如何获取数组mongodb中字段的值

假设我的数据库结构为

{
    name: "alex",
    age: 21,
    school: [
       {grade: 45, class: "elementary school", name: "blabla"},
       {grade: 89, class: "high school", name: "blabla2"},
       {grade: 12, class: "college", name: "blabla3"}
    ]
},
{
    name: "felix",
    age: 45,
    school: [
       {grade: 12, class: "elementary school", name: "bla"},
       {grade: 45, class: "high school", name: "bla2"},
       {grade: 16, class: "college", name: "bla3"}
    ]
}

我想做的是得到每个名字的学校名称

{
    name: alex,
    school: ["blabla", "blabla2", "blabla3"]
},
{
    name: felix,
    school: ["bla", "bla2", "bla3"]
}

要做到这一点,

db.collection.aggregate([
    {group: {_id: "$name", school: {$addToSet: {"$school.name"}}}}
])

但无法实现。我应该修改什么？

在使用$unwind阶段进行分组之前，只需解开学校数组，如下所示：

db.collection.aggregate([
  {$unwind: "$school"},
  {$group: {_id: "$name", school: {$addToSet: "$school.name"}}}
])

输出：

{ "_id" : "felix", "school" : [ "bla3", "bla2", "bla" ] }
{ "_id" : "alex", "school" : [ "blabla3", "blabla2", "blabla" ] }

---

可以使用投影从数组中获取字段值。类似于db.collection.find{}，{u id:0，name:1，school.name:1}；或者使用aggregate db.collection.aggregate{$project:{{u id:0，name:1，school:$school.name}；。对于每个学校文档中的不同名称值，可以使用db.collection.aggregate{$project:{{u id:0，name:1，学校：{$setUnion:[$school.name，[]}；