Hello all - I’m a UK based software engineer at AWS. I’ve been wanting to experiment with Pants for a while to improve developer experience around software that runs in distributed systems, as well as benefit from other perks of a strong set of build tooling - I’m attempting to consolidate a service I own that’s deployed with CDK from a multi-repo setup into a relatively small monorepo. If anyone has combined Pants/AWS CDK and deployment I’d love to hear about it!

We use both CDK and pants at work but haven't been able to really combine them well together. Most of the stuff we are using pants for is related to CI part of Codepipeline.

If you haven't yet opened a ticket, please do let us know more about your respective use cases and what changes to Pants could help you use CDK and Pants together more effectively.

Thanks Carina - I’m still at the “figuring out what I can do and what I’m trying to do” stage before having any specific asks!

Yes, I do! I have it generate cloudasm zip artifacts, then use this script below to take those artifacts and deploy them (so I'm not running my CDK code at deploy time & artifacts are cacheable)

#!/usr/bin/env bash
# unzip a zip to a temporary directory and output the tmpdir path
# useful for cloudasm deploys with cdk
# cdk deploy -a "$(build-support/inline-cloudassembly.sh dist/whatever.zip)/sub/path"

set -euo pipefail

# You can change these constants.
if [[ ! -f "${1}" ]] ; then
  echo "Could not find cloudassembly zip '${1}'"
  exit 1
fi

DIR=$(mktemp -d -t cloudasm)

unzip -q -d "${DIR}" "${1}" 1>&2

echo "${DIR}"