glacierclient.py

#! /usr/bin/env python3
from multipart import upload_large_file, CHUNK_SIZE
from datetime import date
from os.path import getsize
import boto3
import pprint
import argparse
import time
import json


LOG = "glacier.log"


def refresh_inventory(vault, jobid=None):
    """                                                                                                                                                                                       
    You can get all archive IDs by running a vault inventory job
    :param vault: the vault to retrieve inventory for
    :param jobid: optional job id to continue a previous inventory job
    """
    if not jobid:
        job_req = client.initiate_job(vaultName=vault,
                                      jobParameters={'Type': 'inventory-retrieval'})
        pprint.pprint(job_req)
        jobid = job_req['jobId']

    while True:
        status = client.describe_job(vaultName=vault,
                                     jobId=jobid)
        pprint.pprint(status)
        if status['Completed']:
            print("Job complete")
            break
        sleeptime = 300
        print("Sleeping {}...".format(sleeptime))
        time.sleep(sleeptime)

    job_resp = client.get_job_output(vaultName=vault,
                                     jobId=jobid)

    # first download the output and then parse the JSON         
    output = job_resp['body'].read()
    parsedoutput = json.loads(output)
    pprint.pprint(parsedoutput)

    archive_list = parsedoutput['ArchiveList']
    # persist archive_list
    return archive_list


def write_log(data):
    """
    Record glacier object upload
    :param data: information to write out
    :return:
    """
    with open(LOG, "a") as fi:
        fi.write(str(data))
        fi.write("\n")

# read arguments
parser = argparse.ArgumentParser()

parser.add_argument("-u", help="Specifies file to upload")
parser.add_argument("-i", action='store_true', help="View vault inventory (run inventory job), this may take a while")
parser.add_argument("-j", default=None, help="Optional inventory job id to continue")
parser.add_argument("-l", action='store_true', help="List all jobs")
parser.add_argument("-v", required=True, help="Specifies the vault to access")        # vault name

args = parser.parse_args()

client = boto3.client('glacier')
vaultname = args.v
to_upload = args.u      # the file name to upload

# if the file size is less than the chunk size
# upload in one part, otherwise upload multi part
if to_upload:
    if getsize(to_upload) <= CHUNK_SIZE:
        print("Uploading {}...".format(to_upload))

        # upload file
        with open(to_upload, 'rb') as f:
            response = client.upload_archive(vaultName=vaultname,
                                             archiveDescription="{}-{}".format(to_upload, date.today()),
                                             body=f)
        pprint.pprint(response)
        write_log(response)

    else:
        print("Uploading {}...".format(to_upload))

        response = upload_large_file(vaultname, to_upload, "{}-{}".format(to_upload, date.today()))
        write_log(response)

if args.i is True:
    pprint.pprint(refresh_inventory(vaultname, args.j))

if args.l is True:
    response = client.list_jobs(vaultName=vaultname)
    pprint.pprint(response)

print("Done")