MOBIN created FLINK-37132:
-----------------------------
Summary: Add schema validation in Multi Transform
Key: FLINK-37132
URL: https://issues.apache.org/jira/browse/FLINK-37132
Project: Flink
Issue Type: Bug
Components: Flink CDC
Affects Versions: cdc-3.2.1, cdc-3.2.0
Reporter: MOBIN
The following scenarios should throw an exception of [different column count]
{code:java}
void testMultiTransformSchemaColumnsCompatibilityWithDiffColumnCount(
ValuesDataSink.SinkApi sinkApi) {
assertThatThrownBy(
() ->
runGenericTransformTest(
sinkApi,
Arrays.asList(
new TransformDef(
"default_namespace.default_schema.mytable2",
"*",
"age < 18",
null,
null,
null,
null,
null),
new TransformDef(
"default_namespace.default_schema.mytable2",
// reference part column
"id,UPPER(name) AS name",
"age >= 18",
null,
null,
null,
null,
null)),
Collections.emptyList()))
.rootCause()
.isExactlyInstanceOf(IllegalStateException.class)
.hasMessage(
"Unable to merge schema columns={`id` BIGINT NOT
NULL,`name` VARCHAR(255),`age` TINYINT,`description` STRING}, primaryKeys=id,
options=() "
+ "and columns={`id` BIGINT NOT NULL,`name`
STRING}, primaryKeys=id, options=() with different column counts.");
} {code}
In Multi Transform, metadata fields like primaryKeys, partitionKeys, and
options also need to be consistent.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
