Unfortunately, so far I have not been able to reproduce the issue with a
unit test, neither on Github CI [1] nor in a local environment.
In any case, analyzing the code we can see that the risk is in theory
present; and my original message shows that it can manifest in practice
under certain
Thanks for the feedback.
FYI, I have been able to workaround the issue in the application by
tweaking its code to force some Calcite classes to be loaded at different
points.
Nevertheless, it's true it would be nice to break the circular dependency
on Calcite's side; I'll work on the proposed
DEFAULT_CALL could also be converted to a method. There’s no harm if such
SqlCall objects are created each time we need them.
> On Nov 6, 2020, at 11:05 AM, Vladimir Sitnikov
> wrote:
>
> Just in case, I've added a GitHub job that runs Calcite tests with OpenJ9
> JVM 1.8 in
>
Just in case, I've added a GitHub job that runs Calcite tests with OpenJ9
JVM 1.8 in
https://github.com/apache/calcite/commit/d226a9456a8cfdb108a50744d702e915b46c7ffa
Julian>The problem occurs at InferTypes line 41, which is a lambda
I believe lambda/reference is irrelevant since
The problem occurs at InferTypes line 41, which is a lambda:
public static final SqlOperandTypeInference FIRST_KNOWN =
(callBinding, returnType, operandTypes) -> {
…
};
And the error is to do with method handles, which are related to lambdas. Can
you instead convert it
Thanks for the answers.
Indeed, WebSphere uses IBM's own JVM, and my hypothesis is that its class
loader follows a different (and unlucky) order to load Calcite classes.
I will continue digging into it, see if I can isolate the problem, and open
a Jira if there is in fact something to be addressed
As you’re using Websphere, it’s possible you’re using IBM’s Java runtime. Is
that so?
Julian
> On Nov 5, 2020, at 2:23 PM, Stamatis Zampetakis wrote:
>
> Hi Ruben,
>
> Since you mentioned WebSphere I am wondering if it is related to some weird
> problem in the ClassLoader.
> If it was only
Hi Ruben,
Since you mentioned WebSphere I am wondering if it is related to some weird
problem in the ClassLoader.
If it was only a JVM problem I guess it could be reproduced outside of
WebSphere just by referencing a few Calcite classes in some specific order.
Other than that, it is better to
Hello everyone,
I'm getting a strange error on WebSphere 9 (on Ubuntu) when trying to start
an application that uses Calcite (shaded) library. The exception message
says:
** error [-]: Unexpected error.
** ...
** Caused by: java.lang.BootstrapMethodError: