1.表与表的三种关系

一对一、一对多、多对多

2. aggregate聚合管道

聚合管道可以对集合中的问那个进行变换和组合

以下例子的表全部如下：

> db.order.find()
{ "_id" : ObjectId("5f81817f2501fd0ea0437220"), "order_id" : "1", "uid" : 10, "trade_no" : "111", "all_price" : 100, "all_num" : 2 }
{ "_id" : ObjectId("5f81818a2501fd0ea0437221"), "order_id" : "2", "uid" : 7, "trade_no" : "222", "all_price" : 90, "all_num" : 2 }
{ "_id" : ObjectId("5f8181932501fd0ea0437222"), "order_id" : "3", "uid" : 9, "trade_no" : "333", "all_price" : 20, "all_num" : 6 }


> db.order_item.find()
{ "_id" : ObjectId("5f81829c2501fd0ea0437226"), "order_id" : "1", "title" : "商品鼠标 1", "price" : 50, "num" : 1 }
{ "_id" : ObjectId("5f81829c2501fd0ea0437227"), "order_id" : "1", "title" : "商品键盘 2", "price" : 50, "num" : 1 }
{ "_id" : ObjectId("5f81829e2501fd0ea0437228"), "order_id" : "1", "title" : "商品键盘 3", "price" : 0, "num" : 1 }
{ "_id" : ObjectId("5f8182cf2501fd0ea0437229"), "order_id" : "2", "title" : "牛奶", "price" : 50, "num" : 1 }
{ "_id" : ObjectId("5f8182d02501fd0ea043722a"), "order_id" : "2", "title" : "酸奶", "price" : 40, "num" : 1 }
{ "_id" : ObjectId("5f8182e42501fd0ea043722b"), "order_id" : "3", "title" : "矿泉水", "price" : 2, "num" : 5 }
{ "_id" : ObjectId("5f8182e52501fd0ea043722c"), "order_id" : "3", "title" : "毛巾", "price" : 10, "num" : 1 }

• $project

修改文档的结构，可以用来重命名、增加或删除文档中的字段

//只显示order_id、trade_no、all_price
db.order.aggregate([
  {
    $project: { order_id: 1, trade_no: 1, all_price: 1 }
  }
])
//输出
{ "_id" : ObjectId("5f8182692501fd0ea0437223"), "order_id" : "1", "trade_no" : "111", "all_price" : 100 }
{ "_id" : ObjectId("5f81826f2501fd0ea0437224"), "order_id" : "2", "trade_no" : "222", "all_price" : 90 }
{ "_id" : ObjectId("5f8182732501fd0ea0437225"), "order_id" : "3", "trade_no" : "333", "all_price" : 20 }

• $match

用法类似于 find() 方法中的参数。

//只显示trade_no、all_price，且>=90的数据
db.order.aggregate([
  {
    $project: { trade_no: 1, all_price: 1 }
  },
  {
    $match: { "all_price": { $gte: 90 } }
  }
])
//输出
{ "_id" : ObjectId("5f8182692501fd0ea0437223"), "trade_no" : "111", "all_price" : 100 }
{ "_id" : ObjectId("5f81826f2501fd0ea0437224"), "trade_no" : "222", "all_price" : 90 }

• $group

将集合中的文档进行分组，可用于统计结果

//假设id相同统计他的价格
db.order_item.aggregate(
  [{
    $group: { _id: "$order_id", total: { $sum: "$price" } }
  }]
)
//输出
{ "_id" : "2", "total" : 90 }
{ "_id" : "3", "total" : 12 }
{ "_id" : "1", "total" : 100 }

• $sort

//只显示trade_no、all_price，逆序排列，
db.order.aggregate([
  {
    $project: { trade_no: 1, all_price: 1 }
  },
  {
    $sort: { "all_price": -1 }//正序参数填1
  }
])
//输出
{ "_id" : ObjectId("5f8182692501fd0ea0437223"), "trade_no" : "111", "all_price" : 100 }
{ "_id" : ObjectId("5f81826f2501fd0ea0437224"), "trade_no" : "222", "all_price" : 90 }
{ "_id" : ObjectId("5f8182732501fd0ea0437225"), "trade_no" : "333", "all_price" : 20 }

