sharding-jdbc如何配置分析Configuration

sharding-jdbc如何配置分析Configuration，很多新手对此不是很清楚，为了帮助大家解决这个难题，下面小编将为大家详细讲解，有这方面需求的人可以来学习下，希望你能有所收获。

Sharding核心配置主要如下（官网）：

• 分片规则

  分片规则配置的总入口。包含数据源配置、表配置、绑定表配置以及读写分离配置等

• 数据源配置

  真实数据源列表

• 表配置

  逻辑表名称、数据节点与分表规则的配置

• 数据节点配置

  用于配置逻辑表与真实表的映射关系。可分为均匀分布和自定义分布两种形式

• 分片策略配置

  对于分片策略存有数据源分片策略和表分片策略两种维度

• 数据源分片策略：

  对应于DatabaseShardingStrategy。用于配置数据被分配的目标数据源

• 表分片策略
  对应于TableShardingStrategy。用于配置数据被分配的目标表，该目标表存在与该数据的目标数据源内。故表分片策略是依赖与数据源分片策略的结果的

• 自增主键生成策略

  通过在客户端生成自增主键替换以数据库原生自增主键的方式，做到分布式主键无重复。

接下来对各个核心配置进行分析：

1. 以多主多从读写分离、表分片为例

   public final class ShardingMasterSlaveConfigurationPrecise implements ExampleConfiguration {
       
       @Override
       public DataSource getDataSource() throws SQLException {
           ShardingRuleConfiguration shardingRuleConfig = new ShardingRuleConfiguration();
           //分片表规则配置
           shardingRuleConfig.getTableRuleConfigs().add(getOrderTableRuleConfiguration());
           shardingRuleConfig.getTableRuleConfigs().add(getOrderItemTableRuleConfiguration());
           //绑定分片表，主要用来路由
           shardingRuleConfig.getBindingTableGroups().add("t_order, t_order_item");
           //设置默认数据源分片策略
           shardingRuleConfig.setDefaultDatabaseShardingStrategyConfig(new StandardShardingStrategyConfiguration("user_id", new PreciseModuloShardingDatabaseAlgorithm()));
           //设置默认表分片策略
           shardingRuleConfig.setDefaultTableShardingStrategyConfig(new StandardShardingStrategyConfiguration("order_id", new PreciseModuloShardingTableAlgorithm()));
           //主从配置，支持多主多从
           shardingRuleConfig.setMasterSlaveRuleConfigs(getMasterSlaveRuleConfigurations());
           //创建ShardingDataSource数据源
           return ShardingDataSourceFactory.createDataSource(createDataSourceMap(), shardingRuleConfig, new Properties());
       }
       
       private static TableRuleConfiguration getOrderTableRuleConfiguration() {
           //分片表配置
           TableRuleConfiguration result = new TableRuleConfiguration(/*逻辑表*/"t_order", /*数据源名.真实表*/"ds_${0..1}.t_order_${[0, 1]}");
           //自定义主键生成配置
           result.setKeyGeneratorConfig(new KeyGeneratorConfiguration("SNOWFLAKE", "order_id", getProperties()));
           return result;
       }
       
       private static TableRuleConfiguration getOrderItemTableRuleConfiguration() {
           TableRuleConfiguration result = new TableRuleConfiguration("t_order_item", "ds_${0..1}.t_order_item_${[0, 1]}");
           result.setKeyGeneratorConfig(new KeyGeneratorConfiguration("SNOWFLAKE", "order_item_id", getProperties()));
           return result;
       }
       
       private static List<MasterSlaveRuleConfiguration> getMasterSlaveRuleConfigurations() {
           MasterSlaveRuleConfiguration masterSlaveRuleConfig1 = new MasterSlaveRuleConfiguration("ds_0", "demo_ds_master_0", Arrays.asList("demo_ds_master_0_slave_0", "demo_ds_master_0_slave_1"));
           MasterSlaveRuleConfiguration masterSlaveRuleConfig2 = new MasterSlaveRuleConfiguration("ds_1", "demo_ds_master_1", Arrays.asList("demo_ds_master_1_slave_0", "demo_ds_master_1_slave_1"));
           return Lists.newArrayList(masterSlaveRuleConfig1, masterSlaveRuleConfig2);
       }
       
