Process tar files in parallel

ptars.py

#! /usr/bin/env python2.7

from multiprocessing import Pool, Queue
import os
from sys import argv
import tarfile 

BLACKLIST = ['.git']
EXTENSIONS = ['.tgz', '.tar', '.tar.gz', '.tar.bz2']
N_PARALLEL = 10

def extract_goodness(directory, basename):
    if basename.startswith('bob'):
        print 'd({}) b({})'.format(directory, basename)
        with tarfile.open(os.path.join(directory, basename), 'r') as tar:
            for tinfo in tar:
                print tinfo.name

def main():
    if len(argv) < 2:
        return
    pool = Pool(N_PARALLEL)
    for start_dir in argv[1:]:
        for root, dirs, files in os.walk(start_dir):
            for skipdir in BLACKLIST:
                try:
                    dirs.remove(skipdir)
                except ValueError:
                    pass
            for fnam in files:
                for ext in EXTENSIONS:
                    if fnam.endswith(ext):
                        pool.apply_async(
                            extract_goodness,
                            (root, fnam))
    pool.close()
    pool.join()

if __name__ == '__main__':
    main()