spring-data-mongodb的Aggregation详解

1. 概述

本教程要向大家介绍spring-data-mongodb的Aggregation(聚合)操作，也就是类似于关系型数据库的group操作!内容包括其使用实例、应用技巧、基本知识点总结和需要注意事项，具有一定的参考价值，需要的朋友可以参考一下。

在spring-data-mongodb框架中Aggregation分组统计基本的操作包括：

• $group - 分组操作
• $match - 过滤操作
• $project - 从输出中提取字段，可以重命名字段
• $limit - 返回结果集中的前n个文档
• $skip - 丢弃结果集中的前n个文档
• $count,$avg, $sum 等等函数…
• $sort - 排序

2. 需求

例如现在有一张各地城市的饭馆表store_detail_info

MySql:

MongoDB:

现在有一个统计北上广的饭馆的个数,平均信誉分数,平均消费价格的需求,那么

MySql:

查询语句如下：

select
	CITY as CITY_NAME,
	count(*) as COUNT_STORE,
	sum(AVGSCORE) as AVG_SCORE,
	sum(AVGPRICE) as AVG_PRICE
from
	store_detail_info
where
	CITY in ("上海", "北京", "广州")
group by
	CITY
order by
	COUNT_STORE desc
1
2
3
4
5
6
7
8
9
10
11
12
13

MongoDB:

原始的聚合(aggregate)语句如下:

db.store_detail_info.aggregate([{
    "$match": {
      "CITY": {
        "$in": ["北京", "上海", "广州"]
      }
    }
  }, {
    "$group": {
      "_id": "$CITY",
      "CITY_NAME": {
        "$first": "$CITY"
      },
      "COUNT_STORE": {
        "$sum": 1
      },
      "AVG_SCORE": {
        "$avg": "$AVGSCORE"
      },
      "AVG_PRICE": {
        "$avg": "$AVGPRICE"
      }
    }
  }, {
    "$project": {
      "_id": 0,
      "CITY_NAME": 1,
      "COUNT_STORE": 1,
      "AVG_SCORE": 1,
      "AVG_PRICE": 1
    }
  }, {
    "$sort": {
      "COUNT_STORE": -1
    }
  }]
);
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36

$match: 只匹配 北京,上海, 广州 这3个城市

$group: 根据CITY分组，输出城市名称(CITY_NAME), 然后统计COUNT_STORE(饭馆个数)，用$sum函数，统计AVGSCORE(饭馆平均信誉分)，用$avg函数，统计AVGPRICE(饭馆平均消费价格)，用$avg函数

$project: 定义要显示的字段,1为显示，0为不显示

$sort: 按照COUNT_STORE(饭馆个数)降序

下面详解在spring-data-mongodb中我们要怎么去实现这个功能

3. 实现

依赖

首先，我们需要添加使用 MongoDB 和 Spring Boot 所需的依赖项。

让我们将 spring-boot-starter-data-mongodb 添加到我们的 pom.xml 中：

<dependency>
    <groupId>org.springframework.bootgroupId>
    <artifactId>spring-boot-starter-data-mongodbartifactId>
dependency>
1
2
3
4

配置

现在，我们通过将以下配置添加到 application.yml 文件来设置连接：

# MongoDB数据库
spring:
  data:
    mongodb:
      uri: "mongodb://user:password@localhost:27017/test"
1
2
3
4
5

让我们运行应用程序来测试我们是否可以连接到数据库。 我们应该在日志中看到与此类似的消息：

[info,71] - Opened connection [connectionId{localValue:1, serverValue:237}] to localhost:27017
1

这意味着应用程序可以成功连接到 MongoDB。

实现

• 使用框架里的 Aggregation

@SpringBootTest
class MyMongoApplicationTests {
  @Autowired
  MongoTemplate mongoTemplate;

  @BeforeEach
  void contextLoads() {
    System.out.println(mongoTemplate.getDb().getName());
  }

  void aggregation() {
    MatchOperation      match   = Aggregation.match(Criteria.where("CITY").in("上海", "北京", "广州"));
    GroupOperation      group   = Aggregation.group("CITY").first("CITY").as("CITY_NAME").count().as("COUNT_STORE").avg("AVGSCORE").as("AVG_SCORE").avg("AVGPRICE").as("AVG_PRICE");
    ProjectionOperation project = Aggregation.project().andExclude("_id").andInclude("CITY_NAME", "COUNT_STORE", "AVG_SCORE", "AVG_PRICE");
    SortOperation       sort    = Aggregation.sort(Direction.DESC, "COUNT_STORE");

    //@wjw_note: 这里多个Aggregation 是按先后顺序执行的，即后面一个在前面一个的基础上作筛选等操作，所以顺序不一致可能会报错
    Aggregation aggregation = Aggregation.newAggregation(match, group, project, sort);
    System.out.println(aggregation.toString());  // <1>

    AggregationResults<Document> aa        = mongoTemplate.aggregate(aggregation, "store_detail_info", Document.class);
    List<Document>               documents = aa.getMappedResults();
    documents.forEach(document -> {
      System.out.println(document.toJson(indentJsonWriterSettings));
    });
  }

