Nested Data Structures: Printing and Coalescing Data

Learning Goals

• Display the cells in an Array of Arrays
• Traverse Array of Arrays to produce a single value
• Traverse Array of Arrays to produce a new nested data structure

Introduction

When we started this module, we mentioned that we often use nested data structures as a "base" from which to do data processing. Arrays of Arrays is our first milestone in learning to work with nested data.

We're going to pick out three specific types of processing here:

1. Displaying the nested structure
2. Transforming the nested structure into a new structure (a collection)
3. Transforming the nested structure into a result (a single thing usually a number)

In each of these sections, we're going to give you a bit of "reference code." Reference code is code that's a little generic and probably doesn't solve any real problem, but which is kept intentionally very simple so that you can see how it might be adapted to your particular need. Some people call these "reference implementations."

Display the Cells in an Array of Arrays

It's not often we say this, but this following bit of code is worth memorizing. If you want to learn to speak any human language, it's a good idea to learn the grammar, grow your vocabulary, and speak with native speakers as much as possible.

But

That takes years, and, in a pinch, you should really know how to ask for food, water, emergency services, and a bathroom. Even if you don't understand what's happening grammatically, having a few "stock phrases" in times of need is helpful. The following code is a good "stock phrase."

Assuming the following AoA:

spice_rack = [
  ["Posh", "Scary", "Sporty"],
  ["Paprika", "Fajita Mix", "Coriander"],
  ["Parsley", "Sage", "Rosemary"]
]

We display it like:

spice_rack = [
  ["Posh", "Scary", "Sporty"],
  ["Paprika", "Fajita Mix", "Coriander"],
  ["Parsley", "Sage", "Rosemary"]
]

row_index = 0
while row_index < spice_rack.count do
  element_index = 0
  while element_index < spice_rack[row_index].count do
    puts spice_rack[row_index][element_index]
    element_index += 1
  end
  row_index += 1
end #=> nil

This prints out:

Posh
Scary
Sporty
Paprika
Fajita Mix
Coriander
Parsley
Sage
Rosemary

This "formula" can be expanded and changed. Maybe you want something more descriptive than row_index or element_index index. Or, perhaps you need to adjust behavior slightly. That's fine! But if you're not sure how to start, this is always a good start and will get you un-stuck!

Traverse Array of Arrays to Produce a New Nested Data Structure

The same "formula" that we suggested you use above can easily be adapted for doing work other than puts-ing things. Let's say we wanted to gather all the elements that start with P.

spice_rack = [
  ["Posh", "Scary", "Sporty"],
  ["Paprika", "Fajita Mix", "Coriander"],
  ["Parsley", "Sage", "Rosemary"]
]


outer_results = []
row_index = 0
while row_index < spice_rack.count do
  element_index = 0
  inner_results = []
  while element_index < spice_rack[row_index].count do
    # How to read the following line of code:
    #   Array at row_index
    #   Element of the inner array at element_index
    #   The first character of that element...
    if spice_rack[row_index][element_index][0] == "P"
      inner_results << spice_rack[row_index][element_index]
    end
    element_index += 1
  end
  outer_results << inner_results
  row_index += 1
end

outer_results #=> [["Posh"], ["Paprika"], ["Parsley"]]

We might call this filtering. When you learn Ruby's Enumerable methods in just a little while, remembering this name will help a lot!

Traverse Array of Arrays to Produce a Single Value

Another variant on traversing nested Arrays is traversing the matrix and accumulating all the values. Let's imagine that someone created a grid representing a guessing game. In each cell of the grid the games makers have put some amount of cash. Let's sum up all the possible values to determine how much money is in the grid.

guessing_game_grid = [
 [1, 2, 1, 7, 3],
 [2, 100, 15, 4, 18],
 [15, 16, 99, 1, 2, 11]
]

total = 0
row_index = 0
while row_index < guessing_game_grid.count do
  element_index = 0
  while element_index < guessing_game_grid[row_index].count do
    total += guessing_game_grid[row_index][element_index]
    element_index += 1
  end
  row_index += 1
end
total #=> 297

This pattern is known as reduction because you're going to reduce a nested data structure to a single value. As you learn Ruby's Enumerable methods, remembering this name will help a lot!

Conclusion

Congratulations! You've learned three important data processing tasks with matrices and Ruby:

1. Print every element in the grid
2. Gather some of the elements that match a given criterion into a new Array
3. Gather all the elements to create a single value

In the lab, we're going to ask you to adapt these "reference implementations" to do some drills. It's awesome to know that you don't need anything more complex than while loops and assignments to process complex structures into insight!

These drills are designed to help you have the tools you're going to need to work with complex, nested Arrays of Arrays. Building these skills means that when you encounter "Arrays of Hashes of Arrays of Arrays," you'll have trained to meet that challenge.

Keep in mind, you might need to mix-and-match: you might filter some inner Arrays; and, of the ones that passed, you might average their numbers and accumulate those numbers to a new Array or a new total.