Access Ruby Hash Using Dotted Path Key String

Question

The Rails I18n library transforms a YAML file into a data structure that is accessible via a dotted path call using the t() function.

t('one.two.three.four')

Does anyone know how to do this with a Ruby Hash? Or is it only possible directly via a YAML object?

Answer 1

This code not only allows dot notation to traverse a Hash but also square brackets to traverse Arrays with indices. It also avoids recursion for efficiency.

class Hash

  def key_path(dotted_path)
    result = self
    dotted_path.split('.').each do |dot_part|
      dot_part.split('[').each do |part|
        if part.include?(']')
          index = part.to_i
          result = result[index] rescue nil
        else
          result = result[part] rescue nil
        end
      end
    end

    result
  end

end

Example:

a = {"b" => {"c" => [0, [1, 42]]}}
a.key_path("b.c[-1][1]") # => 42

Answer 2

I would suggest taking a look at this gist:
https://gist.github.com/potatosalad/760726

It adds implode and explode methods to Hash object that transforms nested keys to single-level dotted path keys, and vice versa.

Answer 3

Yeah, I don't think that's built-in, anywhere else. But I use something like this in one of my projects:

class Hash
  def dig(dotted_path)
    parts = dotted_path.split '.', 2
    match = self[parts[0]]
    if !parts[1] or match.nil?
      return match
    else
      return match.dig(parts[1])
    end
  end
end

And then call it like

my_hash = {'a' => {'b' => 'a-b', 'c' => 'a-c', 'd' => {'e' => 'a-d-e'}}, 'f' => 'f'}
my_hash.dig('a.d.e') # outputs 'a-d-e' (by calling my_hash['a']['d']['e'])

Answer 4

There is also HashDot.

HashDot allows dot notation syntax use on hashes. It is faster, and more traversable than an object created with OpenStruct.

a = {b: {c: {d: 1}}}
a.b.c.d => 1

Answer 5

Just split on a dot in the path and iterate over this to find the right hash?

path.split(".").inject(hash) { |hash, key| hash[key] }

Alternatively you can build a new hash by iterating recursively over the whole structure:

def convert_hash(hash, path = "")
  hash.each_with_object({}) do |(k, v), ret|
    key = path + k

    if v.is_a? Hash
      ret.merge! convert_hash(v, key + ".")
    else
      ret[key] = v
    end
  end
end

Answer 6

Ruby 2.3 introduces the dig method that looks into nested arrays/hashes, it returns nil when no data is found.

For example:

test_data = {a: {b: {c: {d: 1}, e: 2}}}
path = 'a.b.c.d'.split('.').map(&:to_sym)
# path => [:a, :b, :c, :d]
test_data.dig(*path)

Of course if your nested use string keys, the to_sym step is not needed.