通过Shell脚本生成Terraform HCL格式变量文件的问题

输入输出示例 #

CSV输入示例 (input.csv) #

env,region,vpc_id,subnet_ids
dev,us-east-1,vpc-12345,"subnet-67890,subnet-abcde"
prod,eu-west-1,vpc-xyzab,"subnet-cdefg,subnet-hijkl"

目标HCL变量文件 (variables.tfvars) #

environment_config = {
  dev = {
    region     = "us-east-1"
    vpc_id     = "vpc-12345"
    subnet_ids = ["subnet-67890", "subnet-abcde"]
  }
  prod = {
    region     = "eu-west-1"
    vpc_id     = "vpc-xyzab"
    subnet_ids = ["subnet-cdefg", "subnet-hijkl"]
  }
}

API JSON输入示例 #

[
  {
    "env": "dev",
    "region": "us-east-1",
    "vpc_id": "vpc-12345",
    "subnet_ids": ["subnet-67890", "subnet-abcde"]
  },
  {
    "env": "prod",
    "region": "eu-west-1",
    "vpc_id": "vpc-xyzab",
    "subnet_ids": ["subnet-cdefg", "subnet-hijkl"]
  }
]

方案1：CSV转HCL的Shell脚本 #

直接用awk处理CSV，解析表头和数据行，生成符合HCL语法的变量文件：

#!/bin/bash

awk -F ',' '
BEGIN {
  print "environment_config = {"
}
NR == 1 {
  # 提取表头字段，去除引号
  for (i=1; i<=NF; i++) {
    gsub(/"/, "", $i)
    headers[i] = $i
  }
  next
}
{
  # 提取环境名，清理引号
  env_name = $1
  gsub(/"/, "", env_name)
  printf "  %s = {\n", env_name
  
  # 遍历每个字段生成HCL属性
  for (i=2; i<=NF; i++) {
    field = headers[i]
    value = $i
    gsub(/^"|"$/, "", value)  # 移除字段首尾的引号
    
    # 处理数组类型字段（比如subnet_ids）
    if (field == "subnet_ids") {
      split(value, subnets, /,/)
      printf "    %s = [", field
      for (j=1; j<=length(subnets); j++) {
        gsub(/^ +| +$/, "", subnets[j])  # 清理子网ID前后空格
        printf "%s\"%s\"", (j>1 ? ", " : ""), subnets[j]
      }
      printf "]\n"
    } else {
      printf "    %s = \"%s\"\n", field, value
    }
  }
  printf "  }\n"
}
END {
  print "}"
}' input.csv > variables.tfvars

方案2：API JSON转HCL的Shell脚本 #

先用jq将JSON转成结构化CSV，再复用上述awk逻辑处理，避免直接解析复杂JSON：

#!/bin/bash

# 替换成你的API请求命令，比如 curl https://api.example.com/env-config
curl https://api.example.com/env-config | jq -r '
  .[] | [.env, .region, .vpc_id, ([.subnet_ids[]] | join(","))] | @csv
' | awk -F ',' '
BEGIN {
  print "environment_config = {"
}
{
  env_name = $1
  gsub(/"/, "", env_name)
  printf "  %s = {\n", env_name
  
  printf "    region = \"%s\"\n", $2
  printf "    vpc_id = \"%s\"\n", $3
  
  # 处理subnet_ids数组
  split($4, subnets, /,/)
  printf "    subnet_ids = ["
  for (j=1; j<=length(subnets); j++) {
    gsub(/^"|"$/, "", subnets[j])
    printf "%s\"%s\"", (j>1 ? ", " : ""), subnets[j]
  }
  printf "]\n"
  
  printf "  }\n"
}
END {
  print "}"
}' > variables.tfvars

脚本关键说明 #

• 自动处理字段的引号和空格，避免HCL语法错误
• 针对数组类型字段做特殊解析，生成HCL列表格式
• CSV脚本支持动态表头，新增字段无需修改核心逻辑
• JSON脚本依赖jq工具，需提前安装（apt install jq / brew install jq）

Terraform引用示例 #

variable "environment_config" {
  type = map(object({
    region     = string
    vpc_id     = string
    subnet_ids = list(string)
  }))
  description = "Environment-specific infrastructure config"
}

# 示例：按环境创建VPC标签
resource "aws_vpc" "main" {
  for_each = var.environment_config
  cidr_block = "10.${index(keys(var.environment_config), each.key)}.0.0/16"
  tags = {
    Name = "${each.key}-main-vpc"
  }
}

内容的提问来源于stack exchange，提问作者L_sama