github-actions[bot] commented on code in PR #65846:
URL: https://github.com/apache/doris/pull/65846#discussion_r3714329211
##########
regression-test/suites/nereids_rules_p0/set_preagg/set_preagg.groovy:
##########
@@ -67,8 +79,125 @@ suite("set_preagg") {
aggregate key (k1,k2,k3,k4,k5,k6)
distributed BY hash(k1) buckets 3
properties("replication_num" = "1");
+ create table preagg_t4(
+ k1 int null,
+ k2 int null,
+ k3 int null,
+ k4 int null,
+ k5 int null,
+ k6 int null,
+ v7 bigint SUM,
+ v9 bigint MAX
+ )
+ aggregate key (k1,k2,k3,k4,k5,k6)
+ distributed BY hash(k1) buckets 3
+ properties("replication_num" = "1");
+ create table preagg_t5(
+ k1 int null,
+ v double MAX
+ )
+ aggregate key (k1)
+ distributed BY hash(k1) buckets 1
+ properties("replication_num" = "1");
+ create table preagg_f_l(
+ k1 int null,
+ v9 bigint MAX,
+ v9m bigint MIN
+ )
+ aggregate key (k1)
+ distributed BY hash(k1) buckets 1
+ properties("replication_num" = "1");
+ create table preagg_f_r(
+ k1 int null,
+ v9 bigint MAX,
+ v9m bigint MIN
+ )
+ aggregate key (k1)
+ distributed BY hash(k1) buckets 1
+ properties("replication_num" = "1");
+ create table preagg_asof_l(
+ grp int null,
+ ts datetime null
+ )
+ aggregate key (grp, ts)
+ distributed BY hash(grp) buckets 1
+ properties("replication_num" = "1");
+ create table preagg_asof_r(
+ grp int null,
+ ts datetime MIN,
Review Comment:
[P2] Make q37 reach the ASOF-specific fence
Declaring `ts` as `MIN` makes `MATCH_CONDITION(l.ts >= r.ts)` reference a
value slot on `r`. `LogicalJoin.getExpressions()` includes that match
condition, so the pre-existing join-value check already keeps `preagg_asof_r`
OFF; removing the new `asofSelectedSideRelationIds` logic would leave both this
EXPLAIN assertion and result 200 green. This is distinct from the earlier
result ambiguity: the revised result now distinguishes the chosen partial, but
the plan no longer reaches the mechanism under test. Please keep the
selected-side match columns in the aggregate key, use duplicate identical full
keys with a SUM value (100 and 200) and expect merged 300 versus either faulty
partial, and assert the ASOF-specific OFF reason.
--
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]
