Let me squash the commits. They both seem ready - at least, they both have at least one approval, and I have addressed all the comments I received.
Mihai ________________________________ From: Julian Hyde <jh...@apache.org> Sent: Wednesday, October 9, 2024 11:46 AM To: dev@calcite.apache.org <dev@calcite.apache.org> Subject: Re: [DISCUSS] Towards Calcite 1.38.0 For RC2 we need fixes for 6616, 6617 and 6620. I know 6616 is good. Mihai, are the PRs 6617 and 6620 ready to merge? If so, I'll merge them, adjust the release notes and launch RC2. Julian On Thu, Sep 26, 2024 at 11:00 AM Julian Hyde <jh...@apache.org> wrote: > > Where are we on 1.38? Down to 2: 4512 and 5156. Please review and merge today. > > I have draft release notes. Please review them in the PR: > https://github.com/apache/calcite/pull/3979. > > Julian > > Case status: > > 1. https://issues.apache.org/jira/browse/CALCITE-6161 equalsDeep > > No longer in 1.38 > > 2. https://issues.apache.org/jira/browse/CALCITE-5156 implicit number type > cast > > No longer in 1.38 > > 3. https://issues.apache.org/jira/browse/CALCITE-6560 (supportsNegativeScale) > > I don't think this one is going to make it. I am seeing a classloader > deadlock when I run the tests, and that makes me very nervous. Also, I > would like to fix the related bug > https://issues.apache.org/jira/browse/CALCITE-6598 at the same time. > > 4. https://issues.apache.org/jira/browse/CALCITE-6300 Function > MAP_VALUES/MAP_KEYS gives exception > > No longer in 1.38 because the proposed solution was hacky > > 5. https://issues.apache.org/jira/browse/CALCITE-4512 group by expression > > https://github.com/apache/calcite/pull/3929 needs review > > 6. https://issues.apache.org/jira/browse/CALCITE-5156 Support implicit > number type cast for IN Sub-query > > On Wed, Sep 18, 2024 at 1:21 AM Ruben Q L <rube...@gmail.com> wrote: > > > > I'll try to look at a few PRs in the next days. > > > > I've also updated our Jira dashboard [1] (version: 1.38) to have a clear > > overview about the status of the issues for the next release. > > > > [1] > > https://issues.apache.org/jira/secure/Dashboard.jspa?selectPageId=12333950 > > > > > > On Tue, Sep 17, 2024 at 8:16 PM Mihai Budiu <mbu...@gmail.com> wrote: > > > > > I am now at Current, the confluent conference, but I hope to find some > > > time on Friday to do a few. > > > ________________________________ > > > From: Julian Hyde <jh...@apache.org> > > > Sent: Tuesday, September 17, 2024 12:08:00 PM > > > To: dev@calcite.apache.org <dev@calcite.apache.org> > > > Subject: Re: [DISCUSS] Towards Calcite 1.38.0 > > > > > > I merged fixes for 4 cases yesterday. Can I get some help reviewing > > > and merging PRs (especially those in category 1 and 2)? > > > > > > There are about 6 regular committers who merge a few PRs before each > > > release. (You know who you are -- and I am profoundly grateful! Your > > > work keeps the project going.) If you could each do 2 or 3 PRs by the > > > end of the weekend, we can get this release done. > > > > > > Julian > > > > > > > > > On Mon, Sep 16, 2024 at 10:56 AM Julian Hyde <jhyde.apa...@gmail.com> > > > wrote: > > > > > > > > Yeah, sorry, I don’t often use the GitHub ‘approve’ button. My verbal > > > approvals are often conditional. The ‘approve’ button doesn’t allow > > > nuance. > > > > > > > > > On Sep 15, 2024, at 3:26 PM, Mihai Budiu <mbu...@gmail.com> wrote: > > > > > > > > > > You have not listed CALCITE-4918, > > > https://issues.apache.org/jira/browse/CALCITE-4918 for which I have a PR, > > > which I think you have approved using comments > > > https://github.com/apache/calcite/pull/3947, but not using the official > > > "approve" button. > > > > > > > > > > I was hoping other people would weigh on this design and > > > implementation. > > > > > > > > > > Mihai > > > > > > > > > > ________________________________ > > > > > From: Julian Hyde <jh...@apache.org> > > > > > Sent: Sunday, September 15, 2024 3:12 PM > > > > > To: dev@calcite.apache.org <dev@calcite.apache.org> > > > > > Subject: Re: [DISCUSS] Towards Calcite 1.38.0 > > > > > > > > > > I've logged https://issues.apache.org/jira/browse/CALCITE-6582 to > > > > > track the release. > > > > > > > > > > There are 26 open bugs [1] marked to fix for 1.38. I have organized > > > > > them into 5 categories. Can people please review and merge those in > > > > > category 1? To get a release vote next week, so let's get busy > > > > > reviewing and merging bugs this week. > > > > > > > > > > Category 1. The following have a PR available that seems close to > > > > > ready. If these are good, let's get these merged! > > > > > * CALCITE-6572 Add more tests for NULL arguments to TO_CHAR functions > > > > > * CALCITE-6554 nested correlated sub-query in aggregation does not > > > > > have inner correlation variable bound to inner projection > > > > > * CALCITE-6550 Improve SQL function overloading > > > > > * CALCITE-6343 AS alias operator strips MEASUREness from