Generate a population following a distribution

example

$ mkpop -i pokemon 100 zipf | sort | uniq -c | sort -rn
  68 hoopa
  12 diancie
   7 zygarde
   3 yveltal
   2 xerneas
   1 volcanion
   1 trevenant
   1 pumpkaboo
   1 noivern
   1 noibat
   1 gourgeist
   1 bergmite
   1 avalugg

mkpop.py

#!/usr/bin/env python
import argparse
import sys
import numpy


def zipf_command(args):
    for item in numpy.random.zipf(args.a, args.size):
        yield item


def main(args):
    values = args.infile.readlines()
    mapping = {}
    for item in args.func(args):
        if item not in mapping:
            mapping[item] = values.pop()
        args.outfile.write(mapping[item])

if __name__ == '__main__':
    parser = argparse.ArgumentParser()
    parser.add_argument('-i', '--input', dest='infile', nargs='?', type=argparse.FileType('r'), default=sys.stdin)
    parser.add_argument('-o', '--output', dest='outfile', nargs='?', type=argparse.FileType('w'), default=sys.stdout)
    parser.add_argument('size', type=int)
    subparsers = parser.add_subparsers()
    parser_zipf = subparsers.add_parser('zipf')
    parser_zipf.add_argument('-a', type=float, default=2)
    parser_zipf.set_defaults(func=zipf_command)
    args = parser.parse_args()
    main(args)