Skip to main content

Python SX client-side library

Project description


sxclient is a library which implements client-side methods for communicating with an SX Cluster. Using the provided objects and functions, it is possible to prepare and send a query as per the API documentation at

Internally, sxclient uses requests library ( and currently requires Python 2.7.


In order to run an operation provided by the library, you must:

  • prepare a Cluster object, containing cluster location data;

  • prepare a UserData object, containing user credentials used to authorize operations;

  • prepare either a ClusterSession object or SXController object which serves as a context for the connections with the cluster.

Afterwards, you can run a series of operations using the previously created ClusterSession object as a context.

Initializing Cluster object

The most basic way of initializing the Cluster object is to pass the cluster name:

cluster = sxclient.Cluster('')

If the passed name is not a FQDN, you should pass an IP address too. It will be used to communicate with the cluster in place of name.

cluster = sxclient.Cluster('clustername', ip_addresses='')

You can also pass a list of IP addresses.

cluster = sxclient.Cluster('clustername', ip_addresses=['','',''])

In case you don’t want the connection to be secured by SSL, set is_secure to False:

cluster = sxclient.Cluster('', is_secure=False)

You can also pass a custom port number:

cluster = sxclient.Cluster('', port=8000)

In order to use a custom CA certificate for verification, pass a path to CA bundle in verify_ssl_cert parameter:

cluster = sxclient.Cluster('', verify_ssl_cert='/path/to/ca/bundle')

In case you don’t want to verify SSL certificates at all, set verify_ssl_cert to False.

Initializing UserData object

There are multiple initialization methods for UserData. You can provide a path to the key file:

user_data = sxclient.UserData.from_key_path('/path/to/keyfile')

The key itself can be provided too — either encoded in base64:

user_data = sxclient.UserData.from_key('ZP1rHyR0QB6zEvCwYexGl9SF1G143C/D2hG9rEisLL2zJV3kWQvtAwAA')

or in its binary form:

user_data = sxclient.UserData('d\xfdk\x1f$t@\x1e\xb3\x12\xf0\xb0a\xecF\x97\xd4\x85\xd4mx\xdc/\xc3\xda\x11\xbd\xacH\xac,\xbd\xb3%]\xe4Y\x0b\xed\x03\x00\x00')

You can also initialize the object with username and password (and cluster UUID):

user_data = sxclient.UserData.from_userpass_pair('a_user', 'a_password', '10ca10ca-10ca-10ca-10ca-10ca10ca10ca')

Initializing and working with SXController

After preparing Cluster and UserData objects you may create an SXController object:

sx = sxclient.SXController(cluster, user_data)

Then get all available operations from

print sx.available_operations

You may call any operation via

The return value is a HTTP response object holding the response from SX server. If a command supports the JSON format (as most of them do) you may call it directly:


After you are done working with SXController gracefully close it with:


High level operations

Uploading and downloading files using the aforementioned operations requires some low level knowledge of the underlying SX protocol. To make your life easier, we added three dedicated helpers.

For a given SXController if you wish to upload a file use:

import os
file_size = os.stat('myfile.txt').st_size
uploader = sxclient.SXFileUploader(sx)
with open('myfile.txt', 'r') as fo:
   uploader.upload_stream('my-volume', file_size, 'my_new_file_name.txt', fo)

and if you wish to download a file use:

with sxclient.SXFileDownloader(sx) as downloader:
   content = downloader.get_file_content('my-volume', 'my_new_file_name.txt')

There is another downloader available called SXFileCat. You can use it like this:

downloader = sxclient.SXFileCat(sx)
content = downloader.get_file_content('my-volume', 'my_new_file_name.txt')

The difference between these two downloaders is that SXFileCat streams files block-by-block from SX Cluster and thus is memory, disk and network efficient but not time efficient.

On the contrary, SXFileDownloader streams every file to a temporary file and then yields the content of that file. This is done on multiple threads and connections, therefore it is time efficient but neither memory nor disk, nor network efficient.

SXFileCat in the example isn’t used as a context manager since there is no need to initialize and clean its context. You can still use it with the with statement though.

Additional documentation

For more information regarding usage of a specific object see its docstring. For example, to see the description of listVolumes, use Python built-in help function (note that the first letter is capitalized):

>>> help(sxclient.operations.ListVolumes)

or run pydoc in your favourite shell:

$ pydoc sxclient.operations.ListVolumes

Alternatively, in case you have already initialized an SXController object in the interpreter, you can use a shortcut:

>>> help(sx.listVolumes)

Project details

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

sxclient-0.19.0.tar.gz (32.3 kB view hashes)

Uploaded source

Built Distribution

sxclient-0.19.0-py2-none-any.whl (49.2 kB view hashes)

Uploaded py2

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page