This URL hosts a Git repository (ie. git clone https://steady.supply/git/λ-demo will work). For your convenience, the file README.md from this repository is rendered below.

λ demo

Local prequisites
What does it do?
How do I use it?
How does it work?

Local prequisites

GNU Make
Docker
aws/env file containing AWS IAM and MFA ARNs (example in aws/env-example).

What does it do?

Offers a CLI for calling Pythong modules that deploy AWS CloudFormation templates and package/publish artifacts for AWS Lambda functions.

How do I use it?

Establish the prerequisites mentioned above and run make -C aws guest in the repo root to spin up the container (see aws/Dockerfile). Give an OTP and you’ll be dumped onto an Amazon-authenticated shell in the container. From there you can call the publish and deploy commands.

Example

From the guest, call eg.

# publish version-tracker v0.0.42

Update aws/yaml/parameters.yml with the version mentioned above (this is the part that will be taken care of by the version tracker service/macro for other packages). From the guest again, call eg.

# deploy version-tracker test

The test argument is the stage, not currently used.

How does it work?

An OTP is requested on the host (see aws/scripts/developer/aws-sts-assume-role) for the MFA device specified in aws/env. Still on the host, an IAM session is created for the IAM role specified in aws/env. This session is passed to the guest (see aws/Makefile) via an environment variable. The guest is a Docker container that holds required packages and Python libraries for interacting with AWS along with gubbins for using the session it is passed.

On the guest, there are entrypoint scripts (see scripts/bin) which invoke Click commands, which in turn call Python modules (see aws/script/function) to carry out the work. The pattern is designed so these Python modules can be reused, eg. for more specialised entrypoints.