寻找在mongodb 4.2.x上工作的替代结果投影

Question

根据我的上一个问题(按多个范围查询文档计数，返回具有匹配元素计数的范围开始/结束(，我构建了一个查询来检查多个可能重叠的日期范围中的文档计数。

该查询在MongoDB 4.4上运行，但我也需要在4.2上运行它。在MongoDB 4.2上，我得到以下错误：

Mongo Server error (MongoCommandException): Command failed with error 168 (InvalidPipelineOperator): 'Unrecognized expression '$first'' on server localhost:27017. 
The full response is:
{ 
"ok" : 0.0, 
"errmsg" : "Unrecognized expression '$first'", 
"code" : 168.0, 
"codeName" : "InvalidPipelineOperator"
}

您将如何编写聚合投影以实现相同的结果结构。这是带有数据设置的完整代码

db.createCollection("object_location_tracking");
db.getCollection("object_location_tracking").insertMany([
{
_id: "1",
locationId: "locationA",
objectId: "objectA",
timestamp: ISODate("2020-01-01T00:00:00Z")
},
{
_id: "2",
locationId: "locationB",
objectId: "objectA",
timestamp: ISODate("2020-01-01T00:00:00Z")
},
{
_id: "3",
locationId: "locationA",
objectId: "objectB",
timestamp: ISODate("2019-01-01T00:00:00Z")
},
{
_id: "4",
locationId: "locationB",
objectId: "objectB",
timestamp: ISODate("2020-01-01T00:00:00Z")
}
]);

db.getCollection("object_location_tracking").aggregate([
{$facet: {
"first_bucket_id": [
{$match: {"objectId":"objectA",
"locationId":"locationA",
"timestamp": {$gte: new ISODate('2020-01-01'),
$lt: new ISODate('2020-12-31')}
}},
{$count: "N"}
],
"second_bucket_id": [
{$match: {"objectId":"objectA",
"locationId":"locationA",
"timestamp": {$gte: new ISODate('2020-01-01'),
$lt: new ISODate('2022-12-31')}
}},
{$count: "N"}
],
"third_bucket_id": [
{$match: {"objectId":"objectA",
"locationId":"locationB",
"timestamp": {$gte: new ISODate('2022-01-01'),
$lt: new ISODate('2022-12-31')}
}},
{$count: "N"}
]
}},
{
$set: {
first_bucket_id: { $first: "$first_bucket_id.N"},
second_bucket_id: { $first: "$second_bucket_id.N"},
third_bucket_id: { $first: "$third_bucket_id.N"}
}
}
, {
$project: {
first_bucket_id: 1,
second_bucket_id: 1,
third_bucket_id: 1

}
}
]);

Answer 1

您使用的是第一个数组元素，可以看出这是为版本4.4 添加的新运算符

幸运的是，这很容易克服，只需使用$arrayElemAt，就像这样：

{
$set: {
first_bucket_id: {$arrayElemAt: ["$first_bucket_id.N", 0]},
second_bucket_id: {$arrayElemAt: ["$second_bucket_id.N", 0]},
third_bucket_id: {$arrayElemAt: ["$third_bucket_id.N", 0]}
}
}