带散列的和数组

Question

如何在这两个散列数组之间执行加法，例如，我想从第一个数组中获取第一个哈希和和number, s_number, l_number，从第二个数组获取第一个散列，和将取决于created_at，如果第一个数组中的created_at与第二个数组中的created_at相等，则执行+。请看一看我希望得到的回报，因为我不知道如何更好地解释这一点。

谢谢

raw_data = [
        [
            {"created_at"=>"2013-11-17","number"=>0,"s_number"=>0,"l_number"=>0},
            {"created_at"=>"2013-11-18","number"=>6,"s_number"=>1,"l_number"=>4},
            {"created_at"=>"2013-11-19","number"=>1,"s_number"=>1,"l_number"=>1},
            {"created_at"=>"2013-11-20","number"=>0,"s_number"=>0,"l_number"=>0}
        ],
        [
            {"created_at"=>"2013-11-17","number"=>2,"s_number"=>3,"l_number"=>4},
            {"created_at"=>"2013-11-18","number"=>1,"s_number"=>41,"l_number"=>1},
            {"created_at"=>"2013-11-19","number"=>5,"s_number"=>45,"l_number"=>7},
            {"created_at"=>"2013-11-20","number"=>7,"s_number"=>8,"l_number"=>2}    
        ]
        ]

应该归还什么？

[
        {"created_at"=>"2013-11-17","number"=>2,"s_number"=>3,"l_number"=>4},
        {"created_at"=>"2013-11-18","number"=>7,"s_number"=>42,"l_number"=>5},
        {"created_at"=>"2013-11-19","number"=>6,"s_number"=>46"l_number"=>8},
        {"created_at"=>"2013-11-20","number"=>7,"s_number"=>8,"l_number"=>2}    
]

更新散列格式与我发布的内容不同：

{:created_at=>Sun, 17 Nov 2013, :number=>0, :s_number=>0, :l_number=>0}

Answer 1

首先，您需要对数组进行求和：

raw_data.sum

这将归还如下的东西：

    [
        {"created_at"=>"2013-11-17","number"=>0,"s_number"=>0,"l_number"=>0},
        {"created_at"=>"2013-11-18","number"=>6,"s_number"=>1,"l_number"=>4},
        {"created_at"=>"2013-11-19","number"=>1,"s_number"=>1,"l_number"=>1},
        {"created_at"=>"2013-11-20","number"=>0,"s_number"=>0,"l_number"=>0}
        {"created_at"=>"2013-11-17","number"=>2,"s_number"=>3,"l_number"=>4},
        {"created_at"=>"2013-11-18","number"=>1,"s_number"=>41,"l_number"=>1},
        {"created_at"=>"2013-11-19","number"=>5,"s_number"=>45,"l_number"=>7},
        {"created_at"=>"2013-11-20","number"=>7,"s_number"=>8,"l_number"=>2}    
    ]

现在，我们需要按'created_at‘键对元素进行分组：

raw_data.sum.group_by{|element| element['created_at']}

这将导致：

{
    "2013-11-17" => [
        {"created_at"=>"2013-11-17","number"=>0,"s_number"=>0,"l_number"=>0},
        {"created_at"=>"2013-11-17","number"=>2,"s_number"=>3,"l_number"=>4},
      ],
    "2013-11-18" => [
        {"created_at"=>"2013-11-18","number"=>6,"s_number"=>1,"l_number"=>4},
        {"created_at"=>"2013-11-18","number"=>1,"s_number"=>41,"l_number"=>1}
      ], ...           
}

我们只需要这个散列的值：

raw_data.sum.group_by{|element| element['created_at']}.values

# =>
    [
      [
        {"created_at"=>"2013-11-17","number"=>0,"s_number"=>0,"l_number"=>0},
        {"created_at"=>"2013-11-17","number"=>2,"s_number"=>3,"l_number"=>4},
      ],
      [
        {"created_at"=>"2013-11-18","number"=>6,"s_number"=>1,"l_number"=>4},
        {"created_at"=>"2013-11-18","number"=>1,"s_number"=>41,"l_number"=>1}
      ], ...           
}

现在，我们需要将每个散列数组映射到一个散列。为此，我们将使用带有块的注入和合并函数：

raw_data.sum.group_by{|element| element[:created_at]}.values.map{|a| a.inject({}) {|result, hash| result.merge(hash) {|key, old_value, new_value| key == :created_at ? old_value : old_value + new_value}}}



# =>

[
    {"created_at"=>"2013-11-17","number"=>2,"s_number"=>3,"l_number"=>4},
    {"created_at"=>"2013-11-18","number"=>7,"s_number"=>42,"l_number"=>5},
    {"created_at"=>"2013-11-19","number"=>6,"s_number"=>46"l_number"=>8},
    {"created_at"=>"2013-11-20","number"=>7,"s_number"=>8,"l_number"=>2}    
]

Answer 2

我会这样做：

result = raw_data.flatten.each_with_object({}) do |hash, result|
  date = hash.delete("created_at")
  result[date] = Hash.new(0) unless result[date]
  hash.each{|key, val| result[date][key] += val}
end.map{|date, hash| hash["created_at"] = date; hash}

p result #=> [{"number"=>2, "s_number"=>3, "l_number"=>4, "created_at"=>"2013-11-17"}, ...

或者也许更好，因为它将保持散列的顺序：

result = raw_data.flatten.group_by{|x|x["created_at"]}.values.map do |x| 
  x.inject do |result, hash| 
    result.merge(hash){|key, old, new| old.is_a?(String) ? old : old + new }
  end
end

p result #=> [{"created_at"=>"2013-11-17", "number"=>2, "s_number"=>3, "l_number"=>4},  {"created_at"=>"2013-11-18", "number"=>7, "s_number"=>42, "l_number"=>5} ...

Answer 3

output = []

raw_data.flatten.group_by {|x| x[:created_at].to_s}.each do |key, values|
  temp_hash = { :created_at => key.to_date }
  [:number, :s_number, :l_number].each do |attr|
    temp_hash[attr] = values.collect{|w| w[attr]}.sum
  end
  output << temp_hash
end

# output will contain your result