Github user marmbrus commented on the pull request:
https://github.com/apache/spark/pull/2475#issuecomment-61338760
Okay, thanks for all the comments.
I've done the following:
- Changed the various base relations into `abstract class` instead of
`trait`.
- Removed `Expression`, `Attribute`, `Seq` from all function signatures
instead using `Filter`, `String` and `Array` (a simple case class that is used
only in this package) respectively. This should make binary compatibility /
java interop easier, and also takes care of some things like `a = 1` vs `1 = a`
for the user. Right now this only supports equality, but that will be easy to
extend. This comes at the cost of expressivity but its probably worth it for
long term compatibility.
- Added test cases that verify that pushdown of filters and column pruning
are happening as expected.
- Pulled `Strategy` out of the Planner so it is now possible to integrate
very tightly if so desired. This interface is appropriately commented as being
unstable and only for experimentation.
- Addressed a bunch of other small comments
---
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]
