Project Stage B - Data Structures

You can "discover" data items and groups by looking at the nouns in your user-stories. Consider the Fair Trade example:

Fair Trade is an organisation that sells products from partner organisations in Developing Countries. It attempts to give the suppliers a price that will enable them to live a decent, if poor, life rather than exploiting them as some multi-national clothing companies have notoriously done.

Here is the same "story" with all the nouns highlighted:

Fair Trade is an organisation that sells products from partner organisations in Developing Countries. It attempts to give the suppliers a price that will enable them to live a decent, if poor, life rather than exploiting them as some multi-national clothing companies have notoriously done.

You won't need variables for every noun, but the nouns give you some ideas for variables:

price product supplier organisation country

If you were actually writing a project for this scenario, you would have already done Stage A. But you can still reflect on some ideas:

An organisation might need an account so they can log-in and search for suppliers
A supplier will exist in a specific country (or countries in a global economy)
A supplier might have a LIST of products that they sell
Each product will have a name and a price
A product might have several prices depending on the size of the order (discounts for larger orders)
etc

Looking at the thoughts above, we see that each item has a name and connections/relationships to other data items.

Enter OOP

In OOP, we try to group together connected ideas. Compare an old-fashioned program to an OOP program:

Traditional Approach

OOP

String organization = "Starbucks";
String password = "starry night";
String supplier = "Coffee Growers Union";
String country = "Honduras";
String product = "coffee";
double price = 3.50;

// ******************************

// What if the supplier has a list of
// various products that they sell?

// What if we want a list of all the suppliers
// who sell coffee?

// Traditional programming methods
// make it difficult to create lists.

class Organization
{ String name;
    String password;
}

class Supplier
{ String name;
   String country;
   Product product;
}

class Product
{ String name;
double price;
}

//*** Now making lists is easy *****************

Organization[] orgs = new Organization[1000];

Supplier[] sups = new Supplier[1000];

//*** We should rewrite the Supplier class *********

class Supplier
{ String name;
String country;
Product[] product = new Product[1000];
}

//*** Not all lists should be 1000 items,
//*** but that will probably work for most projects.

The OOP version requires some extra code ( the class command ), but is a lot more flexible and saves work in the long run.

Discovering Data for Your Project

Look back at the user-stories and sample data you collected in stage A or your project. Then do a similar analysis to the one above. Make sure you find examples for all the following :

a String variable
a numeric variable
several data items that are connected
some data item that needs to be stored in a list
a variable that needs a class to store several related items
sample data for each item, including at least 3 samples for any lists and a complete sample for any class