MongoDB 按多个字段分组，不会相互影响结果

【问题标题】：MongoDB group by multiple fields in a way to not affect each other resultMongoDB 按多个字段分组，不会相互影响结果
【发布时间】：2018-07-17 07:57:57
【问题描述】：

我有以下查询，我想要的是拥有一组自定义组字段名称和字段值。

db.getCollection('mycollection').aggregate([
  {"$match":{
     "expireDate":{"$gte":"2018-02-06T00:00:00.000Z"},
     "publishDate":{"$lte":"2018-02-06T00:00:00.000Z"},
     "isPublished":true,"isDrafted":false,
     "deletedAt":{"$eq":null},"deleted":false
  }},
  {"$group":{
     "twentyFourHourAgo":{
        "$sum":{
           "$cond":[
             {"$gt":["$publishDate","2018-02-04T08:48:16.892Z"]},1,0
           ]
        }
      },
      "fortyEightHourAgo":{
        "$sum":{
            "$cond":[
               {"$gt":["$publishDate","2018-02-01T08:48:16.892Z"]},1,0
            ]
        }
      },
      "thirtyDaysAgo":{
         "$sum":{
            "$cond":[
               {"$gt":["$publishDate","2017-12-31T08:48:16.892Z"]},1,0
            ]
         }
      },
      "_id":{
        "position":{"$ifNull":["$position","Unknown"]},
        "workType":{"$ifNull":["$workType","Unknown"]},
        "functionalArea":{"$ifNull":["$functionalArea","Unknown"]},
        "minimumEducation":{"$ifNull":["$minimumEducation","Unknown"]},
        "gender":{"$ifNull":["$gender","Unknown"]},
        "contractType":{"$ifNull":["$contractType","Unknown"]},
        "locations":{"$ifNull":["$locations","Unknown"]},
        "requiredLanguages":{"$ifNull":["$requiredLanguages","Unknown"]},
        "company":{"$ifNull":["$company.name","Unknown"]}},"count":{"$sum":1}
     }
 },
 {"$group":{
     "_id":null,
     "twentyFourHourAgo":{
        "$sum":"twentyFourHourAgo"
     },
     "fortyEightHourAgo":{
        "$sum":"$fortyEightHourAgo"
     },
     "thirtyDaysAgo":{
        "$sum":"$thirtyDaysAgo"
     },
     "position":{"$addToSet":{"Name":"$_id.position","Count":"$count"}},
     "workType":{"$addToSet":{"Name":"$_id.workType","Count":"$count"}},
     "functionalArea":{
        "$addToSet":{"Name":"$_id.functionalArea","Count":"$count"}
     },
     "minimumEducation":{
        "$addToSet":{"Name":"$_id.minimumEducation","Count":"$count"}
     },
     "gender":{"$addToSet":{"Name":"$_id.gender","Count":"$count"}},"contractType":{"$addToSet":{"Name":"$_id.contractType","Count":"$count"}},"locations":{"$addToSet":{"Name":"$_id.locations","Count":"$count"}},"requiredLanguages":{"$addToSet":{"Name":"$_id.requiredLanguages","Count":"$count"}},"company":{"$addToSet":{"Name":"$_id.company","Count":"$count"}}}}]  
)

我在集合模式中的文档是这样的：

/* 1 */
{
    "_id" : ObjectId("59e4540bf14f1607b90ffb81"),
    "vacancyNumber" : "1",
    "position" : "Software Tester",
    "publishDate" : ISODate("2018-01-02T00:00:00.000Z"),
    "expireDate" : ISODate("2018-05-29T00:00:00.000Z"),
    "yearsOfExperience" : 40,
    "minimumEducation" : "Doctorate",
    "functionalArea" : "Education",
    "company" : {
        "id" : ObjectId("59e453fbf14f1607b90ffb80"),
        "name" : "First Company",
        "profile" : "profile",
        "logo" : {
            "container" : "companyFiles",
            "name" : "abbbff58cd3fda2c59ab2ee620ea5aa0",
            "mime" : ".png",
            "size" : 5806
        }
    },
    "durations" : {
        "years" : 3,
        "months" : 4
    },
    "probationPeriod" : {
        "duration" : 34,
        "unit" : "month"
    },
    "salary" : {
        "minSalary" : 1000,
        "maxSalary" : 2000,
        "currency" : "USD",
        "period" : "monthly",
        "isNegotiable" : true
    },
    "locations" : [ 
        "Germany", 
        "Itly", 
        "Iran"
    ],
    "canApplyOnline" : true,
    "skills" : [ 
        "Skill1", 
        "Skill2", 
        "Skill3", 
        "Skill4"
    ],
    "requiredLanguages" : [ 
        "Arabic", 
        "English", 
        "Russian", 
        "Dari", 
        "French"
    ],
    "keywords" : [ 
        "Key1", 
        "Key2"
    ],
    "deleted" : false,
    "deletedAt" : null,
    "isDrafted" : false,
    "isPublished" : true,
    "requiresTravel" : true,
    "gender" : "male",
    "nationalities" : [ 
        "afghan"
    ],
    "workType" : "Full Time",
    "contractType" : "Permanent",
}

