@whatskit/node-stage v0.1.0

@whatskit/node-stage

Library to manage environment variables and stages.

Setup

yarn add @whatskit/node-stage

# update
yarn up @whatskit/node-stage

Config.yaml Example

stages:
  myapp-dev: # one stage per deploy
    
    moduleName:
      variableKey: variableValue
  
    ## dotenv overrides
    # envFiles: [ '.env.myapp-dev.secrets' ]
    ## environment overrides
    # environment:
    #  app__spaDeploy__s3__bucket: myapp-dev-website

Library

import { loadConfig } from "@whatskit/node-stage";

const config = await loadConfig("/path/to/config.yaml", process.env.STAGE);

CLI

Options

--pwd

Root from where to fetch config.yaml and the base for all relative paths.

--stage

The slug of the deployment (ie. myapp-prd/mystage-stg/mystage-dev). This is the base of the config set in conifg.yaml.

--release

Release of the build (ie the git sha) and is unique per code. Autodetected from most CI services or Git.

--releaseStrategy

Release is the id that services like Docker and Sentry will use to identify the build. Most of the time this is independent of the stage, but in some cases, we have a separate build per stage.

• gitsha - (default) make the same build for all stages
• gitsha-stage - make a build based on the stage and git sha in cases where the build is different per stage

--service

If the stage has multiple services, you can define the one you want to deploy here.

Example configuration

stages:
  myapp-prd: &myapp-prd
    yaml_local_override: correct

  myapp-prd-worker:
    <<: *myapp-prd
    stage: myapp-prd

Overriding config and global prefix

CONFIG_PREFIX=app CONFIG_FILE=config.yaml

Command node-stage env

node-stage env var [path-to-variable]

Fetches the variable at the specified path and returns it.

Example:

node-stage env var moduleName.variableKey

Development

Test CLI locally

Set up ./test/secrets.env with credentials to do a E2E test.

# test with ts-node
yarn test:ts-node:cli --help

# build new version
yarn build

# test build
yarn test:cli --help