Magic Modules

Overview

Magic Modules is a tool used to autogenerate support in a variety of open source DevOps tools for Google Cloud Platform. GCP "resource" definitions are encoded in a shared data file, and that data is used to fill in tool-specific templates across each of the tools Magic Modules generates.

Magic Modules generates GCP support for:

• Terraform
• Ansible
• InSpec

In addition, Magic Modules generates support for several companion features/tools:

• Terraform Google Inventory Mapper
• Terraform in Cloud Shell

Importantly, Magic Modules isn't full code generation. Every change is made manually; more than a code generator, Magic Modules is a force multiplier for development. While many Magic Modules resources are defined exactly based on the GCP API, we use Magic Modules to preemptively solve issues across each tool by encoding our field-tested learnings from other tools in those definitions. In effect, an issue solved in one tool will be solved for each other tool.

Getting Started with Magic Modules

You can try out Magic Modules immediately with Open in Cloud Shell below; if you're getting set up on a local workstation, this guide serves as a reference to help you get it set up.

Requirements

To get started, you'll need:

• Ruby 2.6.0
  • You can use rbenv to manage your Ruby version(s)
• Bundler
  • This can be installed with gem install bundler
• If you are getting "Too many open files" ulimit needs to be raised.
  • Mac OSX: ulimit -n 1000

Preparing Magic Modules / One-time setup

Important: Compiling Magic Modules can be done directly from the mmv1 directory within this repository. In the future we will add hybrid generation with multiple generators. All the information below pertains only to the contents of the mmv1 directory, and commands should be executed from that directory.

To get started right away, use the bootstrap script with:

./tools/bootstrap

---

Otherwise, follow the manual steps below:

If you're generating the Terraform providers (google and google-beta), you'll need to check out the repo(s) you're generating in your GOPATH. For example:

git clone https://github.com/hashicorp/terraform-provider-google.git $GOPATH/src/github.com/hashicorp/terraform-provider-google
git clone https://github.com/hashicorp/terraform-provider-google-beta.git $GOPATH/src/github.com/hashicorp/terraform-provider-google-beta

Magic Modules won't work with old versions of the Terraform provider repos. If you're encountering issues with vendoring and paths, make sure both MM and the Terraform provider are running on up to date copies of master.

Once you've prepared the target folders for the tools, run the following to finish getting Magic Modules set up by installing the Ruby gems it needs to run:

cd mmv1
bundle install

Now, you can verify you're ready with:

./tools/doctor

Generating the Terraform Providers

Before making any changes, you can compile the Terraform provider you're working on by running the following command. If Magic Modules has been installed correctly, you'll get no errors.

The following commands should be run from the root directory of the repository. OUTPUT_PATH should be set to the location of your provider repository, which is recommended to be inside your GOPATH.

make terraform VERSION=ga OUTPUT_PATH="$GOPATH/src/github.com/hashicorp/terraform-provider-google"
make terraform VERSION=beta OUTPUT_PATH="$GOPATH/src/github.com/hashicorp/terraform-provider-google-beta"

# Only generate a specific product (plus all common files)
make terraform VERSION=ga OUTPUT_PATH="$GOPATH/src/github.com/hashicorp/terraform-provider-google" PRODUCT=dataproc

It's worth noting that Magic Modules will only generate new files when run locally. The "Magician"- the Magic Modules CI system- handles deletion of old files when creating PRs.

Generating terraform-google-conversion

You can compile terraform-google-conversion by running the following command. If Magic Modules has been installed correctly, you'll get no errors.

make validator OUTPUT_PATH="/path/to/your/terraform-google-conversion"

Making changes to resources

Once again, see the Open in Cloud Shell example above for an interactive example of making a Magic Modules change; this section will serve as a reference more than a specific example.

Magic Modules mirrors the GCP REST API; there are products such as Compute or Container (GKE) that contains resources, GCP resources such as Compute VM Instances or GKE Clusters.

Products are separate folders under [products/], and each folder contains a file named api.yaml that contains the resources that make up the API definition.

Resources are made up of some metadata like their "name" in the API such as Address or Instance, some additional metadata (see the fields in resource.rb), and the meat of a resource, its fields. They're represented by properties in Magic Modules, an array of types.

Adding a new field to a resource in Magic Modules is often as easy as adding a type to the properties array for the resource. See this example where a field was added to all the tools (currently only Terraform) that support beta fields.

