如何将metadata_hash键值全排列合并到哈希数组的每个元素中

需求描述 #

给定由哈希（键值对对象）组成的数组array_of_hashes，需要为数组中的每个元素合并另一个metadata_hash的键值组合。规则如下：

• metadata_hash的所有值均为数组类型
• 首先对metadata_hash的所有值数组取笛卡尔积
• 将每一组笛卡尔积结果转为对应的键值对，再合并到array_of_hashes的每个元素中，最终生成新的结果数组

示例说明 #

示例1 #

输入：

array_of_hashes = [{"name"=>"bob", "age"=>30}, {"name"=>"bart", "age"=>35}]
metadata_hash = {"likes"=>["apples", "oranges"], "dislikes"=>["cold"]}

预期输出：

[{"name"=>"bob", "age"=>30, "likes"=>"apples", "dislikes"=>"cold"},
 {"name"=>"bob", "age"=>30, "likes"=>"oranges", "dislikes"=>"cold"},
 {"name"=>"bart", "age"=>35, "likes"=>"apples", "dislikes"=>"cold"},
 {"name"=>"bart", "age"=>35, "likes"=>"oranges", "dislikes"=>"cold"}]

示例2 #

输入：

array_of_hashes = [{"name"=>"bob", "age"=>30}, {"name"=>"bart", "age"=>35}]
metadata_hash = {"likes"=>["apples", "oranges"], "dislikes"=>["cold"], "courses"=>["science", "history"]}

预期输出：

[{"name"=>"bob", "age"=>30, "likes"=>"apples", "dislikes"=>"cold", "courses"=>"science"},
 {"name"=>"bob", "age"=>30, "likes"=>"oranges", "dislikes"=>"cold", "courses"=>"history"},
 {"name"=>"bob", "age"=>30, "likes"=>"oranges", "dislikes"=>"cold", "courses"=>"science"},
 {"name"=>"bob", "age"=>30, "likes"=>"apples", "dislikes"=>"cold", "courses"=>"history"},
 {"name"=>"bart", "age"=>35, "likes"=>"apples", "dislikes"=>"cold", "courses"=>"science"},
 {"name"=>"bart", "age"=>35, "likes"=>"oranges", "dislikes"=>"cold", "courses"=>"history"},
 {"name"=>"bart", "age"=>35, "likes"=>"oranges", "dislikes"=>"cold", "courses"=>"science"},
 {"name"=>"bart", "age"=>35, "likes"=>"apples", "dislikes"=>"cold", "courses"=>"history"}]

实现方案（Ruby语言） #

核心逻辑分为两步：首先生成metadata_hash所有值的笛卡尔积，转为对应的元数据哈希列表；再将原哈希数组的每个元素与所有元数据哈希合并，扁平化后得到最终结果。
实现代码：

def merge_with_metadata(array_of_hashes, metadata_hash)
  # 提取元数据的键列表
  meta_keys = metadata_hash.keys
  # 对元数据的所有值数组取笛卡尔积，转为哈希列表
  meta_combinations = metadata_hash.values.inject(&:product).map do |combination|
    meta_keys.zip(combination.flatten).to_h
  end
  # 遍历原数组每个元素，与所有元数据组合合并
  array_of_hashes.flat_map do |item|
    meta_combinations.map { |meta| item.merge(meta) }
  end
end

调用示例 #

# 测试示例1
array_of_hashes = [{"name"=>"bob", "age"=>30}, {"name"=>"bart", "age"=>35}]
metadata_hash = {"likes"=>["apples", "oranges"], "dislikes"=>["cold"]}
result = merge_with_metadata(array_of_hashes, metadata_hash)
puts result.inspect # 输出与示例1预期一致

说明 #

• 代码使用merge方法生成新哈希，不会修改原数组中的原始哈希对象
• 支持任意数量键的metadata_hash，只要所有值都是数组即可自动计算笛卡尔积
• 最终结果长度为：原数组长度 × 元数据所有值数组长度的乘积

内容的提问来源于stack exchange，提问作者simoni