gustavodemorais commented on code in PR #28025:
URL: https://github.com/apache/flink/pull/28025#discussion_r3195803548
##########
flink-table/flink-table-planner/src/test/java/org/apache/flink/table/planner/plan/nodes/exec/stream/FromChangelogTestPrograms.java:
##########
@@ -172,6 +172,71 @@ public class FromChangelogTestPrograms {
+ "op => DESCRIPTOR(operation))")
.build();
+ //
--------------------------------------------------------------------------------------------
+ // Set semantics with PARTITION BY
+ //
--------------------------------------------------------------------------------------------
+
+ /**
+ * Verifies that {@code FROM_CHANGELOG(TABLE t PARTITION BY id)} produces
the same logical
+ * output as the row-semantic call. The conditional {@code
SET_SEMANTIC_TABLE} trait switches
+ * the execution to a co-located parallel mode but must not change
row-level semantics.
+ */
+ public static final TableTestProgram SET_SEMANTICS_PARTITION_BY =
+ TableTestProgram.of(
+ "from-changelog-set-semantics-partition-by",
+ "PARTITION BY enables set semantics without
altering output rows")
+ .setupTableSource(
+ SourceTestStep.newBuilder("cdc_stream")
+ .addSchema(SIMPLE_CDC_SCHEMA)
Review Comment:
Can you use a schema where the id is not the first field instead of
SIMPLE_CDC_SCHEMA so we test the partition by adjustment logic?
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
