Rails sum hashes inside an array - ruby-on-rails

Question

Basically I have an array of hashes like so :
[
{ :id => 20, :total => 1, :total2 => 0 },
{ :id => 21, :total => 1, :total2 => 0 },
{ :id => 22, :total => 2, :total2 => 0 },
{ :id => 23, :total => 1, :total2 => 0 },
{ :id => 20, :total => 1, :total2 => 0 },
{ :id => 21, :total => 1, :total2 => 0 },
{ :id => 22, :total => 1, :total2 => 1 },
{ :id => 23, :total => 1, :total2 => 0 }
]
I want the array to sum the last two hash columns like so, keeping the first (:id) as an identifier:
[
{ :id => 20, :total => 2, :total2 => 0 },
{ :id => 21, :total => 2, :total2 => 0 },
{ :id => 22, :total => 3, :total2 => 1 }
]
I have looked around and it seems that the .inject() method is used in this instance but I cannot really figure out the syntax/how to use this.
What I am looking for is to keep the first column (:id) as an ID field; if there is another hash with this ID, like in my example above, the two hashes should be added together.

Answer 1

Can you try this?
array = [{:stemp=>20, :vtotal=>1, :avg=>0}, {:stemp=>21, :vtotal=>1, :avg=>0},{:stemp=>22, :vtotal=>2, :avg=>0}, {:stemp=>23, :vtotal=>1, :avg=>0}, {:stemp=>20, :vtotal=>1, :avg=>0}, {:stemp=>21, :vtotal=>1, :avg=>0}, {:stemp=>22, :vtotal=>1, :avg=>1}, {:stemp=>23, :vtotal=>1, :avg=>0}]
result = array.group_by{|h| h[:stemp]}.map do |stemp, hashes|
{ stemp: stemp, vtotal: hashes.map{|h| h[:vtotal]}.inject(:+), avg: hashes.map{|h| h[:avg]}.inject(:+) }
end
Just copy-pasted it in the IRB console with ruby 1.9.3, outputs this:
[
{:stemp=>20, :vtotal=>2, :avg=>0},
{:stemp=>21, :vtotal=>2, :avg=>0},
{:stemp=>22, :vtotal=>3, :avg=>1},
{:stemp=>23, :vtotal=>2, :avg=>0}
]

Answer 2

I reformatted the data in the question and in my answer so that it is a little easier for others to see what is going on.
data = [
{ :stemp => 20, :vtotal => 1, :avg => 0 },
{ :stemp => 21, :vtotal => 1, :avg => 0 },
{ :stemp => 22, :vtotal => 2, :avg => 0 },
{ :stemp => 23, :vtotal => 1, :avg => 0 },
{ :stemp => 20, :vtotal => 1, :avg => 0 },
{ :stemp => 21, :vtotal => 1, :avg => 0 },
{ :stemp => 22, :vtotal => 1, :avg => 1 },
{ :stemp => 23, :vtotal => 1, :avg => 0 }
]
First, group your hashes by the stemp.
data = data.group_by { |datum| datum[:stemp] }
Then iterate over each stemp and its entries.
data = data.map do |stemp, entries|
# this pulls out each hash's :vtotal entry and then combines it with the + operator
vtotal = entries.map { |entry| entry[:vtotal] }.inject(&:+)
# this does the same as above, but for the avg entry
avg = entries.map { |entry| entry[:avg] }.inject(&:+)
{ :stemp => stemp, :vtotal => vtotal, :avg => avg }
end
This outputs
[
{ :stemp => 20, :vtotal => 2, :avg => 0 },
{ :stemp => 21, :vtotal => 2, :avg => 0 },
{ :stemp => 22, :vtotal => 3, :avg => 1 },
{ :stemp => 23, :vtotal => 2, :avg => 0 }
]

Answer 3

This solution suffers from readability but I wanted to provide it for reference.
Hash#merge accepts a block that will be executed when a colliding key is found.
arr = [ {:id => 20, :total => 1, :total2 => 0} ... ]
arr.group_by{ |h| h[:id] }.map do |_, hash|
hash.reduce do |hash_a, hash_b|
hash_a.merge(hash_b){ |key, old, new| key == :id ? old : old + new }
end
end

Answer 4

This one also works
arr.group_by{|t| t[:stemp]}.map {|key, value| value.inject({}) { |hash, values| values.merge(hash){ |key, v1, v2| key == :stemp ? v1 : v1+v2 }}}
change to :id
arr.group_by{|t| t[:id]}.map {|key, value| value.inject({}) { |hash, values| values.merge(hash){ |key, v1, v2| key == :id ? v1 : v1+v2 }}}

Answer 5

[{"4"=>"20.0"}, {"4"=>"20.0"}, {"4"=>"10.0"}, {"4"=>"10.0", "5"=>"10.0"}, {"4"=>"10.0", "5"=>"0.0"}, {"4"=>"10.0"}, {"4"=>"10.0"}, {"4"=>"0.0", "5"=>"10.66"}, {"4"=>"20.0"}, {"4"=>"10.0"}, {"4"=>"10.0"}, {"4"=>"0.0"}].map{|m| m.map{|k,v| v.to_f}.sum()}.sum()