Re: Tackling Query Federation Use Case using Calcite

2024-12-11 Thread Aditya Mahajan
Thanks for the help!! Just subscribed to the list. :) I will check out the attached sources in the thread given by Charles and Julian. Will follow-up on the thread for any clarifications... Regards, Aditya M. On Thu, Dec 12, 2024 at 2:36 AM Mihai Budiu wrote: > You should subscribe to the mail

Re: Meaning of date subtraction

2024-12-11 Thread Julian Hyde
Looks fine to me: > SELECT >(DATE '2020-03-01' - DATE '2020-01-01') YEAR AS y, >(DATE '2020-03-01' - DATE '2020-01-01') MONTH AS m, >(DATE '2020-03-01' - DATE '2020-01-01') YEAR TO MONTH AS ym; +++---+ | Y | M | YM | +++---+ | +0 | +2 | +0-02 | +++-

Re: Tackling Query Federation Use Case using Calcite

2024-12-11 Thread Mihai Budiu
You should subscribe to the mailing list if you want to get the replies. https://lists.apache.org/thread/c1pmqp240qr7cy4knp5rpmkmomj4lkgz Mihai From: Aditya Mahajan Sent: Wednesday, December 11, 2024 3:29 AM To: dev@calcite.apache.org Subject: Re: Tackling Query

Re: Tackling Query Federation Use Case using Calcite

2024-12-11 Thread Aditya Mahajan
Hello Calcite Team, May I please get any kind of update on this? Does this use-case fit Apache Calcite's functionality? Any kind of help would be greatly appreciated. Thanks, Aditya M. On Wed, Dec 4, 2024, 12:34 PM Aditya Mahajan wrote: > Hello Calcite Team, > > Hope this email finds you wel

[jira] [Created] (CALCITE-6729) {{TypedValue.toLocal()}} mishandles OBJECT column types.

2024-12-11 Thread Chris Dennis (Jira)
Chris Dennis created CALCITE-6729: - Summary: {{TypedValue.toLocal()}} mishandles OBJECT column types. Key: CALCITE-6729 URL: https://issues.apache.org/jira/browse/CALCITE-6729 Project: Calcite

Re: Deriving Calcite queries' execution statistics

2024-12-11 Thread Mihai Budiu
If you can put everything you need in moveNext it's great, since that will involve no code generation for instrumentation, which makes debugging much easier. You may not even need to change the RelNodes at all - after all, all implementations that you want to instrument are already there, in En

Re: Deriving Calcite queries' execution statistics

2024-12-11 Thread Austin Richardson
Indeed, our ultimate goal is to achieve something similar to what EXPLAIN ANALYZE provides for Postgres. Regarding achieving this using the Enumerable convention: one idea we’ve started exploring at a high level is whether adding instrumentation to the moveNext() method would work. For example, gi

[jira] [Created] (CALCITE-6728) Lazy loading of database tables and schemas

2024-12-11 Thread Ulrich Kramer (Jira)
Ulrich Kramer created CALCITE-6728: -- Summary: Lazy loading of database tables and schemas Key: CALCITE-6728 URL: https://issues.apache.org/jira/browse/CALCITE-6728 Project: Calcite Issue Typ

Re: literal "type" issues in my database

2024-12-11 Thread stanilovsky evgeny
Just use a cast here. If I do a "create table as select 2" in my database, it creates a TINYINT column. I hope that answers your question. So I need the RelDataType to be of type TINYINT I do have a mechanism to deal with this in my current code a little bit similar to what you're suggest