commodity-namer

Consistent structured resource names for automated infra:

• prefix all resource names
• add context with optional "resource type"
• truncate name parts proportionally to ensure max length
• enforce RFC 1035:
  • Must start with a letter
  • Can contain letters, digits, and hyphens as interior characters
  • Must end with a letter or digit (cannot end with a hyphen)
  • Maximum length of 63 characters
• option WithReplace to auto replace common characters and conver to lowercase

See:

• https://cloud.google.com/compute/docs/naming-resources

Getting started

import (
  namer "github.com/davidmontoyago/commodity-namer"
)

type MyInfra struct {
  // Embed the namer
  namer.Namer
}

func NewMyInfra() *MyInfra {
  return &MyInfra{
    // Set the base name
    Namer: namer.New("my-prod-stack"),
  }
}

func (y *MyInfra) deploy() {
  name := y.NewResourceName("orders", "bucket", 63) // my-prod-stack-orders-bucket
  ...
  name = y.NewResourceName("orders", "cache", 63) // my-prod-stack-orders-cache
  ...
  name = y.NewResourceName("pending-work", "queue", 63) // my-prod-stack-pending-work-queue
  ...
  name = y.NewResourceName("backend-processor", "service-account", 30) // my-prod-backend-pr-service-a
  ...
  name = y.NewResourceName("ingestor", "generic-service", 25) // my-prod-inges-generic-s
  ...
  name = y.NewResourceName("require-https", "", 20) // my-pro-require-https
}

Install

go get github.com/davidmontoyago/commodity-namer

Name structure

"[base name]-<resource name>-[resource type or group]"

1. base name:: optional prefix to set for all resources. E.g.: gcp-
2. resource name:: required resource name. E.g.: document-store, task-backlog, assets-cache, inference-endpoint
3. resource type: optional resource type or group. E.g: secret, bucket, service, version