WARNING: This repository is no longer maintained ⚠️

The generated content is now available in our Go-lang static application.

This repository will not be updated. The repository will be kept available in read-only mode.

GoServer Generator

Overview

This generator produces a Go + Gin server project with all the ingredients you need for a good start at building a cloud native application. You can choose between either a simple web app, microservice pattern, or a simple blank app.

You can also bring your own optional OpenAPI 2.0 document to direct code generation for top-down development.

Monitoring and Health

The generated projects are pre-wired for monitoring and health checks. The app includes:

1. Prometheus endpoint

2. Hystrix fault tolerance

3. /health endpoint returning JSON payload for UP and DOWN

Deployment Enablement

The generated projects include deployment configuration for the following environments:

1. Docker

   The projects include Dockerfiles to build images for both release and development.

2. Kubernetes

   The projects include a Helm chart for deployment to Kubernetes.

3. Cloud Foundry

   The projects include a manifest for deployment to Cloud Foundry.

4. Dev-ops Pipeline

   The projects include a toolchain and pipeline definition for CI/CD deployment to the IBM Cloud.

Pre-requisites

• Install Yeoman

npm install -g yo

Installation

npm install -g generator-goserver

Usage

yo goserver

Prompting

When running the generator with prompts, make sure you set up a GOPATH. Your application directory will be generated at $GOPATH/src/<application-name>.

When you run 'yo goserver', it will prompt you for the following:

1. Application name

   Specify the application name. It defaults to the current directory name. This is a required value.

2. Application type

   Specify the type of application you want to create. The options are basic web app, microservice, or a simple blank app.

3. OpenAPI 2.0 doc file name (if blank application type selected)

   Specify the relative or absolute file name of a OpenAPI 2.0 document to direct the project's code generation. A route stub will be scaffolded and registered for each route defined in the OpenAPI 2.0 document. This is an optional value.

4. Docker registery

   Specify a docker registry. This is an optional value.

5. Deployment type

   Specifies the environment to which you would want to deploy you app to. Options are Cloud Foundry or Kubernetes.

6. IBM Cloud Service Enablement

   Specify Y|N whether or not you want to scaffold IBM Cloud service enablement into your project. If you specify 'Y', you will be able to select one or more services from a checklist. For each service you select, configuration and access scaffolding code is generated. IBM Cloud service enablement is optional.

Bluemix mode (without prompting)

Use bluemix mode to create an app without having to use the UI. This is useful when you want to use this generator to build a project by only running a script instead of interactively.

Bluemix usage

To specify the name of the project use:

yo goserver --bluemix='{"name":"your-app-name"}'

To specify the type of your application:

yo goserver --bluemix='{"name":"your-app-name"}' --spec='{"applicationType":"your-app-type"}'

The three valid application types are: WEBAPP, MS, and BLANK

To specify which services to add use:

yo goserver --bluemix='{"services":["service1", "service2"]}'

For valid services see below.

Full usage:

yo goserver --bluemix='{"name":"your-app-name","swaggerFilePath":"path-to-your-file","services":["service1", "service2"]}' --spec='{"applicationType":"your-app-type"}'

Note: you can only provide a swagger path if it is a blank application.

Valid Services

• 'watson assistant'
• 'watson discovery'
• 'watson language translator'
• 'watson natural language classifier'
• 'watson language understanding'
• 'watson personality insights'
• 'watson speech to text'
• 'watson text to speech'
• 'watson tone analyzer'
• 'watson visual recognition'

Project Build/Run

Build your generated project one of two ways:

1. If you want to run the project locally, first make sure you have dep for dependency management. Then run the following:

   dep ensure
   go run server.go
   

2. Containerized, using IBM Cloud Developer Tools.

   • Install IBM Cloud Developer Tools on your machine
   • Install the dev plugin: ibmcloud plugin install dev

   Note that a containerized approach is supported through the tooling in special consideration of Kubernetes as a deployment environment, following the dev/prod parity principle of 12 Factor Apps.

