使用mongoose.js按月和按年分组

Question

我在mongodb的收藏如下:

{
    "AccountID" : "87f7fd60-d1ad-11e2-98bb-795730bce125",
    "userId" : ObjectId("51b59fbec46916e60d00000c"),
    "_id" : ObjectId("51b6e603e3efef161b000003"),
    "accessDate" : ISODate("2013-06-11T08:55:31.957Z"),
    "__v" : 0
}
{
        "AccountID" : "47f7fd60-d1ad-11e2-98bb-795730bce125",
        "userId" : ObjectId("51b59fbec46916e60d00000d"),
        "_id" : ObjectId("51b6e603e3efef161b000003"),
        "accessDate" : ISODate("2013-05-1T08:05:31.957Z"),
        "__v" : 0
}

我写了一个查询结果得到以下结果:这是按月和年分组的结果和每天的计数.

{
  "usage": [
    {
      "year": 2013,
      "monthlyusage": [
        {
          "month": 1,
          "dailyusage": [
            {
              "day": 1,
              "count": 205
            },
            {
              "day": 2,
              "count": 1109
            },
            {
              "day": 4,
              "count": 455
            }
          ]
        },
        {
          "month": 2,
          "dailyusage": [
            {
              "day": 11,
              "count": 256
            },
            {
              "day": 2,
              "count": 1001
            },
            {
              "day": 5,
              "count": 65
            }
          ]
        }
      ]
    },
    {
      "year": 2012,
      "monthlyusage": [
        {
          "month": 12,
          "dailyusage": [
            {
              "day": 1,
              "count": 78
            },
            {
              "day": 2,
              "count": 7009
            },
            {
              "day": 28,
              "count": 55
            }
          ]
        },
        {
          "month": 11,
          "dailyusage": [
            {
              "day": 11,
              "count": 800
            },
            {
              "day": 2,
              "count": 5094
            },
            {
              "day": 25,
              "count": 165
            }
          ]
        }
      ]
    }
  ]
}

我怎么能用mongoose.js框架做到这一点

Answer 1

Mongoose为MongoDB聚合框架提供了一个轻量级的包装器.如果您不熟悉聚合,可以从MongoDB文档中了解更多信息:http://docs.mongodb.org/manual/aggregation/

要将数据按照上面描述的形式进行处理,您可以使用具有一系列$ group操作的聚合管道.这里是使用mongoose框架:

var dateSchema = mongoose.Schema({…});
var DateItem = mongoose.model('DateItem', dateSchema);

DateItem.aggregate(
      { $group : { 
           _id : { year: { $year : "$accessDate" }, month: { $month : "$accessDate" },day: { $dayOfMonth : "$accessDate" }}, 
           count : { $sum : 1 }}
           }, 
      { $group : { 
           _id : { year: "$_id.year", month: "$_id.month" }, 
           dailyusage: { $push: { day: "$_id.day", count: "$count" }}}
           }, 
      { $group : { 
           _id : { year: "$_id.year" }, 
           monthlyusage: { $push: { month: "$_id.month", dailyusage: "$dailyusage" }}}
           }, 
      function (err, res)
           { if (err) ; // TODO handle error 
             console.log(res); 
           });
});

第一个$ group将生成此表单的文档,每天一个:

{ 
  "_id" : { "year" : 2013, "month" : 8, "day" : 15 },
  "count" : 1
}

第二个$ group将生成按月分组的文档:

{
  "_id" : { "year" : 2012, "month" : 11 },
 "dailyusage" : [
          { "day" : 6, "count" : 1 },
          { "day" : 9, "count" : 1 },
          ... ]
},

而第三个$ group将产生更大的文档,每年一个.

此查询将您的数据聚合到大型分层文档中.但是,如果您计划在聚合后对此数据运行查询,则这可能不是数据最有用的形式.请考虑如何使用聚合数据.涉及更小文档的模式(可能每月一个,甚至每天一个)可能更方便.