Amazon Document DB中的Mongodb聚合功能 插入文档查询输出

问题描述

我在运行于Amazon document DB的mongodb中有以下文档，即mongodb 3.6

{"item":"journal","instock":[{"warehouse":"A","qty":[2,4,5] },{"warehouse":"C","qty":[8,5,2] },{"warehouse":"F","qty":[3] },{"warehouse":"K",8] },{"warehouse":"P","qty":[3,7,9] } ] }

使用此数据，我只想提取数量为5的库存对象。 我在本地mongodb（4.2）上运行的查询如下：

db.inventory.aggregate({$match:{"item": "journal"}},{$project:{instock: {$filter: {input: "$instock",cond: {$in: [5,"$$this.qty"] } } } }})

给出预期结果：

{ "_id" : ObjectId("5f59a6bd85a366711c35fc7b"),"instock" : [ { "warehouse" : "A","qty" : [ 2,5 ] },{ "warehouse" : "C","qty" : [ 8,2 ] } ] }

但是问题是当我在DocumentDB中运行相同的查询时，如果给出错误： “ errmsg”：“ $ in需要一个数组作为第二个参数，找到：string”，

根据mongodb文档，自版本3.2起就有$ filter操作。所以我不太确定这是什么问题。有什么想法吗？

解决方法

我正在研究为什么此查询无法针对Amazon DocumentDB使用。同时，这是一个重写的查询，可在Amazon DocumentDB上运行，并且应执行原始查询试图执行的操作：

插入文档

db.coll.insert({
    "item": "journal","instock": [{
        "warehouse": "A","qty": [2,4,5]
    },{
        "warehouse": "C","qty": [8,5,2]
    },{
        "warehouse": "F","qty": [3]
    },{
        "warehouse": "K",8]
    },{
        "warehouse": "P","qty": [3,7,9]
    }]
});

查询

db.coll.aggregate([
    { $match: { "item": "journal" } },{ $unwind: "$instock" },{ $match: { "instock.qty": 5 } },{
    "$group": {
        "_id": { "id": "$id","item": "$item" },"instock": {
            "$push": {
                "warehouse": "$instock.warehouse","qty": "$instock.qty"
            }
        }
    }
}]).pretty()

输出

{ 
    "_id" : {
        "item" : "journal"
    },"instock" : [
        {
            "warehouse" : "A","qty" : [
                2.0,4.0,5.0
            ]
        },{
            "warehouse" : "C","qty" : [
                8.0,5.0,2.0
            ]
        }
    ]
}