GitHub user HyukjinKwon opened a pull request:
https://github.com/apache/spark/pull/10502
[SPARK-12355][SQL] Implement unhandledFilter interface for Parquet
https://issues.apache.org/jira/browse/SPARK-12355
This is similar with https://github.com/apache/spark/pull/10427.
As discussed here https://github.com/apache/spark/pull/10221, this PR
implemented `unhandledFilter` to remove duplicated Spark-side filtering.
In case of Parquet, the columns referenced in pushed down filters should be
given to `org.apache.spark.sql.parquet.row.requested_schema` whereas general
datasources such as JDBC do not require the columns.
However, the current interface of `DataSourceStrategy` removes the columns
only referenced in pushed down filters. Therefore, this PR resolved this
problem by manually generating the columns referenced in pushed down filters.
You can merge this pull request into a Git repository by running:
$ git pull https://github.com/HyukjinKwon/spark SPARK-12355
Alternatively you can review and apply these changes as the patch at:
https://github.com/apache/spark/pull/10502.patch
To close this pull request, make a commit to your master/trunk branch
with (at least) the following in the commit message:
This closes #10502
----
commit a55ad54a820f095e5116df05979f169c0fe8e0cf
Author: hyukjinkwon <[email protected]>
Date: 2015-12-29T04:34:53Z
Implement unhandled filters for Parquet
commit cf331a453c3f99ee40ee5ca6f5029dadee3d07f6
Author: hyukjinkwon <[email protected]>
Date: 2015-12-29T04:36:42Z
Correct existing tests
----
---
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]
