1. Understand return values for enumerators.
2. Use true/false expressions within a block.
3. Use the #all? enumerator to create a true/false return value.
4. Use the #none? enumerator to create a true/false return value.
5. Use the #any? enumerator to create a true/false return value.
6. Use the #include? enumerator to create a true/false return value.

When we are iterating over objects in a collection like with #each we generally don't care about the return values.

["Red", "Yellow", "Blue"].each do |color|
  puts "There are #{color.length} letters in #{color}"
end #=> ["Red", "Yellow", "Blue"]

If you run this code in IRB, you'll see:

001:0 > ["Red", "Yellow", "Blue"].each do |color|
?>        puts "There are #{color.length} letters in #{color}"
003:1 > end

There are 3 letters in Red
There are 6 letters in Yellow
There are 4 letters in Blue
=> ["Red", "Yellow", "Blue"]

You can see the block did what we intended it to do, it printed our output. You'll notice the last line also indicates that the #each method also returned a value. All expressions in ruby must return a value. When you use #each on a collection, the return value is always the original collection. Nothing you do inside the block you pass #each will ever change the return value. But that's not always the case. With other enumerator methods, the return value of the method is very much dependent on the block.

Imagine wanting to know if all the numbers in an array are odd. You could use each with something like:

all_odd = true
[1,2,3].each do |number|
  if number.even? # Will evaluate to false for 1, true for 2, false for 3
    all_odd = false
  end
end
all_odd #=> false

That works, the end value of all_odd will be false because 2 flipped the all_odd variable to false. However, something so simple - checking if all the elements in this array are odd - isn't being expressed clearly. Worse than our code not expressing our intention is that our code requires us to maintain variable state, all_odd, which can easily lead to errors (say if some other piece of code accidentally changes that variable value).

Consider the following example using #all?:

all_odd = [1,3].all? do |number|
  number.odd? # Will evaluate to true for 1, true for 3
end #=> true
all_odd #=> true

The rule for the #all? enumerator is that the block passed to it must return true for every iteration for the entire #all? expression or method to return true. If we introduce an even number to the collection, the return value will change.

all_odd = [1,2,3].all? do |number|
  number.odd? # Will evaluate to true for 1, false for 2, true for 3
end #=> false
all_odd #=> false

That's the rule for #all? - every iteration, every loop of the block must return true. When the block encounters the value 2 for number, it will run the expression 2.odd? which will return false. Because there was at least one iteration of the block that had a false return value, the entire #all? expression returns false.

Imagine the opposite of #all?, a method #none?, where we are interested in none of the elements in a collection producing a true expression within the block passed to #none?.

[1,3].none?{|i| i.even?} #=> true

The entire expression #none? returns true because none of those numbers will produce a true expression when asked within the block if they are even. Compare the code above to the code required to test that condition using #each:

none_even = true
[1,3].each do |i|
  if i.even?
    none_even = false
  end
end #=> [1,3] because `#each` always returns the original collection
none_even #=> true

These high-level boolean enumerators like #all? and #none? are way cleaner for evaluating elements in a collection for true/false conditions.

To recap, if any of the elements in the collection evaluate to true when passed to the block, #none? will return false. If none of the elements evaluate to true, #none? will return true.

Sometimes you want to be a bit more forgiving than #all? or #none? and just ensure that at least one element in a collection will create a true expression within the block passed. #any? is perfect for this. The #any? enumerator will return true if at least one iteration of the block evaluates to true, but false if none of them do.

[1,2,100].any?{|i| i > 99} #=> true

The #any? expression above will return true because at least one element, 100, will produce a true evaluation in the block.

Whereas #any? is useful for evaluating the truthiness of the logic of a block, #include? is helpful if you'd like to merely compare actual contents of a known value.

#include? will return true if the given object exists in the element. If it doesn't find a match, it will return false.

the_numbers = [4,8,15,16,23,42]
the_numbers.include?(42)   #=> true
the_numbers.include?(6)   #=> false

The #include? expression first returns true because the_numbers[5] == 42. When it is run with 6, it will evaluate to false since that item is not present in the array.

Please open a GitHub issue or pull-request. Provide a detailed description that explains the issue you have found or the change you are proposing. Then "@" mention your instructor on the issue or pull-request, and send them a link via Connect.

PHRG Ruby Boolean Enumerables