Let’s discuss in the jira case. Can you add more description to the case about what happens and what you think should happen.
> On Jul 18, 2018, at 2:50 PM, [email protected] wrote: > > Hello, > > I've just found a bug. I've placed a test case for it here - > https://issues.apache.org/jira/browse/CALCITE-2420 > > It seems that MIN/MAX aggregate functions does not support Timestamp > properly and RexImpTable.MinMaxImplementor.implementNotNullAdd creates an > implementation in which argument is casted into primitive where it should > be converted. > > Im willing to patch this, but I need some hint on which level this should > be fixed. Why Date/Time/Timestamp are converted here to int/long - ther are > comparable and can be let as is... > > On Wed, Jul 18, 2018 at 11:44 PM Piotr Bojko (JIRA) <[email protected]> wrote: > >> Piotr Bojko created CALCITE-2420: >> ------------------------------------ >> >> Summary: Aggregate functions MIN/MAX does not work for >> Timestamp >> Key: CALCITE-2420 >> URL: https://issues.apache.org/jira/browse/CALCITE-2420 >> Project: Calcite >> Issue Type: Bug >> Components: core >> Affects Versions: 1.17.0 >> Reporter: Piotr Bojko >> Assignee: Julian Hyde >> >> >> >> >> >> >> -- >> This message was sent by Atlassian JIRA >> (v7.6.3#76005) >> > > > -- > Piotr Bojko > http://about.me/ptr.bojko