Tool-specific overrides

While most small changes won't require fiddling with overrides, each tool has "overrides" when it needs to deviate from the definition in api.yaml. This is often minor differences- the naming of a field, or whether it's required or not.

You can find them under the folder for a product, with the name {{tool}}.yaml. For example, Ansible's overrides for Cloud SQL are present at products/sql/ansible.yaml

You can find a full reference for each tool under overrides/{{tool}}/resource_override.rb and overrides/{{tool}}/property_override.rb, as well as some other tool-specific functionality.

Making changes to handwritten files

The Google providers for Terraform have a large number of handwritten files, written before Magic Modules was used with them. While conversion is ongoing, many resources are still managed by hand. You can modify handwritten files under the third_party/terraform directory.

Features that are only present in certain versions need to be "guarded" by wrapping those lines of code in version guards;

<% unless version == 'ga' -%>
  // beta-only code
<% end -%>

Testing your changes

Once you've made changes to resource definition, you can run Magic Modules to generate changes to your tool; see "Generating the Terraform Providers" above if you need a refresher. Once it's generated, you should run the tool-specific tests as if you were submitting a PR against that tool.

You can run tests in the {{output_folder}} you generated the tool in. See the following tool-specific documentation for more details on testing that tool;

Tool	Testing Guide
ansible	instructions
inspec	testing inspec-gcp
terraform	google provider testing guide
terraform (beta)	google-beta provider testing guide

Don't worry about testing every tool, only the primary tool you're making changes against. The Magic Modules maintainers will ensure your changes work against each tool.

If your changes have unintended consequences in another tool, a reviewer will instruct you to mark the field excluded or provide specific feedback on what changes to make to the tool-specific overrides in order for them to work correctly.

Submitting a PR

Before creating a commit, if you've modified any .rb files, make sure you run rake test! That will run rubocop to ensure that the code you've written will pass Travis.

To run rubocop automatically before committing, add a Git pre-commit hook with:

cp .github/pre-commit .git/hooks

Once you've created your commit(s), you can submit the code normally as a PR in the GitHub UI. The PR template includes some instructions to make sure we generate good PR messages for the tools' repo histories.

Once your PR is submitted, one of the Magic Modules maintainers will review it. They'll look over the code before running the "Magician", the Magic Modules CI system that generates PRs against each tool. Each review pass, your reviewer will run the Magician again to update the PRs against the tools.

If there are multiple tools affected, that first reviewer will be the "primary" reviewer, and for each other affected tool a maintainer for that specific tool will make a pass. The primary reviewer will make it clear which other maintainers need to review, and prompt them to review your code; you will communicate primarily with the first reviewer.

Even when multiple tools are affected, this will generally be a quick look by that maintainer with no changes needing to be made.

Once you've gotten approvals from the primary reviewer and the reviewers for any affected tools, the primary reviewer will merge your changes.

Glossary

The maintainers of the repository will tend to use specific jargon to describe concepts related to Magic Modules; here's a quick reference of what some of those terms are.

Term	Definition
tool	One of the OSS DevOps projects Magic Modules generates GCP support in
provider	Synonym for tool as referred to inside the codebase
downstream(s)	A PR created by the Magician against a tool
upstream	A PR created against Magic Modules or the Magic Modules repo
The Magician	The Magic Modules CI system that drives the GitHub robot modular-magician

Compiling MMv1 + tpgtools

We are currently developing a new generation tool for Terraform called tpgtools. This relies on a declarative client library that handles the actuation of GCP resources. We plan to gradually move resources from being generated by the existing Magic Modules Ruby code (mmv1) to using tpgtools. While we move resources over there will be a period of time when both generators are in use. To assist with generation we have a series of make targets that will run the compilers in tandem to generate the Terraform provider.

Sample Usage to compile at beta: make OUTPUT_PATH=/path/to/terraform-provider-google-beta VERSION=beta

Target single product: make OUTPUT_PATH=/path/to/terraform-provider-google VERSION=ga PRODUCT=compute

Target single resource make OUTPUT_PATH=/path/to/terraform-provider-google VERSION=ga PRODUCT=compute RESOURCE=image

For more advanced usage of mmv1 compiler flags, please execute the compiler directly from within the mmv1 directory.