MongoDB Group data different fields distinct problem solving method

Recently by the aggregation of MongoDB baffled, conditions:

1, the table has the field Mac,pv,ip,dates,type

2, requirements, based on IP to group, calculate PV Total, non-duplicated Mac number, non-duplicated IP number

3, the following is the solution, Description: MongoDB authoritative guide is used in the object, that is, "Macs": {}, and then in the reduce with in to find, prev.macs[obj.mac]= "" To add, easy to appear unexpected number error

Now it's easy to change the array and clear it in finalize

1 Db.test.group ({2"Key": "IP",3"Initial": {"Mac": 0, "PV": 0, "IP": 0, "Macs": [], "IPs":[]},4"$reduce":function(Obj,prev) {5         if(Prev.macs.indexOf (OBJ.MAC) <0){6Prev.mac =Prev.macs.push (OBJ.MAC);7         }8         if(Prev.ips.indexOf (OBJ.IP) <0){9Prev.ip =Prev.ips.push (OBJ.IP);Ten         } Oneprev.pv+=OBJ.PV; A     }, -"Finalize":function(prev) { -prev.macs=[]; theprev.ips=[]; -     }, -"condition": {"dates": "2014-05-09", "type": "Type1"} -});

If there is a better way, please leave a message and discuss together