A method for seeing how a disk (on MS Windows) is used.

disk_use_analysis.py

#! python

import os
import shutil
import argparse
import json

parser = argparse.ArgumentParser(description='Creates a report about the distribution over folders of disk use.')
parser.add_argument('drive_letter', nargs=1, metavar='drive letter', help='The drive that should be analysed.')

args = parser.parse_args()

def analyse(drive_letter):
    print('Analysing', drive_letter, 'drive...')
    base_dir = '{drive_letter}:\\'.format(drive_letter=drive_letter)
    dir_size = 4096
    total_space, space_used, space_free = shutil.disk_usage(base_dir)
    bytes_remaining = space_used
    print('total space: {tot}GB, used space: {used}GB, free space: {free}GB'.format(tot=total_space//2**30, used=space_used//2**30, free=space_free//2**30))
    output = {base_dir: {'total_space': total_space, 'used_space': space_used, 'free_space': space_free}}
    skips = 0
    current_tick = '/'
    for dir_name, dir_list, file_list in os.walk(base_dir):
        if is_junction(dir_name):
            continue
        file_size = 0
        files = {}
        for file_name in file_list:
            try:
                if os.path.islink(os.path.join(dir_name, file_name)):
                    continue
                size = os.path.getsize(os.path.join(dir_name, file_name))
                files[file_name] = size
                bytes_remaining -= size
                print('Bytes left to analyse:', bytes_remaining, end='\r')
                file_size += size
            except OSError:
                print('can\'t get the size of', os.path.join(dir_name, file_name))
                skips += 1
        the_path = dir_name.split('\\')
        cur_dict = output[base_dir]
        for dire in the_path[1:]:
            try:
                cur_dict = cur_dict[dire]
            except KeyError:
                cur_dict[dire] = {'file_size': file_size}
                cur_dict[dire]['files'] = files
        bytes_remaining -= dir_size
    with open('disk_use.json', 'w') as json_output:
        json_output.write(json.dumps(output))
    print('finished analysis')

def is_junction(path):
    mode = os.lstat(path).st_mode
    return mode == 16895

if __name__ == '__main__':
    analyse(args.drive_letter[0])