２つの配列を比較して、重複している値を取り出す(回答)

It is answer1

ruby

arr1 = [2, 4, 5, 6, 7]
arr2 = [7, 8, 9, 7, 4, 3, 2]

def extract_value(arr1, arr2)
  arr3 = arr1.uniq + arr2.uniq
  arr4 = arr3.inject(Hash.new(0)){| init, v | init[v]+=1; init;}
  answer = arr4.map{|k, v|  k if v >= 2}.compact
end

extract_value(arr1, arr2)

It is answer2

ruby

arr1 = [2, 4, 5, 6, 7]
arr2 = [7, 8, 9, 7, 4, 3, 2]

require 'set'
def extract_value(arr1, arr2)
  set_val, set_val2 = Set.new, Set.new
  arr1.map{|v| set_val << v}
  arr2.each {|v| set_val2 << v if set_val.include?(v) }
  set_val2.to_a
end

extract_value(arr1, arr2)

It is answer3

ruby

arr1 & arr2

Which is a better algorithm?

Let me check

ruby

arr1, arr2 = [], []
1.upto(100000000).each {|v| arr1 << rand(1..500)}
1.upto(100000000).each {|v| arr2 << rand(1..1000)}

embed time

ruby

def extract_value1(arr1, arr2)
  puts Time.now
  arr3 = arr1.uniq + arr2.uniq
  arr4 = arr3.inject(Hash.new(0)){| init, v | init[v]+=1; init;}
  answer = arr4.map{|k, v|  k if v >= 2}.compact
  puts Time.now
  answer
end

It takes 8 seconds

text

2017-02-18 15:55:40 +0900
2017-02-18 15:55:48 +0900

embed time

ruby

require 'set'
def extract_value2(arr1, arr2)
  puts Time.now
  set_val, set_val2 = Set.new, Set.new
  arr1.map{|v| set_val << v}
  arr2.each {|v| set_val2 << v if set_val.include?(v) }
  puts Time.now
  set_val2.to_a
end

It takes 44 seconds

text

2017-02-18 15:54:49 +0900
2017-02-18 15:55:33 +0900

It looks that first algorithm is better.
I think it is because of uniq function.

I will change value of arr and function also

ruby

arr1, arr2 = [], []
1.upto(100000000).each {|v| arr1 << rand(1..5000000)}
1.upto(100000000).each {|v| arr2 << rand(1..10000000)}

I will use uniq function in second one and name extract_value3

ruby

require 'set'
def extract_value3(arr1, arr2)
  puts Time.now
  set_val, set_val2 = Set.new, Set.new
  arr1 = arr1.uniq
  arr2 = arr2.uniq
  arr1.map{|v| set_val << v}
  arr2.each {|v| set_val2 << v if set_val.include?(v) }
  puts Time.now
  set_val2.to_a
end

This is the result of extract_value1

2017-02-18 16:26:37 +0900
2017-02-18 16:29:22 +0900

This is the result of extract_value2

2017-02-18 16:45:04 +0900
2017-02-18 16:48:12 +0900

This is the result of extract_value3

2017-02-18 16:39:18 +0900
2017-02-18 16:41:20 +0900

Conclusion

3 is a best algorithm.