Find a unique element in a compound array

Question

I am trying to solve a problem where I need to find the airport code in an array of arrays of that represents the starting point of a multi-city flight plan. For example: Given

Answer 1

If you think with Ruby's syntax in mind, just take a transpose and take out all arrivals from departures.

def flight_origin(arr)
  plan = arr.transpose
  plan[0].index((plan[0] - plan[1])[0])
end

flight_origin([['LAX', 'BWI'], ['BOS', 'SEA'], ['HNL', 'LAX'], ['SEA', 'HNL']]) # => 1

HTH

Answer 2

You could do this:

legs = [['LAX', 'BWI'], ['BOS', 'SEA'], ['HNL', 'LAX'], ['SEA', 'HNL']]

airports = legs.flatten 
  #=> ["LAX", "BWI", "BOS", "SEA", "HNL", "LAX", "SEA", "HNL"] 
legs.map(&:first).find { |ap| airports.count(ap) == 1 }
  #=> "BOS"

This could instead be written using Array#transpose (as has been done in other answers) or Enumerable#zip since:

legs.map(&:first)
  #=> ["LAX", "BOS", "HNL", "SEA"] 
legs.transpose.first
  #=> ["LAX", "BOS", "HNL", "SEA"] 
legs.first.zip(*legs[1..-1]).first
  #=> ["LAX", "BOS", "HNL", "SEA"] 

My main reason for answering, however, is to make a plug for the method Array#difference, which I'd like to see in some future Ruby version. It is defined here.

With it, we can write:

airports = legs.flatten 
  #=> ["LAX", "BWI", "BOS", "SEA", "HNL", "LAX", "SEA", "HNL"] 
(legs.map(&:first) - airports.difference(airports.uniq)).first
  #=> "BOS" 

Note that:

airports.difference(airports.uniq)
  #=> ["LAX", "SEA", "HNL"]

contains all airports that appear more than once in the legs; that is, all all the "intermediate" airports.

Answer 3

trips = [['LAX', 'BWI'], ['BOS', 'SEA'], ['HNL', 'LAX'], ['SEA', 'HNL']]

arrivals = trips.map(&:last)

p trips.find{|fligth| ! arrivals.include? fligth[0] } #=> ["BOS", "SEA"]

Answer 4

sorry i, have no enough time to explain my code - but i think nothing is such difficult not to figure out what is happening :)

here is the working example

def find_start_point(list)
    start = []
    finish = []

    list.each do |l|
        start.push(l[0])
        finish.push(l[1])
    end

    start.each do |st|
        if !finish.include? st
            return start.index(st)
        end
    end

end

Answer 5

Building on dwenzel's idea:

airports = [['LAX', 'BWI'], ['BOS', 'SEA'], ['HNL', 'LAX'], ['SEA', 'HNL']]

departures, arrivals = airports.transpose

first_departure_index = departures.index{|dep| !arrivals.include?(dep)}

Answer 6

I see what you're attempting. The originating airport would be the only one included in a 0 index of a sub array but not a 1 index of any sub array.

However, in you're solution, you are comparing each combination of 2 sub arrays (including a sub array and itself, by the way...), and then returning the index of list if a[0] != b[1] is ever true for sub array a. This is returning too many results, and you'll always end up returning the last index. For example, 'SEA', the 0 index of the 3rd sub array, does not equal 'BWI', the 1 index of the 0 sub array, so your start_point now equals 3.

I won't do all of your work for you :), but let me suggest this: When you're going through your iterations, keep track of which sub arrays' index 0 ever equals a different sub array's index 1. Your answer will be the only one not included in this list.

Edit: Continue to work through my above suggestion for good practice, but here's a real quick and short solution:

def find_start_point(list)
  list.each_with_index do |sub, idx|
    return idx if list.flatten.count(sub[0]) == 1
  end
end

This works by returning the index of the sub array with an index 0 where there are no other occurrences of that airport (by flattening the entire array and using #count)