AWS Boto3: Clients vs Resources - DynamoDB

Recently, my colleague brought up the difficulty of using the AWS SDK for Python (Boto3) while working with DynamoDB, especially the cumbersome mapping of AttributeValue objects on the Table operations. One of the easiest ways to get around this difficulty is to switch from the clients interface to the resources interface.

It is important to know the key difference between clients and resources. From the definition of the Boto3 documentation:

Clients provide a low-level interface to AWS whose methods map close to 1:1 with service APIs.

Resources represent an object-oriented interface to Amazon Web Services (AWS). They provide a higher-level abstraction than the raw, low-level calls made by service clients.

Let's quickly write a script to demonstrate the difference in the responses:

import boto3
import simplejson

def print_items(desc, response):
    items = response.get('Items')
    items_json = simplejson.dumps(items, indent=4)
    print(f'{desc}:', items_json, sep='\n')

ddb_table_name = 'test'

# DynamoDB client
ddb_client = boto3.client('dynamodb')

# Scan the table
response = ddb_client.scan(TableName=ddb_table_name)
print_items('client', response)

print('\n')

# DynamoDB resource
ddb_resource = boto3.resource('dynamodb')

# Scan the table
response = ddb_resource.Table(ddb_table_name).scan()
print_items('resource', response)

The output:

client:
[
    {
        "datetime": {
            "S": "2024-01-20T00:01:02"
        },
        "id": {
            "N": "1"
        },
        "data": {
            "L": [
                {
                    "M": {
                        "amount": {
                            "N": "40"
                        }
                    }
                },
                {
                    "M": {
                        "amount": {
                            "N": "56"
                        }
                    }
                }
            ]
        }
    }
]


resource:
[
    {
        "datetime": "2024-01-20T00:01:02",
        "id": 1,
        "data": [
            {
                "amount": 40
            },
            {
                "amount": 56
            }
        ]
    }
]

Notice the simplicity of the second output. The abstraction of the resources interface deserialized the low-level raw response of the AWS service API.

In most cases, the resources interface can be used over the low-level clients for the abstraction benefits. However, as stated at the top of the resources interface documentation:

The AWS Python SDK team does not intend to add new features to the resources interface in boto3".

Therefore, it is essential to compare both interfaces to determine whether the resources interface is sufficient for your use case.

In a future post, I will explore another method to work with the AttributeValue maps in Boto3.

Further reading

• DynamoDB - Boto3 1.34.23 documentation
• More info about resource deprecation? · boto/boto3 · Discussion #3563
• simplejson · PyPI