Mongodb Aggregate nested subdocuments

Question

Having list of documents

[
{
    "__v" : 21,
    "_id" : ObjectId("546330dbb8926d177052e9ff"),
    "code" : "WfvCc",
    "description" : "",
    "elements" : [ 
        {
            "_id" : ObjectId("546471f61e13b76a0b20ccaf"),
            "comments" : [],
            "meta" : {
                "createdBy" : "545ab39ef1b0c88a695fcf8d",
                "modifiedAt" : "1415868918045",
                "createdAt" : "1415868918045"
            },
            "title" : "awesome title",
            "votes" : {
                "count" : 3,
                "meta" : [ 
                    {
                        "createdBy" : "545ab39ef1b0c88a695fcf8d",
                        "_id" : ObjectId("546473831e13b76a0b20ccb7"),
                        "createdAt" : "1415869315618"
                    }, 
                    {
                        "createdBy" : "545aaddcf1b0c88a695fcf84",
                        "_id" : ObjectId("546473d71e13b76a0b20ccbc"),
                        "createdAt" : "1415869399584"
                    }, 
                    {
                        "createdBy" : "5461c0e2c9c39a192c44226c",
                        "_id" : ObjectId("546474041e13b76a0b20ccbe"),
                        "createdAt" : "1415869444056"
                    }
                ]
            }
        }
    ]
},
{
    "__v" : 21,
    "_id" : ObjectId("546330dbb8926d177052e9ff"),
    "code" : "WfvCc",
    "description" : "",
    "elements" : [ 
        {
            "_id" : ObjectId("546471f61e13b76a0b20ccaf"),
            "comments" : [],
            "meta" : {
                "createdBy" : "545ab39ef1b0c88a695fcf8d",
                "modifiedAt" : "1415868918045",
                "createdAt" : "1415868918045"
            },
            "title" : "awesome title",
            "votes" : {
                "count" : 3,
                "meta" : [ 
                    {
                        "createdBy" : "545ab39ef1b0c88a695fcf8d",
                        "_id" : ObjectId("546473831e13b76a0b20ccb7"),
                        "createdAt" : "1415869315618"
                    }, 
                    {
                        "createdBy" : "545aaddcf1b0c88a695fcf84",
                        "_id" : ObjectId("546473d71e13b76a0b20ccbc"),
                        "createdAt" : "1415869399584"
                    }, 
                    {
                        "createdBy" : "5461c0e2c9c39a192c44226c",
                        "_id" : ObjectId("546474041e13b76a0b20ccbe"),
                        "createdAt" : "1415869444056"
                    }
                ]
            }
        }
    ]
}
]

I would like to aggregate the list of users aka. elements.votes.meta.createdBy across the documents and count the total numbers of occurrences across the document. *Note that the elements.votes.meta.createdBy is unique per document, so in theory this should make is simpler.

So far, I've ended up with a query:

db.sessions.aggregate(
  { $project: { 
      meta: "$elements.votes.meta"
  }},
  { $unwind: "$meta" },
  { $group: {
      _id: "voters",
      voters: {
          $addToSet: "$meta.createdBy"
      }
  }}
)

Just to get completely stuck again. I know I need a double grouping, just can't seem to be able to figure it out. Any help appreciated.

marcinn · Accepted Answer · 2015-02-07 22:59:57Z

14

First you should get the totals for each 'user'. (i.e. { $group: {_id: '$user', count: {'$sum': 1} }})

Then just group by null to create a document with the result, add each user to set and push the result from first grouping onto array field. (2nd grouping)

db.test5.aggregate(
  { $unwind: "$elements" },
  { $unwind: "$elements.votes.meta" },
  { $project: {_id: '$_id', user: '$elements.votes.meta.createdBy'} },
  { $group: {_id: '$user', count: {'$sum': 1} }},
  { $group: {
        _id: null, 
        users: {$addToSet: '$_id'}, 
        occurances: {$push: {'user': '$_id', count: '$count'}}
        }
   }
)

Result:

{
    "result" : [ 
        {
            "_id" : null,
            "users" : [ 
                "545ab39ef1b0c88a695fcf8d", 
                "545aaddcf1b0c88a695fcf84", 
                "5461c0e2c9c39a192c44226c"
            ],
            "occurances" : [ 
                {
                    "user" : "5461c0e2c9c39a192c44226c",
                    "count" : 2
                }, 
                {
                    "user" : "545aaddcf1b0c88a695fcf84",
                    "count" : 2
                }, 
                {
                    "user" : "545ab39ef1b0c88a695fcf8d",
                    "count" : 2
                }
            ]
        }
    ],
    "ok" : 1
}

answered Feb 7, 2015 at 22:59

marcinn

1,78611 silver badges13 bronze badges

Sign up to request clarification or add additional context in comments.

1 Comment

Jacek Dominiak Over a year ago

awesome, with slight modification it is exactly what I was looking for. The key was to double $unwind the set. Cheers!

Collectives™ on Stack Overflow

Mongodb Aggregate nested subdocuments

1 Answer 1

1 Comment

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

1 Comment

Your Answer

Sign up or log in

Post as a guest

Linked

Related