luoyuxia commented on code in PR #20016:
URL: https://github.com/apache/flink/pull/20016#discussion_r1036606895
##########
flink-connectors/flink-connector-hive/src/main/java/org/apache/flink/connectors/hive/HiveTableSource.java:
##########
@@ -257,6 +279,12 @@ public void applyProjection(int[][] projectedFields,
DataType producedDataType)
this.projectedFields = Arrays.stream(projectedFields).mapToInt(value
-> value[0]).toArray();
}
+ @Override
+ public Result applyFilters(List<ResolvedExpression> filters) {
+ this.filters = new ArrayList<>(filters);
+ return Result.of(new ArrayList<>(filters), new ArrayList<>(filters));
Review Comment:
Yes, you're right. We just pushdown the orc filter to orc input format.
The reason is some formats don't implement filter pushdown or don't support
filter pushdown. So, we still make the plan keep all the filters, which can
make sure the result is correct.
I think i'll be better to prun these filters in here, but it's
hard/impossible to do that.
--
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]