}
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28

<1> 打印输出原生的MongodbDB aggregate 命令,帮助理解

运行后输出:

{ "aggregate" : "__collection__", "pipeline" : [{ "$match" : { "CITY" : { "$in" : ["上海", "北京", "广州"]}}}, { "$group" : { "_id" : "$CITY", "CITY_NAME" : { "$first" : "$CITY"}, "COUNT_STORE" : { "$sum" : 1}, "AVG_SCORE" : { "$avg" : "$AVGSCORE"}, "AVG_PRICE" : { "$avg" : "$AVGPRICE"}}}, { "$project" : { "_id" : 0, "CITY_NAME" : 1, "COUNT_STORE" : 1, "AVG_SCORE" : 1, "AVG_PRICE" : 1}}, { "$sort" : { "COUNT_STORE" : -1}}]}
2022-11-07 19:10:08.692 [main] INFO  org.mongodb.driver.connection - [info,71] - Opened connection [connectionId{localValue:3, serverValue:242}] to localhost:27017
{
  "CITY_NAME": "北京",
  "COUNT_STORE": 2214,
  "AVG_SCORE": 4.0862691960252935,
  "AVG_PRICE": 85.12420957542909
}
{
  "CITY_NAME": "上海",
  "COUNT_STORE": 1950,
  "AVG_SCORE": 3.781025641025641,
  "AVG_PRICE": 90.47435897435898
}
{
  "CITY_NAME": "广州",
  "COUNT_STORE": 1483,
  "AVG_SCORE": 3.6790289952798383,
  "AVG_PRICE": 60.956169925826025
}

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21

• 使用原生java驱动 Document

@SpringBootTest
class MyMongoApplicationTests {
  @Autowired
  MongoTemplate mongoTemplate;

  @BeforeEach
  void contextLoads() {
    System.out.println(mongoTemplate.getDb().getName());
  }

  @Test
  void aggregation2() {
    List<Document> pipeline = new ArrayList<Document>();
    Document       group    = new Document();
    group.put("$group",
        new Document("_id", "$CITY")
            .append("CITY_NAME", new Document("$first", "$CITY"))
            .append("COUNT_STORE", new Document("$sum", 1))
            .append("AVG_SCORE", new Document("$sum", "$AVGSCORE"))
            .append("AVG_PRICE", new Document("$sum", "$AVGPRICE")));

    Document project = new Document();
    project.append("$project",
        new Document("CITY_NAME", 1)
            .append("COUNT_STORE", 1)
            .append("AVG_SCORE", 1)
            .append("AVG_PRICE", 1)
            .append("_id", 0));

    Document match = new Document();
    match.append("$match", new Document("CITY_NAME", new Document().append("$in", Arrays.asList("上海", "北京", "广州"))));

    Document sort = new Document();
    sort.append("$sort", new Document("COUNT_STORE", -1));

    pipeline.add(group);
    pipeline.add(project);
    pipeline.add(match);
    pipeline.add(sort);

    System.out.println(org.springframework.data.mongodb.core.query.SerializationUtils.serializeToJsonSafely(pipeline));  // <1>

    AggregateIterable<Document> output = mongoTemplate.getCollection("store_detail_info").aggregate(pipeline);
    output.forEach(document -> {
      System.out.println(document.toJson(indentJsonWriterSettings));
    });
  }
}
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48

<1> 打印输出原生的MongodbDB aggregate 命令,帮助理解

运行后输出:

[{ "$group" : { "_id" : "$CITY", "CITY_NAME" : { "$first" : "$CITY"}, "COUNT_STORE" : { "$sum" : 1}, "AVG_SCORE" : { "$sum" : "$AVGSCORE"}, "AVG_PRICE" : { "$sum" : "$AVGPRICE"}}}, { "$project" : { "CITY_NAME" : 1, "COUNT_STORE" : 1, "AVG_SCORE" : 1, "AVG_PRICE" : 1, "_id" : 0}}, { "$match" : { "CITY_NAME" : { "$in" : ["上海", "北京", "广州"]}}}, { "$sort" : { "COUNT_STORE" : -1}}]
2022-11-07 19:12:40.534 [main] INFO  org.mongodb.driver.connection - [info,71] - Opened connection [connectionId{localValue:3, serverValue:245}] to localhost:27017
{
  "CITY_NAME": "北京",
  "COUNT_STORE": 2214,
  "AVG_SCORE": 9047,
  "AVG_PRICE": 188465
}
{
  "CITY_NAME": "上海",
  "COUNT_STORE": 1950,
  "AVG_SCORE": 7373,
  "AVG_PRICE": 176425
}
{
  "CITY_NAME": "广州",
  "COUNT_STORE": 1483,
  "AVG_SCORE": 5456,
  "AVG_PRICE": 90398
}
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20

结尾

至此,我们使用框架和原生api实现了MongoDB的聚合功能.

---

<<<<<<<<<<<< [完] >>>>>>>>>>>>