What's new in 2021?
A lot. 🎉
On Thursday, January 14th, 2021, Neo4j had a big announcement in their GraphQL Community Call.
As part of the initial GRANDstack (GraphQL, React, Apollo, Neo4j Database) offering, Neo4j labs had created https://github.com/neo4j-graphql/neo4j-graphql-js - enabling developers to build a GraphQL API layer on top of an already existing Neo4j database and the Cypher query language.
Neo4j has announced this effort is graduating from a labs project into a fully supported Neo4j product for the GraphQL layer - @neo4j/graphql
This was a huge announcement; and I would encourage you to review the community call and materials below.
@neo4j/graphql is currently in early alpha release and is available NOW.
Useful resources
- Neo4j GraphQL Community Call - January 2021
- Slide deck for the Community Call
- @neo4j/graphql on npm
- https://github.com/neo4j/graphql-tracker-temp - This is the temporary repo where you can submit issues, etc.
- Neo4j on Slack
This project is a starter for building a GRANDstack (GraphQL, React, Apollo, Neo4j Database) application using Next.js and TypeScript instead of the original create-react-app example available at https://github.com/grand-stack/grand-stack-starter.
To run this application as intended, you will need to:
- Build and run the Dockerized project
- Seed your Neo4j database with sample data
A demo application has been deployed using a free Neo4j Sandbox as the backend database at [https://nextjs-grandstack-starter-typescript.vercel.app/]
Please note that these sandboxes do not last forever - ten (10) days at most with active renewal and upkeep.
TL:DR If the demo app is broken, that's why. You can run this locally to your heart's content. 🤣
If you would like to have your Next.js application and Neo4j Database running in a Docker environment, you can quickly build, start, and stop versions of Neo4j Database to your heart's content!
To run this example, all you need to have installed on your system is Docker and npm installed on your development system.
If you do not have Docker installed on your development system, go to freely available Docker Desktop and get that installed and configured on your development machine.
If you already have npm and Docker installed on your development system, run:
# Run the project using Neo4j v3.5.x
$ npm run dev
# ALTERNATIVE: Run the project using Neo4j v4.x.x
$ npm run dev:v4You should be able to access the following URLs:
- http://localhost:7474/browser/ - This is the Neo4j Browser application that you can use to explore your Neo4j database - as well as run Cypher commands to seed your database with example data
- http://localhost:3000 - The frontend for our Next.js application
- http://localhost:3000/api/graphql - The GraphIQL explorer for our backend Next.js API which will be a serverless GraphQL function on Vercel
- http://localhost:3000/api/ping - A sample API route that will be a serverless function on Vercel
Additionally, the following scripts have been created for managing your Docker environment based on the version of Neo4j you are using.
Neo4j v3.5.x:
dev- This starts the Dockerized project with services defined indocker-compose.ymldev:clean- This builds fresh images and containers for the Dockerized project with services defined indocker-compose.ymldev:stop- This stops the Dockerized project with services defined indocker-compose.yml
Neo4j v4.x.x:
dev:v4- This starts the Dockerized project with services defined indocker-compose-neo4j-v4.x.ymldev:v4:clean- This builds fresh images and containers for the Dockerized project with services defined indocker-compose-neo4j-v4.x.ymldev:v4:stop- This stops the Dockerized project with services defined indocker-compose-neo4j-v4.x.yml
Docker
docker:destroy:global- WARNING: This command will delete all Docker images, containers, networks, and volumes for ALL Dockerized applications on your system
Once your Dockerized project is running, you can navigate to the Neo4j Browser at http://0.0.0.0:7474/browser/.
Open up ./app/neo4j/__seed__/db.cypher so you can copy and paste the example Cypher statements into the Cypher window and press play to seed your database with example data.
This project uses Jest and Enzyme for unit and integration tests.
The following scripts are available at both the top-level project directory as well as the app directory:
npm run testornpm test- This runs a single run of the Jest tests for our project.npm run test:ci- Useful for running Jest tests in a continuous integration (CI) environment- See https://jestjs.io/docs/en/cli#--ci for more details
npm run test:coverage- Generates a code coverage report of the Jest tests
npm run test:coverage:view- Generates a code coverage report of the Jest tests and automatically launches a web browser on macOS/Linux to interactively see what code is and is not being covered in tests
-
npm run test:verbose- This displays verbose output during the execution of the Jest tests -
npm run test:watch- This is the only test script that runs locally. It runs Jest in--watchmode; running tests related to current code changes and not the entire suite
This project uses the Next.js + Webpack Bundle Analyzer to generate bundle analysis of both client and server bundles by running the npm run analyze script:
The public folder in your Next.js app can be used to serve whatever assets you desire (e.g. app/public/logo.png would be available at http://localhost:3000/logo.png)
One benefit of developing with Next.js and deploying to Vercel is that any files contained within your pages/api folder will be separate serverless functions.
In our case, we have two serverless functions that we are exposing:
graphql- This is the GraphQL endpoint our application and our GraphIQL IDE will useping- Displays a simple message to verify our serverless functions are online
For more details, please refer to Vercel Serverless Functions
You can explore your GraphQL schema using the GraphIQL IDE - available at http://localhost:3000/api/graphql
To verify your back-end API is running, you should be able to visit http://localhost:3000/api/ping and see a response.
This project has an example Apollo GraphQL server up and running at /api/graphql
If you are running this project locally, you can view the GraphIQL IDE at http://localhost:3000/api/graphql
This project uses Google Analytics to track user interactions and evaluate the application's usage.
The Neo4j Database is the cornerstone of our GRANDstack application.
Once you have defined the environment variables - either in app/.env for local development or the appropriate places for your environment - you can have a Neo4j Database hosted anywhere you'd like.
Neo4j has three options to get you up and running quickly:
- Neo4j Desktop - FREE - Perfect for exploring Neo4j on your local machine
- Neo4j Sandbox - FREE - This allows you to run short-lived Neo4j projects in the cloud for free. By default, Neo4j sandboxes will terminate after three (3) days; however you do have the ability to optionally extend for another seven (7) days if desired.
- Neo4j Aura - $ - This is Neo4j's flagship offering with a developer-friendly graph database as a service.
All you will need is the bolt URL and the credentials to access your database.
The original https://github.com/grand-stack/grand-stack-starter had a helper script to seed your Neo4j Database with example data. That same example database is available for you in a series of Cypher commands.
Once you have connected to your Neo4j Database with Neo4j Browser, simply copy the text as is from app/neo4j/__seed__/db.cypher and execute the Cypher commands in a single action:
You can verify that your Neo4j Database has been successfully created:
This project is ready to be configured for deployment to Vercel, as well as optionally using the Vercel for GitHub integration for automatic deployment.
We can define the environment variables we would like to use for our Production, Preview, and Development environments. Navigate to your Vercel project settings and define the following environment variables:
GOOGLE_ANALYTICS_TRACKING_ID=UA-156456153-7
NEO4J_URI=bolt://54.167.150.120:32844
NEO4J_USER=neo4j
NEO4J_PASSWORD=
# Enable encrypted driver connection for Neo4j
#NEO4J_ENCRYPTED=true
# Specify a specific Neo4j database (v4.x+ only)
#NEO4J_DATABASE=neo4jPlease refer to app/.env.sample to see the latest environment variables that you will need to declare if the above list is incomplete.
Please see https://vercel.com/docs/v2/build-step#environment-variables and/or https://nextjs.org/docs/basic-features/environment-variables for details on defining environment variables for your application on Vercel.