flink-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Fabian Hueske <fhue...@gmail.com>
Subject Re: Preparing Table API & SQL for Flink 1.1.0
Date Wed, 25 May 2016 15:04:36 GMT
Hi everybody,

thanks for the feedback so far.

I just went over JIRA and increased the priority of some of the issues I
listed (the TOP 6, out of which 1 is fixed, + FLINK-3723) to CRITICAL to
distinguish them from the rest.

I found two more critical issue that should be fixed:
- FLINK-3971: Incorrect handling of null values in aggregation functions
- FLINK-3944: Reordering Cartesian products and joins to resolve the
Cartesian products into equi-joins.

@Vasia: I think that FLINK-3656 is important to resolve, but I see this
independent of the release.

Most of these issues are currently work in progress which is very good.
It would be great if those who are looking for an issue to work on would
consider to pick a critical one.

Please do also open new issues if you find bugs or important improvements
that should be fixed for 1.1.0.

Thanks, Fabian

2016-05-20 16:09 GMT+02:00 Vasiliki Kalavri <vasilikikalavri@gmail.com>:

> Hey all,
>
> @Fabian: thanks for compiling the list of issues and trying out TPC-H I
> think it would be nice to include the first 5 from your list in 1.1.0. What
> about FLINK-3656 (re-working the tests)? Do we want to do this before the
> release as well?
>
> Great to see Timo willing to coordinate and Yijie willing to help!
> Unfortunately, I won't have much time to offer in the following months, but
> I'll try to keep myself up-to-date :)
>
> Cheers,
> -Vasia.
>
> On 20 May 2016 at 15:59, Ufuk Celebi <uce@apache.org> wrote:
>
> > On Fri, May 20, 2016 at 3:32 PM, Timo Walther <twalthr@apache.org>
> wrote:
> > > What are the plans for 1.1.0 release so far?
> >
> > Hey Timo,
> >
> > following the 3 months release schedule, a release would be due soon
> > (June 8th). My personal opinion is that exact date is probably a
> > little optimistic, but June sounds about right.
> >
> > We should definitely start the general release discussion. Stephan
> > already suggested this in another thread, so you can expect this to
> > happen soon.
> >
> > It's very good timing that you started preparing SQL for 1.1. That
> > will probably be one of the major features of 1.1. :-)
> >
>

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message