       private static Map<String, DataSource> createDataSourceMap() {
           final Map<String, DataSource> result = new HashMap<>();
           result.put("demo_ds_master_0", DataSourceUtil.createDataSource("demo_ds_master_0"));
           result.put("demo_ds_master_0_slave_0", DataSourceUtil.createDataSource("demo_ds_master_0_slave_0"));
           result.put("demo_ds_master_0_slave_1", DataSourceUtil.createDataSource("demo_ds_master_0_slave_1"));
           result.put("demo_ds_master_1", DataSourceUtil.createDataSource("demo_ds_master_1"));
           result.put("demo_ds_master_1_slave_0", DataSourceUtil.createDataSource("demo_ds_master_1_slave_0"));
           result.put("demo_ds_master_1_slave_1", DataSourceUtil.createDataSource("demo_ds_master_1_slave_1"));
           return result;
       }
       
       private static Properties getProperties() {
           Properties result = new Properties();
           result.setProperty("worker.id", "123");
           return result;
       }
   }

   
   

2. ShardingRuleConfiguration分片规则核心配置

   @Getter
   @Setter
   public final class ShardingRuleConfiguration implements RuleConfiguration {
       //表规则配置
       private Collection<TableRuleConfiguration> tableRuleConfigs = new LinkedList<>();
   
       //绑定表配置
       private Collection<String> bindingTableGroups = new LinkedList<>();
   
       //广播表配置
       private Collection<String> broadcastTables = new LinkedList<>();
   
       //默认数据源名称
       private String defaultDataSourceName;
   
       //默认分库策略
       private ShardingStrategyConfiguration defaultDatabaseShardingStrategyConfig;
   
       //默认分片策略
       private ShardingStrategyConfiguration defaultTableShardingStrategyConfig;
   
       //默认主键生成工具类
       private KeyGeneratorConfiguration defaultKeyGeneratorConfig;
   
       //主从规则配置
       private Collection<MasterSlaveRuleConfiguration> masterSlaveRuleConfigs = new LinkedList<>();
   
       //数据脱敏规则配置
       private EncryptRuleConfiguration encryptRuleConfig;
   }

   
   

    

3. 分析ShardingDataSourceFactory#createDataSource中的ShardingRule配置

   /**
    * Sharding data source factory.
    * 
    * @author zhangliang 
    */
   @NoArgsConstructor(access = AccessLevel.PRIVATE)
   public final class ShardingDataSourceFactory {
       
       /**
        * Create sharding data source.
        *
        * @param dataSourceMap data source map
        * @param shardingRuleConfig rule configuration for databases and tables sharding
        * @param props properties for data source
        * @return sharding data source
        * @throws SQLException SQL exception
        */
       public static DataSource createDataSource(
               final Map<String, DataSource> dataSourceMap, final ShardingRuleConfiguration shardingRuleConfig, final Properties props) throws SQLException {
           //创建ShardingDataSource数据源，同时创建sharding rule配置
           return new ShardingDataSource(dataSourceMap, new ShardingRule(shardingRuleConfig, dataSourceMap.keySet()), props);
       }
   }

   
    

