github-actions[bot] commented on code in PR #69122:
URL: https://github.com/apache/doris/pull/69122#discussion_r4282049992
##########
regression-test/suites/nereids_syntax_p0/inpredicate_with_struct.groovy:
##########
@@ -15,27 +15,33 @@
// specific language governing permissions and limitations
// under the License.
-suite("inpredicate_with_struct") {
+suite("inpredicate_with_complex_type") {
sql """ set enable_nereids_planner = true;"""
sql """ set enable_fallback_to_original_planner=false;"""
- // support struct type
- order_qt_in_predicate_11 """
- select struct(1,"2") in (struct(1,3), null);
- """
- order_qt_in_predicate_12 """
- select struct(1,"2") not in (struct(1,3), null);
- """
- order_qt_in_predicate_13 """
- select struct(1,"2") in (struct(1,3), struct(1,2));
- """
- order_qt_in_predicate_14 """
- select struct(1,"2") not in (struct(1,3), struct(1,2));
- """
- order_qt_in_predicate_15 """
- select struct(1,"2") in (struct(1,3), struct(1,"2"), struct(1,1));
- """
- order_qt_in_predicate_16 """
- select struct(1,"2") not in (struct(1,3), struct(1,"2"),
struct(1,1));
- """
+
+ test {
+ sql """ select struct(1, "2") in (struct(1, 2)); """
+ exception "in predicate could not contains complex type"
+ }
+
+ test {
+ sql """ select struct(1, "2") not in (struct(1, 2)); """
+ exception "in predicate could not contains complex type"
+ }
+
+ test {
+ sql """ select [1, 2] in ([1, 2]); """
Review Comment:
[P2] Update the remaining complex-IN test expectations
The new ARRAY error case leaves `inpredicate_with_list.groovy:36-54`
expecting six ARRAY IN queries to succeed with fallback disabled; its first
`c_array in ([1,2], [1,3])` now throws the complex-type error. FE tests also
retain the old contract:
`TypeCoercionUtilsTest.testInPredicateStructCastKeepsNullableCastResult`
expects STRUCT IN coercion to succeed, and
`VarBinarySqlSupportTest.testBinaryInPredicatesFailDuringAnalysis` requires the
VARBINARY diagnostic for ARRAY/STRUCT operands, which the complex guard now
preempts. Please update these existing expectations along with the new negative
cases; otherwise both regression and FE unit suites fail under the intended
behavior.
--
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]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
