The Chain of Responsibility
Create cascading fallbacks for lookups, ensuring that if a primary query fails, secondary and tertiary sources are consulted.
variable "requested_ami" {
type = string
default = ""
}
# Link 1: Direct user input
# Link 2: SSM Parameter Store
data "aws_ssm_parameter" "default_ami" {
name = "/infrastructure/default_ami"
}
# Link 3: Hardcoded fallback
locals {
hardcoded_fallback = "ami-0abcdef1234567890"
# The Chain resolves the final value
final_ami = coalesce(
var.requested_ami != "" ? var.requested_ami : null,
data.aws_ssm_parameter.default_ami.value != "" ? data.aws_ssm_parameter.default_ami.value : null,
local.hardcoded_fallback
)
}
resource "aws_instance" "server" {
ami = local.final_ami
instance_type = "t3.micro"
}
The Chain of Responsibility: The Cascading Divination
When divining the exact parameters for a spell—such as which Amazon Machine Image to use—relying on a single source of truth is brittle. What if the user forgets to provide it? What if the parameter store is empty?
The Chain of Responsibility pattern establishes a series of handlers. In Terraform, this is elegantly achieved using the coalesce() function. It evaluates a list of expressions and returns the first non-null, non-empty value. The magic flows through the user’s explicit command first, falls back to a central registry if absent, and finally settles on a hardcoded safety net. The spell will not fail due to a single missing link.