Building a GAE+GWT application using the best practices (Part 1)

Building a GAE+GWT application using the best practices series

1. Part 1
2. Part 2
3. Part 3
4. Part 4
5. Part 5

This is the first installment of the series Building a GAE+GWT application using the best practices.

Introduction

In the next few blog posts, I’m going to present my experience building a simple (but non-trivial) web application using GWT and Google App Engine, while applying the best practices introduced by Ray Ryan in his excellent GWT best practices at Google IO last year.

The application I’m going to build is called RateChecker. It’s simply a tool that goes fetches the posted USD/CAD exchange rate from a bank website, persists the rate to the data store, and present the user with the recent rates when requested. It can potentially do more (like generating a histogram of the rates, etc), but for the purpose of blog series, I’m going with the basics to illustrate the implementation pattern without losing my bearings in the embellishing.

Here’s a screenshot of the finished application:

The Goals

I intend to use this minimal application to practice using the following technologies/techniques: * Server/Persistence tier: * Google Guice as dependency injection container on the server side * JDO as the persistence layer * AppEngine Cron task * Command pattern for request handling * Client tier: * Google GIN as dependency injection container on the client side * Model-View-Presenter pattern on the client side * EventBus to decouple components * Command pattern for remote service calls * Unit Testing on both Client/Server side with mock objects

Also I plan to rewrite the service layer with one of the alternate JVM language, such as Scala or Clojure, if I my schedule allows and I have enough motivation.