Starter kit for apollo server using webpack and typescript
1. exported schema as example for GraphQL Schema
2. Working Apollo Server (webpack + tslint + tsloader)
3. Typescript 2.0.0 => ES6
4. Dockerfile to make the apollo-server a container.
5. unit testing (mocha-webpack+chai) + coverage report (mocha-istanbul-spec+istanbul).
6. working with graphql-tools
7. standard-version for auto SemVer.
Please note that you will need to rename the library name in some files:
1. package.json (ofcourse ;))
npm run build - build the library files (Required for start:watch)
npm run build:watch - build the library files in watchmode (Useful for development)
npm start - Start the server
npm run start:watch - Start the server in watchmode (Useful for development)
npm test - run tests once
npm run test:watch - run tests in watchmode (Useful for development)
npm run test:growl - run tests in watchmode with growl notification (even more useful for development)
npm run upver - runs standard-version to update the server version.
npm start
1. src - directory is used for typescript code that is part of the project
1a. main.ts - Main server file. (Starting Apollo server)
1b. main.spec.ts - Tests file for main
1c. schema - Module used to build schema
- index.ts - simple logic to merge all modules into a schema using graphql-tools
- modules/ - directory for modules to be used with graphql-tools
1c. schema.spec.ts - Basic test for schema.
1c. main.test.ts - Main for tests runner.
3. package.json - file is used to describe the library
4. tsconfig.json - configuration file for the library compilation
6. tslint.json - configuration file for the linter
7. typings.json - typings needed for the server
8. webpack.config.js - configuration file of the compilation automation process for the library
9. webpack.config.test.js - configuration file of the compilation when testing
10. Dockerfile - Dockerfile used to describe how to make a container out of apollo server
11. mocha-webpack.opts - Options file for mocha-webpack
1. node_modules - directory npm creates with all the dependencies of the module (result of npm install)
2. dist - directory contains the compiled server (javascript)
3. html-report - output of npm test, code coverage html report.
The person type was added to demonstrate a database like access, parametrized queries, resolvers and drill down. The data is currently hard coded but simulates a storage. Each person has an id, name and sex. It also has a dynamic field called matches. For demonstration purposes, this field will retrieve all members of the other sex by using a resolver.
Since this is a computed field the query can be infinitely nested, for example, try in the graphiql editor this query:
{
getPerson(id: "1") {
id,
name
sex
matches {
id
name
sex
matches {
id
name
sex
matches {
id
name
sex
}
}
}
}
}
It will return a nested, alternating male/femal results.
To list all persons, use the persons
query:
{
persons {
id
name
}
}
There is also an example of a mutation - addPerson(name: String, sex: String)
, to use it:
mutation {
addPerson(name: "kuku", sex: "male"){
id
name
}
}
Note that the query generates a random id and that the added persons are transient, i.e. not persisted and will be gone once you shut down the server.