Question

I am trying to merge two arrays of objects in ruby. The objects have two relevant fields; id and reach_cost.

I want my resultant array to contain unique ids, where each object has the smallest reach_cost in the case of a collision.

Running;

result = a1 | a2;

Yields mixed results, it appears that elements of a1 take precedence over elements of a2.

I could of course iterate over both arrays and make the comparison on element.reach_cost manually, but this is a high performance environment, and this method gets called an awful lot. For that reason I am trying to leverage the native components of the | operator.

Is it possible to direct the | operator to prefer one object over another? Perhaps by overriding <=> or similar?

I have read the source code of the | operator in the docs but it doesn’t appear to make any comparison, merely preferring the first array parameter over the second.

Answer 1

I suggest storing these arrays as hashes. This provides O(1) id lookup. Here’s what the code may look like:

h1 = {'1' => {:val => 2}, '2' => {:val => 3}}
h2 = {'1' => {:val => 5}, '2' => {:val => 1}}

def merge_hashes a, b
  a.reduce({}) do |memo, obj|
    k, v = obj

    # choose element with smallest :val
    memo[k] = if b[k][:val] < v[:val]
                b[k]
              else
                v
              end

    memo
  end
end

merge_hashes h1, h2 # => {"1"=>{:val=>2}, "2"=>{:val=>1}}

This should work pretty fast.

If you like one-liners, here’s one (kindly provided by @steenslag in the comments):

h1.merge(h2){|key,old,new| old[:val] < new[:val] ? old : new}