MongoDB compass - 从文档中的 3 个字段及其值中获取具有最大值的字段名称（键）

Question

我有这个示例 mongodb 文档 -

{
    _id: 5db85ee97d9fb13ead4fc54c
    applId: 5d48f34f7d9fb10ce171f905
    fileId: "dd386cf7-4139-45c2-9853-cbb126621b51"
    job: Object
    country: "US"
    fullName: "abcd xyz"
    htmlWordCount: 2766
    textWordCount: 1867
    rchilliTextWordCount: 2840
    deleted: 0
    dateEntered: 2019-10-29 15:46:49.237
    dateModified: 2019-10-29 15:46:49.237
}

我想在 compass 中构建一个查询，以便在输出中有以下字段 -

{
    _id: 5db85ee97d9fb13ead4fc54c
    country: "US"
    fullName: "abcd xyz"
    htmlWordCount: 2766
    textWordCount: 1867
    rchilliTextWordCount: 2840
    winner: "rchilliTextWordCount"
}

请注意，它有一个名为“winner”的新字段，它始终返回字数最多的列（在 3 个"htmlWordCount", "textWordCount", "rchilliTextWordCount" 列中）。这个新列"winner" 将在运行时查询时生成。此查询也在country = "US" 上过滤。

如何在 MongoDB Compass 中执行此操作，或者聚合管道应该是什么样的？

Answer 1

您可以使用$switch 或$cond

db.collection.aggregate([
  {
    $match: {
      country: "US"
    }
  },
  {
    $project: {
      country: 1,
      fullName: 1,
      htmlWordCount: 1,
      textWordCount: 1,
      rchilliTextWordCount: 1,
      winner: {
        $switch: {
          branches: [
            {
              case: {
                $and: [
                  {
                    $gt: [
                      "$htmlWordCount",
                      "$textWordCount"
                    ]
                  },
                  {
                    $gt: [
                      "$htmlWordCount",
                      "$rchilliTextWordCount"
                    ]
                  }
                ]
              },
              then: "htmlWordCount"
            },
            {
              case: {
                $and: [
                  {
                    $gt: [
                      "$textWordCount",
                      "$htmlWordCount"
                    ]
                  },
                  {
                    $gt: [
                      "$textWordCount",
                      "$rchilliTextWordCount"
                    ]
                  }
                ]
              },
              then: "textWordCount"
            },
            {
              case: {
                $and: [
                  {
                    $gt: [
                      "$rchilliTextWordCount",
                      "$htmlWordCount"
                    ]
                  },
                  {
                    $gt: [
                      "$rchilliTextWordCount",
                      "$textWordCount"
                    ]
                  }
                ]
              },
              then: "rchilliTextWordCount"
            }
          ],
          default: "No winners"
        }
      }
    }
  }
])

MongoPlayground

Answer 2

这是获取结果的另一种方法：

1. 获取文档的字段名称及其值
2. 查找名称在[ "htmlWordCount", "textWordCount", "rchilliTextWordCount" ] 中的字段的最大值。

一般来说，从一个数组中求一个最大值是一种归约；所以我在这种情况下使用了$reduce。注意代码更简单。如果您想添加另一个字段来计算最大值，只需将其添加到数组中即可。

db.winner.aggregate([
  { $match: { country: "US"} },
  { $addFields: { fieldNameValues: { "$objectToArray": "$$ROOT" } } },
  { $project: { _id: 1, country: 1, fullName: 1, htmlWordCount: 1, textWordCount: 1, rchilliTextWordCount: 1, 
                winner: { 
                    $reduce: {
                        input: "$fieldNameValues",
                        initialValue: { },
                        in: {
                            $cond: [
                               { $and: [ 
                                   { $in: [ "$$this.k", [ "htmlWordCount", "textWordCount", "rchilliTextWordCount" ] ] }, 
                                   { $gt: [ "$$this.v", "$$value.v"] } ]
                               },
                               "$$this",
                               "$$value"
                            ]
                        }
                    } 
               } 
  } },
  { $addFields: { winner: "$winner.k" } }
] )

---

[编辑添加]

样本数据和结果：

{
        "_id" : 1,
        "fileId" : "dd386cf7-4139-45c2-9853-cbb126621b51",
        "job" : { },
        "country" : "US",
        "fullName" : "abcd xyz",
        "htmlWordCount" : 2766,
        "textWordCount" : 1867,
        "rchilliTextWordCount" : 2840
}
{
        "_id" : 2,
        "fileId" : "dd386cf7-4139-45c2-9853-cbb126621b51",
        "job" : { },
        "country" : "US",
        "fullName" : "lmn opqrs",
        "htmlWordCount" : 5,
        "textWordCount" : 9,
        "rchilliTextWordCount" : 2
}

输出：

{
        "_id" : 1,
        "country" : "US",
        "fullName" : "abcd xyz",
        "htmlWordCount" : 2766,
        "textWordCount" : 1867,
        "rchilliTextWordCount" : 2840,
        "winner" : "rchilliTextWordCount"
}
{
        "_id" : 2,
        "country" : "US",
        "fullName" : "lmn opqrs",
        "htmlWordCount" : 5,
        "textWordCount" : 9,
        "rchilliTextWordCount" : 2,
        "winner" : "textWordCount"
}