One MongoDB statistic requirement

Demand:

A colleague from the off-duty operation sent a requirement to count the number of occurrences of a field (_ID) in all document in a collection in the database. The range of the _id field is 0-4000.

Suppose collection is like this:

{    123},{    456},{    _id:123}

So the end result: [{_id:123, value:2}, {_id:456, value:1}] is actually a count frequency operation

Scheme:

Rough methods can traverse all document and accumulate, but the better way is to use MongoDB's MapReduce operation: https://docs.mongodb.com/manual/core/map-reduce/

db.collection.mapReduce (    function() {Emit (this. _id, 1)},      function(key, values) {return  array.sum (values);},    1}}    )

It is worth mentioning that emit, value is 1, so that each document is a 1, then reduce when we get a full 1 array.

Next, using the method mentioned by http://www.cnblogs.com/agentgamer/p/4994650.html, can be inherited into the shell, the standard output to other tools to use, the output of the inline is basically a JSON.

One MongoDB statistic requirement