danny0405 commented on a change in pull request #10123:
[FLINK-14665][table-planner-blink] Support computed column for create…
URL: https://github.com/apache/flink/pull/10123#discussion_r344606608
##########
File path:
flink-table/flink-table-planner-blink/src/test/resources/org/apache/flink/table/planner/plan/batch/sql/DagOptimizationTest.xml
##########
@@ -199,37 +189,35 @@
LogicalSink(name=[`default_catalog`.`default_database`.`sink1`], fields=[a, b1])
LogicalSink(name=[`default_catalog`.`default_database`.`sink2`], fields=[a,
b1])
+- LogicalUnion(all=[true])
:- LogicalProject(a1=[$0], b1=[$1])
- : +- LogicalProject(a1=[$0], b1=[$1])
- : +- LogicalFilter(condition=[<=($0, 10)])
- : +- LogicalTableScan(table=[[default_catalog, default_database,
MyTable, source: [TestTableSource(a, b, c)]]])
- +- LogicalProject(a=[$0], b1=[$1])
- +- LogicalProject(a=[$2], b1=[$1])
- +- LogicalFilter(condition=[=($2, $0)])
- +- LogicalJoin(condition=[true], joinType=[inner])
- :- LogicalProject(a1=[$0], b1=[$1])
- : +- LogicalFilter(condition=[<=($0, 10)])
- : +- LogicalTableScan(table=[[default_catalog,
default_database, MyTable, source: [TestTableSource(a, b, c)]]])
- +- LogicalTableScan(table=[[default_catalog, default_database,
MyTable2, source: [TestTableSource(a, b, c, d, e)]]])
+ : +- LogicalFilter(condition=[<=($0, 10)])
+ : +- LogicalTableScan(table=[[default_catalog, default_database,
MyTable, source: [TestTableSource(a, b, c)]]])
+ +- LogicalProject(a=[$2], b1=[$1])
+ +- LogicalFilter(condition=[=($2, $0)])
+ +- LogicalJoin(condition=[true], joinType=[inner])
+ :- LogicalProject(a1=[$0], b1=[$1])
+ : +- LogicalFilter(condition=[<=($0, 10)])
+ : +- LogicalTableScan(table=[[default_catalog,
default_database, MyTable, source: [TestTableSource(a, b, c)]]])
+ +- LogicalTableScan(table=[[default_catalog, default_database,
MyTable2, source: [TestTableSource(a, b, c, d, e)]]])
]]>
</Resource>
<Resource name="planAfter">
<![CDATA[
-Calc(select=[a, b], where=[<=(a, 10)], reuse_id=[1])
-+- TableSourceScan(table=[[default_catalog, default_database, MyTable, source:
[TestTableSource(a, b, c)]]], fields=[a, b, c])
+Calc(select=[a AS a1, b AS b1], reuse_id=[1])
++- Calc(select=[a, b, c], where=[<=(a, 10)])
+ +- TableSourceScan(table=[[default_catalog, default_database, MyTable,
source: [TestTableSource(a, b, c)]]], fields=[a, b, c])
-Calc(select=[a0 AS a, b AS b1], reuse_id=[2])
-+- HashJoin(joinType=[InnerJoin], where=[=(a0, a)], select=[a, b, a0],
build=[right])
- :- Exchange(distribution=[hash[a]])
+Calc(select=[a, b1], reuse_id=[2])
++- HashJoin(joinType=[InnerJoin], where=[=(a, a1)], select=[a1, b1, a, b, c,
d, e], build=[left])
+ :- Exchange(distribution=[hash[a1]])
: +- Reused(reference_id=[1])
+- Exchange(distribution=[hash[a]])
- +- Calc(select=[a])
- +- TableSourceScan(table=[[default_catalog, default_database,
MyTable2, source: [TestTableSource(a, b, c, d, e)]]], fields=[a, b, c, d, e])
+ +- TableSourceScan(table=[[default_catalog, default_database, MyTable2,
source: [TestTableSource(a, b, c, d, e)]]], fields=[a, b, c, d, e])
Sink(name=[`default_catalog`.`default_database`.`sink1`], fields=[a, b1])
+- Reused(reference_id=[2])
Sink(name=[`default_catalog`.`default_database`.`sink2`], fields=[a, b1])
-+- Union(all=[true], union=[a, b])
++- Union(all=[true], union=[a1, b1])
:- Reused(reference_id=[1])
Review comment:
The test query is `SELECT * FROM table1 UNION ALL SELECT * FROM table2` so
the name is actually correct. Because the union fields digest come from
`getRowType.getFieldNames.mkString(", ")`, the `UNION` use
`ReturnTypes.LEAST_RESTRICTIVE` to infer the union type, for which we always
use the field name of the first union input.
----------------------------------------------------------------
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.
For queries about this service, please contact Infrastructure at:
[email protected]
With regards,
Apache Git Services
