xuyangzhong commented on a change in pull request #17652:
URL: https://github.com/apache/flink/pull/17652#discussion_r741892214
##########
File path:
flink-table/flink-table-planner/src/test/resources/org/apache/flink/table/planner/plan/nodes/exec/stream/TableSourceJsonPlanTest_jsonplan/testPartitionPushDown.out
##########
@@ -64,7 +61,7 @@
"b" : "INT"
} ]
},
- "description" : "TableSourceScan(table=[[default_catalog,
default_database, PartitionTable, filter=[], partitions=[{p=A}], project=[a,
b], metadata=[]]], fields=[a, b])",
+ "description" : "TableSourceScan(table=[[default_catalog,
default_database, PartitionTable, partitions=[{p=A}], project=[a, b],
metadata=[]]], fields=[a, b])",
Review comment:
Sorry, I test some cases and find one bad case when the table source has
a watermark. It will miss the rule in optimize period "predicate_pushdown" and
match the "logical". So PushDownProject will be called first in "logical", and
then PushDownWatermark will be called later in "logical_rewrite", and then
PushDownFilter in "logical_rewrite". The source description just like :
`TableSourceScan(table=[[default_catalog, default_database, TestTable,
project=[a, b, c], metadata=[], watermark=[c], filter=[>(a, 1)]]], fields=[a,
b, c])`
Maybe we need to sort the specs in table source to print them in order if
the order is important.
--
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]
