scaffoldingAPI

This project was created with the intention of facilitating the creation of API's using technologies like GraphQL with ApolloServer.

Projetos projects used:

First steps

Download the repository for your machine:

git clone https://github.com/isrmic/scaffoldingAPI myproj
cd myproj
npm install
npm run start

After performing these steps the server should be ready to run at http: // localhost: 3000 /.

At first the playground is enabled but can be disabled later.

Defining schemas

All schemas will remain inside the "schemas" folder having the separation of the files for definition of types and queries in Query.gql and types and mutations in Mutation.gql.

Defining resolvers

In this project so that everything is organized are made some separations regarding the location of each file, for example we have the resolvers (Query, Mutation, Default Resolver) that are separated by folders and identified by the filename according to what was defined in the schematic.

Example: // schema/Query.gql

#query example
type Query {
    example: [Example]
}
#type example for test
type Example {
    field1: String
    field2: String
    field3: String
}

// resolvers/Query/example.js

const example = () => async (root, args, context, info) => {
	const data = [
		{
			field1: 'this is a field1',
			field2: 'this is a field2',
		},
		{
			field1: 'this is a field1 of object 2',
			field2: 'this is a field2 of object 2',
		}
	];

	return data;
};

module.exports = example;

The "example.js" file located in "resolvers / Query / example.js" will automatically be understood as resolving to the "example" query defined in the schema.

Resolvers have this pattern in their writing:

const test = () => async (root, args, context, info) => {
    //definition of resolver
};

module.exports = test;

A function that returns another asynchronous or synchronous function (function, arow function) is exported, and this function that is returned by the test function is the resolver that will return the data as requested. NOTE: this will work for both querys and mutations, the difference is even in the organization by the separation of the files.

Resolvers can also be created from a simple command:

npm run generate <query || mutation> <nameresolver>

Tests

You can do a test of how everything works before even starting the server for tests in the playground, if you want to test the example already you can execute the following command:

npm run test

This way it will display the results returned by resolving example.js.

Result:

 { example:
   [ { field1: 'this is a field1',
       field2: 'this is a field2',
       field3: 'this is a field 3' },
     { field1: 'this is a field1 of object 2',
       field2: 'this is a field2 of object 2',
       field3: 'this is a field 3' } ] }

If you want to take advantage of the features of graphql as a selection of fields and subfields you can pass a string next to the test command that would be the query made to the API.

npm run test "{ example { field1 field2 } }"

You should receive a result similar to this:

Result:

 { example:
   [ { field1: 'this is a field1', field2: 'this is a field2' },
     { field1: 'this is a field1 of object 2',
       field2: 'this is a field2 of object 2' } ] }

License