== API Changes ==
We welcome users to compile Spark applications against 1.0. There are
a few API changes in this release. Here are links to the associated
upgrade guides - user facing changes have been kept as small as
That does seem like a good fix, but right now we are only blocking the
release on major regressions that affect all components. So I don't
think this is sufficient to block it from going forward and cutting a
new candidate. This is because we are in the very late stage of the
We can slot that for the 1.0.1 release and merge it into the 1.0
branch so people can get access to the fix easily.
I think the question for me would be does this only happen when you call partitionBy, or always? And how common do you expect calls to partitionBy to be? If we can wait for 1.0.1 then I’d wait on this one.
I made the necessary interface changes to my apps that use MLLib and tested all of my code against rc11 on Fedora 20 and OS X 10.9.3. (The Fedora Rawhide package remains at 0.9.1 pending some additional dependency packaging work.)
This voting is now closed, and it successfully passes with 13 "+1"
votes and one "0" vote.
Thanks to everyone who tested the RC and voted. Here are the totals:
+1: (13 votes)
0: (1 vote)
-1: (0 vote)
* = binding
Please hold off announcing Spark 1.0.0 until Apache Software
Foundation makes the press release tomorrow. Thank you very much for
On Thu, May 29, 2014 at 9:14 AM, Patrick Wendell <[hidden email]> wrote: