The overlay between rule items in a mixed configuration is associated by the data source name and the table name.
If the previous rule is aggregation-oriented, the next rule needs to use the aggregated logical data source name configured by the previous rule when configuring the data source. Similarly, if the previous rule is table aggregation-oriented, the next rule needs to use the aggregated logical table name configured by the previous rule when configuring the table.
dataSources: # Configure the real data source name.
write_ds:
# ...Omit specific configuration.
read_ds_0:
# ...Omit specific configuration.
read_ds_1:
# ...Omit specific configuration.
rules:
- !SHARDING # Configure data sharding rules.
tables:
t_user:
actualDataNodes: ds.t_user_${0..1} # Data source name 'ds' uses the logical data source name of the readwrite-splitting configuration.
tableStrategy:
standard:
shardingColumn: user_id
shardingAlgorithmName: t_user_inline
shardingAlgorithms:
t_user_inline:
type: INLINE
props:
algorithm-expression: t_user_${user_id % 2}
- !ENCRYPT # Configure data encryption rules.
tables:
t_user: # Table `t_user` is the name of the logical table that uses the data sharding configuration.
columns:
pwd:
cipherColumn: cipher_pwd
encryptorName: encryptor_aes
encryptors:
encryptor_aes:
type: aes
props:
aes-key-value: 123456abc
- !READWRITE_SPLITTING # Configure readwrite-splitting rules.
dataSources:
ds: # The logical data source name 'ds' for readwrite-splitting is used in data sharding.
type: Static
props:
write-data-source-name: write_ds # Use the real data source name 'write_ds'.
read-data-source-names: read_ds_0, read_ds_1 # Use the real data source name 'read_ds_0', 'read_ds_1'.
loadBalancerName: roundRobin
loadBalancers:
roundRobin:
type: ROUND_ROBIN
props:
sql-show: true