measures > > > > > * CALCITE-6161 The equalsDeep of sqlCall should compare sqlOperator's > > > sqlKind > > > > > * CALCITE-6226 Wrong ISOWEEK and no ISOYEAR on BigQuery FORMAT_DATE > > > > > * CALCITE-2067 RexLiteral cannot represent accurately floating point > > > > > values, including NaN, Infinity > > > > > * CALCITE-5156 Support implicit number type cast for IN Sub-query > > > > > * CALCITE-6071 RexCall should carry source position information for > > > > > runtime error reporting > > > > > * CALCITE-6498 Elasticsearch multi-field mappings do not work > > > > > * CALCITE-6522 MAP_KEYS and MAP_VALUES function should throw if a key > > > > > value is null > > > > > * CALCITE-6020 SqlToRelConverter should not replace windowed SUM with > > > > > equivalent expression using SUM0 > > > > > * CALCITE-4512 GROUP BY expression with argument name same with SELECT > > > > > field and alias causes validation error > > > > > > > > > > Category 2. The following cases have a draft PR. Is it ready for 1.38? > > > > > * CALCITE-6431 Implement the SINGLE_VALUE aggregation in > > > > > HiveSqlDialect And SparkSQLDialect > > > > > * CALCITE-6300 Function MAP_VALUES/MAP_KEYS gives exception when > > > > > mapVauleType and mapKeyType not equals map Biggest mapKeytype or > > > > > mapValueType > > > > > > > > > > Category 3. Need to decide whether to accept or close: > > > > > * CALCITE-6207 Improve connection resource in JDBCUtils > > > > > * CALCITE-6411 Support Collect in ToLogicalConverter > > > > > * CALCITE-6352 The map_contains_key function may return true when the > > > > > key and mapkeytype types are different. > > > > > > > > > > Category 4. The following seem to have no PR. Are there any worth > > > > > holding the train for? > > > > > * CALCITE-6277 About the processing of +infinity parameters and > > > > > -infinity parameters in math function > > > > > * CALCITE-6203 Avg operator test in SparkAdapterTest > > > > > * CALCITE-6326 Add LOG10 and LN function (enabled in Mysql, Spark > > > library) > > > > > * CALCITE-6259 The implementation of the Log library operator does not > > > > > match the actual dialect behavior. > > > > > * CALCITE-5541 Upgrade JavaCC > > > > > * CALCITE-6469 Join on condition generates wrong plan when the > > > > > condition is IN sub-query > > > > > * CALCITE-6463 alias generation issue about correlated subquery > > > handling > > > > > > > > > > Category 5. Are there any more cases with PR ready (i.e. potential > > > category 1)? > > > > > > > > > > Julian > > > > > > > > > > [1] > > > https://issues.apache.org/jira/issues/?jql=project%20%3D%20CALCITE%20AND%20fixVersion%20%3D%201.38.0%20and%20resolution%20%3D%20Unresolved%20 > > > > > > > > > > On Fri, Sep 13, 2024 at 6:20 PM Xiong Duan <xi...@apache.org> wrote: > > > > >> > > > > >> +1 We start voting in about ten days. We will merge the almost-ready > > > PR. > > > > >> > > > > >> Cancai Cai <caic68...@gmail.com> 于2024年9月14日周六 08:55写道： > > > > >>> > > > > >>> +1，There are still many valuable PRs in Calcite. We can try to merge > > > them before 1.38. > > > > >>> > > > > >>> [1] https://github.com/apache/calcite/pulls > > > > >>> > > > > >>> Best wishes, > > > > >>> Cancai Cai > > > > >>> > > > > >>>> 2024年9月14日 04:04，Julian Hyde <jhyde.apa...@gmail.com> 写道： > > > > >>>> > > > > >>>> +1 we’re ready for a release, and I’m happy to be RM. > > > > >>>> > > > > >>>> I think we could have a first RC and vote in about ten days. What > > > do others feel about timing? > > > > >>>> > > > > >>>> Julian > > > > >>>> > > > > >>>> > > > > >>>>> On Sep 13, 2024, at 4:58 AM, Benchao Li <libenc...@apache.org> > > > wrote: > > > > >>>>> > > > > >>>>> Last release date is 2024-05-06, sorry for the typo. > > > > >>>>> > > > > >>>>> Benchao Li <libenc...@apache.org> 于2024年9月13日周五 19:52写道： > > > > >>>>>> > > > > >>>>>> It's been a bit more than 4 months since our last release > > > (2023-05-06) > > > > >>>>>> [1] and there are currently almost 150 new commits in main > > > branch. Per our > > > > >>>>>> tradition of producing one release every 2-3 months, I think it's > > > time > > > > >>>>>> to consider for next release now. > > > > >>>>>> > > > > >>>>>> According to [2], the following release managers would be: > > > > >>>>>> - 1.38.0 Julian Hyde > > > > >>>>>> - 1.39.0 Stamatis Zampetakis (or maybe 1.40.0 depending on the > > > timing) > > > > >>>>>> - 1.40.0 Hongyu Guo > > > > >>>>>> - 1.41.0 Jing Zhang (she said 1.39.0 in the ML, but according to > > > the > > > > >>>>>> order of the email, the next one should be 1.41.0) > > > > >>>>>> > > > > >>>>>> [1] https://calcite.apache.org/news/2024/05/06/release-1.37.0/ > > > > >>>>>> [2] > > > https://lists.apache.org/thread/kx56c4ow50g50hx2y8ybp631jmtnwnj6 > > > > >>>>>> > > > > >>>>>> -- > > > > >>>>>> > > > > >>>>>> Best, > > > > >>>>>> Benchao Li > > > > >>>>> > > > > >>>>> > > > > >>>>> > > > > >>>>> -- > > > > >>>>> > > > > >>>>> Best, > > > > >>>>> Benchao Li > > > > >>>> > > > > >>> > > > > > > >
