List top 20 files + sizes over 300MB in / or specified directory

list.py

import os
import functools
import sys 

from heapq import *
from math import log 

def pretty_size(n,pow=0,b=1024,u='B',pre=['']+[p for p in'KMGTPEZY']):
    pow,n=min(int(log(max(n*b**pow,1),b)),len(pre)-1),n*b**pow
    return "%%.%if %%s%%s"%abs(pow%(-pow-1))%(n/b**float(pow),pre[pow],u)

startpath = "/" 
count = 0 
tsize = 0
if len(sys.argv) > 1:
    startpath = sys.argv[1]

data = []
threshold = 300000000 # 300 megabytes
top = 100 # show top 20 files

for (path, dirs, files) in os.walk(startpath):
    for filename in files:
        try:
            size = os.stat(path+"/"+filename).st_size
	    tsize += size
        except:
	    size = 0
            pass
            #print "Error accessing "+path+"/"+filename
        count += 1
	if count%10000 == 0:
	    print "scanned "+str(count)+" files, "+pretty_size(tsize)+" so far ("+(path+'/'+filename)+")"
        if size >= threshold:
            heappush(data, (-1*size, path+filename))

for x in range(0, top):
#while True:
    try:
        entry = heappop(data)
        print pretty_size(entry[0]*-1)+": "+entry[1]
    except:
        break

print("\nTotal Files: "+str(count))
print("Total Size: "+pretty_size(tsize))