Welcome to OGeek Q&A Community for programmer and developer-Open, Learning and Share
Welcome To Ask or Share your Answers For Others

Categories

0 votes
421 views
in Technique[技术] by (71.8m points)

mongodb - mongo db map reduce sum results as integer

My collections seems like this

{
"bet_session_id": ObjectId("51a60cba6ef215d019000299"),
"date": ISODate("2013-05-29T14:13:14.572Z"),
"player_items": [
{
  "chip": NumberInt(1),
  "item": "x14",
  "ratio": NumberInt(35),
  "total_win": NumberInt(0)
},
{
  "chip": NumberInt(1),
  "item": "x15",
  "ratio": NumberInt(35),
  "total_win": NumberInt(36)
},
{
  "chip": NumberInt(1),
  "item": "121511141013",
  "ratio": NumberInt(5),
  "total_win": NumberInt(6)
},
],
"user_id": "7876010",
}
,
{
"bet_session_id": ObjectId("51a60cba6ef215d019000299"),
"date": ISODate("2013-05-29T14:13:14.572Z"),
"player_items": [
{
  "chip": NumberInt(1),
  "item": "x14",
  "ratio": NumberInt(35),
  "total_win": NumberInt(0)
},
{
  "chip": NumberInt(1),
  "item": "x15",
  "ratio": NumberInt(35),
  "total_win": NumberInt(36)
},
{
  "chip": NumberInt(1),
  "item": "121511141013",
  "ratio": NumberInt(5),
  "total_win": NumberInt(6)
},
{
  "chip": NumberInt(1),
  "item": "12151114",
  "ratio": NumberInt(8),
  "total_win": NumberInt(9)
},
{
  "chip": NumberInt(1),
  "item": "1514",
  "ratio": NumberInt(17),
  "total_win": NumberInt(18)
},
{
  "chip": NumberInt(1),
  "item": "1215",
  "ratio": NumberInt(17),
  "total_win": NumberInt(18)
},
{
  "chip": NumberInt(1),
  "item": "151814171316",
  "ratio": NumberInt(5),
  "total_win": NumberInt(6)
},
{
  "chip": NumberInt(1),
  "item": "151413",
  "ratio": NumberInt(11),
  "total_win": NumberInt(12)
},
{
  "chip": NumberInt(1),
  "item": "15181417",
  "ratio": NumberInt(8),
  "total_win": NumberInt(9)
}
],
"user_id": "9034906623",
}

using above the data i need to calculate player total wins

and My map reduce functions is here :

var mapBetPlayWinStats = function() { 
for (i=0; i<this.player_items.length; i++ ) emit( this.player_items[i].item, this.player_items[i].total_win )};

var reduceBetPlayWinStats = "function(item,values) { var sum = 0; sum += values; return sum; }"

db.bet_results.mapReduce(mapBetPlayWinStats, reduceBetPlayWinStats, {out: "bet_number_play_win_stats"});

RESULTS:
{
"_id": "red",
"value": "20,50,20,NumberLong(20),NumberLong(20),100" 
}
{
"_id": "odd",
"value": "NumberLong(0),NumberLong(0)" 
}
{
"_id": "even",
"value": "0,20,NumberLong(20),NumberLong(20)" 
}

i guess my map reduce functions is correct but i need to sum value items as integer. i also tried to parseInt() and new NumberLong() functions.

See Question&Answers more detail:os

与恶龙缠斗过久,自身亦成为恶龙;凝视深渊过久,深渊将回以凝视…
Welcome To Ask or Share your Answers For Others

1 Reply

0 votes
by (71.8m points)

The following works for me:

In your map function change your emit to

emit( this.player_items[i].item, NumberInt(this.player_items[i].total_win) )};

In your reduce change sum to be:

var sum = NumberInt(0);

Now testing it:

> db.bet_number_play_win_stats.find({value:{$type:16}})
{ "_id" : "121511141013", "value" : 6 }
{ "_id" : "x14", "value" : 0 }
{ "_id" : "x15", "value" : 36 }
Fetched 3 record(s) in 1ms

16 (Hex 10) is the type for 32-bit integer in bson.

If your number of items being aggregated is not too large, I would recommend looking at using aggregation framework:

> var r=db.bet_results.aggregate( [
             {$unwind : "$player_items"}, 
             {$group: { _id:"$player_items.item", 
                        wins:{$sum:"$player_items.total_win"}
                      }
              }
 ] )
> db.bet_number_play_win_agg.save(r.result)
Inserted 1 record(s) in 7ms
> > db.bet_number_play_win_agg.find({wins:{$type:16}})
{ "_id" : "121511141013", "wins" : 6 }
{ "_id" : "x15", "wins" : 36 }
{ "_id" : "x14", "wins" : 0 }
Fetched 3 record(s) in 1ms

Same result, type preserved, and faster too.


与恶龙缠斗过久,自身亦成为恶龙;凝视深渊过久,深渊将回以凝视…
OGeek|极客中国-欢迎来到极客的世界,一个免费开放的程序员编程交流平台!开放,进步,分享!让技术改变生活,让极客改变未来! Welcome to OGeek Q&A Community for programmer and developer-Open, Learning and Share
Click Here to Ask a Question

...