Derive Insight from NDS Using Methods...Repeated

Learning Goals

• Create an "Nth-Order" method

Introduction

In the previous lab we saw how "First-Order" methods make it easier for humans to reason about code.

However, we can create code that is easier to maintain, clearer and more beautiful if we wrap "First-Order" methods inside of other methods. In the same way that we nested "raw" [], Hash, and Array methods inside "First-Order" methods, we can nest "First-Order" methods inside other methods. We're going to call those Nth-Order methods.

"Make Breakfast" is an Nth-Order method in Real Life. It contains "Make toast" and "Make coffee." Each of those have multiple methods like "Get Bread. Toast Bread. Apply butter to toasted bread. Etc."

We brought in this idea of "Order" as a teaching to help encourage you to see how to "wrap" methods. The technique is what you should take away, not the label. Wrapping low-level methods with nicely-named, abstract methods is a good thing.

When code is structured this way, there are many benefits:

• It's easier to debug
• It's more obvious to read
• It's more easy to get help with online

Starting Point

First, let's start with some "starter" code.

vm = [[[{:name=>"Vanilla Cookies", :price=>3}, {:name=>"Pistachio Cookies", :price=>3}, {:name=>"Chocolate Cookies", :price=>3}, {:name=>"Chocolate Chip Cookies", :price=>3}], [{:name=>"Tooth-Melters", :price=>12}, {:name=>"Tooth-Destroyers", :price=>12}, {:name=>"Enamel Eaters", :price=>12}, {:name=>"Dentist's Nightmare", :price=>20}], [{:name=>"Gummy Sour Apple", :price=>3}, {:name=>"Gummy Apple", :price=>5}, {:name=>"Gummy Moldy Apple", :price=>1}]], [[{:name=>"Grape Drink", :price=>1}, {:name=>"Orange Drink", :price=>1}, {:name=>"Pineapple Drink", :price=>1}], [{:name=>"Mints", :price=>13}, {:name=>"Curiously Toxic Mints", :price=>1000}, {:name=>"US Mints", :price=>99}]]]

def total_value_of_spinner(nds, row_index, column_index)
  coordinate_total = 0
  inner_len = nds[row_index][column_index].length
  inner_index = 0
  while inner_index < inner_len do
    coordinate_total += nds[row_index][column_index][inner_index][:price]
    inner_index += 1
  end
  coordinate_total
end

# Main code

grand_total = 0
row_index = 0
while row_index < vm.length do
  column_index = 0
  while column_index < vm[row_index].length do
    grand_total += total_value_of_spinner(vm, row_index, column_index)
    column_index += 1
  end
  row_index += 1
end

# End Main code

p grand_total #=> 1192

Create an "Nth-Order" method

Looking at the starter code, the main loop is still not so easy to reason about. Let's zoom in to it:

# Non-runnable
grand_total = 0
row_index = 0
while row_index < vm.length do
  column_index = 0
  while column_index < vm[row_index].length do
    grand_total += total_value_of_spinner(vm, row_index, column_index)
    column_index += 1
  end
  row_index += 1
end

That inner-most while means "go to each spinner on the row, and total it up". That is, "total up the row based on its spinners by use of total_value_of_spinner" Wouldn't it be GREAT if there were a method called total_snack_value_of_row? Again, there's nothing stopping us from making it! All we have to do to total a row's snack count is add together the snack counts of its spinners.

Let's update the code:

vm = [[[{:name=>"Vanilla Cookies", :price=>3}, {:name=>"Pistachio Cookies", :price=>3}, {:name=>"Chocolate Cookies", :price=>3}, {:name=>"Chocolate Chip Cookies", :price=>3}], [{:name=>"Tooth-Melters", :price=>12}, {:name=>"Tooth-Destroyers", :price=>12}, {:name=>"Enamel Eaters", :price=>12}, {:name=>"Dentist's Nighmare", :price=>20}], [{:name=>"Gummy Sour Apple", :price=>3}, {:name=>"Gummy Apple", :price=>5}, {:name=>"Gummy Moldy Apple", :price=>1}]], [[{:name=>"Grape Drink", :price=>1}, {:name=>"Orange Drink", :price=>1}, {:name=>"Pineapple Drink", :price=>1}], [{:name=>"Mints", :price=>13}, {:name=>"Curiously Toxic Mints", :price=>1000}, {:name=>"US Mints", :price=>99}]]]

def total_snack_price_on_spinner(nds, row_index, column_index)
  coordinate_total = 0
  inner_len = nds[row_index][column_index].length
  inner_index = 0
  while inner_index < inner_len do
    coordinate_total += nds[row_index][column_index][inner_index][:price]
    inner_index += 1
  end
  coordinate_total
end

