MongoDB mapReduce操作

mapReduce函数原型

db.collection.mapReduce(
                         <map>,
                         <reduce>,
                         {
                           out: <collection>,
                           query: <document>,
                           sort: <document>,
                           limit: <number>,
                           finalize: <function>,
                           scope: <document>,
                           jsMode: <boolean>,
                           verbose: <boolean>
                         }
                       )

mapReduce函数关键是map和reduce函数的编写！！
map函数一般比较简单，仅仅包含emit(key,value) 函数
key、value都要是json格式！！
如果key或value仅包含一个字段，可以直接返回。

{
_id:ObjectId("XXXXXXXXXXXXXXXXXXXXXX"),
name:"Jack",
age:28,
sex:"male",
address:"XXXX",
income:10000
}

仅仅返回单个字段：

emit(sex,age)

返回多个字段：

emit({"sex":this.sex,"address":this.address},{"age":this.age,"income":this.income,"count":1})

value中字段可以自己任意定义，count：1用于统计数目

function reduce(key,values)
reduce函数默认有两个传入参数，
- key即map阶段的key
- values是map阶段emit对具有相同key的记录进行合并后的结果，是数组
reduce函数只负责处理values内容，不对key进行处理，返回的处理结果也必须符合json格式，不含key内容

function reduce(key, values) {
    var reduced = {
        age : 0,
        income : 0,
        count : 0
    };
    values.forEach(function (val) {
        reduced.age += val.age;
        reduced.income += val.income;
        reduced.count += val.count;
    });
    return reduced;
}

finalize函数在Reduce处理结束，返回结果之前再做一次处理，optional字段

function Finalize(key, reduced) {
    reduced.avgAge = reduced.age / reduced.count;
    reduced.avgIncome = reduced.income / reduced.count;
    return reduced;
}

其他字段不详细解释，可自行查看官网相关文档
本文实现的mapReduce函数是根据性别和地址，统计平均年龄和收入