mongo-mapreduce测试（7）——使用js存储过程
mongo-mapreduce测试（1）——count/sum/where条件
mongo-mapreduce测试（2）——列转行与finalize函数
mongo-mapreduce测试（3）——group by having
mongo-mapreduce测试（7）——使用js存储过程
mongo-mapreduce测试（9）——python调用
mongo-mapreduce测试（10）——阶段总结（2）
可以将mr中常用的代码编写成js函数，存储到mongo中，这样可以使得mr代码更简洁些，而且js函数还可以重复使用。
1. sum
#将sum功能写成函数，保存到js存储过程中。
> db.system.js.save( { _id : "Sum" ,
... value : function(key,values)
... {
... var total = 0;
... for(var i = 0; i < values.length; i++)
... total += values[i];
... return total;
... }});
#执行MR，在reduce过程中调用sum函数（注意根据sum函数，可知map中的value必须是一个数值类型，不能使文档型）。
> var map = function() {
... emit(this.name, this.dic);
... };
> var reduce = function(key, values) {
... return Sum(key, values);
... };
> var xjbu = db.tianyc_test3.mapReduce(map, reduce, {out: "tianyc_test3_result2"});
> db.tianyc_test3_result2.find()
{ "_id" : "neu", "value" : 10 }
{ "_id" : "xtt", "value" : 15 }
{ "_id" : "yct", "value" : 3 }
2. count
> db.system.js.save( { _id : "Count" ,
... value : function(key,values)
... {
... return values.length;
... }});
> var map = function() {
... emit(this.name, 1);
... };
> var reduce = function(key, values) {
... return Count(key, values);
... };
> var xjbu = db.tianyc_test3.mapReduce(map, reduce, {out: "tianyc_test3_result2"});
> db.tianyc_test3_result2.find()
{ "_id" : "neu", "value" : 4 }
{ "_id" : "xtt", "value" : 5 }
{ "_id" : "yct", "value" : 2 }
3. avg
> db.system.js.save( { _id : "Avg" ,
... value : function(key,values)
... {
... return Sum(key, values)/values.length;
... }});
> var map = function() {
... emit(this.name, this.dic);
... };
> var reduce = function(key, values) {
... return Avg(key, values);
... };
> db.tianyc_test3.mapReduce(map, reduce, {out: "tianyc_test3_result2"});
{
"result" : "tianyc_test3_result2",
"timeMillis" : 9,
"counts" : {
"input" : 11,
"emit" : 11,
"reduce" : 3,
"output" : 3
},
"ok" : 1,
}
> db.tianyc_test3_result2.find()
{ "_id" : "neu", "value" : 2.5 }
{ "_id" : "xtt", "value" : 3 }
{ "_id" : "yct", "value" : 1.5 }
4. max
> db.system.js.save( { _id : "Max" ,
... value : function(key,values)
... {
... var result=values[0];
... for (var i=0;i<values.length;i++){
... if (result<values[i]) result=values[i];
... }
... return result;
... }});
> var map = function() {
... emit(this.name, this.dic);
... };
> var reduce = function(key, values) {
... return Max(key, values);
... };
> db.tianyc_test3.mapReduce(map, reduce, {out: "tianyc_test3_result2"});
{
"result" : "tianyc_test3_result2",
"timeMillis" : 10,
"counts" : {
"input" : 11,
"emit" : 11,
"reduce" : 3,
"output" : 3
},
"ok" : 1,
}
> db.tianyc_test3_result2.find()
{ "_id" : "neu", "value" : 4 }
{ "_id" : "xtt", "value" : 5 }
{ "_id" : "yct", "value" : 2 }
5. min
> db.system.js.save( { _id : "Min" ,
... value : function(key,values)
... {
... var result=values[0];
... for (var i=0;i<values.length;i++){
... if (result>values[i]) result=values[i];
... }
... return result;
... }});
> var map = function() {
... emit(this.name, this.dic);
... };
> var reduce = function(key, values) {
... return Min(key, values);
... };
> db.tianyc_test3.mapReduce(map, reduce, {out: "tianyc_test3_result2"});
{
"result" : "tianyc_test3_result2",
"timeMillis" : 10,
"counts" : {
"input" : 11,
"emit" : 11,
"reduce" : 3,
"output" : 3
},
"ok" : 1,
}
> db.tianyc_test3_result2.find()
{ "_id" : "neu", "value" : 1 }
{ "_id" : "xtt", "value" : 1 }
{ "_id" : "yct", "value" : 1 }
>
当前共写了5个小函数：
> db.system.js.find({},{_id:1})
{ "_id" : "Sum" }
{ "_id" : "Count" }
{ "_id" : "Avg" }
{ "_id" : "Max" }
{ "_id" : "Min" }
参考这里。