4. ShardingRule分片规则

   public ShardingRule(final ShardingRuleConfiguration shardingRuleConfig, final Collection<String> dataSourceNames) {
           Preconditions.checkArgument(null != shardingRuleConfig, "ShardingRuleConfig cannot be null.");
           Preconditions.checkArgument(null != dataSourceNames && !dataSourceNames.isEmpty(), "Data sources cannot be empty.");
           this.shardingRuleConfig = shardingRuleConfig;
           //sharding数据源名称
           shardingDataSourceNames = new ShardingDataSourceNames(shardingRuleConfig, dataSourceNames);
           //创建表规则集合
           tableRules = createTableRules(shardingRuleConfig);
           //创建分组绑定表
           //1.从tableRules查找
           //2.是否广播表，如果是广播表，dataSourceNames.size() == 1 ? dataSourceNames.iterator().next() : shardingRuleConfig.getDefaultDataSourceName();
           //3.已上两种情况都不满足，则使用默认数据源创建tableRule
           bindingTableRules = createBindingTableRules(shardingRuleConfig.getBindingTableGroups());
           //广播表
           broadcastTables = shardingRuleConfig.getBroadcastTables();
           //默认分库策略
           defaultDatabaseShardingStrategy = createDefaultShardingStrategy(shardingRuleConfig.getDefaultDatabaseShardingStrategyConfig());
           //默认表分片策略
           defaultTableShardingStrategy = createDefaultShardingStrategy(shardingRuleConfig.getDefaultTableShardingStrategyConfig());
           //默认主键生成工具类
           defaultShardingKeyGenerator = createDefaultKeyGenerator(shardingRuleConfig.getDefaultKeyGeneratorConfig());
           //创建主从规则
           masterSlaveRules = createMasterSlaveRules(shardingRuleConfig.getMasterSlaveRuleConfigs());
           //数据脱敏规则
           encryptRule = createEncryptRule(shardingRuleConfig.getEncryptRuleConfig());
       }

   
   

    

5. TableRule表规则

   public TableRule(final TableRuleConfiguration tableRuleConfig, final ShardingDataSourceNames shardingDataSourceNames, final String defaultGenerateKeyColumn) {
           //获取逻辑表
           logicTable = tableRuleConfig.getLogicTable().toLowerCase();
           //inline表达式解析出真实表，比如：ds_${0..1}.t_order_${[0, 1]}
           //解析出来为：
           //ds_0.t_order_0
           //ds_0.t_order_1
           //ds_1.t_order_0
           //ds_1.t_order_1
           List<String> dataNodes = new InlineExpressionParser(tableRuleConfig.getActualDataNodes()).splitAndEvaluate();
   
           dataNodeIndexMap = new HashMap<>(dataNodes.size(), 1);
   
           //真实表数据节点
           //1.如果没有配置真实表，则根据逻辑表、数据源来生成对应真实表数据节点
           //2.否则根据inline表达式解析出来的真实表来生成真实数据节点
           actualDataNodes = isEmptyDataNodes(dataNodes)
               ? generateDataNodes(tableRuleConfig.getLogicTable(), shardingDataSourceNames.getDataSourceNames()) : generateDataNodes(dataNodes, shardingDataSourceNames.getDataSourceNames());
           //真实表
           actualTables = getActualTables();
           //数据源->分片策略
           databaseShardingStrategy = null == tableRuleConfig.getDatabaseShardingStrategyConfig() ? null : ShardingStrategyFactory.newInstance(tableRuleConfig.getDatabaseShardingStrategyConfig());
           //表->分片策略
           tableShardingStrategy = null == tableRuleConfig.getTableShardingStrategyConfig() ? null : ShardingStrategyFactory.newInstance(tableRuleConfig.getTableShardingStrategyConfig());
           //主键字段
           generateKeyColumn = getGenerateKeyColumn(tableRuleConfig.getKeyGeneratorConfig(), defaultGenerateKeyColumn);
           //生成主键工具类，比如SNOWFLAKE
           shardingKeyGenerator = containsKeyGeneratorConfiguration(tableRuleConfig)
                   ? new ShardingKeyGeneratorServiceLoader().newService(tableRuleConfig.getKeyGeneratorConfig().getType(), tableRuleConfig.getKeyGeneratorConfig().getProperties()) : null;
       }

   
   

    

看完上述内容是否对您有帮助呢？如果还想对相关知识有进一步的了解或阅读更多相关文章，请关注亿速云行业资讯频道，感谢您对亿速云的支持。