zstan commented on code in PR #13389:
URL: https://github.com/apache/ignite/pull/13389#discussion_r3853305708
##########
modules/calcite/src/test/java/org/apache/ignite/internal/processors/query/calcite/integration/QueryBlockingTaskExecutorIntegrationTest.java:
##########
@@ -61,8 +61,7 @@ public void testJoinRehash() throws Exception {
assertQuery(sql)
.withParams("region0")
- .matches(QueryChecker.containsSubPlan("IgniteMergeJoin"))
-
.matches(QueryChecker.containsSubPlan("IgniteExchange(distribution=[affinity"))
+
.matches(QueryChecker.matches(".*IgniteMergeJoin.*IgniteExchange\\(distribution=\\[affinity.*"))
Review Comment:
such kind of enumeration in tests are potential bug )
look plan for this test lokk like :
```
IgniteColocatedHashAggregate(group=[{}], EXPR$0=[SUM($0)]): rowcount = 1.0,
cumulative cost = IgniteCost [rowCount=249.75, cpu=304.8283137373023,
memory=11.0, io=2.0, network=806.0], id = 19270
IgniteProject(AMOUNT=[$1]): rowcount = 15.0, cumulative cost = IgniteCost
[rowCount=234.75, cpu=289.8283137373023, memory=6.0, io=2.0, network=806.0], id
= 19269
IgniteMergeJoin(condition=[=($2, $0)], joinType=[inner],
variablesSet=[[]], leftCollation=[[0 ASC-nulls-first]], rightCollation=[[0
ASC-nulls-first]]): rowcount = 15.0, cumulative cost = IgniteCost
[rowCount=219.75, cpu=274.8283137373023, memory=6.0, io=2.0, network=806.0], id
= 19268
IgniteExchange(distribution=[single]): rowcount = 100.0, cumulative
cost = IgniteCost [rowCount=201.0, cpu=201.0, memory=1.0, io=1.0,
network=801.0], id = 19265
IgniteIndexScan(table=[[PUBLIC, ORDER_ITEMS]],
index=[ORDER_ITEMS_ORDERID], requiredColumns=[{3, 4}], inlineScan=[false],
collation=[[3 ASC-nulls-first, 0 ASC-nulls-first]]): rowcount = 100.0,
cumulative cost = IgniteCost [rowCount=101.0, cpu=101.0, memory=1.0, io=1.0,
network=1.0], id = 15638
IgniteExchange(distribution=[single]): rowcount = 1.0, cumulative cost
= IgniteCost [rowCount=3.75, cpu=13.8283137373023, memory=5.0, io=1.0,
network=5.0], id = 19267
IgniteSort(sort0=[$0], dir0=[ASC-nulls-first]): rowcount = 1.0,
cumulative cost = IgniteCost [rowCount=2.75, cpu=12.8283137373023, memory=5.0,
io=1.0, network=1.0], id = 19266
IgniteIndexScan(table=[[PUBLIC, ORDERS]], index=[ORDERS_REGION],
filters=[=($t1, ?0)], rowType=[RecordType(INTEGER ID)], pro
```
and if i write smth like :
```
.matches(QueryChecker.containsSubPlan("IgniteIndexScan"))
.matches(QueryChecker.containsSubPlan("IgniteProject"))
```
it also passed - i.e. such kind of enumeration not take sequence into
account,
```
.matches(QueryChecker.containsSubPlan("IgniteIndexScan"))
.matches(QueryChecker.containsSubPlan("IgniteProject"))
```
same as
```
.matches(QueryChecker.containsSubPlan("IgniteProject"))
.matches(QueryChecker.containsSubPlan("IgniteIndexScan"))
```
my approach - is sequence aware.
--
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]