/* 2 */
{
    "_id" : ObjectId("59f9402e05d04ebe5653d98f"),
    "vacancyNumber" : "1",
    "position" : "Software Engineer",
    "publishDate" : ISODate("2018-01-03T00:00:00.000Z"),
    "expireDate" : ISODate("2018-11-10T00:00:00.000Z"),
    "yearsOfExperience" : 40,
    "minimumEducation" : "Doctorate",
    "functionalArea" : "Education",
    "company" : {
        "id" : ObjectId("59e453fbf14f1607b90ffb80"),
        "name" : "First Company",
        "profile" : "profile",
        "logo" : {
            "container" : "logo container",
            "name" : "logo name",
            "mime" : "logo mime type",
            "size" : 1
        }
    },    
    "durations" : {
        "years" : 3,
        "months" : 4
    },
    "probationPeriod" : {
        "duration" : 34,
        "unit" : "month"
    },
    "salary" : {
        "minSalary" : 1000,
        "maxSalary" : 2000,
        "currency" : "USD",
        "period" : "monthly",
        "isNegotiable" : true
    },
    "locations" : [ 
        "Afghanistan", 
        "Itly", 
        "Iran"
    ],
    "skills" : [ 
        "Skill1", 
        "Another Skill"
    ],
    "requiredLanguages" : [ 
        "Arabic", 
        "English", 
        "Russian", 
        "Dari", 
        "French"
    ],
    "keywords" : [ 
        "Keyword", 
        "Key1"
    ],
    "deleted" : false,
    "deletedAt" : null,
    "isDrafted" : false,
    "isPublished" : true,
    "gender" : "male",
    "nationalities" : [ 
        "afghan", 
        "iranian"
    ],
    "workType" : "Full Time",
    "contractType" : "Short-Term",
}

/* 3 */
{
    "_id" : ObjectId("5a03235234f7504f13970abd"),
    "vacancyNumber" : "1",
    "position" : "Software Tester",
    "publishDate" : ISODate("2017-10-10T00:00:00.000Z"),
    "expireDate" : ISODate("2018-11-25T00:00:00.000Z"),
    "yearsOfExperience" : 40,
    "minimumEducation" : "Doctorate",
    "functionalArea" : "IT Software",
    "company" : {
        "id" : ObjectId("59e453fbf14f1607b90ffb80"),
        "name" : "My First Company",
        "profile" : "profile",
        "logo" : {
            "container" : "logo container",
            "name" : "logo name",
            "mime" : "logo mime type",
            "size" : 1
        }
    },
    "durations" : {
        "years" : 3,
        "months" : 4
    },
    "probationPeriod" : {
        "duration" : 34,
        "unit" : "month"
    },
    "salary" : {
        "minSalary" : 1000,
        "maxSalary" : 2000,
        "currency" : "USD",
        "period" : "monthly",
        "isNegotiable" : true
    },
    "locations" : [ 
        "Germany", 
        "Itly", 
        "Iran"
    ],
    "skills" : [ 
        "Skill1", 
        "Test Skill"
    ],
    "requiredLanguages" : [ 
        "Arabic", 
        "English", 
        "Russian", 
        "Dari", 
        "French"
    ],
    "keywords" : [ 
        "Test Key", 
        "Keyword"
    ],
    "deleted" : false,
    "deletedAt" : null,
    "isDrafted" : false,
    "isPublished" : true,
    "gender" : "female",
    "nationalities" : [ 
        "afghan"
    ],
    "workType" : "Part Time",
    "contractType" : "Permanent",
}

现在我想通过自定义表达式检查“twentyFourHourAgo、fortyEightHourAgo、30DaysAgo”以及字段的值 (functionalArea, position, locations, keywords, workType) 来计算数据组。

我当前的查询结果是

{
    "_id" : null,
    "twentyFourHourAgo" : 0,
    "fortyEightHourAgo" : 0.0,
    "thirtyDaysAgo" : 2.0,
    "position" : [ 
        {
            "Name" : "Software Engineer",
            "Count" : 1.0
        }, 
        {
            "Name" : "Software Tester",
            "Count" : 1.0
        }
    ],
    "workType" : [ 
        {
            "Name" : "Full Time",
            "Count" : 1.0
        }, 
        {
            "Name" : "Part Time",
            "Count" : 1.0
        }
    ],
    "functionalArea" : [ 
        {
            "Name" : "Education",
            "Count" : 1.0
        }, 
        {
            "Name" : "IT Software",
            "Count" : 1.0
        }
    ],
    "minimumEducation" : [ 
        {
            "Name" : "Doctorate",
            "Count" : 1.0
        }
    ],
    "gender" : [ 
        {
            "Name" : "male",
            "Count" : 1.0
        }, 
        {
            "Name" : "female",
            "Count" : 1.0
        }
    ],
    "contractType" : [ 
        {
            "Name" : "Short-Term",
            "Count" : 1.0
        }, 
        {
            "Name" : "Permanent",
            "Count" : 1.0
        }
    ],
    "locations" : [ 
        {
            "Name" : [ 
                "Afghanistan", 
                "Itly", 
                "Iran"
            ],
            "Count" : 1.0
        }, 
        {
            "Name" : [ 
                "Germany", 
                "Itly", 
                "Iran"
            ],
            "Count" : 1.0
        }
    ],
    "requiredLanguages" : [ 
        {
            "Name" : [ 
                "Arabic", 
                "English", 
                "Russian", 
                "Dari", 
                "French"
            ],
            "Count" : 1.0
        }
    ],
    "company" : [ 
        {
            "Name" : "First Company",
            "Count" : 1.0
        }, 
        {
            "Name" : "My First Company",
            "Count" : 1.0
        }
    ]
}

