λ# - GraphQL with AWS's AppSync

Congratulations! Welcome to Lambda Sharp, LLC! A fictitious company that you just started working at! AppSync will be the focus of your work.

SUBJECT: Data project that must be completed!

FROM: Dave aka DBA <mailto:[email protected]>

We just acquirered Northwind Traders! We need an API ASAP and your the only developer available right now. I've created a partial GraphQL schema and a (not so pretty) mock datasource lambda function to start querying the data. We need this in production TONIGHT!

Thanks,

Dave aka "The DBA"

Level 0 - Setup

• Create a new lambda function from this repo. northwindTradersMockDb.js
  1. Navigate to your Lambda Console and click Create function.
  2. Keep the default Author from scratch selected.
  3. Name your function.
  4. Use the Runtime (NodeJs 10.x).
  5. Set a timeout of 10 seconds.
  6. Select the role you created from the Existing Role dropdown.
  7. Click Create function to create function
• Create a new api in AppSync
  1. Click Build from Scratch option with the name HackathonApp.
  2. Click Edit Schema. Copy/paste schema.graphql contents in to the Schema window. Save.
• Create new Data Source in AppSync
  1. Data source name: NorthwindTraders.
  2. Data source type: AWS Lambda Function.
  3. Region: sames as the lambda function created previously (likely US-EAST-1)
  4. Data source: Reference the lambda function created previously.
  5. Use New role option.

Helpful hint: Be sure you can access the CloudWatch trail events for this lambda function.

Level 1 - Query

AWS Resources

• Quick Start
• Configure Resolvers
• Lambda Resolvers

• Add a resolver to get all customers. In Schema, right pane (no scroll bar but it will scroll) > Query > allCustomers. Use the lambda function NorthwindTraders > Click Attach. Add the request mapping template.

{
  "version": "2017-02-28",
  "operation": "Invoke",
  "payload": {
    "field": "allCustomers"
  }
}

• Add a resolver to get a single customer. In Schema, right pane (no scroll bar but it will scroll) > Query > getCustomer. Use the lambda function NorthwindTraders > Click Attach. Add the request mapping template.

{
  "version": "2017-02-28",
  "operation": "Invoke",
  "payload": {
    "field": "getCustomer",
    "arguments": $util.toJson($context.arguments)
  }
}

• In Queries, query for customers.

query {
  allCustomers {
    CustomerID
    ContactName
  }
}

Level 2 - Query Related data

• Get a single customer order using getCustomer

query {
  getCustomer(id: "ADD A CUSTOMER ID HERE") {
    __typename
    ContactName
    ContactTitle
    Country
  }
}

• Add a resolver to get orders of a single customer. In Schema, right pane (no scroll bar but it will scroll) > Customer > getCustomerOrders. Use the lambda function NorthwindTraders > Click Attach. Add the request mapping template.

{
  "version": "2017-02-28",
  "operation": "Invoke",
  "payload": {
    "field": "getCustomerOrders",
    "source": $utils.toJson($context.source)
  }
}

• Get a single customer's orders

query {
  getCustomer(id: "ANTON") {
    ContactName
    getCustomerOrders {
      # lambda invocation
      __typename
      OrderID
    }
  }
}

Level 3 - Schema

AWS Resource - Schema design

Based on the previous levels, build the following:

• Add Employee type in the schema.
• Add a resolver. data source field getEmployee can be resolved.
• An Employee type is connected to an Order type.
• Construct a query that will show the employee's name on a customer's order.

Example of employee data

{
  "EmployeeID": 9,
  "LastName": "Dodsworth",
  "FirstName": "Anne",
  "Title": "Sales Representative",
  "TitleOfCourtesy": "Ms.",
  "BirthDate": "1966-01-27 00:00:00.000",
  "HireDate": "1994-11-15 00:00:00.000",
  "Address": "7 Houndstooth Rd.",
  "City": "London",
  "Region": "NULL",
  "PostalCode": "WG2 7LT",
  "Country": "UK",
  "HomePhone": "(71) 555-4444",
  "Extension": 452,
  "Photo": "",
  "Notes": "Anne has a BA degree in English from St. Lawrence College.  She is fluent in French and German.",
  "ReportsTo": 5,
  "PhotoPath": "http://accweb/emmployees/davolio.bmp"
}

Level 4 - Mutations

Helpful Resources

* [AWS Resource](https://docs.aws.amazon.com/appsync/latest/devguide/designing-your-schema.html#adding-a-mutation-type) * [Quick start writing queries](https://docs.aws.amazon.com/appsync/latest/devguide/quickstart-write-queries.html)

• Add a mutation type for updating an order.
• Resolve to lambda function method updateOrder
• Write a mutation in the query window to update the ShippedDate field.

type Mutation {
  updateOrder(OrderID: Int, ShippedDate: String): Order
}

Add mutation: Mutation to the type Schema. Construct an update order in the Queries. This resource is helpful: AWS Run Queries and Mutations

You know you did it right when you send a new ShippedDate and returns with the order object as expected.

Boss

• Add another data source for AppSync (DynamoDB is a good option).
• Extend a type in the schema to use the new data source.
• Build a query and show us customer data with new information!

Suggested Data Sources

• JSON Currency Exchange
• Google Trends

Additional Resources

• Resolver Context Reference
• Building a client
• Common Mistakes

Demo Time!

Show us your Schema and run some queries in front of us!

Data Sources

https://github.com/tmcnab/northwind-mongo

https://www.csvjson.com/csv2json