Import

The import block allows you to merge multiple pipelines together, into a flatter pipeline. import blocks are evaluated before the dependency tree is generated, and hence, variables or expressions are not permitted. If you need to statisfy a more dynamic evaluation, use macro instead.

import block accepts a single parameter without any id. The block specifies the source, which should be a constant value, and could be a local file,

Example usage

togomak {
  version = 2
}

import {
  source = "./module"
}

stage "main" {
  script = "echo script from the main file"
}

import {
  source = "git::https://github.com/srevinsaju/togomak-first-module"
}

Argument reference

• source - URL to fetch the pipeline from, could be local file path, relative file path, absolute file path, a Git URL, a HTTP URL, GCS bucket, AWS s3 bucket or mercurial.

The URL for source may be link to the required pipeline inclusive of the protocol, or you may omit in certain cases, where detectors would be run on top of it.

For example:

• github.com/srevinsaju/togomak would automatically evaluate to a SSH URL
• ./file.hcl would evaluate a relative file in the Togomak current working directory.
• git::https://github.com/srevinsaju/togomak//examples/docker would specifically checkout the examples/docker folder from the GitHub repository https://github.com/srevinsaju/togomak.git over https using git.

Additionally, you may add query strings like ?ref=abcdefg to checkout a specific commit of the repository. Check out the documentation of the underlying library, go-getter which powers terraform modules here: hashicorp/go-getter

What's the difference between macro and import?

Great question, the following diagram may help you filter out the differences between them.