spring boot使用sharding jdbc的配置方式

时间：2021-07-16 05:50:43

本文介绍了spring boot使用sharding jdbc的配置方式，分享给大家，具体如下：

说明

要排除DataSourceAutoConfiguration，否则多数据源无法配置

?
1
2
3
4
5
6
7
8
9
@SpringBootApplication
@EnableAutoConfiguration(exclude={DataSourceAutoConfiguration.class})
public class Application {
 
  public static void main(String[] args) {
   SpringApplication.run(Application.class, args);
  }
 
}

配置的多个数据源交给sharding-jdbc管理，sharding-jdbc创建一个DataSource数据源提供给mybatis使用

官方文档：http://shardingjdbc.io/index_zh.html

步骤

配置多个数据源，数据源的名称最好要有一定的规则，方便配置分库的计算规则

?
1
2
3
4
5
6
7
8
9
10
11
@Bean(initMethod="init", destroyMethod="close", name="dataSource0")
@ConfigurationProperties(prefix = "spring.datasource")
public DataSource dataSource0(){
  return new DruidDataSource();
}
 
@Bean(initMethod="init", destroyMethod="close", name="dataSource1")
@ConfigurationProperties(prefix = "spring.datasource2")
public DataSource dataSource1(){
  return new DruidDataSource();
}

配置数据源规则，即将多个数据源交给sharding-jdbc管理，并且可以设置默认的数据源，当表没有配置分库规则时会使用默认的数据源

?
1
2
3
4
5
6
7
8
@Bean
public DataSourceRule dataSourceRule(@Qualifier("dataSource0") DataSource dataSource0,
    @Qualifier("dataSource1") DataSource dataSource1){
  Map<String, DataSource> dataSourceMap = new HashMap<>();
  dataSourceMap.put("dataSource0", dataSource0);
  dataSourceMap.put("dataSource1", dataSource1);
  return new DataSourceRule(dataSourceMap, "dataSource0");
}

配置数据源策略和表策略，具体策略需要自己实现

?
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
@Bean
public ShardingRule shardingRule(DataSourceRule dataSourceRule){
  //表策略
  TableRule orderTableRule = TableRule.builder("t_order")
      .actualTables(Arrays.asList("t_order_0", "t_order_1"))
      .tableShardingStrategy(new TableShardingStrategy("order_id", new ModuloTableShardingAlgorithm()))
      .dataSourceRule(dataSourceRule)
      .build();
  TableRule orderItemTableRule = TableRule.builder("t_order_item")
      .actualTables(Arrays.asList("t_order_item_0", "t_order_item_1"))
      .tableShardingStrategy(new TableShardingStrategy("order_id", new ModuloTableShardingAlgorithm()))
      .dataSourceRule(dataSourceRule)
      .build();
  //绑定表策略，在查询时会使用主表策略计算路由的数据源，因此需要约定绑定表策略的表的规则需要一致，可以一定程度提高效率
  List<BindingTableRule> bindingTableRules = new ArrayList<BindingTableRule>();
  bindingTableRules.add(new BindingTableRule(Arrays.asList(orderTableRule, orderItemTableRule)));
  return ShardingRule.builder()
      .dataSourceRule(dataSourceRule)
      .tableRules(Arrays.asList(orderTableRule, orderItemTableRule))
      .bindingTableRules(bindingTableRules)
      .databaseShardingStrategy(new DatabaseShardingStrategy("user_id", new ModuloDatabaseShardingAlgorithm()))
      .tableShardingStrategy(new TableShardingStrategy("order_id", new ModuloTableShardingAlgorithm()))
      .build();
}

创建sharding-jdbc的数据源DataSource，MybatisAutoConfiguration会使用此数据源

?
1
2
3
4
@Bean("dataSource")
public DataSource shardingDataSource(ShardingRule shardingRule){
  return ShardingDataSourceFactory.createDataSource(shardingRule);
}

需要手动配置事务管理器（原因未知）

?
1
2
3
4
5
//需要手动声明配置事务
@Bean
public DataSourceTransactionManager transactitonManager(@Qualifier("dataSource") DataSource dataSource){
  return new DataSourceTransactionManager(dataSource);
}

分库策略的简单实现，接口：DatabaseShardingAlgorithm

?
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
49
50
import java.util.Collection;
import java.util.LinkedHashSet;
 
import com.dangdang.ddframe.rdb.sharding.api.ShardingValue;
import com.dangdang.ddframe.rdb.sharding.api.strategy.database.SingleKeyDatabaseShardingAlgorithm;
import com.google.common.collect.Range;
 
/**
 * Created by fuwei.deng on 2017年5月11日.
 */
public class ModuloDatabaseShardingAlgorithm implements SingleKeyDatabaseShardingAlgorithm<Long> {
 
  @Override
  public String doEqualSharding(Collection<String> databaseNames, ShardingValue<Long> shardingValue) {
   for (String each : databaseNames) {
      if (each.endsWith(shardingValue.getValue() % 2 + "")) {
        return each;
      }
    }
    throw new IllegalArgumentException();
  }
  
