Question

我的收藏品有以下格式 -

{
 "title" : "Seasonal", 
 "state" : "Maharashtra",
 "fruits" : [ "banana", "orange", "jackfruit", "mango", "tamarind"]
} 
{
 "title" : "Seasonal", 
 "state" : "Karnataka",
 "fruits" : [ "banana", "avacado", "blackberry", "coconut", "grape"]
}
{
 "title" : "Alltime", 
 "state" : "Karnataka",
 "fruits" : [ "banana", "lemon", "pomegranate", "pineapple"]
}

我的查询就像

db.fruits.aggregate([{"$match": {"state": "Karnataka", "fruits": {"$elemMatch": {"$in": ["banana","coconut"]}}}},{ $unwind : "$fruits" },{"$group" : {"_id" : "$fruits","count" : {"$sum":1}}}])

将输出设为 -

{ "_id" : "pineapple", "count" : 1 }
{ "_id" : "lemon", "count" : 1 }
{ "_id" : "grape", "count" : 1 }
{ "_id" : "pomegranate", "count" : 1 }
{ "_id" : "coconut", "count" : 1 }
{ "_id" : "blackberry", "count" : 1 }
{ "_id" : "avacado", "count" : 1 }
{ "_id" : "banana", "count" : 2 }

相反，我想要的是输入水果名称及其出现次数。即。

{ "_id" : "coconut", "count" : 1 }
{ "_id" : "banana", "count" : 2 }

请您告诉我，如何对这些数组元素进行分组？

Answer 1

使用以下聚合。您在$match之后缺少$unwind阶段。

初始匹配仅过滤文档，其中存在水果数组，其中至少有一个水果来自匹配的输入水果数组。

一旦你$unwind，它就会解开所有包含其他水果的水果。因此，当您使用第二个匹配时，您过滤水果只包含与输入水果数组匹配的水果。

db.fruits.aggregate([
  {"$match":{"state":"Karnataka","fruits":{"$elemMatch":{"$in":["banana","coconut"]}}}},
  {"$unwind":"$fruits"},
  {"$match":{"fruits":{"$in":["banana","coconut"]}}},
  {"$group":{"_id":"$fruits","count":{"$sum":1}}}
])

Mongo $ group数组元素基于输入数组

1 个答案: