MongoDB Pizza Shop

The MongoDB Pizza Shop is a small Sample Application that demonstrates the usage of MongoDB in a Java project using Spring Data.

System Requirements

The following requirements have to be met for setting up and compiling the project:

• Gradle
• Groovy
• Git
• Java
• Mongo DB

If you're on Mac OS, the easiest way to meet all requirements is to use the package manager Homebrew:

1. Install XCode via Apple's App Store.
2. In XCode, install 'Command Line Tools'.
3. Install Homebrew with
    $ ruby -e "$(curl -fsSkL raw.github.com/mxcl/homebrew/go)"
    (see http://mxcl.github.com/homebrew/ for checking if the command is still valid)
4. Install all required packages directly in the shell, e.g. via
    $ brew install git

Alternatively, you can set the following system variables in '.bash_profile' after manually installing Gradle, Groovy, Git, Java, and Mongo DB:

$ export GRADLE_HOME<Your Gradle Home>/gradle-VERSION
$ export MONGO_HOME=<Your Mongo Home>/mongodb-VERSION
$ export GROOVY_HOME=<Your Groovy Home>/groovy-VERSION
$ export JAVA_HOME=<Your Java Home>
$ export CLASSPATH=<Your Java Home>/lib/

$ export PATH=$PATH:$GRADLE_HOME/bin
$ export PATH=$PATH:$MONGO_HOME/bin
$ export PATH=$PATH:$JAVA_HOME/bin
$ export PATH=$PATH:$GROOVY_HOME/bin

Setting up the Project

In the console, select any directory on your computer and clone this repository from github. That's it! The project is ready to be build for your favorite IDE:

$ gradle eclipse
$ gradle idea

Starting the Application

There are two options to start the application:

1. Command "$ gradle jettyRun" in the console for running on port 8080.
2. Run JettyStart.main from the IDE for running on the port specified as a program argument.

For the latter option, make sure to set the working directory to ./shop/ui. In both cases, the application will be available at http://localhost:port/pizza.

Data Generator

On startup, the application automatically checks if the database "pizza" exists and generates users, products, and random orders for the last 3 years. Hence, the easiest way to re-initialize the data is to delete the entire database "pizza" and restart the application.

General Hints

• The Login "admin" / "admin123" can be used for accessing a list of all orders.
• For convenience, a random customer is automatically logged in as soon as the user proceeds to the checkout.
• Any particular customer can be logged in with the password "customer123"

Recommendations

All recommendations except for "You may also like" are randomized in this application. The "You may also like" recommendations are implemented with mahout and its collaborative filtering functionality. To be more precise, a combination of item-based and user-based similarity measures is used to suggest unknown products a certain user is likely to be interested in.

MongoDB Collections

order
    stores all submitted orders
product
    the product catalog
recommendation.mapping
    mapping table used and maintained by mahout
recommendation.preference
    user-item-preference (counts the number of orders for each user-item-pair)
sequence.order
    incremental counter for providing order IDs
user
    stores all registered users