Lambda powertools python layer

Why this project exists

This is a custom construct that will create AWS Lambda Layer with AWS Powertools for Python library. There are different ways how to create a layer and when working with CDK you need to install the library, create a zip file and wire it correctly. With this construct you don't have to care about packaging and dependency management, just create a construct and add it to your function. The construct is an extension of the existing LayerVersion construct from the CDK library, so you have access to all fields and methods.

See the API for details.

import { LambdaPowertoolsLayer } from 'cdk-lambda-powertools-python-layer';

const powertoolsLayer = new LambdaPowertoolsLayer(this, 'TestLayer');

Python

from cdk_lambda_powertools_python_layer import LambdaPowertoolsLayer

powertoolsLayer = LambdaPowertoolsLayer(self, 'PowertoolsLayer')

The layer will be created during the CDK synth step and thus requires Docker.

Install

TypeSript/JavaScript:

npm i cdk-lambda-powertools-python-layer

Python:

pip install cdk-lambda-powertools-python-layer

Usage

Python

A single line will create a layer with powertools for python:

from cdk_lambda_powertools_python_layer import LambdaPowertoolsLayer

powertoolsLayer = LambdaPowertoolsLayer(self, 'PowertoolsLayer')

You can then add the layer to your funciton:

from aws_cdk import aws_lambda

aws_lambda.Function(self, 'LambdaFunction',
                            code=aws_lambda.Code.from_asset('function'),
                            handler='app.handler',
                            runtime=aws_lambda.Runtime.PYTHON_3_9,
                            layers=[powertoolsLayer])

You can specify the powertools version by passing the optional version paramter, otherwise the construct will take the latest version from pypi repository.

LambdaPowertoolsLayer(self, 'PowertoolsLayer', version='1.24.0')

Additionally, powertools have extras depenedncies such as Pydantic, documented here. This is not included by default, and you have to set this option in the construct definition if you need it:

LambdaPowertoolsLayer(self, 'PowertoolsLayer', include_extras=True)

Full example:

from aws_cdk import Stack, aws_lambda
from cdk_lambda_powertools_python_layer import LambdaPowertoolsLayer
from constructs import Construct


class LayerTestStack(Stack):

    def __init__(self, scope: Construct, construct_id: str, **kwargs) -> None:
        super().__init__(scope, construct_id, **kwargs)
        
        powertoolsLayer = LambdaPowertoolsLayer(
            self, 'PowertoolsLayer', include_extras=True, version='1.24.0')

        aws_lambda.Function(self, 'LambdaFunction',
                            code=aws_lambda.Code.from_asset('function'),
                            handler='app.handler',
                            runtime=aws_lambda.Runtime.PYTHON_3_9,
                            layers=[powertoolsLayer])

TypeScript

Full example for TypeScript:

import { Stack, StackProps } from 'aws-cdk-lib';
import { Construct } from 'constructs';
import { LambdaPowertoolsLayer } from 'cdk-lambda-powertools-python-layer';
import { Code, Function, Runtime } from 'aws-cdk-lib/aws-lambda';
import * as path from 'path';

export class CdkPowertoolsExampleStack extends Stack {
  constructor(scope: Construct, id: string, props?: StackProps) {
    super(scope, id, props);

    const powertoolsLayer = new LambdaPowertoolsLayer(this, 'TestLayer', {
      version: '1.22.0',
      includeExtras: true
    });

    new Function(this, 'LambdaFunction', {
      code: Code.fromAsset(path.join('./function')),
      handler: 'app.handler',
      runtime: Runtime.PYTHON_3_9,
      layers: [powertoolsLayer],
    });
  }
}