Re: Discussion around CALCITE-2489 (risks linked to Class.getFields())

Hi Alessandro, Many thanks for pushing this forward. I had a quick look at the diff and the changes look reasonable. I am not sure I got entirely the problem with the constructors and the order of the parameters but if we are talking about JavaBeans couldn't we enforce the same limitations which

Re: Discussion around CALCITE-2489 (risks linked to Class.getFields())

Alessandro, thanks for pushing this. Frankly speaking, I don't like @CalciteField(order=42) approach. It looks like annotating the constructor parameters makes more sense, and it is more flexible. A similar case is covered in Jackson with "parameter names" module: https://github.com/FasterXML/jack

[HELP] ExceptionInInitializerError caused by ImmutableBeans?

Hello everyone, While testing an application that uses Calcite, I'm getting an error when I try to run it on a Websphere9 environment (IBM J9 VM, Java8, Ubuntu) [1]. It would seem there is an issue around org.apache.calcite.util.ImmutableBeans and the reflection mechanisms that it uses, in the con

Re: [HELP] ExceptionInInitializerError caused by ImmutableBeans?

Can you please log it? Link it to https://issues.apache.org/jira/browse/CALCITE-4200 (if it’s not the same issue) and of course FLINK-19820. > On Feb 9, 2021, at 8:28 AM, Ruben Q L wrote: > > Hello everyone, > > While testing an applicatio

Re: Discussion around CALCITE-2489 (risks linked to Class.getFields())

A very important use case is when you are basing a table on third-party class and you are not able to add annotations. People using these classes should be able to tell the reflective adapter (that’s not its real name, but you know what I mean) how to derive a stable ordering for the fields. Th

Re: Discussion around CALCITE-2489 (risks linked to Class.getFields())

Stamatis, Vladimir, thanks for your comments. The empty constructor Stamatis is suggesting would probably be enough, but what Vladimir is suggesting would also improve on the annotation front (in particular, it's easier to annotate just the constructor rather than each field in the whole hierarchy

Re: [HELP] ExceptionInInitializerError caused by ImmutableBeans?

Definitely worth logging/fixing but given that java9 is already EOL [1] I don't think there are going to be many people impacted by this. [1] https://www.oracle.com/java/technologies/java-se-support-roadmap.html On Tue, Feb 9, 2021 at 10:38 PM Julian Hyde wrote: > Can you please log it? Link it

Re: [DISCUSS] Release Managers

That’s reasonable, we did make many changes in year 2020. Hope the good news for 2021. Julian Hyde 于2021年2月8日 周一上午2:17写道： > Release 2.0 is largely symbolic. We develop Calcite incrementally, so it’s > more important that there are regular releases, and that they are backwards > compatible (within