I’ve just committed 707 and 2061. That leaves the following:

https://issues.apache.org/jira/browse/CALCITE-2018 AssertionError: rel has 
lower cost (julian to review PR #552)
https://issues.apache.org/jira/browse/CALCITE-2054 UPDATE with parameters 
(julian to review PR #568)
https://issues.apache.org/jira/browse/CALCITE-1861 Spatial indexes
https://issues.apache.org/jira/browse/CALCITE-1616 Data profiler

Also seeing some minor issues in Windows, and possibly a performance regression 
in RelMetadataTest.testPullUpPredicatesForExprsItr.

First RC probably tomorrow. Not all of the above will make the cut.

Has anyone had chance to test any of the adapters?

We’re trying to get to stable, so please don’t make commits without my +1 (I 
have been running exhaustive tests before committing).

Julian



> On Nov 27, 2017, at 11:10 AM, Julian Hyde <jh...@apache.org> wrote:
> 
> Thanks for reviewing, Michael. Yes, it’s a lot.
> 
> Materialized views were a big motivation for doing this. You currently have 
> to do some tricky work with APIs to define materialized views (mappings 
> inside the planner, predicates, statistics). I wanted to demonstrate it 
> working using a few lines of SQL.
> 
> I hadn’t really thought of DDL + adapters. Yes, I guess someone could do
> 
>  CREATE TABLE MyCassandraSchema.MyTable (int i, date j)
> 
> and Calcite would create a table in Cassandra. We’d need to add SPIs to 
> Schema to make that work.
> 
> Julian
> 
> 
> 
>> On Nov 23, 2017, at 9:00 AM, Michael Mior <mm...@uwaterloo.ca> wrote:
>> 
>> Reviewed 707. It's a lot of code, so I can't say I did a particularly
>> thorough job, but no clear issues stuck out to me and the approach seems
>> sound. Glad to see materialized views are included in there. Not suggesting
>> this should be part of 707, but looking forward to having some adapters
>> implement basic DDL in the future. It won't make sense for all of them, but
>> I can certainly see it being useful. Awesome work!
>> 
>> --
>> Michael Mior
>> mm...@apache.org
>> 
>> 2017-11-22 18:28 GMT-05:00 Julian Hyde <jh...@apache.org>:
>> 
>>> An update on release 1.15. We’re still on track for first RC on Monday.
>>> 
>>> We never seem to do a good job of verifying Calcite’s adapters. Can I have
>>> volunteers to validate the Cassandra, Mongo, and Druid adapters? I plan to
>>> test Calcite on Windows.
>>> 
>>> The following issues remain for the release:
>>> https://issues.apache.org/jira/browse/CALCITE-707 <
>>> https://issues.apache.org/jira/browse/CALCITE-707> Server and DDL (julian)
>>> https://issues.apache.org/jira/browse/CALCITE-2018 <
>>> https://issues.apache.org/jira/browse/CALCITE-2018> AssertionError: rel
>>> has lower cost (julian to review PR #552)
>>> https://issues.apache.org/jira/browse/CALCITE-2041 <
>>> https://issues.apache.org/jira/browse/CALCITE-2041> Adding the ability to
>>> turn off nullability matching (julian to review PR #563 and #570)
>>> https://issues.apache.org/jira/browse/CALCITE-2054 <
>>> https://issues.apache.org/jira/browse/CALCITE-2054> UPDATE with
>>> parameters (julian to review PR #568)
>>> https://issues.apache.org/jira/browse/CALCITE-2061 <
>>> https://issues.apache.org/jira/browse/CALCITE-2061> Parameters in OFFSET
>>> and FETCH (julian to review PR #569)
>>> https://issues.apache.org/jira/browse/CALCITE-1861 <
>>> https://issues.apache.org/jira/browse/CALCITE-1861> Spatial indexes
>>> https://issues.apache.org/jira/browse/CALCITE-1616 <
>>> https://issues.apache.org/jira/browse/CALCITE-1616> Data profiler
>>> 
>>> All are “best effort”; i.e. they may not make it into the release.
>>> 
>>> Can someone please review 707? It is an important change.
>>> 
>>> Julian
>>> 
>>> 
>>>> On Nov 14, 2017, at 12:37 PM, Michael Mior <mm...@uwaterloo.ca> wrote:
>>>> 
>>>> I looks like there are significant version upgrades available for
>>> libraries
>>>> in use within adapters. Those would probably be good to include as well.
>>>> (The MongoDB driver in particular is three years out of date.)
>>>> 
>>>> --
>>>> Michael Mior
>>>> mm...@apache.org
>>>> 
>>>> 2017-11-14 14:51 GMT-05:00 Julian Hyde <jh...@apache.org>:
>>>> 
>>>>> I have logged https://issues.apache.org/jira/browse/CALCITE-2049 <
>>>>> https://issues.apache.org/jira/browse/CALCITE-2049>, “Release Calcite
>>>>> 1.15.0”. Please follow that to get updates on the state of the release.
>>>>> 
>>>>> Vova, I would like to include https://issues.apache.org/
>>>>> jira/browse/CALCITE-2018 <https://issues.apache.org/
>>>>> jira/browse/CALCITE-2018> too. I will definitely review your pull
>>>>> request. Sorry it’s taken me a while.
>>>>> 
>>>>> Everyone, please look at https://issues.apache.org/
>>>>> jira/browse/CALCITE-2027 <https://issues.apache.org/
>>>>> jira/browse/CALCITE-2027>, specifically the list of other technologies I
>>>>> would like to upgrade (guava, jetty, jackson, panino, h2, hsqldb,
>>> javacc,
>>>>> maven) or change the “preferred” version. If there are other libraries
>>> we
>>>>> should upgrade, let’s discuss. At this stage I would like to be
>>> ambitious,
>>>>> and upgrade as far as we can.
>>>>> 
>>>>> Julian
>>>>> 
>>>>> 
>>>>>> On Nov 14, 2017, at 7:29 AM, Michael Mior <mm...@uwaterloo.ca> wrote:
>>>>>> 
>>>>>> Sounds good to me. I have nothing in particular I'm hoping to include.
>>>>>> Looking forward to ditching JDK 7 :)
>>>>>> 
>>>>>> --
>>>>>> Michael Mior
>>>>>> mm...@apache.org
>>>>>> 
>>>>>> 2017-11-13 22:46 GMT-05:00 Julian Hyde <jh...@apache.org>:
>>>>>> 
>>>>>>> Last release (1.14) was on 2nd October. I think we should aim for a
>>>>>>> release 1.15 in early December (which means starting an RC vote around
>>>>>>> November 27), and I volunteer to be release manager.
>>>>>>> 
>>>>>>> Does this timing work for everyone?
>>>>>>> 
>>>>>>> Any particular features that people would like to include?
>>>>>>> 
>>>>>>> I would like to announce that this will be the last release that will
>>>>>>> support JDK 7[1].
>>>>>>> 
>>>>>>> Julian
>>>>>>> 
>>>>>>> [1] https://issues.apache.org/jira/browse/CALCITE-2027 <
>>>>>>> https://issues.apache.org/jira/browse/CALCITE-2027>
>>>>> 
>>>>> 
>>> 
>>> 
> 

Reply via email to