按MongoDB中具有相等字段的文档数进行排序/分组

Question

What I want to achieve is grouping these documents,

{voteid: A, userid: x}, {voteid: C, userid: x},
{voteid: A, userid: y}, {voteid: A, userid: z},
{voteid: C, userid: z}, {voteid: D, userid: z}

in objects with number of votes (users are irreleveant here) and also sorting them based on voteid :

{voteid: A, votes: 3},
{voteid: C, votes: 2},
{voteid: D, votes: 1}

How should I approach that?

I'm working with MongoDB 3.0 Node.js driver

Answer 1

Okay, so this will help you do your query (count) and sort them for voteid ,

db.yourCollection.aggregate([{
     $group: {
        _id: '$voteid',
        votes: {$sum: 1}
     }
   },
   {
     $sort: {
       _id: 1
     }
   },
   {
     $project: {
      _id: 0,
      voteid: '$_id',
      votes: 1
     }
   }
])

What we are doing here is first apply $group operator on the records and find all the occurences of the voteid by assigining it to _id as accumulator object and subsequently storing count values -- {$sum: 1} in votes .

Then, we sort based on _id we just created with votesid and then project the result in the form we needed -- eliminating _id and remapping the previous group _id to voteid and allowing votes in our result.

Hope this helps.