gistfile1.txt

#!/usr/bin/env python

import sys,math

buckets = {}

for f in open(sys.argv[1]).readlines():
    f.strip()
    s,p = f.split(' ', 1)
    sz = int(s)
    sz_mb = sz / (1024*1024)
    [n,s] = buckets.get(sz_mb, [0,0])
    buckets[sz_mb] = (n+1, s+sz)

for v in sorted(buckets.keys()):
    [n,s] = buckets[v]
    print "%s => %s (%s MB)" % (v, n, s/(1024*1024))