Github user jkbradley commented on the pull request:
https://github.com/apache/spark/pull/3637#issuecomment-66509244
Thanks everyone for all of the comments!
@shivaram No problem, thanks for checking out the design doc! The 2 main
use cases you listed are correct. There is a remaining question about whether
the strongly typed API should remain public; my initial plan was to make it
public, but the above discussion is making me wonder if part should be private.
@Lewauthe I agree about discussing whether the typed API should be public
or private. (But I'm not quite sure what you meant about the optimization API;
please clarify.) I hope the updates to the design doc below help.
@srowen It sounds like we're converging to a solution, and I hope some use
cases in the design doc will help.
I've updated the design doc for
[https://issues.apache.org/jira/browse/SPARK-3702], which is the parent task of
[https://issues.apache.org/jira/browse/SPARK-4789]. (I'm asking someone to add
the subtask connection.) The design doc is linked from the main JIRA and also
here:
[https://docs.google.com/document/d/1I-8PD0DSLEZzzXURYZwmqAFn_OMBc08hgDL1FZnVBmw/edit?usp=sharing]
Perhaps we can discuss more on the JIRA rather than here.
---
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]
