aws-samples/boto-session-manager-project

Provides an alternative, or maybe a more user friendly way to use the native boto3 API.

---

Welcome to boto_session_manager Documentation

About boto_session_manager

boto_session_manager is a light weight, zero dependency python library that simplify managing your AWS boto3 session in your application code. It bring auto complete and type hint to the default boto3 SDK, and provide smooth development experience with the following goodies:

• boto3 Client auto complete
• Cached boto3 Client
• Assume IAM role in application code
• Set temporary credential for AWS Cli

Additionally, if you use boto3-stubs and you did pip install "boto3-stubs[all]", then boto_session_manager comes with the auto complete and type hint for all boto3 methods out-of-the-box, without any extra configuration (such as explicit type annotations)

Feature

Boto Client Auto Complete

Provide an Enum class to access the aws service name to create boto client.

from boto_session_manager import BotoSesManager, AwsServiceEnum

bsm = BotoSesManager()
s3_client = bsm.s3_client

One click to jump to the documentation:

Client method auto complete:

Arguments type hint:

Note: you have to do pip install "boto3-stubs[all]" to enable "Client method auto complete" and "Arguments type hint" features.

Cached Client

Once an boto session is defined, each AWS Service client should be created only once in most of the case. boto_session_manager.BotoSesManager.get_client(service_name) allow you to fetch the client object from cache if possible.

from boto_session_manager import BotoSesManager, AwsServiceEnum

bsm = BotoSesManager()
s3_client1 = bsm.get_client(AwsServiceEnum.S3)
s3_client2 = bsm.get_client(AwsServiceEnum.S3)
assert id(s3_client1) = id(s3_client2)

Or you can just do:

bsm.s3_client.list_buckets() # it cache the client when needed

Assume Role

Create another boto session manager based on an assumed IAM role. Allow you to check if it is expired and maybe renew later.

bsm_assumed = bsm.assume_role("arn:aws:iam::111122223333:role/your-assume-role-name")
sts_client = bsm_assumed.get_client(AwsServiceEnum.sts)
print(sts_client.get_caller_identity())

print(bsm_assumed.is_expired())

From 1.5.1, it adds support for auto-refreshable assumed role (Beta). Note that it is using AssumeRoleCredentialFetcher and DeferredRefreshableCredentials from botocore, which is not public API officially supported by botocore. This API may be unstable.

bsm_assumed = bsm.assume_role(
    "arn:aws:iam::111122223333:role/your-assume-role-name",
    duration_seconds=900,
    auto_refresh=True,
)

# even though the duration seconds is only 15 minutes,
# but it can keep running for 1 hour.
tick = 60
sleep = 60
for i in range(tick):
    time.sleep(sleep)
    print("elapsed {} seconds".format((i + 1) * sleep))
    print("Account id = {}".format(bsm_new.sts_client.get_caller_identity()["Account"]))

AWS CLI context manager

You explicitly defined a boto session manager that is not the same as the default one used by your AWS CLI. The boto_session_manager.BotoSesManager.awscli() context manager can temporarily set your default AWS CLI credential as the same as the one you defined, and automatically revert it back.

# explicitly define a boto session manager
bsm = BotoSesManager(
    profile_name="my_aws_profile",
)

with bsm.awscli():
    # now the default AWS CLI credential is the same as the ``bsm`` you defined

Here's a more detailed example:

import os
from boto_session_manager import BotoSesManager

def print_default_aws_cli_credential():
    print("AWS_ACCESS_KEY_ID =", os.environ.get("AWS_ACCESS_KEY_ID"))
    print("AWS_SECRET_ACCESS_KEY =", os.environ.get("AWS_SECRET_ACCESS_KEY"))
    print("AWS_SESSION_TOKEN =", os.environ.get("AWS_SESSION_TOKEN"))
    print("AWS_REGION =", os.environ.get("AWS_REGION"))

print("--- before ---")
print_default_aws_cli_credential()

bsm = BotoSesManager(profile_name="aws_data_lab_open_source_us_east_1")
with bsm.awscli():
    print("--- within awscli() context manager ---")
    print_default_aws_cli_credential()

print("--- after ---")
print_default_aws_cli_credential()

# --- before ---
# AWS_ACCESS_KEY_ID = None
# AWS_SECRET_ACCESS_KEY = None
# AWS_SESSION_TOKEN = None
# AWS_REGION = None
# --- within awscli() context manager ---
# AWS_ACCESS_KEY_ID = ABCDEFG...
# AWS_SECRET_ACCESS_KEY = ABCDEFG...
# AWS_SESSION_TOKEN = ABCDEFG...
# AWS_REGION = us-east-1
# --- after ---
# AWS_ACCESS_KEY_ID = None
# AWS_SECRET_ACCESS_KEY = None
# AWS_SESSION_TOKEN = None
# AWS_REGION = None

Install

boto_session_manager is released on PyPI, so all you need is:

$ pip install boto_session_manager

To upgrade to latest version:

$ pip install --upgrade boto_session_manager