mongoose 查询返回按单个字段分组的文档的最新版本答案

【问题标题】：mongoose query to return most recent versions of documents grouped by single fieldmongoose 查询返回按单个字段分组的文档的最新版本
【发布时间】：2016-12-14 19:34:12
【问题描述】：

我有这些文件，

{
    "_id" : 1,
    "fId" : "test",
    "pId" : "test",
    "url" : "A",
    "subfolder" : "a"
}

,
{
    "_id" : 3,
    "fId" : "test",
    "pId" : "test", 
    "url" : "A",
    "subfolder" : "b"
}

,
{
    "_id" : 5,
    "fId" : "test",
    "pId" : "test"
    "url" : "A",
    "subfolder" : "c"
}

,
{
    "_id" : 2,
    "fId" : "test",
    "pId" : "test",
    "url" : "B",
    "subfolder" : "a"
}

,
{
    "_id" : 4,
    "fId" : "test",
    "pId" : "test",
    "url" : "B",
    "subfolder" : "b"
}

,
{
    "_id" : 7,
    "fId" : "test",
    "pId" : "test",
    "url" : "A",
    "subfolder" : "d"
}

,
{
    "_id" : 8,
    "fId" : "test",
    "pId" : "test",
    "url" : "B",
    "subfolder" : "d"
}

,
{
    "_id" : 9,
    "fId" : "test",
    "pId" : "test",
    "url" : "A",
    "subfolder" : "e"
}

,
{
    "_id" : 10,
    "fId" : "test",
    "pId" : "test",
    "url" : "B",
    "subfolder" : "e"
}

,
{
    "_id" : 6,
    "fId" : "test",
    "pId" : "test",
    "url" : "B",
    "subfolder" : "c"
}

如何编写 mongoose 查询以根据 _id 对这些文档进行排序，按子文件夹分组，并返回最近的 5 个组。

预期结果是

group 1:
[{
    "_id" : 10,
    "fId" : "test",
    "pId" : "test",
    "url" : "B",
    "subfolder" : "e"
},
{
    "_id" : 9,
    "fId" : "test",
    "pId" : "test",
    "url" : "A",
    "subfolder" : "e"
}]
,
group 2:
[{
    "_id" : 8,
    "fId" : "test",
    "pId" : "test",
    "url" : "B",
    "subfolder" : "d"
}
,
{
    "_id" : 7,
    "fId" : "test",
    "pId" : "test",
    "url" : "A",
    "subfolder" : "d"
}]
,

group 3:
[{
    "_id" : 6,
    "fId" : "test",
    "pId" : "test",
    "url" : "B",
    "subfolder" : "c"
}
,
{
    "_id" : 5,
    "fId" : "test",
    "pId" : "test"
    "url" : "A",
    "subfolder" : "c"
}]
,
group 4:
[{
    "_id" : 4,
    "fId" : "test",
    "pId" : "test",
    "url" : "B",
    "subfolder" : "b"
}
,
{
    "_id" : 3,
    "fId" : "test",
    "pId" : "test", 
    "url" : "A",
    "subfolder" : "b"
}]
,

group 5:
[{
    "_id" : 2,
    "fId" : "test",
    "pId" : "test",
    "url" : "B",
    "subfolder" : "a"
}
,
{
    "_id" : 1,
    "fId" : "test",
    "pId" : "test",
    "url" : "A",
    "subfolder" : "a"
}]

我尝试了一个查询

model.aggregate(
   [      
     {$match : {fId : "test" ,  pId : "test"}},
     {$group : {_id : "$subfolder", recording: { $push: "$$ROOT" } }},     
     {$skip : 5*(pageNumber-1)},
     {$sort : {"_id": -1}},    
     {$limit : 5 }
   ]
  )

我得到随机的5组，不是最近的，结果也不是按照_id排序的。

【问题讨论】：

标签： javascript node.js mongodb mongoose

【解决方案1】：

您在这里误解了跳过和限制的使用。试试这个：

model.aggregate(
   [      
     {$match : {fId : "test" ,  pId : "test"}},
     {$group : {_id : "$subfolder", recording: { $push: "$$ROOT" } }},     
     {$sort : {"recording._id": -1}},
     {$limit: 5}
   ]
  )

提供示例文档的输出：

       {
    "_id" : "5",
    "recording" : [
        {
            "_id" : 9,
            "fId" : "test",
            "pId" : "test",
            "url" : "A",
            "subfolder" : "5"
        },
        {
            "_id" : 10,
            "fId" : "test",
            "pId" : "test",
            "url" : "B",
            "subfolder" : "5"
        }
    ]
}
{
    "_id" : "4",
    "recording" : [
        {
            "_id" : 7,
            "fId" : "test",
            "pId" : "test",
            "url" : "A",
            "subfolder" : "4"
        },
        {
            "_id" : 8,
            "fId" : "test",
            "pId" : "test",
            "url" : "B",
            "subfolder" : "4"
        }
    ]
}
{
    "_id" : "3",
    "recording" : [
        {
            "_id" : 5,
            "fId" : "test",
            "pId" : "test",
            "url" : "A",
            "subfolder" : "3"
        },
        {
            "_id" : 6,
            "fId" : "test",
            "pId" : "test",
            "url" : "B",
            "subfolder" : "3"
        }
    ]
}
{
    "_id" : "2",
    "recording" : [
        {
            "_id" : 3,
            "fId" : "test",
            "pId" : "test",
            "url" : "A",
            "subfolder" : "2"
        },
        {
            "_id" : 4,
            "fId" : "test",
            "pId" : "test",
            "url" : "B",
            "subfolder" : "2"
        }
    ]
}
{
    "_id" : "1",
    "recording" : [
        {
            "_id" : 1,
            "fId" : "test",
            "pId" : "test",
            "url" : "A",
            "subfolder" : "1"
        },
        {
            "_id" : 2,
            "fId" : "test",
            "pId" : "test",
            "url" : "B",
            "subfolder" : "1"
        }
    ]
}

【讨论】：

@Sabreena 它适用于您在问题中提供的数据...查看编辑
分组很好，但我需要根据文档的 _id 排序的组。也就是说，我需要最近的两个文档，其子文件夹名称与组 1 相同。依此类推（ _id : 10 是最近的文件）
{$match : {fId : "test" , pId : "test"}}, {$group : {_id : "$subfolder", 录音: { $push: "$$ROOT" } }}, {$sort : {"recording._id": -1}}, {$skip : 5*(pageNumber-1)}, {$limit: 5}
感谢 felix，它工作得很好！！！我添加了 $skip 以向我的应用添加分页。