drone-plugin-hcltf

a drone plugin to execute cdktf within a harness pipeline

images are tagged with node<version>-tofu<version>-cdktf<version> and you can browse published versions on dockerhub

Options

PLUGIN_CODE_DIR: directory in your repo where the CDKTF code is located

PLUGIN_APP: CDKTF app to synth

PLUGIN_OUTPUT_DIR (optional): directory to store resulting TF code

PLUGIN_NPM_INSTALL_OPTIONS (optional): additional flags to pass to npm i command

PLUGIN_HCL (optional): if set, generate output in HCL instead of JSON (default)

preperation

this plugin uses opentofu, as such to work with cdktf you will need to set the providers in your cdktf.json to reference the opentofu registry:

for the aws provider:

  "terraformProviders": [
    "registry.opentofu.org/hashicorp/aws@5.33.0"
  ],

usage

in an iacm pipeline, this plugin must be used before any tofu steps, as they all need the tf code to exist locally:

- step:
    type: Plugin
    name: cdktf
    identifier: cdktf
    spec:
      connectorRef: account.harnessImage
      image: harnesscommunity/drone-plugin-cdktf-ts
      settings:
          CODE_DIR: <+workspace.envVars.CDKTF_DIR>
          APP: <+workspace.envVars.CDKTF_APP>
          OUTPUT_DIR: /harness/<+workspace.folderPath.split("/stacks")[0]>

where you will need to set CDKTF_DIR and an environment variable in your workspace to be the relitive location of your CDKTF code in your repo, and CDKTF_APP is set to the name of the app you want to synthisize for your pipeline.

the folderPath on your workspace should be set and end in stacks/<app name> to follow the standard output scheme of CDKTF.

testing

from the location of your cdftf code:

rm -rf .gen cdktf.out node_modules;
docker run -v ${PWD}:/harness -e PLUGIN_CODE_DIR=/harness -e PLUGIN_APP=app -it drone-plugin-cdktf-ts