Dan Burkert has posted comments on this change.

Change subject: Adding kudu datasource for spark
......................................................................


Patch Set 6:

Hey Chris,

I was looking at this today, and ended up making a bunch of small changes in a 
local copy.  I've put the changes up here: 
https://github.com/danburkert/kudu/commit/9651525e3061446dc07be6295490f0a28de7c2c0.
  It looks like since then you've made some of the same changes in the newer 
revisions.  Take a look at my commit, and feel free to pull any of the changes 
in to this review.  Alternatively I can rebase against your latest and push a 
new revision if you want to take them wholesale.

The biggest structural change is that I added a KuduRow class so that 
individual RowResults don't have to be unpacked every time.  I'm not sure how 
this is going to interact with serialization, if at all.  If it has bad 
consequences we can add some explicit serialization.

-- 
To view, visit http://gerrit.cloudera.org:8080/2848
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings

Gerrit-MessageType: comment
Gerrit-Change-Id: I0f91772f58e9eee9de45901866867e9a5014cfbe
Gerrit-PatchSet: 6
Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-Owner: Chris George <[email protected]>
Gerrit-Reviewer: Dan Burkert <[email protected]>
Gerrit-Reviewer: Kudu Jenkins
Gerrit-HasComments: No

Reply via email to