如您所见，我有三个具有以下属性的文档：

  1. 两个文档具有相同的位置Software Tester，但查询返回1 Software Tester（这意味着如果我有多个文档在特定列中有一些共同的值，它们的计数结果是错误的）。其他字段“contractType, workType, etc...”也存在同样的问题。

  2. 在locations等数组类型字段中，我的第一个文档在locations数组中有Germany, Italy, Iran值，我的第二个文档有Afghanistan, Italy, Iran，我的第三个文档有Germany, Italy, Iran。但是查询结果是这样的：

    "locations" : [ 
  {
    "Name" : [ 
      "Afghanistan", 
      "Itly", 
      "Iran"
    ],
    "Count" : 1.0
  }, 
  {
    "Name" : [ 
      "Germany", 
      "Itly", 
      "Iran"
    ],
    "Count" : 1.0
  }
 ],

这应该是：Germany => 2, Italy,Iran => 3, and Afghanistan => 1 其他数组类型字段也存在同样的问题。

【问题讨论】：

  • 您需要在$group 之前的每个数组元素上使用$unwind。这将使您能够计算每个数组中的所有元素
  • @N3i1 我已经尝试过了，但是对于 $unwind，其他字段会影响并返回其他字段的错误结果，我已经尝试过 locations 字段。
  • 我在您上面的 OP 中看不到正在使用的 $unwind。请看我下面的答案

标签： mongodb mongoose aggregation-framework


【解决方案1】：

抱歉，我之前误解了您的问题。为了能够$unwindlocation 数组，但不影响您的twentyFourHourAgo 等，您可以查看使用$first。

如果您希望对单个元素进行计数/求和，则需要 $unwind 任何数组。

$first的使用示例。

db.getCollection('foo').aggregate([
{ $unwind : "$locations" },
{ "$group" : { "_id" : "$_id", 
        "twentyFourHourAgo":{ $first : {
            "$sum" : { "$cond":[
                {"$gt":["$publishDate", ISODate("2016-10-10T00:00:00.000Z")]},1,0 ] } } },
        "fortyEightHourAgo" : { $first : {
            "$sum" : { "$cond" : [ 
                { "$gt" : [ "$publishDate","2018-01-02T00:00:00.000Z"]},1,0 ] } } },
        "thirtyDaysAgo" : { $first : { 
            "$sum" : { "$cond" : [
                { "$gt" : [ "$publishDate","2017-12-31T08:48:16.892Z"]},1,0 ] } } },
        } },
{ "$group" : { "_id" : null,
     "twentyFourHourAgo" : { "$sum" : "$twentyFourHourAgo" },
     "fortyEightHourAgo" : { "$sum" : "$fortyEightHourAgo" },
     "thirtyDaysAgo" : { "$sum" : "$thirtyDaysAgo" },
 }}
])

输出：

"_id" : null,
"twentyFourHourAgo" : 0,
"fortyEightHourAgo" : 3.0,
"thirtyDaysAgo" : 3.0,

请参阅此处$first，了解我认为它可能有用的更多信息。我在开头贴了$unwind，以帮助证明它可以解决您的 OP 中的问题。

【讨论】：

  • 如您所见，此更改导致其他字段，我有三个文档，但结果类似于"twentyFourHourAgo" : 0, "fortyEightHourAgo" : 9.0, "thirtyDaysAgo" : 9.0,
  • 对不起，我错过了。我试图强调$unwind 的必要性，以便解决您的count 问题！您是否尝试过$projection 来投影twentyFourHourAgo、fortyEightHourAgo 和thirtyDaysAgo 的值，以免干扰您的分组？
  • 是的，我试过了，但我无法在群组中使用它。
猜你喜欢
  • 2023-03-18
  • 1970-01-01
  • 2014-05-20
  • 1970-01-01
  • 1970-01-01
  • 1970-01-01
  • 2018-09-19
相关资源
最近更新 更多
热门标签
Java Python linux javascript Mysql C# Docker 算法 前端 SpringBoot Redis Vue spring 设计模式 .net core .net kubernetes c++ 数据库 数据结构 大数据 js 机器学习 微服务 Android Go 程序员 面试 JVM ASP.net core 云原生 人工智能 后端 PHP git CSS golang k8s Nginx Django mybatis 深度学习 多线程 React 架构 devops 爬虫 云计算 Spring Boot LeetCode