mongodb的mapreduce使用
先插入数据,插入的数据如下
> db.books.find()
{ "_id" : ObjectId("533ee1e8634249165a819cd0"), "name" : "apue", "pagenum" : 1023 }
{ "_id" : ObjectId("533ee273634249165a819cd1"), "name" : "clrs", "pagenum" : 2000 }
{ "_id" : ObjectId("533ee2ab634249165a819cd2"), "name" : "python book", "pagenum" : 600 }
{ "_id" : ObjectId("533ee2b7634249165a819cd3"), "name" : "golang book", "pagenum" : 400 }
{ "_id" : ObjectId("533ee2ca634249165a819cd4"), "name" : "linux book", "pagenum" : 1500 }
写map函数
var map = function() {
var category;
if ( this.pagenum >= 1000 )
category = 'Big Books';
else
category = "Small Books";
emit(category, {name: this.name});
};
map说明: 里面会调用emit(key, value)，集合会按照指定的key进行映射分组, 类似group by
上面map后的结果为: (按照category分组, 分组结果是{name: this.name}的list)
{"big books",[{name: "apue"}, {name : "linux book"}, {name : "clrs"}]]);
{"small books",[{name: "python book"}, {name : "golang book"}]);
写reduce函数
var reduce = function(key, values) {
var sum = 0;
values.forEach(function(doc) {
sum += 1;
});
return {books: sum};
};
reduce说明:简化函数，会对map分组后的数据进行分组简化，注意：在reduce(key,value)中的key就是emit中的key，value为emit分组后的emit(value)的集合, 这里是{name: this.name}的list
写mapReduce函数
> var count = db.books.mapReduce(map, reduce, {out: "book_results"});
> db[count.result].find()
{ "_id" : "big books", "value" : { "books" : 3 } }
{ "_id" : "small books", "value" : { "books" : 2 } }
如下运行可以看详细信息
> db.books.mapReduce(map, reduce, {out: "book_results"});
{
"result" : "book_results",
"timeMillis" : 107,
"counts" : {
"input" : 5,
"emit" : 5,
"reduce" : 2,
"output" : 2
},
"ok" : 1,
}
> db.book_results.find()
{ "_id" : "big books", "value" : { "books" : 3 } }
{ "_id" : "small books", "value" : { "books" : 2 } }
官网doc http://docs.mongodb.org/manual/core/map-reduce/