serverless-plugin-for-each

Serverless plugin that adds $forEach syntax to reduce code duplication and allow creating dynamic templates.

Configuration

$forEach block requires two fields:

iterator

A collection to iterate over. Can be:

• array

  iterator:
    - first
    - second

• object

  iterator:
    first: first-value
    second: second-value

• environment variable regular expression

  iterator:
    $env: ENVIRONMENT_VARIABLE_NAME_REGEX

template

An array or object to replace $forEach block with. Template can contain $forEach.key and $forEach.value variables that are interpolated during the replacement. Depending on the iterator type, these variables are replaced with:

• array:
  • $forEach.key - item index in the array
  • $forEach.value - item value
• object:
  • $forEach.key - field name
  • $forEach.value - field value
• environment variable regular expression:
  • $forEach.key - environment variable name
  • $forEach.value - environment variable value

Examples

Populate environment variables based on the list

Config

service: my-service

provider:
  environment:
    LOG_LEVEL: info
    REGION: us-east-1
    $forEach:
      iterator: ${self:custom.queues}
      template:
        $forEach.key_QUEUE_URL:
          Fn::ImportValue: my-service-$forEach.value-url

custom:
  # this list does not need to be hardcoded here and can come from a file, for example
  queues:
    FIRST: first-queue
    SECOND: second-queue

Result

service: my-service

provider:
  environment:
    LOG_LEVEL: info
    REGION: us-east-1
    FIRST_QUEUE_URL:
      Fn::ImportValue: my-service-first-queue-url
    SECOND_QUEUE_URL:
      Fn::ImportValue: my-service-second-queue-url

custom:
  # this list does not need to be hardcoded here and can come from a file, for example
  queues:
    FIRST: first-queue
    SECOND: second-queue

Multiple $forEach blocks in the same object

Config

service: my-service

provider:
  environment:
    LOG_LEVEL: info
    REGION: us-east-1
    $forEach_queues: # A "_suffix" can be added to avoid duplicated keys
      iterator: ${self:custom.queues}
      template:
        $forEach.key_QUEUE_URL:
          Fn::ImportValue: my-service-$forEach.value-url
    $forEach_tables:
      iterator: ${self:custom.tables}
      template:
        $forEach.key_TABLE_ARN:
          Fn::ImportValue: my-service-$forEach.value-arn

custom:
  queues:
    FIRST: first-queue
    SECOND: second-queue
  tables:
    ONE: table-one
    TWO: table-two

Result

service: my-service

provider:
  environment:
    LOG_LEVEL: info
    REGION: us-east-1
    FIRST_QUEUE_URL:
      Fn::ImportValue: my-service-first-queue-url
    SECOND_QUEUE_URL:
      Fn::ImportValue: my-service-second-queue-url
    ONE_TABLE_ARN:
      Fn::ImportValue: my-service-one-table-arn
    TWO_TABLE_ARN:
      Fn::ImportValue: my-service-two-table-arn

custom:
  queues:
    FIRST: first-queue
    SECOND: second-queue
  tables:
    ONE: table-one
    TWO: table-two

Attach multiple streams to lambda using environment variables

Config

service: my-service

functions:
  helloWorld:
    handler: ./src/hello-world.js
    events:
      - $forEach:
          iterator:
            $env: STREAM_ARN$ # matches all env vars with this suffix
          template:
            - stream:
                arn: $forEach.value
                startingPosition: TRIM_HORIZON
                enabled: true

Result

Assuming you have FIRST_STREAM_ARN=<first-stream-arn> and SECOND_STREAM_ARN=<second-stream-arn> set, the configuration above would be converted into

service: my-service

functions:
  helloWorld:
    handler: ./src/hello-world.js
    events:
      - stream:
          arn: <first-stream-arn>
          startingPosition: TRIM_HORIZON
          enabled: true
      - stream:
          arn: <second-stream-arn>
          startingPosition: TRIM_HORIZON
          enabled: true