(num, suffix='B'):
    for unit in ['', 'Ki', 'Mi', 'Gi', 'Ti', 'Pi', 'Ei', 'Zi']:
        if abs(num) < 1024.0:
            return "%3.1f%s%s" % (num, unit, suffix)
        num /= 1024.0
    return "%.1f%s%s" % (num, 'Yi', suffix)


def clean_up(path):
    duplicate_dirs = []
    duplicate_files = []
    size = 0
    last_size = 0
    patterns = [
        r"[0-9]{4}-[0-9]{2}-[0-9]{2}T[0-9]{6}\.[0-9]{6}Z",
        r"[0-9]{4}-[0-9]{2}-[0-9]{2}T[0-9]{2} [0-9]{2} [0-9]{2}\.[0-9]{3}Z",
        r"\([0-9]*\)"
    ]
    for root, dirs, files in os.walk(path):
        for dir in dirs:
            delete = False

            for pattern in patterns:
                if re.search(pattern, dir):
                    delete = True
                    break

            if delete:
                path = os.path.join(root, dir)
                duplicate_dirs.append(path)
                size += get_size(path)

        for file in files:
            delete = False

            for pattern in patterns:
                if re.search(pattern, file):
                    delete = True
                    break

            if not delete and file.endswith(".tmp"):
                delete = True

            if delete:
                path = os.path.join(root, file)
                duplicate_files.append(path)
                size += get_size(path)

        if last_size != size:
            print(sizeof_fmt(size))
            last_size = size

    for path in duplicate_files:
        try:
            os.remove(path)
        except Exception:
            print(f"Could not remove {path}.")

    for path in duplicate_dirs:
        try:
            shutil.rmtree(path)
        except Exception:
            print(f"Could not remove {path}.")


if __name__ == "__main__":
    clean_up(sys.argv[1])