This GitHub Action allows you to test a single Terraform module. It provides options for linting, validating, formatting, planning, and testing the module. The action uses Terraform commands and TFLint to perform these tasks.
The following table describes the inputs for this action:
| Name | Description | Required | Default Value |
|---|---|---|---|
module |
Working directory | Yes | N/A |
lint |
Lint the module | No | true |
validate |
Validate the module | No | true |
fmt |
Format the module | No | true |
test |
Test the module | No | true |
test_dir |
Test directory, defaults to 'tests' | No | tests |
review |
Do review checks on module. If false only init runs. | No | true |
prefix |
tfstate s3 bucket prefix | No | duplo-tfstate |
- name: Validate My Terraform Module
uses: duplocloud/actions/setup-terraform@main
with:
module: ./path/to/module
lint: true
validate: true
fmt: true
test: true
test_dir: tests
review: false
prefix: duplo-tfstateIn the above example, the action is being used to validate a Terraform module located in the ./path/to/module directory. All of the available options (lint, validate, fmt, plan, and test) are set to their default values of "true", which means all tasks will be performed.
When using Azure, you will need to set this environment variable in your workflow file: RESOURCE_GROUP. This will allow the azure cli to retrieve the ARM_ACCESS_KEY value and set the variable in the environment. This is needed by Azure TF provider to authenticate to Azure to use the storage bucket backend. The variable may be set at any level as long as it is available to the action.
Set the DUPLO_TF_BUCKET environment variable to override the default bucket name. This is useful when you want to use a different bucket for your Terraform state files. The action will use this bucket name instead of the default one.