Github user MLnick commented on the pull request:
https://github.com/apache/spark/pull/455#issuecomment-44519603
Hi @mateiz
Thanks for taking a look and the comments. I'll make the necessary changes.
On the inference of key/value classes, I will take a deeper look and see
what the best approach may be for non-SF formats. Tests with incorrect classes
is a good idea to stress test it and see where it breaks.
I had a similar idea in mind for the ```PythonConverter``` interface. I can
try to flesh that out a bit for this PR, but I sort of wanted to get something
in that is clearly marked as VERY experimental so that we could perhaps get
some feedback from usage "in the wild". I suspect there will be many edge cases
with this.
Thoughts?
As per the mailing list discussion on HBase, I will try to get an HBase
example working to iron out some kinks
(http://apache-spark-user-list.1001560.n3.nabble.com/Python-Spark-and-HBase-td6142.html#a6507).
At a minimum, I'd like to have this work relatively easily with ES (already
does), HBase, Cassandra.
---
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.
---
