使用 Python API 创建文件后，如何立即从 Google Vault 导出下载文件？

Question

Using Python API, I have created an export. How do I download the .zip file in the export using the same authorized service? When creating the export, I can see the bucketName and objectNames of the cloudStorageSink, however I cannot find any documentation on how to download them to my host using the existing service that created the export

#!/usr/bin/env python
from __future__ import print_function

import datetime
import json
import time

from googleapiclient.discovery import build
from httplib2 import Http
from oauth2client import file, client, tools

# If modifying these scopes, delete the file token.json.
SCOPES = 'https://www.googleapis.com/auth/ediscovery'

def list_exports(service, matter_id):
    return service.matters().exports().list(matterId=matter_id).execute()


def get_export_by_id(service, matter_id, export_id):
    return service.matters().exports().get(matterId=matter_id, exportId=export_id).execute()

def get_service():
    '''
    Look for an active credential token, if one does not exist, use credentials.json
    and ask user for permission to access.  Store new token, return the service object
    '''
    store = file.Storage('token.json')
    creds = store.get()
    if not creds or creds.invalid:
        flow = client.flow_from_clientsecrets('credentials.json', SCOPES)
        creds = tools.run_flow(flow, store)
    service = build('vault', 'v1', http=creds.authorize(Http()))

    return service


def create_drive_export(service, matter_id, export_name, num_days):
    """
    once we have a matter_id , we can create an export under it with the relevant files we are looking for.

    """
    # set times for beginning and end of query:
    today = datetime.datetime.now()
    print("creating a drive export at {}".format(today))
    start_time = today - datetime.timedelta(days=num_days)

    drive_query_options = {'includeTeamDrives': True}
    user_list = ['me@gmail.com']
    drive_query = {
        'corpus': 'DRIVE',
        'dataScope': 'ALL_DATA',
        'searchMethod': 'ACCOUNT',
        'accountInfo': {
            'emails': user_list
        },
        'driveOptions': drive_query_options,
        # end time is more recent date, start time is older date
        'endTime': '{}-{}-{}T00:00:00Z'.format(today.year, today.month, today.day),
        'startTime': '{}-{}-{}T00:00:00Z'.format(start_time.year, start_time.month, start_time.day),
        'timeZone': 'Etc/GMT'
    }

    wanted_export = {
        'name': export_name,
        'query': drive_query,
        'exportOptions': {
            'driveOptions': {}
        }
    }

    return service.matters().exports().create(matterId=matter_id, body=wanted_export).execute()


def get_export(service, matter_id, export_id):
    return service.matters().exports().get(matterId=matter_id, exportId=export_id).execute()


def main():
    service = get_service()
    matter_id = '<known_matter_id>'
    timestamp = datetime.datetime.now().strftime("%Y%m%d.%H%M%s")
    export = create_drive_export(service, matter_id, "code_gen_export.{}".format(timestamp), 1)

    # check every 5 seconds until export is done being created:
    while export['status'] == 'IN_PROGRESS':
        export = get_export(service, matter_id, export['id'])
        print('...')
        time.sleep(5)

    # print(json.dumps(export, indent=2))
    print(json.dumps(export['cloudStorageSink']['files'], indent=2))


if __name__ == '__main__':
    main()

and running the above code produces:

creating a drive export at 2018-09-20 17:12:38.026402
...
...
...
...
...
...
[
  {
    "md5Hash": "hash_value",
    "bucketName": "bucket_string",
    "objectName": "object1_string/code_gen_export.20180920.17121537481558-custodian-docid.csv",
    "size": "1684"
  },
  {
    "md5Hash": "hash_value",
    "bucketName": "bucket_string",
    "objectName": "object2_string/code_gen_export.20180920.17121537481558-metadata.xml",
    "size": "10600"
  },
  {
    "md5Hash": "hash_value",
    "bucketName": "bucket_string",
    "objectName": "object3_string/code_gen_export.20180920.17121537481558_0.zip",
    "size": "21599222"
  }
]

Can I download the .zip file using the service object I created in get_service()?

Answer 1

After a long struggle with the above, I found the right approach with the aid of one of Googles' API support agents.

Notice that you will need to create a new service using:

build('storage', 'v1', credentials=credentials)

where cradintials is:

service_account.Credentials.from_service_account_file(
        SERVICE_ACCOUNT_FILE,
        scopes=SCOPES, 
        subject='user@domain.com'
)

(it may be that the same argument you used for your credentials: http=creds.authorize(Http()) will work as weel - I did not try that)

In addition you will need to use a byte stream library such as: io and import googleapiclient.http as well.

The full code:

import io
from google.oauth2 import service_account
from googleapiclient.discovery import build
import googleapiclient.http


SCOPES = ['https://www.googleapis.com/auth/devstorage.full_control']
SERVICE_ACCOUNT_FILE = 'yourServiceAccountFile.json'
bucket_name = 'yourBucketName'
object_name = 'yourObjectName.zip'

credentials = service_account.Credentials.from_service_account_file(
        SERVICE_ACCOUNT_FILE,
        scopes=SCOPES, 
        subject='user@domain.com'
)

service = build('storage', 'v1', credentials=credentials)

req = service.objects().get_media(bucket=bucket_name, object=object_name)

out_file = io.BytesIO()
downloader = googleapiclient.http.MediaIoBaseDownload(out_file, req)

done = False
while done is False:
    status, done = downloader.next_chunk()
    print("Download {}%.".format(int(status.progress() * 100)))

file_name = '/Users/myUser/Downloads/new_file.zip'
open(file_name, "w").write(out_file.getvalue())

Answer 2

The above answer is great, but causes issue with large files, as BytesIO holds the data in memory. In a low RAM environment a 2GB can kill your download. Suggest using FileIO instead.

Change the following bit of code:

out_file = io.BytesIO()
downloader = googleapiclient.http.MediaIoBaseDownload(out_file, req)

done = False
while done is False:
    status, done = downloader.next_chunk()
    print("Download {}%.".format(int(status.progress() * 100)))

file_name = '/Users/myUser/Downloads/new_file.zip'
open(file_name, "w").write(out_file.getvalue())

To:

file_name = '/myfilepath/myfilename.ext'
with io.FileIO(file_name, mode='wb') as out_file:
    downloader = googleapiclient.http.MediaIoBaseDownload(out_file, req)
    done = False
    while not done:
        status, done = downloader.next_chunk()