Github user rdblue commented on the issue:
https://github.com/apache/spark/pull/13701
@gatorsmile, sorry for the delay, I was evidently not getting notifications
until I changed some settings yesterday.
There are a few tests in Parquet that generate files with test data that
would be appropriate. The simplest one is
[TestReadWriteEncodingStats](https://github.com/apache/parquet-mr/blob/master/parquet-hadoop/src/test/java/org/apache/parquet/hadoop/TestReadWriteEncodingStats.java#L85-94)
where you can see how to build a file with at least one column that has stats
you can verify filters with. To get it working for row groups, you'd just need
to set the row group size to something small, like 8k, and bump up the number
of records until you get a few groups.
Unfortunately, we don't have any pre-built test cases you can use for a
performance baseline, though. I think the best you can do with the data
generation approach is validate that row groups are being filtered using an
approach like you already have, counting the number of rows.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
