split large mongodumps into files usable for import

splitjson.py

#!/usr/bin/env python

from itertools import izip_longest
import json
import subprocess
import sys

def grouper(iterable, n, fillvalue=None):
    args = [iter(iterable)] * n
    return izip_longest(fillvalue=fillvalue, *args)
        

def split_pretty_array(infh, outfilebase, numperfile):
    for i, group in enumerate(grouper(json.loads(infh), numperfile, '')):
        with open('%s_%s.json' % (outfilebase, i), 'w') as outputfile:
            json.dump(group, outputfile)


if __name__ == '__main__':
    split_pretty_array(sys.stdin, sys.argv[1], 10)