| Index: trunk/tools/wsor/contribution_inequality/topcontributors.py |
| — | — | @@ -0,0 +1,32 @@ |
| | 2 | +#!/usr/bin/python |
| | 3 | + |
| | 4 | +''' lists top contributors by year and namespace ''' |
| | 5 | + |
| | 6 | +import os |
| | 7 | +import sys |
| | 8 | +import csv |
| | 9 | + |
| | 10 | +from itertools import groupby |
| | 11 | +from contextlib import closing |
| | 12 | +from argparse import ArgumentParser |
| | 13 | +from collections import deque |
| | 14 | + |
| | 15 | +parser = ArgumentParser(description=__doc__) |
| | 16 | +parser.add_argument('data_path', metavar='data') |
| | 17 | +parser.add_argument('maxlen', metavar='number', type=int) |
| | 18 | + |
| | 19 | +if __name__ == '__main__': |
| | 20 | + |
| | 21 | + ns = parser.parse_args() |
| | 22 | + |
| | 23 | + with closing(open(ns.data_path)) as f: |
| | 24 | + reader = csv.DictReader(f, delimiter='\t', quoting=csv.QUOTE_NONE) |
| | 25 | + groupfunc = lambda row : (row['namespace'], row['year']) |
| | 26 | + for key, subiter in groupby(reader, groupfunc): |
| | 27 | + # smart way to keep only the tail |
| | 28 | + users = deque((row['user_id'] for row in subiter ), maxlen=ns.maxlen) |
| | 29 | + print '\t'.join(key + tuple(users)) |
| | 30 | + sys.stdout.flush() |
| | 31 | + |
| | 32 | + |
| | 33 | + |
| Property changes on: trunk/tools/wsor/contribution_inequality/topcontributors.py |
| ___________________________________________________________________ |
| Added: svn:executable |
| 1 | 34 | + * |