4
In my mongodb database,there is only one document below:
  {
      "_id" : ObjectId("5226efc5ccf23796f9590f67"),
      "count" : NumberLong(126045),
      "eissn" : "1466-5034",
      "grade" : 0.0,
      "issn" : "1466-5026",
      "journalAlias" : ["International journal of systematic and evolutionary microbiology", "Int J Syst Evol Microbiol", "Int. J. Syst. Evol. Microbiol."],
      "journalName" : "International journal of systematic and evolutionary microbiology",
      "ccNameCount" : [{
          "ccName" : "ATCC",
          "paperCount" : 1,
          "strainCount" : 2
        }, {
          "ccName" : "ATCC",
          "paperCount" : 3,
          "strainCount" : 4
        }, {
          "ccName" : "BCCM/DCG",
          "paperCount" : 5,
          "strainCount" : 6
        }, {
          "ccName" : "BCCM/IHEM",
          "paperCount" : 0,
          "strainCount" : 0
       }
    }

What I want ,is to find out how many different ccNameCout.paperCount whose ccName is "ATCC".So ,in my java code ,I write like this:

coll.distinct("ccNameCount.paperCount",new BasicDBObject("ccNameCount.ccName","ATCC")),

but the result is 4, instead of 2.

I know the process of function distinct.Firstly , it search the collection whose ccNameCount.ccName is "ATCC", and then ,it will do a distinct() in these found documents.

But what I need, is to find how many different paperCount whose ccName is "ATCC". Anyone can help me ?

  • Do you need to know the different paperCount numbers or just how many are different? – joao Feb 26 '14 at 7:05
  • Do you still have any problems with this? If my answer solves your problem as you intended mark it as accepted. – joao Feb 28 '14 at 12:43
4

I believe the best way to achieve that is by using the aggregation framework:

db.collection.aggregate(
    {$unwind : "$ccNameCount"},
    {$match : {
                "ccNameCount.ccName" : "ATCC"
    }},
    {$group: {
        _id : "$ccNameCount.paperCount"
    }},
    {$group: {
        _id : "count",
        total : {"$sum" : 1}
    }}
);

This will give you the distinct count of paperCount for ccName="ATCC" in the results total field.

Also, If you need to know each of the distinct values besides the total count you can use the $addToSet operator:

db.collection.aggregate(
    {$unwind : "$ccNameCount"},
    {$match : {
                "ccNameCount.ccName" : "ATCC"
    }},
    {$group: {
        _id : "$ccNameCount.paperCount"
    }},
    {$group: {
        _id : "count",
        total : {"$sum" : 1},
        distinctValues : {$addToSet : "$_id"}
    }}
);

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.