I saw you introduced a config parameter to disable previous behavior. That’s fine, but let’s do it in such a way that the parameter can disappear in a (near) future release. I hate config parameters.
Julian > On Jul 3, 2018, at 1:26 AM, Zoltan Haindrich <[email protected]> wrote: > > Hello, > > I plan to submit a patch to restore earlier performance today - I think I've > came up with a plan to make a more complete fix; but that will definetly take > more time than a few days...and of course there is a risk that I'm wrong :) > > cheers, > Zoltan > > > On 07/02/2018 09:13 PM, Volodymyr Vysotskyi wrote: >> I think I can finish CALCITE-2392 on Wed or Thursday. >> Kind regards, >> Volodymyr Vysotskyi >> пн, 2 лип. 2018 о 21:22 Julian Hyde <[email protected]> пише: >>> OK, so 2384 and 2392 are the critical path. >>> >>> Zoltan, when do you think you can finish 2392? >>> >>> Volodymr, when do you think you can finish 2384? >>> >>> When we have these answers we can set a target date for the RC. >>> >>> Julian >>> >>> >>>> On Jul 2, 2018, at 11:13 AM, Volodymyr Vysotskyi <[email protected]> >>> wrote: >>>> >>>> Hi, >>>> >>>> Here is the list of the Jiras, which in my opinions should be fixed >>> before >>>> the release: >>>> - https://issues.apache.org/jira/browse/CALCITE-2384 Performance issue >>>> - https://issues.apache.org/jira/browse/CALCITE-2392 Regression >>>> appeared after 1.16 (not sure about blocker status for this Jira) >>>> - https://issues.apache.org/jira/browse/CALCITE-2379 CVSS - I will >>> merge PR >>>> soon >>>> >>>> Nice to have fixes (most of them from the previous list): >>>> - https://issues.apache.org/jira/browse/CALCITE-194 Arrays in MongoDB - >>>> AFAIK, Igor Kryvenko is working on the moving calcite-test-dataset to >>>> docker, so after that, we can merge it. >>>> - https://issues.apache.org/jira/browse/CALCITE-2259 Java8 syntax >>>> - https://issues.apache.org/jira/browse/CALCITE-2280 Babel (blocked by >>> 2259) >>>> - https://issues.apache.org/jira/browse/CALCITE-2339 JDBC adapter >>> timestamp >>>> arithmetic >>>> >>>> Depending on the time, required to fix CALCITE-2384 >>>> <https://issues.apache.org/jira/browse/CALCITE-2384> and CALCITE-2392 >>>> <https://issues.apache.org/jira/browse/CALCITE-2392>, we can set a >>> target >>>> date for the first RC, but I don't know the complexity of these issues >>> and >>>> current status for CALCITE-2384 >>>> <https://issues.apache.org/jira/browse/CALCITE-2384>. >>>> >>>> Kind regards, >>>> Volodymyr Vysotskyi >>>> >>>> >>>> пн, 2 лип. 2018 о 20:42 Julian Hyde <[email protected]> пише: >>>> >>>>>> On Jun 29, 2018, at 7:29 PM, Francis Chuang <[email protected] >>>>> <mailto:[email protected]> wrote: >>>>>> >>>>>> Can someone please confirm that the following PRs can be closed? >>>>>> https://github.com/apache/calcite/pull/180 < >>>>> https://github.com/apache/calcite/pull/180> >>>>>> https://github.com/apache/calcite/pull/17 < >>>>> https://github.com/apache/calcite/pull/17> >>>>>> https://github.com/apache/calcite/pull/422 < >>>>> https://github.com/apache/calcite/pull/422> >>>>> >>>>> I reviewed those PRs. They can be closed. I added comments asking the >>>>> original contributors to close the PRs. This is always the preferred >>> option. >>>>> >>>>> Are we agreed that adding ‘Close #NNN’ to another commit - while not >>>>> perfect - is the least bad option for closing abandoned PRs? If so I’ll >>> do >>>>> that for these PRs in about 24 hours. >>>>> >>>>> Volodymyr, >>>>> >>>>> Do we have a list of what’s left for 1.17? I think we can set a target >>>>> date for the first RC. >>>>> >>>>> Julian >>>>> >>>>> >>> >>>