def total_snack_value_of_row(nds, row_index)
  grand_row_total = 0
  column_index = 0
  while column_index < nds[row_index].length do
    grand_row_total += total_snack_price_on_spinner(nds, row_index, column_index)
    column_index += 1
  end
  grand_row_total
end

grand_total = 0
row_index = 0
while row_index < vm.length do
  grand_total += total_snack_value_of_row(vm, row_index)
  row_index += 1
end

p grand_total #=> 1192

Wrap an Additional Method

Suddenly all that's left in our original kinda-messy "raw" implementation is the following:

# Non-Runnable

grand_total = 0
row_index = 0
while row_index < vm.length do
  grand_total += total_snack_value_of_row(vm, row_index)
  row_index += 1
end

But this bit of code has a purpose. What does this code mean? What is it trying to accomplish?

Isn't it "sum up the values of all the rows" the same thing as seeking "total_value_of_snacks_in_machine?" Just as before, we're going to "wrap" total_snack_value_of_row into a new method: total_value_of_snacks_in_machine.

vm = [[[{:name=>"Vanilla Cookies", :price=>3}, {:name=>"Pistachio Cookies", :price=>3}, {:name=>"Chocolate Cookies", :price=>3}, {:name=>"Chocolate Chip Cookies", :price=>3}], [{:name=>"Tooth-Melters", :price=>12}, {:name=>"Tooth-Destroyers", :price=>12}, {:name=>"Enamel Eaters", :price=>12}, {:name=>"Dentist's Nighmare", :price=>20}], [{:name=>"Gummy Sour Apple", :price=>3}, {:name=>"Gummy Apple", :price=>5}, {:name=>"Gummy Moldy Apple", :price=>1}]], [[{:name=>"Grape Drink", :price=>1}, {:name=>"Orange Drink", :price=>1}, {:name=>"Pineapple Drink", :price=>1}], [{:name=>"Mints", :price=>13}, {:name=>"Curiously Toxic Mints", :price=>1000}, {:name=>"US Mints", :price=>99}]]]

def total_snack_price_on_spinner(nds, row_index, column_index)
  coordinate_total = 0
  inner_len = nds[row_index][column_index].length
  inner_index = 0
  while inner_index < inner_len do
    coordinate_total += nds[row_index][column_index][inner_index][:price]
    inner_index += 1
  end
  coordinate_total
end

def total_snack_value_of_row(nds, row_index)
  grand_row_total = 0
  column_index = 0
  while column_index < nds[row_index].length do
    grand_row_total += total_snack_price_on_spinner(nds, row_index, column_index)
    column_index += 1
  end
  grand_row_total
end

def total_value_of_snacks_in_machine(nds)
  grand_total = 0
  row_index = 0
  while row_index < nds.length do
    grand_total += total_snack_value_of_row(nds, row_index)
    row_index += 1
  end
  grand_total
end

p total_value_of_snacks_in_machine(vm) #=> 1192

Wow! That's really handy! Instead of one triply-nested while loop, we now have three helpful methods that will help hide away complexity and keep our code readable and maintainable.

From the perspective of generating insight, our code knows what insight we want and what insight it is ready to share. This code makes it easy to find by calling a method whose name immediately suggest the insight we want: total_value_of_snacks_in_machine. The method then, internally, uses other methods to help keep itself clean and simple:

total_value_of_snacks_in_machine
 \-> total_snack_value_of_row
    \-> total_snack_price_on_spinner

Growing the Code

Looking at the three methods we have to process this NDS, if we wanted to figure out the insight of average_price_per_snack we could take total_value_of_snacks_in_machine and divide it by the (not-yet written) total_number_of_snacks. That really wouldn't be that much work at all.

By creating clean, abstract, well-named, and small methods that use each other, we can grow the features of our code while keeping it clean and understandable.

Lab

In the lab you're going to write a method that uses First-Order methods to do its work.

Conclusion

Whenever you have complex data that needs to be "transformed," "massaged," or "changed," let methods help your human brain keep track of the constituent actions that need to happen to create the insights you need. Compare this last bit of code with the first "raw [] and Array methods" implementation. Which version would you rather debug? Which is easier to talk about and ask for help with?

What's even more interesting, is that as your collection of "First-Order" methods grows, you'll find yourself writing methods that mix their outputs together to create new insights. Eventually, you might have so many well-written, tiny, clear "First-Order" methods that reasoning about, explaining, and extending your code is simple. That's the goal.