Designing Tests with ColumnFixture

Think of each row of a ColumnFixture as one or more function calls. The values of the named variables are set, and then the functions are called.

For example, lets say you were writing an inventory system, and you wanted to test its ability to keep track of stocked items. You might write tables like this:

inventory.AddStockFixture
partNumber	quantity	valid?
324	28	true
586	14	true

This test adds two stock items. The valid() function calls the parts of the system that cause a stock item to be added. The code for the AddStockFixture might look like this:

package inventory;
import fit.ColumnFixture;
public class AddStockFixture extends ColumnFixture {
  public int partNumber;
  public int quantity;
  public boolean valid() {
    try {
      StockItem item = new StockItem(partNumber);
      Inventory.addStockItem(item,quantity);
      return true;
    } catch (Exception e) {
      return false;
    }
  }
}

Next, we might write a table that simulates some purchases that reduce the stock of an item.

inventory.PurchaseItemFixture
partNumber	quantity	valid?
324	3	true
586	2	true

It should be clear how this fixture would be written.

Finally we'd write a table that checked to make sure that the inventory had been properly updated.

inventory.CheckInventoryFixture
partNumber	quantity()
324	25
586	12

See how easy this is?