gshpychka
commented
2 years ago The Infrastructure and the code that runs on the infrastructure should be separate.
FYI the CDK best practices suggest the opposite:
The AWS CDK not only generates AWS CloudFormation templates for deploying infrastructure, it also bundles runtime assets like Lambda functions and Docker images and deploys them alongside your infrastructure. So it's not only possible to combine the code that defines your infrastructure and the code that implements your runtime logic into a single construct— it's a best practice. These two kinds of code don't need to live in separate repositories or even in separate packages.
A construct that is self-contained, in other words that completely describes a piece of functionality including its infrastructure and logic, makes it easy to evolve the two kinds of code together, test them in isolation, share and reuse the code across projects, and version all the code in sync.
Describe the feature
A LambdaCodeDeploymentAction. A new action for a deploy stage of CodePipelines that allows either a .zip file from an S3 bucket or an Artifact from CodeBuild to be deployed to a Lambda function.
Use Case
Its frustrating that my Lambda logic and my CDK have to be in the same repository so that when I update my Lambda it updates the code deployed to the Lambda function. The Infrastructure and the code that runs on the infrastructure should be separate. By doing this, I can use a source action to gather my lambda code from a separate repository to my CDK, build the code to create a .ZIP and then deploy that code to Lambda. All without having to update the CDK repository.
Proposed Solution
As described above, creating a LambdaCodeDeploymentAction Action to fulfil this deployment. A feature that is available through the console using the 'Upload from' button.
Other Information
No response
Acknowledgements
CDK version used
2.35.0
Environment details (OS name and version, etc.)
Mac OS Monterey 12.5.1