How to connect different AWS services using Boto3 library in Python?

In this article, we will see how to use the Boto3 library in Python to connect with different AWS services. Boto3 is the official AWS SDK for Python that provides both low-level client interfaces and high-level resource interfaces.

Understanding Client vs Resource

Boto3 offers two main ways to interact with AWS services:

• Client ? Provides low-level service access with all service operations

• Resource ? Provides high-level object-oriented interface (limited services)

Step-by-Step Approach

Step 1 ? Create an AWS session using Boto3 library.

Step 2 ? Pass the AWS service name to client() for low-level access or resource() for high-level interface.

Connecting with AWS Client

The client interface provides access to all AWS service operations ?

import boto3

# Function to get AWS Client connection
def get_aws_client(service_name):
    session = boto3.session.Session()
    # User can pass customized access_key_id, secret_access_key and session_token
    client = session.client(service_name)
    return client

# Connect to different AWS services
s3_client = get_aws_client('s3')
glue_client = get_aws_client('glue')
sqs_client = get_aws_client('sqs')

print("S3 Client:", type(s3_client).__name__)
print("Glue Client:", type(glue_client).__name__)
print("SQS Client:", type(sqs_client).__name__)

S3 Client: S3
Glue Client: Glue
SQS Client: SQS

Connecting with AWS Resource

The resource interface provides high-level object-oriented access ?

import boto3

# Function to get AWS Resource connection
def get_aws_resource(service_name):
    session = boto3.session.Session()
    # User can pass customized access_key_id, secret_access_key and session_token
    resource = session.resource(service_name)
    return resource

# Connect to AWS services that support resource interface
s3_resource = get_aws_resource('s3')
sqs_resource = get_aws_resource('sqs')

print("S3 Resource:", s3_resource)
print("SQS Resource:", sqs_resource)

S3 Resource: s3.ServiceResource()
SQS Resource: sqs.ServiceResource()

Resource Limitations

Not all AWS services support the resource interface. For example, AWS Glue only supports client interface ?

import boto3

try:
    glue_resource = boto3.resource('glue')
except Exception as e:
    print(f"Error: {e}")
    print("Use boto3.client('glue') instead")

Error: boto3.exceptions.ResourceNotExistsError: The 'glue' resource does not exist.
Use boto3.client('glue') instead

Services Supporting Resource Interface

The following AWS services support the resource interface:

Complete Example with Authentication

Here's a complete example showing how to connect with custom credentials ?

import boto3

# Method 1: Using default credentials (from ~/.aws/credentials or environment)
def connect_with_defaults(service_name, interface_type='client'):
    if interface_type == 'client':
        return boto3.client(service_name)
    else:
        return boto3.resource(service_name)

# Method 2: Using custom credentials
def connect_with_credentials(service_name, access_key, secret_key, region='us-east-1'):
    session = boto3.Session(
        aws_access_key_id=access_key,
        aws_secret_access_key=secret_key,
        region_name=region
    )
    return session.client(service_name)

# Example usage
s3_client = connect_with_defaults('s3', 'client')
print(f"Connected to S3: {type(s3_client).__name__}")

# For services supporting resource interface
s3_resource = connect_with_defaults('s3', 'resource')
print(f"S3 Resource: {s3_resource}")

Connected to S3: S3
S3 Resource: s3.ServiceResource()

Conclusion

Use boto3.client() for complete access to all AWS service operations. Use boto3.resource() for high-level object-oriented interface when supported. Always check service compatibility before choosing the interface type.