Question

在我的mongodb中，我收藏的很少，我想通过使用pymongo比较集合1和集合2来创建一个新的集合。

    Collection 1 :
    Object id       timestamp                          Prof_Name   subjects1 
    abc67478898k  ISODate("2018-01-03T09:26:37.541Z")   ABDC      "sub1, sub2, sub3"
    jjjjjjjjjj    ISODate("2018-01-03T09:26:37.541Z")   XYZ       "sub2, sub4, sub8"

    Collection 2 :
    Object id   timestamp               UUID   subjects2            rating score
    3333333    ISODate("2018-01-03TZ")  7897  "sub1,sub4, sub7"     7      10
    444444     ISODate("2018-01-03TZ")  4532   "sub2"               4      6
    777777     ISODate("2018-01-03TZ")  7876  "sub1,sub2,sub3"      8      8
    1111111    ISODate("2018-01-03TZ")   654   "sub1,sub3"          7      8

我正在获取我的3个集合，Prof_name为每个主题找到collection2中的匹配主题，并且某个时间戳和我的mongo查询之间的UUID和UUID_count如下：

    db.data1.aggregate([
      {"$lookup":{
        "from":"data2",
        "let":{"subject":{"$split":["$SUBJECT",", "]}},
        "pipeline":[
          {"$match": {"expr":{"$and":[{"$eq":[{"$year":"$timestamp"}, 2016]}, {"$eq":[{"$month":"$timestamp"}, 1]}]}}},
          {"$addFields":{"SUBJECT_ID":{"$split":["$SUBJECT_ID",", "]},"SUBJECT":"$$subject"}},
          {"$unwind":"$SUBJECT"},
          {"$match":{"$expr":{"$in":["$SUBJECT","$SUBJECT_ID"]}}},
          {"$facet":{
            "UUID":[{"$group":{"_id":{"id":"$_id","UUID":"$UUID"}}},{"$count":"UUID_Count"}],
            "REST":[
              {"$group":{"_id":null,"subjects_list":{"$addToSet":"$SUBJECT"},"UUID_distinct_list":{"$addToSet":"$UUID"}}},
              {"$addFields":{"subject_count":{"$size":"$subjects_list"},"UUID_distinct_count":{"$size":"$UUID_distinct_list"}}},
              {"$project":{"_id":0}}
             ]
          }},
          {"$replaceRoot":{"newRoot":{"$mergeObjects":[{"$arrayElemAt":["$UUID",0]},{"$arrayElemAt":["$REST",0]}]}}}
        ],
        "as":"ref_data"
      }},
      {"$unwind":{"path":"$ref_data","preserveNullAndEmptyArrays":true}},
      {"$addFields":{"ref_data.Prof_Name":"$Prof_Name"}},
      {"$replaceRoot":{"newRoot":"$ref_data"}},
      {"$out":"data3"}
    ])

以上查询为我提供了下面的收藏。

    Collection 3 :

     objectid  Prof_name subjects_list  UUID_list           UUID-count subject_count
        12       ABDC      sub1,sub2,sub3 7897,4532,7876,654  4            3
        34       XYZ       sub2,sub4,sub8 7897,4532,7876      2            3

现在我想为我的集合3获取另一个列，其中列出了每个主题的计数列表以及与每个主题相关的UUID，如下所示收集3：

     objectid  Prof_name subjects_list  UUID_list           UUID-count subject_count each_sub_count             UUID-assocaited_sub
        12       ABDC      sub1,sub2,sub3 7897,4532,7876,654  4           3         sub1:3,sub2:2,sub3:2    [sub1:7897,7876,654, sub2:4532,7876, sub3:7876]
        34       XYZ       sub2,sub4,sub8 7897,4532,7876      2           3         sub2:2,sub4:1,sub8:0    [sub2:4532,7876, sub4:7897,sub8:0]

最后2列是我需要的，我如何实现这一点，可以修改上面的查询并获取它或什么是新查询来获取这些列。

Answer 1

在$facet中包含另一个管道。

{"$facet":{
  "UUID":[{"$group":{"_id":{"id":"$_id","UUID":"$UUID"}}},{"$count":"UUID_Count"}],
  "COUNT":[
     {"$group":{"_id":null,"subjects_list":{"$addToSet":"$SUBJECT"},"UUID_distinct_list":{"$addToSet":"$UUID"}}},
     {"$addFields":{"subject_count":{"$size":"$subjects_list"},"UUID_distinct_count":{"$size":"$UUID_distinct_list"}}},
     {"$project":{"_id":0}}
  ],
  "SUB":[
     {"$group":{"_id":"$SUBJECT","count":{"$sum":1}," UUID_list":{"$push":"$UUID"}}},
     {"$group":{"_id":null,"each_sub_count":{"$push":{"sub":"$_id", "count":"$count"}},"UUID-assocaited_sub":{"$push":{"sub":"$_id", uuids:"$UUID_list"}}}},
     {"$project":{"_id":0}}
  ]
}},
{"$replaceRoot":{"newRoot":{"$mergeObjects":[{"$arrayElemAt":["$UUID",0]},{"$arrayElemAt":["$COUNT",0]}, {"$arrayElemAt":["$SUB",0]}]}}}

MongoDB聚合查询以获取每个元素的列表及其每个记录的计数

1 个答案: