Making open cloud easily accessible and managed.
Table of Content
- Introduction
- Supported Service Providers
- Getting Started
- Overriding Providers
- Service Types
- Development setup
- Important Notes for Developers
- Test Changes
- NodeCloud Code Generation tool
- License
NodeCloud ☁️
is a standard library to get a single API on the open cloud with multiple providers. It is a NodeJs library which comes with plugins for each cloud provider. NodeCloud's aim is to abstract away the differences between different cloud providers. It provides an easy to use API for developers in order to interact with different cloud providers.
NodeCloud will be useful to you if:
- you work on a project which uses multiple cloud providers
- you are looking for an abstract cloud API which can switch between cloud providers with fewer code changes
- you are an open-source enthusiast who is into cloud engineering or code generation
- you want to improve your skills in NodeJS, Typescript and cloud service providers
- Amazon Web Services (AWS)
- Azure
- Google Cloud Platform (GCP)
- DigitalOcean
- AliCloud
📢 if your required cloud provider plugin is not listed here, we'd love your help to add it :)
In order to use NodeCloud
, you will need to follow the following steps:
1️⃣ Download NodeCloud common module
Head on over to NPM or Yarn to download the latest version of NodeCloud.
npm i @nodecloud/common
or
yarn add @nodecloud/common
2️⃣ Download at least one NodeCloud plugin
Once @nodecloud/common
is installed, you need to install the plugins to interact with different cloud providers. The below table shows the available plugins maintained by ScoRe Lab. There can be other community-driven plugins that you will be able to use with NodeCloud.
Plugin | Installation |
---|---|
AWS plugin | yarn add @nodecloud/aws-plugin or npm i @nodecloud/aws-plugin |
Azure plugin | yarn add @nodecloud/gcp-plugin or npm i @nodecloud/gcp-plugin |
Google Cloud plugin | yarn add @nodecloud/azure-plugin or npm i @nodecloud/azure-plugin |
Alibaba plugin | yarn add nodecloud-ali-plugin or npm i nodecloud-ali-plugin |
Digital Ocean plugin | yarn add nodecloud-digitalocean-plugin or npm i nodecloud-digitalocean-plugin |
3️⃣ Create the NodeCloud config file
Create the .nc.config.js
file in the project root in the following format.
Content of .nc.config.js file is assumed as the following structure. It is an array of supported providers.
name
: Provider identifier, this can be used to identify the plugin at a glance.tag
: Tag name that will be used to load the given provider internally.plugin
: Plugin moduleconfigPath
: Provider configuration file
This config file can contain an array of objects for all providers and all will be loaded. Supported values for name : aws, azure, alicloud, digitalocean, google
const nodeCloudAwsPlugin = require("@nodecloud/aws-plugin");
const nodeCloudGcpPlugin = require("@nodecloud/gcp-plugin");
const nodeCloudAzurePlugin = require("@nodecloud/azure-plugin");
const providers = [
{
name: "aws",
tag: "aws",
plugin: nodeCloudAwsPlugin,
configPath: "C:\\Users\\Rajitha\\opensource\\aws_cred.json",
},
{
name: "google",
tag: "google",
plugin: nodeCloudGcpPlugin,
configPath: {
projectId: "astral-hold-276807",
keyFilename: "C:\\Users\\Rajitha\\opensource\\gcp_cred.json",
},
},
{
name: "azure",
tag: "azure",
plugin: nodeCloudAzurePlugin,
},
];
module.exports = providers;
4️⃣ Enjoy the awesomeness of NodeCloud
Congratulations! You just configured NodeCloud in your project. Let's start with some cloud interactions.