  @Override
  public Collection<String> doInSharding(Collection<String> databaseNames, ShardingValue<Long> shardingValue) {
   Collection<String> result = new LinkedHashSet<>(databaseNames.size());
    for (Long value : shardingValue.getValues()) {
      for (String tableName : databaseNames) {
        if (tableName.endsWith(value % 2 + "")) {
          result.add(tableName);
        }
      }
    }
    return result;
  }
  
  @Override
  public Collection<String> doBetweenSharding(Collection<String> databaseNames, ShardingValue<Long> shardingValue) {
   Collection<String> result = new LinkedHashSet<>(databaseNames.size());
    Range<Long> range = (Range<Long>) shardingValue.getValueRange();
    for (Long i = range.lowerEndpoint(); i <= range.upperEndpoint(); i++) {
      for (String each : databaseNames) {
        if (each.endsWith(i % 2 + "")) {
          result.add(each);
        }
      }
    }
    return result;
  }
 
}

分表策略的基本实现，接口：TableShardingAlgorithm

?
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
49
50
import java.util.Collection;
import java.util.LinkedHashSet;
 
import com.dangdang.ddframe.rdb.sharding.api.ShardingValue;
import com.dangdang.ddframe.rdb.sharding.api.strategy.table.SingleKeyTableShardingAlgorithm;
import com.google.common.collect.Range;
 
/**
 * Created by fuwei.deng on 2017年5月11日.
 */
public class ModuloTableShardingAlgorithm implements SingleKeyTableShardingAlgorithm<Long> {
 
  @Override
  public String doEqualSharding(Collection<String> tableNames, ShardingValue<Long> shardingValue) {
   for (String each : tableNames) {
      if (each.endsWith(shardingValue.getValue() % 2 + "")) {
        return each;
      }
    }
    throw new IllegalArgumentException();
  }
  
  @Override
  public Collection<String> doInSharding(Collection<String> tableNames, ShardingValue<Long> shardingValue) {
   Collection<String> result = new LinkedHashSet<>(tableNames.size());
    for (Long value : shardingValue.getValues()) {
      for (String tableName : tableNames) {
        if (tableName.endsWith(value % 2 + "")) {
          result.add(tableName);
        }
      }
    }
    return result;
  }
  
  @Override
  public Collection<String> doBetweenSharding(Collection<String> tableNames, ShardingValue<Long> shardingValue) {
   Collection<String> result = new LinkedHashSet<>(tableNames.size());
    Range<Long> range = (Range<Long>) shardingValue.getValueRange();
    for (Long i = range.lowerEndpoint(); i <= range.upperEndpoint(); i++) {
      for (String each : tableNames) {
        if (each.endsWith(i % 2 + "")) {
          result.add(each);
        }
      }
    }
    return result;
  }
 
}

至此，分库分表的功能已经实现

读写分离

读写分离需在创建DataSourceRule之前加一层主从数据源的创建

?
1
2
3
4
5
6
7
8
9
10
11
12
13
// 构建读写分离数据源, 读写分离数据源实现了DataSource接口, 可直接当做数据源处理.
// masterDataSource0, slaveDataSource00, slaveDataSource01等为使用DBCP等连接池配置的真实数据源
DataSource masterSlaveDs0 = MasterSlaveDataSourceFactory.createDataSource("ms_0",
          masterDataSource0, slaveDataSource00, slaveDataSource01);
DataSource masterSlaveDs1 = MasterSlaveDataSourceFactory.createDataSource("ms_1",
          masterDataSource1, slaveDataSource11, slaveDataSource11);
 
// 构建分库分表数据源
Map<String, DataSource> dataSourceMap = new HashMap<>(2);
dataSourceMap.put("ms_0", masterSlaveDs0);
dataSourceMap.put("ms_1", masterSlaveDs1);
 
// 通过ShardingDataSourceFactory继续创建ShardingDataSource

强制使用主库时

?
1
2
3
HintManager hintManager = HintManager.getInstance();
hintManager.setMasterRouteOnly();
// 继续JDBC操作

强制路由

  1. 使用ThreadLocal机制实现，在执行数据库操作之前通过HintManager改变用于计算路由的值
  2. 设置HintManager的时候分库和分表的策略必须同时设置，并且设置后需要路由的表都需要设置用于计算路由的值。比如强制路由后需要操作t_order和t_order_item两个表，那么两个表的分库和分表的策略都需要设置
?
1
2
3
4
5
HintManager hintManager = HintManager.getInstance();
hintManager.addDatabaseShardingValue("t_order", "user_id", 1L);
hintManager.addTableShardingValue("t_order", "order_id", order.getOrderId());
hintManager.addDatabaseShardingValue("t_order_item", "user_id", 1L);
hintManager.addTableShardingValue("t_order_item", "order_id", order.getOrderId());

事务

  1. sharding-jdbc-transaction实现柔性事务（默认提供了基于内存的事务日志存储器和内嵌异步作业），可结合elastic-job（sharding-jdbc-transaction-async-job）实现异步柔性事务
  2. 没有与spring结合使用的方式，需要自己封装

以上就是本文的全部内容，希望对大家的学习有所帮助，也希望大家多多支持服务器之家。

原文链接：https://my.oschina.net/dengfuwei/blog/1595151

标签：

相关文章