yux created FLINK-35272: --------------------------- Summary: Pipeline Transform job supports omitting / renaming calculation column Key: FLINK-35272 URL: https://issues.apache.org/jira/browse/FLINK-35272 Project: Flink Issue Type: Improvement Components: Flink CDC Affects Versions: 3.1.0 Reporter: yux
Currently, pipeline transform rules require all columns used by expression calculation must be present in final projected schema, and shall not be renamed or omitted. The reason behind this is any column not directly present in projection rules will be filtered out in the PreProjection step, and then the PostProjection process could not find those non-present but indirectly depended columns. -- This message was sent by Atlassian Jira (v8.20.10#820010)