+1 This is a very good idea.
Best, Jark On Wed, 9 Dec 2020 at 10:43, Xingbo Huang <hxbks...@gmail.com> wrote: > +1 > > This is a very good proposal.In release-1.12, many newly added features are > only supported on the blink planner. For example, the newly added features > of PyFlnk in FLIP-137[1] and FLIP-139[2] are only available on the blink > planner. > > [1] > > https://cwiki.apache.org/confluence/display/FLINK/FLIP-137%3A+Support+Pandas+UDAF+in+PyFlink > [2] > > https://cwiki.apache.org/confluence/display/FLINK/FLIP-139%3A+General+Python+User-Defined+Aggregate+Function+Support+on+Table+API > > Best, > Xingbo > > Leonard Xu <xbjt...@gmail.com> 于2020年12月9日周三 上午9:46写道： > > > +1 > > > > Very good proposal, blink planner has been the default planner since > Flink > > 1.11.0, and many new features are not supported in legacy planner. > > > > Best, > > Leonard > > > > > 在 2020年12月9日，05:15，Arvid Heise <ar...@ververica.com> 写道： > > > > > > +1, add a small info box about legacy planner and point to 1.11 doc > > > (nothing should have changed) > > > > > > Ideally, "legacy" and "blink" does not appear anywhere in the doc > (except > > > for that info box) > > > > > > On Tue, Dec 8, 2020 at 6:35 PM Marta Paes Moreira <ma...@ververica.com > > > > > wrote: > > > > > >> +1, this is confusing (esp. for new users) and also creates more and > > more > > >> "annotation clutter" as new features are added. > > >> > > >> On Tue, Dec 8, 2020 at 5:30 PM Aljoscha Krettek <aljos...@apache.org> > > >> wrote: > > >> > > >>> +1 > > >>> > > >>> Yes, please! > > >>> > > >>> On 08.12.20 16:52, David Anderson wrote: > > >>>> I agree -- I think separating out the legacy planner info should > make > > >>>> things clearer for everyone, and then some day we can simply drop > it. > > >>> Plus, > > >>>> doing it now will make it easier to make improvements to the docs > > going > > >>>> forward. > > >>>> > > >>>> David > > >>>> > > >>>> On Tue, Dec 8, 2020 at 4:38 PM Timo Walther <twal...@apache.org> > > >> wrote: > > >>>> > > >>>>> Hi Seth, > > >>>>> > > >>>>> this is a very good idea. We might not be able to remove the legacy > > >>>>> planner immediately but at least we can make the docs easier for > > >> current > > >>>>> and future users of the Blink planner. > > >>>>> > > >>>>> Making the SQL docs Blink-only with a dedicated legacy planner page > > >>>>> sounds good to me. > > >>>>> > > >>>>> Regards, > > >>>>> Timo > > >>>>> On 08.12.20 16:36, Seth Wiesman wrote: > > >>>>>> Hi Everyone, > > >>>>>> > > >>>>>> I've been spending a lot of time recently working on the SQL > > >>>>> documentation > > >>>>>> and I'm finding it very difficult to explain semantics as the two > > >> table > > >>>>>> planners continue to diverge. As Blink has been the default > planner > > >> for > > >>>>>> some time, and 1.12 now offers bounded data stream support, how > does > > >>> the > > >>>>>> community feel about making the documentation "blink only"? > > >>>>>> > > >>>>>> We would update the documentation to assume users are always using > > >> the > > >>>>>> Blink planner. As the legacy planner still exists we would create > a > > >>>>>> dedicated legacy planner page for users that have not migrated for > > >>>>> whatever > > >>>>>> reason - likely dataset interop. On this page, we would clearly > list > > >>> the > > >>>>>> features that are not supported by the legacy planner and any > > >> semantics > > >>>>>> that differ from the Blink planner. > > >>>>>> > > >>>>>> Seth > > >>>>>> > > >>>>> > > >>>>> > > >>>> > > >>> > > >>> > > >> > > > > > > > > > -- > > > > > > Arvid Heise | Senior Java Developer > > > > > > <https://www.ververica.com/> > > > > > > Follow us @VervericaData > > > > > > -- > > > > > > Join Flink Forward <https://flink-forward.org/> - The Apache Flink > > > Conference > > > > > > Stream Processing | Event Driven | Real Time > > > > > > -- > > > > > > Ververica GmbH | Invalidenstrasse 115, 10115 Berlin, Germany > > > > > > -- > > > Ververica GmbH > > > Registered at Amtsgericht Charlottenburg: HRB 158244 B > > > Managing Directors: Timothy Alexander Steinert, Yip Park Tung Jason, Ji > > > (Toni) Cheng > > > > >