• $limit

//只显示一个搜索结果
db.order.aggregate([
  {
    $project: { trade_no: 1, all_price: 1 }
  },
  {
    $sort: { "all_price": -1 }
  },
  {
    $limit:1
  }
])
//输出
{ "_id" : ObjectId("5f8182692501fd0ea0437223"), "trade_no" : "111", "all_price" : 100 }

• $skip

//跳过第一个结果
db.order.aggregate([
  {
    $project: { trade_no: 1, all_price: 1 }
  },
  {
    $sort: { "all_price": -1 }
  },
  {
    $skip: 1
  }
])
//输出
{ "_id" : ObjectId("5f81826f2501fd0ea0437224"), "trade_no" : "222", "all_price" : 90 }
{ "_id" : ObjectId("5f8182732501fd0ea0437225"), "trade_no" : "333", "all_price" : 20 }

• $lookup

db.order.aggregate([
  {
    $lookup:
    {
      from: "order_item",
      localField: "order_id",//目前表order索引
      foreignField: "order_id",//目的表order_item索引
      as: "items"//新生的键名
    }
  }
])
//输出
{
    "_id": ObjectId("5f8182692501fd0ea0437223"),
    "order_id": "1",
    "uid": 10,
    "trade_no": "111",
    "all_price": 100,
    "all_num": 2,
    "items": [{
        "_id": ObjectId("5f81829c2501fd0ea0437226"),
        "order_id": "1",
        "title": "商品鼠标 1",
        "price": 50,
        "num": 1
    }, {
        "_id": ObjectId("5f81829c2501fd0ea0437227"),
        "order_id": "1",
        "title": "商品键盘 2",
        "price": 50,
        "num": 1
    }, {
        "_id": ObjectId("5f81829e2501fd0ea0437228"),
        "order_id": "1",
        "title": "商品键盘 3",
        "price": 0,
        "num": 1
    }]
} {
    "_id": ObjectId("5f81826f2501fd0ea0437224"),
    "order_id": "2",
    "uid": 7,
    "trade_no": "222",
    "all_price": 90,
    "all_num": 2,
    "items": [{
        "_id": ObjectId("5f8182cf2501fd0ea0437229"),
        "order_id": "2",
        "title": "牛奶",
        "price": 50,
        "num": 1
    }, {
        "_id": ObjectId("5f8182d02501fd0ea043722a"),
        "order_id": "2",
        "title": "酸奶",
        "price": 40,
        "num": 1
    }]
} {
    "_id": ObjectId("5f8182732501fd0ea0437225"),
    "order_id": "3",
    "uid": 9,
    "trade_no": "333",
    "all_price": 20,
    "all_num": 6,
    "items": [{
        "_id": ObjectId("5f8182e42501fd0ea043722b"),
        "order_id": "3",
        "title": "矿泉水",
        "price": 2,
        "num": 5
    }, {
        "_id": ObjectId("5f8182e52501fd0ea043722c"),
        "order_id": "3",
        "title": "毛巾",
        "price": 10,
        "num": 1
    }]
}

3.SQL 和 和 NOSQL 对比

4.数据库备份及还原

以下操作不需要启动mongodb服务

数据库导出备份：

mongodump -h dbhost -d dbname -o dbdirectory

-h：一般为127.0.0.1，也可以指定端口号

-d： 数据库名

-o： 要提取到的目录

数据库恢复及导入：

mongorestore -h dbhost -d dbname dbdirectory

--drop：恢复时候先删除当前数据，然后恢复备份数据，恢复后备份后添加修改的数据都会被删除，慎用

如果导入的数据库是有密码的

mongodump -h dbhost -d dbname -u username -p password -o dbdirectory

mongorestore -h dbhost -d dbname dbdirectory -u username -p password

windows10下需要先下载mongodb-database-tools-windows-x86_64-100.1.1这个包

cmd打开进入C:\Software-Download\mongodb\mongodb-database-tools-windows-x86_64-100.1.1\bin，假设安装在其他地方填写mongodb-database-tools安装的路径找到bin文件，cd进入

mongodump -h 127.0.0.1 -d wenera -o C:\222-test\node\bf 导出

mongorestore -h 127.0.0.1 -d test C:\222-test\node\bf导入