数据分片

背景信息

数据分片 Java API 规则配置允许用户直接通过编写 Java 代码的方式，完成 ShardingSphereDataSource 对象的创建，Java API 的配置方式非常灵活，不需要依赖额外的 jar 包就能够集成各种类型的业务系统。

参数解释

配置入口

类名称：org.apache.shardingsphere.sharding.api.config.ShardingRuleConfiguration

可配置属性：

分片表配置

类名称：org.apache.shardingsphere.sharding.api.config.ShardingTableRuleConfiguration

可配置属性：

自动分片表配置

类名称：org.apache.shardingsphere.sharding.api.config.ShardingAutoTableRuleConfiguration

可配置属性：

分片策略配置

标准分片策略配置

类名称：org.apache.shardingsphere.sharding.api.config.strategy.sharding.StandardShardingStrategyConfiguration

可配置属性：

复合分片策略配置

类名称：org.apache.shardingsphere.sharding.api.config.strategy.sharding.ComplexShardingStrategyConfiguration

可配置属性：

Hint 分片策略配置

类名称：org.apache.shardingsphere.sharding.api.config.strategy.sharding.HintShardingStrategyConfiguration

可配置属性：

不分片策略配置

类名称：org.apache.shardingsphere.sharding.api.config.strategy.sharding.NoneShardingStrategyConfiguration

可配置属性：无

算法类型的详情，请参见内置分片算法列表。

分布式序列策略配置

类名称：org.apache.shardingsphere.sharding.api.config.strategy.keygen.KeyGenerateStrategyConfiguration

可配置属性：

算法类型的详情，请参见内置分布式序列算法列表。

分片审计策略配置

类名称：org.apache.shardingsphere.sharding.api.config.strategy.audit.ShardingAuditStrategyConfiguration

可配置属性：

算法类型的详情，请参见内置分片审计列算法列表。

操作步骤

1. 创建真实数据源映射关系，key 为数据源逻辑名称，value 为 DataSource 对象；
2. 创建分片规则对象 ShardingRuleConfiguration，并初始化对象中的分片表对象 ShardingTableRuleConfiguration、绑定表集合、广播表集合，以及数据分片所依赖的分库策略和分表策略等参数；
3. 调用 ShardingSphereDataSourceFactory 对象的 createDataSource 方法，创建 ShardingSphereDataSource。

配置示例

public final class ShardingDatabasesAndTablesConfigurationPrecise implements ExampleConfiguration {
    @Override
    public DataSource getDataSource() throws SQLException {
        return ShardingSphereDataSourceFactory.createDataSource(createDataSourceMap(), Collections.singleton(createShardingRuleConfiguration()), new Properties());
    }
    private ShardingRuleConfiguration createShardingRuleConfiguration() {
        ShardingRuleConfiguration result = new ShardingRuleConfiguration();
        result.getTables().add(getOrderTableRuleConfiguration());
        result.getTables().add(getOrderItemTableRuleConfiguration());
        result.getBindingTableGroups().add(new ShardingTableReferenceRuleConfiguration("foo", "t_order, t_order_item"));
        result.getBroadcastTables().add("t_address");
        result.setDefaultDatabaseShardingStrategy(new StandardShardingStrategyConfiguration("user_id", "inline"));
        result.setDefaultTableShardingStrategy(new StandardShardingStrategyConfiguration("order_id", "standard_test_tbl"));
        Properties props = new Properties();
        props.setProperty("algorithm-expression", "demo_ds_${user_id % 2}");
        result.getShardingAlgorithms().put("inline", new AlgorithmConfiguration("INLINE", props));
        result.getShardingAlgorithms().put("standard_test_tbl", new AlgorithmConfiguration("STANDARD_TEST_TBL", new Properties()));
        result.getKeyGenerators().put("snowflake", new AlgorithmConfiguration("SNOWFLAKE", new Properties()));
        result.getAuditors().put("sharding_key_required_auditor", new AlgorithmConfiguration("DML_SHARDING_CONDITIONS", new Properties()));
        return result;
    }
    private ShardingTableRuleConfiguration getOrderTableRuleConfiguration() {
        ShardingTableRuleConfiguration result = new ShardingTableRuleConfiguration("t_order", "demo_ds_${0..1}.t_order_${[0, 1]}");
        result.setKeyGenerateStrategy(new KeyGenerateStrategyConfiguration("order_id", "snowflake"));
        result.setAuditStrategy(new ShardingAuditStrategyConfiguration(Collections.singleton("sharding_key_required_auditor"), true));
        return result;
    }
    private ShardingTableRuleConfiguration getOrderItemTableRuleConfiguration() {
        ShardingTableRuleConfiguration result = new ShardingTableRuleConfiguration("t_order_item", "demo_ds_${0..1}.t_order_item_${[0, 1]}");
        result.setKeyGenerateStrategy(new KeyGenerateStrategyConfiguration("order_item_id", "snowflake"));
        return result;
    }
    private Map<String, DataSource> createDataSourceMap() {
        Map<String, DataSource> result = new HashMap<>();
        result.put("demo_ds_0", DataSourceUtil.createDataSource("demo_ds_0"));
        result.put("demo_ds_1", DataSourceUtil.createDataSource("demo_ds_1"));
        return result;
    }
}

相关参考

• 核心特性：数据分片
• 开发者指南：数据分片