Hypergraph Express Middleware

Fastest way to create Graphql Server with Express

Installation

• yarn add @gql-tools/hypergraph-express-middleware

Usage

middleware

• Add Middleware

import { middleware } from ' @gql-tools/hypergraph-express-middleware';

const context = { dbConnection, otherGlobals };
app.use('/graphql', middleware(context));

• Now place your Resolvers in graphs/resolvers.js.

• And your types in graphs/types.

• You can change default locations by defining new paths in .hypergraphrc in your project root.

{
  "types": "some/other/graphs",
  "resolvers": "src/other/resolvers"
}

playground

import { playground } from ' @gql-tools/hypergraph-express-middleware';

app.use('/graphiql', playground(/* graphql route*/ '/graphql'));

merge

• Merges two graphql resolvers, useful if you want to divide resolver to different files

Why Hypergraph? or Hot-ish Reloading of Resolvers

• Hypergraph seperates your main node application with the graphql resolvers. You can have separate Dev Servers for your resolvers and Node Server so whenever there are changes in Resolvers only the resolver server is reloaded.

• Hypergraph does cache invalidation during development in order to achieve hot-ish reloading for Resolvers, so it relies on NODE_ENV environment variable. Server reloading will not work if there's no NODE_ENV var present.

• Hot Reloading can be achieved with any server, nodemon or webpack-dev-server or both(Required that only one does reloading).

eg with Nodemon

const resolverReload = nodemon({
  script: 'path to root resolver',
  watch: ['root resolver', 'types dir if you want'],
  nodeArgs: process.argv.slice(2),
}).on('quit', process.exit);

const expressReload = nodemon({
  script: 'path to express app',
  watch: ['express dir'],
  nodeArgs: process.argv.slice(2),
}).on('quit', process.exit);