猫鼬mapreduce() [英] mongoose mapreduce()

问题描述

我一直在跟踪玩家的分数，以记录增长情况.我的架构如下:

I am keeping track of players' scores in order to record growth. My schema is as follows:

Schema = new schema({
    pid: { type: Number }
  , name: { type: String }
  , alliance: {
      aid: { type: Number }
    , name: { type: String }
    }
  , city_cnt: { type: Number }
  , score: { type: Number }
  , rank: { type: Number }
  , createdTimeAt: { type: Date, default: Date.now }
  , createdDayAt: { type: Number, default: function() { return Math.floor(Date.now() / (24 * 60 * 60)) } }
})

现在我想每天为每位玩家拉出一条记录(这是该天的最后记录分数).所以我把这张图缩小了:

Now I want to pull out one record per day (being the last recorded score for that day) per player. So I have this map reduce:

growthMap = function(){
  emit('createdTimeAt', this.createdTimeAt.toString())
}

growthReduce = function(prev, doc) { 
  if (doc.createdTimeAt > prev.createdTimeAt) {
    prev.createdTimeAt = doc.createdTimeAt
    prev.score = doc.score

    prev.pid = doc.pid
    prev.name = doc.name
  }
}

mongoose.connect('mongodb://localhost/my_database')
mongoose.model('Players', Schema)

mongoose.connection.db.executeDbCommand({
  mapreduce: 'players'
, map: growthMap.toString()
, reduce: growthReduce.toString()
, out: 'playerGrowth'
}, function(err, ret){ 
  if (err) console.log(err)
  else {
    console.log(ret.documents)
    mongoose.connection.db.collection('playerGrowths', function(err, collection){
      collection.find({}).toArray(function(err, docs){
        console.dir(docs)
      })

    })
  }
})

地图缩小似乎可以正常工作.第一个输出 console.log(ret.documents)产生以下结果:

The map reduce seems to work fine. The first output console.log(ret.documents) produces this result:

[ { result: 'playerGrowth',
    timeMillis: 11268,
    counts: { input: 14635, emit: 14635, reduce: 15, output: 1 },
    ok: 1 } ]

但是，每当我尝试在第二个输出 console.log(docs)中查询新集合时，都会得到一个没有结果的空数组.我确定我做错了什么，但是在mapreduce方面缺少猫鼬文档.

However, whenever I try to query the new collection in the second output console.log(docs), I get an empty array with no results. I'm sure I am doing something wrong, but the mongoose documentation is lacking when it comes to mapreduce.

根据下面Jed的回答，我将 mongoose.connection.db.collection('playerGrowths'更改为 mongoose.connection.db.collection('playerGrowth'，然后我m至少从find({}).toArray()获得结果.但是，这仍然不是我期望或需要的...

As per Jed's answer below I changed mongoose.connection.db.collection('playerGrowths' to mongoose.connection.db.collection('playerGrowth' and I'm at least getting a result from the find({}).toArray(). However, it's still not what I'm expecting or need...

我的新结果

[ { result: 'playerGrowth',
    timeMillis: 5743,
    counts: { input: 14635, emit: 14635, reduce: 15, output: 1 },
    ok: 1 } ]
null
[ { _id: 'createdTimeAt', value: null } ]

推荐答案

generate()函数应返回键/值对.由于您的键"是"createdTimeAt"，因此其汇总结果.您可能应该这样做:

The emit() function should return a key-value pair. Since your "key" is "createdTimeAt", its what it aggregates results by. You should probably do:

emit(this.createdTimeAt.toString(), <your value>)

这篇关于猫鼬mapreduce()的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！