B.DynamoDbCommon

A python library that makes it easier to interact with AWS DynamoDB tables.

Description

This library extends functionality of pynamodb, boto3, json, etc. modules to make it easier to interact with AWS DynamoDB tables. It contains various useful functionalities:

• Attributes (PynamoDB custom attributes);
• Encoders (JSON encoders to work with pynamo/dynamo data types);
• Models (PynamoDB custom models);
• Seeds (Various data seeding functions);
• Utils (Other cool functions);

Remarks

Biomapas aims to modernise life-science industry by sharing its IT knowledge with other companies and the community. This is an open source library intended to be used by anyone. Improvements and pull requests are welcome.

Related technology

• Python3
• Boto3
• AWS DynamoDB
• PynamoDB

Assumptions

This project assumes you know about DynamoDB service, and you prefer using PynamoDB ORM to interact with DynamoDB tables.

• Good Python skills and basis of OOP.
• Good PynamoDB/DynamoDB skills.

Useful sources

• DynamoDB API reference:
  https://docs.aws.amazon.com/amazondynamodb/latest/developerguide/HowItWorks.API.html
• DynamoDB boto3:
  https://boto3.amazonaws.com/v1/documentation/api/latest/reference/services/dynamodb.html
• PynamoDB ORM documentation:
  https://pynamodb.readthedocs.io/en/latest/
• PynamoDB ORM PyPi:
  https://pypi.org/project/pynamodb/

Install

Before installing this library, ensure you have these tools setup:

• Python / Pip

To install this project from source run:

pip install .

Or you can install it from a PyPi repository:

pip install b-dynamodb-common

Usage & Examples

This section shows various examples on how to use this library.

Attributes

Attributes module. Contains various custom PynamoDB ORM attribtues.

• Fernet attribute

Custom PynamoDB attribute that encrypts data in the database. Using Fernet algorithm

class User(Model):
    SECRET_ENCRYPTION_KEY = b'123456'

    first_name = FernetAttribute(SECRET_ENCRYPTION_KEY)
    last_name = FernetAttribute(SECRET_ENCRYPTION_KEY)

• KMS attribute

Custom PynamoDB attribute that encrypts data in the database. Using AWS KMS key to encrypt/decrypt data.

boto_client = boto3.client('kms')
kms_key_arn = 'arn:of:the:custom:kms:key'
    
class User(Model):
    first_name = KmsAttribute(boto_client, kms_key_arn)
    last_name = KmsAttribute(boto_client, kms_key_arn)

Encoders

Encoders module. Contains various encoding functionality.

• DynamoDbEncoder

Custom JSON encoder to handle DynamoDB data types.

data = {
    'key1': 'RandomData',
    'key2': OrderedSet([1, 2, 3]),
    'key3': Decimal(1.1)
}

json.dumps(data, cls=DynamoDbEncoder)

• PynamoDbEncoder

Custom JSON encoder to handle PynamoDB ORM and DynamoDB data types.

data = {
    'key1': MapAttribute(map_key_1='RandomData'),
    'key2': OrderedSet([1, 2, 3]),
    'key3': Decimal(1.1)
}

json.dumps(data, cls=PynamoDbEncoder)

Models

Models module. Contains various PynamoDB-based custom models.

• Permission model

Model that contains permissions attribute.

entity = PermissionsModel()
entity.pk = 'PK'
entity.permissions = ['list', 'of', 'permissions']
entity.save()

# Add more permissions and save.
entity.add_permission('permission')
entity.save()

• Model type factory

Allows to use the same pynamodb Model against multiple databases.

from pynamodb.models import Model

# Create your own model. Example, User model.
class User(Model): pass

# Specify table 1 against which an example user will be saved.
user_model_table_1 = ModelTypeFactory(User).create('user_table_1', 'eu-west-1')
user_model_table_1(hash_key='hash', range_key='range').save()

# Specify another table and save user in different table.
user_model_table_2 = ModelTypeFactory(User).create('user_table_2', 'eu-east-1')
user_model_table_2(hash_key='hash', range_key='range').save()

Seeds

Currently this module is empty.

Utils

Utilities module. Contains lots of cool functions.

• List function

Wraps PynamoDB query and scan functions for better management.

list_function: PynamoDBListFunction[DummyEntity] = PynamoDBListFunction(DummyEntity.query, 'PK')
items = list(list_function())

# You can also specify a transformer function to transform results before returning.
list_function: PynamoDBListFunction[DummyEntity] = PynamoDBListFunction(
    # PynamoDB list function (query).
    DummyEntity.query, 
    # Positional arguments.
    'PK',
    # Transformer function.
    transformer=lambda item: item.pk
)

# Will contain only pks.
ids = list(list_function())

• List results

Wraps PynamoDB query and scan functions to handle recursive last_evaluated_key tokens.

list_function: PynamoDBListFunction[DummyEntity] = PynamoDBListFunction(
    DummyEntity.scan,
    limit=10,
    filter_condition=DummyEntity.pk.is_in([...])
)

result = PynamoDBListResult(list_function)

# Fetch one time.
result.fetch(recursive=False)

# Check whether all results have been fetched.
result.finished

# If not, feel free to call it one more time and not worry about last_evaluated_key.
result.fetch(recursive=False)

# If you want to retrieve absolutely all results in one call:
result.fetch(recursive=True) # Simple!

Testing

This package has integration tests based on pytest. To run tests simply run:

pytest b_dynamodb_common_test/integration/tests

Contribution

Found a bug? Want to add or suggest a new feature? Contributions of any kind are gladly welcome. You may contact us directly, create a pull-request or an issue in github platform. Lets modernize the world together.