Question

If I have two ranges that overlap:

x = 1..10
y = 5..15

When I say:

puts x.include? y 

the output is:

false 

because the two ranges only overlap partially.

But if I want it to be "true" when there is partial overlap between two ranges, how would I write that? In other words I need a way to know when one range contains a subset of another range. I assume there's an elegant way to write this in Ruby but the only solutions I can think of are verbose.

Was it helpful?

Solution

Be careful using this with large ranges but this is an elegant way to do it:

(x.to_a & y.to_a).empty?

OTHER TIPS

The efficient way is to compare the limits

(x.first <= y.last) and (y.first <= x.last)

You could also convert the ranges to sets, since you're basically doing set intersection here. Might be easier if you are dealing with more than two ranges.

x = (1..10).to_set
y = (5..15).to_set
!(x & y).empty? #returns true (true == overlap, false == no overlap)

This method can be used to test overlap between multiple ranges in an efficient way:

def range_overlap?(ranges)
  sorted_ranges = ranges.sort
  sorted_ranges.each_cons(2).each do |r1, r2|
    return true if r2.first <= r1.last
  end
  return false
end


def test(r)
  puts r.inspect, range_overlap?(r)
  puts '================'
  r = r.reverse
  puts r.inspect, range_overlap?(r)
  puts '================'
end


test [[1,9], [10, 33]]
test [[1,10], [5, 8]]
test [[1,10], [10, 33]]

If a range includes either the beginning or the end of a second range, then they overlap.

(x === y.first) or (x === y.last)

is the same as this:

x.include?(y.first) or x.include?(y.last)

But if I want it to be "true" when there is partial overlap between two ranges, how would I write that?

You can convert the ranges to an array, and use the & operator (conjunction). This returns a new array with all the elements occuring in both arrays. If the resulting array is not empty, that means, that there are some overlapping elements:

def overlap?(range_1, range_2)
  !(range_1.to_a & range_2.to_a).empty?
end

If you're checking for overlap, then I'd just do

(x.include? y.first) or (x.include? y.last)

as one range will have to include at least one of the ends of the other. This is more intuitive to me than the accepted conjuction answer, though not quite as efficient as MarkusQ's limit comparison.

Rails has Range#overlaps?

def overlaps?(other)
  cover?(other.first) || other.cover?(first)
end

If you're using Ruby 2.6, you can use Range#cover? with another Range.

(1..5).cover?(2..3)     #=> true
(1..5).cover?(0..6)     #=> false
(1..5).cover?(1...6)    #=> true

Some helpful enumerable methods:

# x is a 'subset' of y
x.all?{|n| y.include? n}
# x and y overlap
x.any?{|n| y.include? n}
# x and y do not overlap
x.none?{|n| y.include? n}
# x and y overlap one time
x.one?{|n| y.include? n}
Licensed under: CC-BY-SA with attribution
Not affiliated with StackOverflow
scroll top