Analyze disk space usage of some common hidden/auto-generated files

find_mac_files.py

import fnmatch, os

def bytesto(bytes, to, bsize=1024):
    """convert bytes to megabytes, etc.
       sample code:
           print('mb= ' + str(bytesto(314575262000000, 'm')))
       sample output: 
           mb= 300002347.946
    """

    a = {'k' : 1, 'm': 2, 'g' : 3, 't' : 4, 'p' : 5, 'e' : 6 }
    r = float(bytes)
    for i in range(a[to]):
        r = r / bsize

    return(r)

def get_size(start_path = '/tank/XXXXXX'):
    txt_file = "/tmp/filelist.csv"
    # Empty file
    open(txt_file, 'w').close()
    fh = open(txt_file,"w")    
    fh.write("FilePath\tFileSize\n")

    total_size = 0

    file_names = ['Thumbs.db', '._.DS_Store', '.DS_Store', '._AppleDouble', '.apdisk', 'Desktop.ini', '.VolumeIcon.icns', '.TemporaryItems', '.Trashes', '._.TemporaryItems']
    #file_names_regex = '*._*'

    skip_dirs = ['.zfs']

    for dirpath, dirnames, filenames in os.walk(os.path.abspath(start_path), followlinks=False):
        if '.zfs' in dirnames:
            dirnames.remove('.zfs')
            continue

        if not dirnames and not filenames:
            fh.write(dirpath + "\t" + str(0) + "\n")

        for f in filenames:
            fp = os.path.join(dirpath, f)
            # Skip Symlinks
            if os.path.islink(fp):
                continue

            file_size = bytesto(os.path.getsize(fp), 'm')

            if file_size > 5000:
                fh.write(fp + "\t" + str(file_size) + "\n")

            if any(f in s for s in file_names):
                  
                total_size += file_size
                fh.write(fp + "\t" + str(file_size) + "\n")
                # os.remove("/tmp/foo.txt")

    fh.close()
    return total_size

size = get_size()
print str(size) + " MB"