   There are ibmcloud dev commands to simplify this for you:

   1. ibmcloud dev build

      Builds Docker image for dev mode and does dep ensure to install dependencies.

   2. ibmcloud dev test

      Runs project unit tests in dev mode Docker container.

   3. ibmcloud dev debug

      Runs the project in debug mode in the dev mode Docker container. The app will start and listen on port 8181 by default for a debug client to attach and take control.

   4. ibmcloud dev run

      Runs the project in the release mode Docker container. The release mode Docker container is built with GIN_MODE set to release.

Project Deployment

Docker

Build a Docker image and run project in a Docker container using Docker commands in the project root directory:

1. docker build -t my-image .
2. docker run -p 8080:8080 --name my-container my-image

Stop and optionally remove the container and image with the following commands:

1. docker stop my-container
2. docker rm my-container
3. docker rmi my-image

Kubernetes Deployment

Deploy to Kubernetes using Helm or the IBM Cloud Developer Tools.

1. Helm

   1. Push your image to a Docker image accessible to your Kubernetes environment, such as Docker Hub or your company's private image registry.

   2. Install your project from the project's root directory, using the included Helm chart:

      helm install chart/<application name> --name=<release name> --set image.repository=<image name> --set image.tag=<tag value> --set image.pullPolicy=<pull policy>

      Where:

      • <application name>

        The name you gave to your application when you generated it.

      • <release name>

        An arbitrary name you give to this install instance.

      • <image name>

        The registry/image name of your release Docker image - e.g. 'registry.ng.bluemix.net/myspace/myimage'

      • <tag value>

        The image tag value of your release Dockerimage - e.g. 'latest' or '1.0.0'

      • <pull policy>

        'Always' or 'IfNotPresent'. See Kubernetes image documentation for further explanation.

   Notes:

   1. If the helm install command above gives you an error about not finding 'tiller', execute helm init --upgrade.

   2. The helm command installs to the Kubernetes environment pointed to by the KUBECONFIG environment variable. Make sure you are in configuration mode for your Kubernetes cluster.

   3. The Helm command is installed when you install the IBM Cloud Developer Tools.

   4. To delete the helm deployment, execute helm del --purge <release name>.

2. IBM Cloud Developer Tools

   ibmcloud dev deploy --target container

   Notes:

   1. The tool will prompt for registry/image name, then push your image and install your Helm chart to the Kubernetes environment pointed to by your KUBECONFIG environment variable.

   2. For IBM Cloud, set KUBECONFIG using ibmcloud cs cluster-config <cluster name>. Note this command is installed as part of IBM Cloud Developer Tools.

Clound Foundry Deployment

1. Add a host entry to the "manifest.yml" file (Ex. host: my-app-name).

2. If you want to connect to a service that is already provisioned in the Cloud, add the instance name in manifest.yml under services or replace the placeholder if you indicated a service on project creation

3. cf push

   Note: if you installed IBM Cloud Developer Tools, you can run the ibmcloud cf push command. Otherwise, install the cf command from Pivotal.

4. IBM Cloud Developer Tools

   ibmcloud dev deploy

Development

Clone this repository and link it via npm

git clone https://github.com/ibm-developer/generator-goserver
cd generator-goserver
npm link

In a separate directory invoke the generator via

yo goserver

Publishing Changes

In order to publish changes, you will need to make a pull request against master by either fork the repository or creating a new branch against master.

Make sure to follow the conventional commit specification before contributing. To help you with commit a commit template is provide. Run config.sh to initialize the commit template to your .git/config or use commitizen

Once you are finished with your changes, run npm test to make sure all tests pass. Do a pull request against master, make sure the build passes. A team member will review and merge your pull request. Once merged to master an auto generated pull request will be created against master to update the changelog. Make sure that the CHANGELOG.md and the package.json is correct before merging the pull request. After the auto generated pull request has been merged to master the version will be bumped and published to npm.