Apache ShardingSphere uses ThreadLocal to manage sharding key value or hint route.
Users can add sharding values to HintManager, and those values only take effect within the current thread.
Apache ShardingSphere is able to add special comments in SQL to hint route too.
Usage of hint:
Hint algorithms require users to implement the interface of org.apache.shardingsphere.api.sharding.hint.HintShardingAlgorithm
.
Apache ShardingSphere will acquire sharding values from HintManager to route.
Take the following configurations for reference:
rules:
- !SHARDING
tables:
t_order:
actualDataNodes: demo_ds_${0..1}.t_order_${0..1}
databaseStrategy:
hint:
algorithmClassName: xxx.xxx.xxx.HintXXXAlgorithm
tableStrategy:
hint:
algorithmClassName: xxx.xxx.xxx.HintXXXAlgorithm
defaultTableStrategy:
none:
defaultKeyGenerateStrategy:
type: SNOWFLAKE
column: order_id
props:
sql-show: true
HintManager hintManager = HintManager.getInstance();
hintManager.addDatabaseShardingValue
to add sharding key value of data source.hintManager.addTableShardingValue
to add sharding key value of table.Users can use
hintManager.setDatabaseShardingValue
to add sharding in hint route to some certain sharding database without sharding tables.
Sharding values are saved in ThreadLocal
, so it is necessary to use hintManager.close()
to clean ThreadLocal
.
HintManager
has implemented AutoCloseable
. We recommend to close it automatically with try with resource
.
// Sharding database and table with HintManager
String sql = "SELECT * FROM t_order";
try (HintManager hintManager = HintManager.getInstance();
Connection conn = dataSource.getConnection();
PreparedStatement preparedStatement = conn.prepareStatement(sql)) {
hintManager.addDatabaseShardingValue("t_order", 1);
hintManager.addTableShardingValue("t_order", 2);
try (ResultSet rs = preparedStatement.executeQuery()) {
while (rs.next()) {
// ...
}
}
}
// Sharding database and one database route with HintManager
String sql = "SELECT * FROM t_order";
try (HintManager hintManager = HintManager.getInstance();
Connection conn = dataSource.getConnection();
PreparedStatement preparedStatement = conn.prepareStatement(sql)) {
hintManager.setDatabaseShardingValue(3);
try (ResultSet rs = preparedStatement.executeQuery()) {
while (rs.next()) {
// ...
}
}
}
Be the same as sharding based on hint.
hintManager.setWriteRouteOnly
to configure primary database route.Be the same as data sharding based on hint.
String sql = "SELECT * FROM t_order";
try (HintManager hintManager = HintManager.getInstance();
Connection conn = dataSource.getConnection();
PreparedStatement preparedStatement = conn.prepareStatement(sql)) {
hintManager.setWriteRouteOnly();
try (ResultSet rs = preparedStatement.executeQuery()) {
while (rs.next()) {
// ...
}
}
}
To use SQL Hint function, users need to set sqlCommentParseEnabled
to true
.
The comment format only supports /* */
for now. The content needs to start with ShardingSphere hint:
, and the attribute name needs to be writeRouteOnly
.
/* ShardingSphere hint: writeRouteOnly=true */
SELECT * FROM t_order;
Be the same as sharding based on hint.
hintManager.setDataSourceName
to configure database route.String sql = "SELECT * FROM t_order";
try (HintManager hintManager = HintManager.getInstance();
Connection conn = dataSource.getConnection();
PreparedStatement preparedStatement = conn.prepareStatement(sql)) {
hintManager.setDataSourceName("ds_0");
try (ResultSet rs = preparedStatement.executeQuery()) {
while (rs.next()) {
// ...
}
}
}
To use SQL Hint function, users need to set sqlCommentParseEnabled
to true
. Currently, only support routing to one data source.
The comment format only supports /* */
for now. The content needs to start with ShardingSphere hint:
, and the attribute name needs to be dataSourceName
.
/* ShardingSphere hint: dataSourceName=ds_0 */
SELECT * FROM t_order;