down load sample

get_sample.py

import os
import logging
import argparse
from boto.s3.connection import S3Connection

def create_dir(type):
    if not os.path.exists(type):
        os.mkdir(type)

def get_aws_credentials():
     return {'key'    : os.environ.get('AWS_ACCESS_KEY'),
             'secret' : os.environ.get('AWS_SECRET_KEY')}

def parse_arg():
    parser = argparse.ArgumentParser()
    parser.add_argument('-m', help='stats month in 2013', dest='month',required=True)
    parser.add_argument('-d', help='stats day', dest='day',required=True)
    return parser.parse_args()

def download_sample(month, day):
    # connect to the bucket
    cred = get_aws_credentials()
    conn = S3Connection(cred['key'],cred['secret'])

    bucket_name = 'mat-redshift-prod'
    bucket = conn.get_bucket(bucket_name)

    interested_types = ('stat_clicks', 'stat_events', 'stat_installs', 'stat_opens')


    bucket_day = '1_1/2013/0%d/0%d/' % (month, day)
    bucket_name = 'qproc_imports/prod/qproc_import_csv/%s/%s'

    for type in interested_types:
        create_dir(type)

        for key in bucket.list(bucket_name % (type, bucket_day)):
            try:
                file_name = key.name.split('/')[-1]
                if not os.path.exists('%s/%s' % (type, file_name)):
                    key.get_contents_to_filename('%s/%s' % (type, file_name))
            except Exception, e:
                print(e)

if __name__ == "__main__":
    args = parse_arg()
    month = int(args.month)
    day   = int(args.day)
    download_sample(month, day)