如何按 _id 和国家/地区分组？

Question

我需要按 _id 和国家/地区进行分组。我已经成功分组 _id, ，但想知道如何对这些国家/地区进行分组 _ids 并返回每个国家/地区的计数。

我正在使用聚合框架。到目前为止，一切都很好。

conn = Mongo::Connection.new
db   = conn['foobar_development']

cmd = {
  aggregate: 'live_daily_stats',
  pipeline: [
    { '$project' => {
      :metacontent => 1,
      :visits => 1,
    } },
    { '$unwind' => '$visits' },
    { '$match' => { 'visits.minute' => { '$gt' => 224 } } },
    { '$sort' => { 'visits.minute' => 1 } },
    { '$group' => { 
      :_id => '$_id', 
      :visits => { '$push' => '$visits' }, 
      :visits_count => { '$sum' => 1 },
      :metacontent => { '$addToSet' => '$metacontent' },
      } 
    },
    { '$sort' => { 'visits_count' => -1 } },
  ]
}

res = db.command(cmd)['result']

返回结果如下：

[
    [0] {
                 "_id" => "20120726/foobar/song/custom-cred",
              "visits" => [
            [0] {
                                              "country_name" => "UK",
                               "iso_two_letter_country_code" => "UK",
                                                   "referer" => "http://localhost:3000/",
                                                    "minute" => 59,
                                                  "token_id" => "134326199711wfryhpdq"
            },
            [1] {
                                              "country_name" => "UK",
                               "iso_two_letter_country_code" => "UK",
                                                   "referer" => "http://localhost:3000/",
                                                    "minute" => 59,
                                                  "token_id" => "134326199711wfryhpdq"
            },
            [2] {
                                              "country_name" => "US",
                               "iso_two_letter_country_code" => "US",
                                                   "referer" => "http://localhost:3000/",
                                                    "minute" => 59,
                                                  "token_id" => "134326199711wfryhpdq"
            }
        ],
        "visits_count" => 1,
         "metacontent" => [
            [0] {
                                     "date" => "20120726"
            }
        ]
    },
    [1] {
                 "_id" => "20120725/foobar/song/test-pg3-long-title-here-test-lorem-ipsum-dolor-lo",
              "visits" => [
            [0] {
                                              "country_name" => "UK",
                               "iso_two_letter_country_code" => "UK",
                                                   "referer" => "http://localhost:3000/",
                                                    "minute" => 58,
                                                  "token_id" => "13432600883knjzcbic"
            }
        ],
        "visits_count" => 1,
         "metacontent" => [
            [0] {
                                     "date" => "20120725"
            }
        ]
    }
]

Answer 1

我改变了 $group 连接两者 _id 和 country_name:

cmd = {
  aggregate: 'live_daily_stats',
  pipeline: [
    { '$project' => {
      :metacontent => 1,
      :visits => 1,
    } },
    { '$unwind' => '$visits' },
    { '$match' => { 'visits.minute' => { '$gt' => 224 } } },
    { '$sort' => { 'visits.minute' => 1 } },
    { '$group' => { 
      :_id => { '$add' => ['$_id', '$visits.country_name']}, 
      :visits => { '$push' => '$visits' }, 
      :visits_count => { '$sum' => 1 },
      :metacontent => { '$addToSet' => '$metacontent' },
      } 
    },
    { '$sort' => { 'visits_count' => -1 } },
  ]
}

Answer 2

从文档

$组组文件一起用于计算的目的 基于文件集合的聚合值。几乎， 小组经常支持每个页面的平均页面视图等任务 每天一个网站。

$组的输出取决于您如何定义组。首先 为您提供的组指定标识符（即_ID字段） 用这个管道创建。您可以从中指定单个字段 管道中的文档，先前计算的值或 从几个传入字段组成的聚合密钥。

每个组表达式都必须指定_id字段。您可以指定 _ID字段作为虚线字段路径参考，包含多个字段的文档，括在括号中（即{}）或常量值。

我会首先尝试分组_id和国家/地区（让您执行您想要的键），然后通过_ID对结果进行分组，以给出所需的结构。

更新：

我在思考这样的东西..但我没有env设置来检查它..

    conn = Mongo::Connection.new
    db   = conn['foobar_development']

    cmd = {
      aggregate: 'live_daily_stats',
      pipeline: [
        { '$project' => {
          :metacontent => 1,
          :visits => 1,
        } },
        { '$unwind' => '$visits' },
        { '$match' => { 'visits.minute' => { '$gt' => 224 } } },
        { '$sort' => { 'visits.minute' => 1 } },
        { '$group' => { 
          :_id => {'$_id','$visits.iso_two_letter_country_code'},
          :page_id => '$_id',
          :visits_count => { '$sum' => 1 },
   .... whatever you want ...
          :metacontent => { '$addToSet' => '$metacontent' },
          } 
        },
        { '$group' => { 
          :_id => '$page_id', 
   .... whatever you want ...
          } 
        },
        { '$sort' => { 'visits_count' => -1 } },
      ]
    }

    res = db.command(cmd)['result']

.