flink-cdc
flink-cdc copied to clipboard
Flink CDC is a streaming data integration tool
https://issues.apache.org/jira/browse/FLINK-35600
There is a circular dependency between the `DataFieldSerializer`, `DataTypeSerializer`, and `RowTypeSerializer` classes. This circular dependency can lead to incomplete initialization of the serializers, causing a `NullPointerException` when you attempt to...
https://issues.apache.org/jira/browse/FLINK-35544
Code optimization extracts constants
Hive metastore Require column names to be lowercase, but field names from upstream tables may not meet the requirements. I've transferred the columnname into lowercase and tested the case with...
Support parsing SQL like following and emit new SchemaChangeEvent to downstream. ``` CREATE TABLE `user_log_copy` LIKE `user_log` ``` and ``` CREATE TABLE `user_log_copy` AS SELECT * FROM `user_log` ```
CDC yaml without pipeline should not throw exception.
Currently, tables that are not present in transform blocks should be treated as if there's such a dummy fallback block: ```yaml transform: - source-table: "\.*.\.*" # capture all unmentioned tables...
Currently, CDC runs (pipeline and source) E2e cases on various Flink versions sequentially. A complete workflow would need almost 80 minutes to finish. This PR runs E2e test cases in...