Simple script to print: human size of a directory tree, number of files in that tree, number of hard links in that tree, number of softlinks in that tree and number of directories.

get_tree_size.py

#!/usr/bin/env python3
# Simple script to print:
# Size of a directory tree
# Number of files in that tree
# Number of hard links in that tree
# Number of softlinks in that tree

import os
import sys
import datetime
from os.path import normpath, basename
import operator
from itertools import starmap

# Make sure we do not "hang" on large directories
max_directory_entries = 100000

# http://stackoverflow.com/a/25613067/194447
def file_size(size):
    from math import log2
    _suffixes = ['bytes', 'KiB', 'MiB', 'GiB', 'TiB', 'EiB', 'ZiB']
    order = int(log2(size) / 10) if size else 0
    return '{:.4g}{}'.format(size / (1 << (order * 10)), _suffixes[order])

# Adapted from https://www.python.org/dev/peps/pep-0471/#notes-on-exception-handling
def get_tree_size(path):
    """Return total size of files in path and subdirs. If
    is_dir() or stat() fails, print an error message to stderr
    and assume zero size (for example, file has been deleted).
    """
    size = 0
    files = 0
    symlinks = 0
    hardlinks = 0
    directories = 0
    skipped_directories = 0
    for entry in os.scandir(path):
        try:
            is_dir = entry.is_dir(follow_symlinks=False)
        except OSError as error:
            print('Error calling is_dir():', error, file=sys.stderr)
            continue
        if is_dir:
            if entry.stat(follow_symlinks=False).st_size < max_directory_entries:
                directories += 1
                # http://stackoverflow.com/a/18132749/194447
                size, files, symlinks, hardlinks, directories, skipped_directories = starmap(operator.iadd, zip((size, files, symlinks, hardlinks, directories, skipped_directories), get_tree_size(entry.path)))
            else:
                skipped_directories += 1
                files += entry.stat(follow_symlinks=False).st_size
        else:
            try:
                size += entry.stat(follow_symlinks=False).st_size
                files += 1
                if entry.is_symlink():
                    symlinks += 1
                if entry.stat(follow_symlinks=False).st_nlink > 1:
                    hardlinks += 1
            except OSError as error:
                print('Error calling stat():', error, file=sys.stderr)
    return (size, files, symlinks, hardlinks, directories, skipped_directories)


if __name__ == "__main__":
    directory = sys.argv[1]
    start_time = datetime.datetime.now()
    size, files, symlinks, hardlinks, directories, skipped_directories = get_tree_size(directory)
    directory = basename(normpath(directory))
    size = file_size(size)
    end_time = datetime.datetime.now()
    difference = str(end_time - start_time)
    print("{:s} : {:s} in {:d} files, with {:d} symlinks, {:d} hardlinks, {:d} directories and {:d} skipped directories (more than {:,} files) in {:s}".format(directory, size, files, symlinks, hardlinks, directories, skipped_directories, max_directory_entries, difference))