What are people finding as a best practice for folder structure in your apollo server?
E.g. I'm wondering if it would be a good idea to structure things like so:
src/
enum/
input/
interface/
mock/
mutation/
query/
resolver/
scalar/
subscription/
type/
server.ts
Then I was thinking I could merge everything together like so:
import http from "http";
import express from "express";
import { ApolloServer } from "apollo-server-express";
import { makeExecutableSchema } from "graphql-tools";
import { mergeResolvers, mergeTypes } from "merge-graphql-schemas";
import path from "path";
import glob from "glob";
import fs from "fs";
(async () => {
// TypeDefs
const typeDefs = glob
.sync(path.join(__dirname, "./**/*.graphql"))
.map(f => fs.readFileSync(f, { encoding: "utf8" }));
// Resolvers
const resolvers = await Promise.all(
glob
.sync(path.join(__dirname, "./**/*.resolver.ts"))
.map(async f => (await import(f)).resolver)
);
// Mocks
const mocks = await Promise.all(
glob
.sync(path.join(__dirname, "./**/*.mock.ts"))
.map(async f => (await import(f)).mock)
);
const schema = makeExecutableSchema({
resolvers: mergeResolvers(resolvers),
resolverValidationOptions: {
requireResolversForResolveType: false
},
typeDefs: mergeTypes(typeDefs)
});
const PORT = 4000;
const app = express();
const server = new ApolloServer({
mocks: Object.assign({}, ...mocks),
schema
});
server.applyMiddleware({ app });
const httpServer = http.createServer(app);
server.installSubscriptionHandlers(httpServer);
httpServer.listen(PORT, () => {
console.log(
`π Server ready at http://localhost:${PORT}${server.graphqlPath}`
);
console.log(
`π Subscriptions ready at ws://localhost:${PORT}${
server.subscriptionsPath
}`
);
});
})();
What do you think?
Top comments (3)
I'd say the best architecture is using Nest with their GraphQL plugin
i second this , been using it , quite a savior
github.com/daniele-zurico/graphql-...