HCL: the language
Blocks, arguments and types
resource "aws_s3_bucket" "logs" { # block type, resource type, local name
bucket = "acme-prod-logs-ap-south-1" # argument
tags = { # map
Environment = "prod"
}
}
A resource's address is aws_s3_bucket.logs; its attributes are referenced as aws_s3_bucket.logs.arn.
Types: string, number, bool, list(type), set(type), map(type), object({...}), tuple([...]), and any.
Variables, locals and outputs
variable "environment" {
description = "Deployment environment"
type = string
validation {
condition = contains(["dev", "staging", "prod"], var.environment)
error_message = "environment must be dev, staging or prod."
}
}
variable "subnets" {
type = map(object({
cidr = string
az = string
}))
}
variable "db_password" {
type = string
sensitive = true # redacted in plan/apply output (still stored in state!)
}
locals {
name_prefix = "acme-${var.environment}"
common_tags = {
Environment = var.environment
ManagedBy = "terraform"
Repo = "acme/platform-infra"
}
}
output "vpc_id" {
description = "ID of the VPC"
value = aws_vpc.main.id
}
Setting variable values (later sources win): defaults → terraform.tfvars / *.auto.tfvars → -var-file=prod.tfvars → -var 'environment=prod'; environment variables TF_VAR_environment=prod sit below files in precedence. Undefined required variables are prompted for interactively (or fail in CI with -input=false).
- locals — named expressions computed once inside a module (avoid repetition).
- outputs — expose values from a module to its caller, or from the root module to the CLI/remote state readers.
Expressions and functions
name = "${local.name_prefix}-web" # interpolation
instance = var.environment == "prod" ? "m6i.large" : "t3.medium" # conditional
azs = slice(data.aws_availability_zones.available.names, 0, 3)
cidrs = [for i, az in local.azs : cidrsubnet(var.vpc_cidr, 4, i)] # for expression → list
by_name = { for s in var.services : s.name => s } # for expression → map
prod_only = [for s in var.services : s.name if s.critical] # with filter
arns = aws_iam_role.app[*].arn # splat
merged = merge(local.common_tags, { Name = "web" })
json = jsonencode({ Version = "2012-10-17", Statement = [] })
file_body = templatefile("${path.module}/user_data.sh.tftpl", { cluster = var.cluster_name })
value = try(var.config.optional_field, "default")
length(var.subnets); lookup(var.amis, var.region, null); coalesce(var.a, var.b)
Useful functions: cidrsubnet (subnet maths — remember the IP topic), merge, concat, flatten, keys/values, zipmap, toset, contains, format, join/split, replace, lower, jsonencode/yamlencode, file, templatefile, base64encode, try, can, timestamp (avoid in resources — causes perpetual diffs). Explore them in terraform console.
Data sources
data "aws_availability_zones" "available" { state = "available" }
data "aws_ami" "al2023" {
most_recent = true
owners = ["amazon"]
filter {
name = "name"
values = ["al2023-ami-*-x86_64"]
}
}
data "aws_caller_identity" "current" {}
data "aws_iam_policy_document" "assume_ec2" {
statement {
actions = ["sts:AssumeRole"]
principals {
type = "Service"
identifiers = ["ec2.amazonaws.com"]
}
}
}
aws_iam_policy_document is the idiomatic way to write IAM JSON in Terraform — validated and composable.
Dynamic blocks
Generate repeated nested blocks:
resource "aws_security_group" "web" {
name = "${local.name_prefix}-web"
vpc_id = aws_vpc.main.id
dynamic "ingress" {
for_each = var.ingress_ports # e.g. [80, 443]
content {
from_port = ingress.value
to_port = ingress.value
protocol = "tcp"
cidr_blocks = ["0.0.0.0/0"]
}
}
}
Use sparingly — overused dynamic blocks make modules hard to read.