springboot integrates mybatis-plus and sharding-jdbc to achieve sub-library, sub-table and separation of reading and writing (including complete project code)
1. Integrate sharding-jdbc
There are 4 ways to integrate springboot into the sharding-jdbc official website, and the .properties files are all used online. Here I mainly use the combination of yml+bean. Bean is mainly used to achieve more flexible sharding key sharding rules.
1.1Introduce maven dependencies
There is a small pit here. I first introduced the latest sharding-jdbc and mybatis-plus dependencies, but an error was reported when starting the project. Therefore, I personally do not recommend citing the latest package.
<dependency>
<groupId>org.apache.shardingsphere</groupId>
<artifactId>sharding-jdbc-spring-boot-starter</artifactId>
<version>4.0.0-RC1</version>
</dependency>
1.2 Configuring sub-databases and tables and separation of reading and writing (core)
spring:
shardingsphere:
datasource:
names: "search-center-00,search-center-00-slave,search-center-01,search-center-01-slave"
search-center-00:
type: com.alibaba.druid.pool.DruidDataSource
driver-class-name: com.mysql.cj.jdbc.Driver
url: jdbc:mysql://localhost:3306/search_center_00?useUnicode=true&characterEncoding=utf8&serverTimezone=UTC
username: root
password: 123456
search-center-00-slave:
type: com.alibaba.druid.pool.DruidDataSource
driver-class-name: com.mysql.cj.jdbc.Driver
url: jdbc:mysql://localhost:3307/search_center_00?useUnicode=true&characterEncoding=utf8&serverTimezone=UTC
username: root
password: 123456
search-center-01:
type: com.alibaba.druid.pool.DruidDataSource
driver-class-name: com.mysql.cj.jdbc.Driver
url: jdbc:mysql://localhost:3306/search_center_01?useUnicode=true&characterEncoding=utf8&serverTimezone=UTC
username: root
password: 123456
search-center-01-slave:
type: com.alibaba.druid.pool.DruidDataSource
driver-class-name: com.mysql.cj.jdbc.Driver
url: jdbc:mysql://localhost:3307/search_center_01?useUnicode=true&characterEncoding=utf8&serverTimezone=UTC
username: root
password: 123456
sharding:
master-slave-rules:
search-center-00:
master-data-source-name: search-center-00
slave-data-source-names: search-center-00-slave
search-center-01:
master-data-source-name: search-center-01
slave-data-source-names: search-center-01-slave
tables:
user:
actual-data-nodes: search-center-$->{(0..1).collect{t ->t.toString().padLeft(2,'0')}}.user_$->{(0..1).collect{t ->t.toString().padLeft(2,'0')}}
database-strategy:
standard:
shardingColumn: id
precise-algorithm-class-name: com.gory.searchcenter.config.DatabaseAlgorithm
table-strategy:
standard:
sharding-column: id
precise-algorithm-class-name: com.gory.searchcenter.config.TableAlgorithm
props:
sql:
show: true
names: Customize database names, and then configure these names specifically, such as: url, password, username, etc.
tables: Divide each table into databases and tables. For example, here the user table is divided into databases and tables.
actual-data-nodes: It is the node allocation of user table sub-database sub-table.
database-strategy: It is the strategy of sub-library. Here I configure the sub-library rules more flexibly based on the ID and through beans.
table-strategy: It is the strategy of sharding tables, and the configuration is roughly the same as that of sharding databases.
master-slave-rules: Set master-slave to achieve read-write separation.
DatabaseAlgorithm
public class DatabaseAlgorithm implements PreciseShardingAlgorithm<Long> {
@Override
public String doSharding(Collection<String> databaseNames, PreciseShardingValue<Long> shardingValue) {
Long shardResult = Math.abs((long) shardingValue.getValue().hashCode()) % 2;
for (String databaseName : databaseNames) {
String shardValue = String.format("%02d", shardResult);
if (databaseName.endsWith(shardValue)){
System.out.println("shardingValue:"+shardingValue+", databaseName:" + databaseName);
return databaseName;
}
}
throw new UnsupportedOperationException();
}
}
TableAlgorithm
public class TableAlgorithm implements PreciseShardingAlgorithm<Long> {
/**
* @param tableNames 有效的数据源 或者 表 的名字 tableNames 就为配置文件中的 配置的数据源信息 -> ds0 , ds1
* @param shardingValue SQL 分片列 对应的实际值
* @return 返回相应的表名
*/
@Override
public String doSharding(Collection<String> tableNames, PreciseShardingValue<Long> shardingValue) {
Long shardResult = Math.abs((long) shardingValue.getValue().hashCode()) / 2 % 2;
for (String tableName : tableNames) {
String shardValue = String.format("%02d", shardResult);
if (tableName.endsWith(shardValue)){
System.out.println("shardingValue:"+shardingValue+", tableName:" + tableName);
return tableName;
}
}
throw new UnsupportedOperationException();
}
}
By the way, the shard primary key id here is set using the snowflake algorithm. It is best to assign values to your sharding keys that correspond to your sharding rules.
2. Integrate mybatis-plus
As mentioned in the previous section, introducing the latest maven dependency will result in an error, so I still do not recommend using the latest one.
2.1 Introduce maven dependencies
<dependency>
<groupId>com.baomidou</groupId>
<artifactId>mybatis-plus-boot-starter</artifactId>
<version>3.2.0</version>
</dependency>
<dependency>
<groupId>com.baomidou</groupId>
<artifactId>mybatis-plus-generator</artifactId>
<version>3.5.1</version>
</dependency>
2.2 Configuration of mybatis-plus
mybatis-plus:
mapper-locations: classpath:/mapper/*.xml
configuration:
log-impl: org.apache.ibatis.logging.stdout.StdOutImpl
Then start the class and add the package scan table annotation
@MapperScan({"com.gory.searchcenter.mapper"})
3. Database master-slave configuration
In the previous two sections, we have been able to complete database and table partitioning, but because the database does not have a master-slave setting, the slave table does not have a database, resulting in no data in the query. Because this article is mainly for learning, the configuration of mysql is in the window environment.
3.1my.ini configuration
main library
#The main library id cannot be repeated
server_id=1
#Open bin-log
log_bin=E:\lib\mysql-5.7.30-winx64\master-bin.log
#mysql The time to clear expired logs is consistent with the slave database setting
expire_logs_days = 1
#Set the database to be synchronized
binlog-do-db=search_center_00
binlog-do-db=search_center_01
--------------------------------- ---------------------------------- |
From the library
server_id=2
#Open bin-log
log_bin=E:\lib\mysql-5.7.30-winx64-slave\slave-bin.log
expire_logs_days = 1
#Set the database to be synchronized
binlog-do-db=search_center_00
binlog-do-db=search_center_01
--------------------------------- ---------------------------------- |
It is worth mentioning that when setting multiple synchronization tables, they cannot be separated by commas.
3.1 sql configuration
main library
Enable the slave server (Slave) to access the master server (Master)
grant replication slave on . to ‘root’@‘127.0.0.1’ identified by ‘123456’;
View authorization information
show master status ;
From the library
Stop sync
stop slave;
Set the information of the main library and link the slave library with the main library
change master to
master_host='127.0.0.1',
master_user='root',
master_password='123456',
master_log_file='log-bin.000008',
master_log_pos=107;
The two configurations master_log_file and master_log_pos are show master with the master server corresponding to status
Start sync
start salve;
Check slave startup status
show slave status；
If it appears, it means the startup is successful.
Slave_IO_Running:Yes
Slave_SQL_Running:Yes
4. Complete project code
Complete code: https://github.com/gorylee/SearchCenter