rdblue commented on code in PR #15461:
URL: https://github.com/apache/iceberg/pull/15461#discussion_r2893132765
##########
core/src/main/java/org/apache/iceberg/util/TableScanUtil.java:
##########
@@ -181,7 +232,8 @@ public static <T extends PartitionScanTask>
List<ScanTaskGroup<T>> planTaskGroup
List<T> groupingKeyTasks = entry.getValue();
Iterables.addAll(
taskGroups,
- toTaskGroupIterable(groupingKey, groupingKeyTasks, splitSize,
lookback, weightFunc));
+ toTaskGroupIterable(
+ groupingKey, groupingKeyTasks, splitSize, lookback, weightFunc,
fileIO));
Review Comment:
I definitely prefer not serializing FileIO per task. This should be
broadcasted once for a scan and reused. FileIO serialization is expensive!
--
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]
