• Overview
• Template Output
• Technical requirements
• Async API Requirements
• Supported protocols
• How to use the template
  • CLI
    • Run the following command to generate a Go module
    • How to use the generated code
      • Pre-requisites
      • Running the code
• Template configuration
• Contribution guide

Overview

This template generates a Go module that uses watermill as library for building event-driven applications. The reason to choose watermill is because it provides an abstraction to messaging supporting multiple protocols. This will help this generator support multiple protocols.

Template Output

The go code generated by this template has the following structure

• asyncapi
  • handlers.go -- handlers for publishers and subscribers
  • payloads.go -- generated go structs using Modelina
  • publishers.go
  • router.go -- configures watermill router, needed only for subscribers
  • server.go -- server config
  • subscribers.go
• go.mod
• go.sum
• main.go

Technical requirements

• 1.1.0 =< Generator < 2.0.0,
• Generator specific requirements

Async API Requirements

To have correctly generated code, your AsyncAPI file MUST define operationId for every operation.

Example:

channels:
  light/measured:
    subscribe:
      operationId: LumenPublish

Supported protocols

| Protocol | Subscriber | Publishers | |---|---|---| | AMQP | Yes | Yes |

How to use the template

This template must be used with the AsyncAPI Generator. You can find all available options here.

CLI

This template has been tested to generate an AMQP subscriber for this asyncapi.yaml file

Run the following command to generate a Go module

npm install -g @asyncapi/generator
# clone this repository and navigate to this repository
ag test/asyncapi.yaml @asyncapi/go-watermill-template -o /path/to/generated-code -p moduleName=your-go-module-name

Following are the options that can be passed to the generator

1. moduleName: name of the go module to be generated

How to use the generated code

The above code currently generates a Go module that has a AMQP subscriber.

Pre-requisites

To run the generated code the following needs to be installed

1. go 1.16 +
2. rabbitmq-server OR docker

Running the code

1. Navigate to the path where the code was generated
2. Run the following commands to download the dependencies

go mod download
go mod tidy

3. Currently the code does not utilize the server bindings to generate the server URI. It is currently hardcoded to point to a local instance of rabbitmq. It is hardcoded as "amqp://guest:guest@localhost:5672/" at <generated-code>/config/server.go. Change it as per your rabbitmq instance requirements
4. Finally to execute the code run

go run main.go

5. Running local instance of rabbitmq, navigate to it using http://localhost:15672/ with username and password guest/ guest (These are default rabbitmq credentials). FYI one can start an instance of rabbitmq using docker as follow

   docker run -d -p 15672:15672 -p 5672:5672 rabbitmq:3-management

6. Create a queue as per the AsyncAPI spec. This can be done either of the following ways
   • Using the UI: Refer to this article that walks through the process of how this can be done in the UI / RabbitMQ Admin
   • cURL request. Default rabbitmq user is guest and password is guest

    curl --user <rabbit-user>:<rabbit-password> -X PUT \
      http://localhost:15672/api/queues/%2f/<queue-name> \
      -H 'cache-control: no-cache' \
      -H 'content-type: application/json' \
      -d '{
    "auto_delete":false,
    "durable":true
    }'

7. Publish a message to the queue as per the AsyncAPI spec. This can be done either of the following ways
   • Using the UI: Refer to this article that walks through the process of how this can be done in the UI / RabbitMQ Admin
   • cURL request. Default rabbitmq user is guest and password is guest

    curl --user <rabbit-user>:<rabbit-password> -X POST \
      http://localhost:15672/api/exchanges/%2f/amq.default/publish \
      -H 'cache-control: no-cache' \
      -H 'content-type: application/json' \
      -d ' {
    "properties":{},
    "routing_key":"light/measured",
    "payload":"{\"id\":1,\"lumens\":2,\"sentAt\":\"2021-09-21\"}",
    "payload_encoding":"string"
    }'

8. Check the output at the terminal where go run main.go was running and the published message should be printed

Template configuration

You can configure this template by passing different parameters in the Generator CLI: -p PARAM1_NAME=PARAM1_VALUE -p PARAM2_NAME=PARAM2_VALUE

|Name|Description|Required|Example| |---|---|---|---| |moduleName|Name for the generated Go module|false|my-app|

Contribution guide

If you are interested in contributing to this repo refer to the contributing docs