The below code is an example of usage in AWS.
const nc = require("@nodecloud/common"); // NodeCloud common module
const optionsProvider = {
overrideProviders: false,
};
const ncProviders = nc.getProviders(optionsProvider);
const options = {
apiVersion: "2017-11-01",
};
const computeModule = ncProviders.aws.compute(options);
function launchInstance() {
const instanceParams = {
ImageId: "ami-07ebfd5b3428b6f4d", // Image of Ubuntu Server 18.04 LTS
InstanceType: "t2.micro",
KeyName: "nodeCloud", // key name of Key pair
MinCount: 1,
MaxCount: 1,
};
// create AWS EC2 instance
computeModule
.create(instanceParams)
.then((res) => {
console.log(`All done ! ${res}`);
})
.catch((err) => {
console.log(`Oops something happened ${err}`);
});
}
function stopInstance() {
const params = {
InstanceIds: ["i-0928af5c626f85da9"],
DryRun: false,
};
// stop AWS EC2 instance
computeModule
.stop(params)
.then((res) => {
console.log(res);
})
.catch((err) => {
console.log(err);
});
}
NodeCloud officially supports AWS, GCP, Azure, DigitalOcean and AliCloud. If you want to use a community-driven plugin override the providers' list as follows.
const nodeCloud = require("nodecloud");
const options = {
overrideProviders: true,
};
const ncProviders = nodeCloud.getProviders(options);
Service Category | Service | AWS | GCP | Azure | DigitalOcean | AliCloud |
---|---|---|---|---|---|---|
Compute | IaaS | EC2 | Compute Engine | Virtual Machine | Droplets | ECS |
Faas | AWS Lambda* | Cloud Functions* | Azure Functions* | - | Function Compute* | |
Containers | ECS, EKS | Google Kubernetes Engine | AKS, Azure Service Fabric* | DO Kubernetes* | Container Service*, Container Service for Kubernetes* | |
Containers (without infrastructure) | AWS Fargate* | Cloud Run* | - | - | ECI* | |
Paas | AWS Elastic Beanstalk | App Engine* | App Service | - | Simple Application Server* | |
Storage | Object Storage | S3 | Cloud Storage | Azure Blob Storage | Spaces* | Bucket (OSS) |
Block Storage | EBS | Persistent Disks | Disk Storage | Volumes | NAS* | |
Networking | Load Balancer | ELB | Cloud Load Balancing* | Azure Load Balancer | DO Load Balancer | SLB |
Peering/Dedicated Interconnect | Direct Connect | Cloud Interconnect* | ExpressRoute* | - | Express Connect* | |
DNS | Route53 | Google Domains, Cloud DNS | Azure DNS | DO DNS* | Alibaba Cloud DNS* | |
Databases | RDBMS | RDS, Amazon Aurora*, Amazon Redshift* | Cloud SQL*, Cloud Spanner | SQL Database, Azure Database for MySQL*, Azure Database for PostgreSQL* | Managed Databases(PostgreSQL* and MySQL*) | ApsaraDB (MySQL, MariaDB TX, SQL Server, PostgreSQL) |
NoSQL: key-value | DynamoDB | Cloud Firestore, Cloud Bigtable* | Table Storage | Managed Databases(Redis)* | ApsaraDB for Redis* | |
NoSQL: indexed | Amazon SimpleDB* | Cloud Firestore | Cosmos DB | - | ApsaraDB for MongoDB* | |
Security/ Authorization | Identity Access Management | AWS IAM | Cloud IAM* | Azure Active Directory*, Azure Role Based Access Control* | - | Resource Access Management* |
*yet to be implemented
NodeCloud relies on the passionate members of its community to keep delivering impactful tools to people all over the world. Contributions of any kind are welcome!
You can help us in many ways. Including new features, bug fixing, error reporting and documentation. Before contributing, be sure to consult NodeClouds's contribution guidelines. As a member of our community, you must abide by our Code Of Conduct.
1️⃣ Fork the leopardslab/nodecloud repository
Follow these instructions on how to fork a repository
2️⃣ Cloning the repository
Once you have set up your fork of the leopardslab/nodecloud
repository, you'll want to clone it to your local machine. This is so you can make and test all of your personal edits before adding it to the master version of leopardslab/nodecloud
.
