Counting results in aggregate selection - mongodb

Question

My MongoDB database have a structure
{
"_id" : ObjectId("5c1ccc20fc0f60769227d455"),
"type" : 0,
"id" : "hwJyzAHyfjXUlrGhblT7txWd",
"userowner" : 1.0,
"campid" : "9548",
"date" : 1545391136,
"useragent" : "mozilla/5.0 (windows nt 10.0; win64; x64; rv:65.0) gecko/20100101 firefox/65.0",
"domain" : "",
"referer" : "",
"country" : "en",
"language" : "en-US",
"languages" : [
"en-US",
"en"
],
"screenres" : [
"1920*1080"
],
"avscreenres" : [
"1080*1858"
],
"webgl" : "angle (nvidia geforce gtx 1060 6gb direct3d11 vs_5_0 ps_5_0)",
"hash" : 123,
"timezone" : -180,
"result" : true,
"resultreason" : "learning",
"remoteip" : "0.0.0.0"
}
Every a document have a vield "result" with a bool value.
I make aggregation selection:
db.getCollection('clicks').aggregate([
{ $match: {userowner: 1, date:{$gte: 0, $lte: 9545392055}} },
{ $group : {_id : "$campid",
number: {$sum: 1}}}
])
and get a Result:
/* 1 */
{
"_id" : "4587",
"number" : 2.0
}
/* 2 */
{
"_id" : "9548",
"number" : 1346.0
}
How can count the amount of value "true" and "false" in a field "result" and get a result like this:
/* 1 */
{
"_id" : "4587",
"number" : 2.0,
"passed":100,
"blocked":120
}
/* 2 */
{
"_id" : "9548",
"number" : 1346.0,
"passed":100,
"blocked":120
}

Answer 1

I hope this works as per your requirement.
db.getCollection('clicks').aggregate(
[
{
$match: {
userowner: 1, date: {
$gte: 0, $lte: 9545392055
}
}
},
{
$group: {
_id: "$campid", passed: {
$sum: {
$cond:
[
{ $eq: ["$result", true] },
1, 0
]
}
},
blocked: {
$sum: {
$cond:
[
{
$eq: ["$result", false]
}
, 1, 0]
}
},
number: { $sum: 1 }
}
},
{
$project: {
_id: 0,
campid: "$_id",
number: 1,
passed: 1,
blocked: 1
}
}
])
Output:-
{
"passed" : 3,
"blocked" : 2,
"number" : 5,
"campid" : "4587"
}
{
"passed" : 2,
"blocked" : 1,
"number" : 3,
"campid" : "9548"
}
Refer $group, $cond, and $eq for more info.

Answer 2

With MongoDb 3.6 and newer, you can leverage the use of $arrayToObject operator within a $replaceRoot pipeline to get the desired result.
You would need to group the documents intially by the campid and the result field, aggregate the sum and pass the results to yet another group pipeline stage. This group stage will prepare the documents in a way that $arrayToObject operator will give you the desired object by creating a key-value array using $push.
The result from this is then fed to the $replaceRoot pipeline to bring the passed and blocked fields to the root of the document.
The following aggregate pipeline describes the above:
db.getCollection('clicks').aggregate([
{ "$match": { "userowner": 1, "date": { "$gte": 0, "$lte": 9545392055 } } },
{ "$group": {
"_id": {
"campid": "$campid",
"result": { "$cond": [ "$result", "passed", "blocked" ] }
},
"count": { "$sum": 1 }
} },
{ "$group": {
"_id": "$_id.campid",
"number": { "$sum": "$count" },
"counts": {
"$push": {
"k": "$_id.result",
"v": "$count"
}
}
} },
{ "$replaceRoot": {
"newRoot": {
"$mergeObjects": [
{ "$arrayToObject": "$counts" },
"$$ROOT"
]
}
} },
{ "$project": { "counts": 0 } }
])