Navigate to the location on your computer where you want to host your code. Once in the appropriate folder, run the following command to clone the repository to your local machine.
git clone https://github.com/your-username/nodecloud.git
3️⃣ Bootstrapping the repository
You'll then want to navigate within the folder that was just created that contains all of the content of the forked repository. There you'll want to run the installation script to get the updated version of all the third-party dependencies.
cd nodecloud
yarn
❇️ This project is based on Lerna and Yarn workspaces where there are multiple projects in the same repository. It's better to get some background knowledge about these before making any changes in the code. Check the lerna.json
for the current configuration.
❇️ @nodecloud/aws-plugin
, @nodecloud/gcp-plugin
, @nodecloud/azure-plugin
and @nodecloud/common
are Lerna
managed yarn workspaces where the generator
is another standalone yarn workspace.
❇️ Never use NPM client to install third-party dependencies. This project does not contain a package-lock.json
instead it contains a yarn.lock
file. Using npm i
to install NPM modules will harm the project structure.
❇️ Use lerna
commands when installing third-party libraries in Lerna managed yarn workspaces. For example, the below command will add the @google-cloud/translate
npm module to the project @nodecloud/gcp-plugin
.
lerna add @google-cloud/translate --scope=@nodecloud/gcp-plugin
❇️ Adding a common dependency to ALL packages
lerna add the-dep-name
❇️ Adding Dev dependencies
If you have common dev dependencies, it’s better to specify them in the workspace root package.json. For instance, this can be dependencies like Jest, Husky, Storybook, Eslint, Prettier, etc.
yarn add husky --dev -W
Adding the -W flag makes it explicit that we’re adding the dependency to the workspace root.
❇️ Removing Dependencies
lerna exec -- yarn remove dep-name
❇️ Use symlinks when you want to make changes while testing them out in a demo project.** Do not use npm link** instead use yarn linking
yarn link
You should see output like:
success Registered "@nodecloud/gcp-plugin". info You can now run yarn link "@nodecloud/gcp-plugin"
in
the projects where you want to use this module and it will be used instead.
Now that our package is symlinked, run the following command to link the package to your project.
yarn link @nodecloud/gcp-plugin
❇️ The API documentation is generated using JSDoc. Be mindful when making any changes to the auto-generated comments in JavaScript classes. If you have generated new classes run the following command to generate the API documentation.
yarn doc or `jsdoc -c jsdoc.json`
❇️ The JSDoc configuration can be found in the jsdoc.json
file. Currently, it is configured with the better-docs template.
❇️ The code generation component ( generator
yarn workspace) is developed using the TypeScript compiler API. It is better to have an understanding of Abstract Syntax trees, Parsers, and transformers when making code changes in the generator
.
❇️ Please run prettier before committing code to put it in a nice looking format.
lerna run prettier or yarn pretty-quick
Mocha
and Chai
is configured to run unit tests in this project. When you start making changes to the code on your local branch, you'll need to test those changes. Before your code can be accepted into the master branch, it will have to pass all of the tests in yarn workspaces. To test changes, run the following commands:
1️⃣ Save Current Changes
When you get to a point when you want to test the functionality of the code, make sure all your changes are saved. They don't necessarily have to be committed changes in order to test them.
2️⃣ Test changes
To make sure that the application is properly updated, run the tests. If you add elements that do not have tests to prove whether they work correctly or not, please include them in your pull request. When you run the below test command Lerna
will execute all the tests in the packages of this monorepo.
yarn test
This is where the magic happens✨. We don't code any JavaScript classes in NodeCloud plugins, instead, they are automatically generated by a code generation tool. This automating process is currently working for AWS, Azure and Google Cloud. Once you add a service in node-cloud.yml
in the required format for the generator, run tsc main && node main
to generate the service. You will immediately see the output in generatedClasses
directory inside the generator
. More information about the generation tool can be found in the ReadME
of generator
workspace.
MIT @ leopardslab