Question about submitting bugs

[Archie Cobbs]

Thanks Jon.

Here are three suspected issues boiled down to simple test cases. Right now
they are all failing with the 1.8.0_60 javac. I'm not up-to-date enough on
current compiler development to know whether these are already fixed in
jdk9 or not.

Lambda generates IllegalAccessError at runtime:
    https://gist.github.com/archiecobbs/0e9a3e460c9581d13cb9

Compilation failure when overriding clone()

http://stackoverflow.com/questions/33023332/why-doesnt-this-java-program-compile

Compilation fails when assigning static final field using qualified name

http://stackoverflow.com/questions/11705814/java-compile-error-depends-on-whether-static-variable-name-is-qualified

Thanks,
-Archie


On Wed, Feb 17, 2016 at 10:56 AM, Jonathan Gibbons <
jonathan.gibbons at oracle.com> wrote:

> On 02/17/2016 08:40 AM, Archie Cobbs wrote:
>
> Hi,
>
> I've got a meta-question regarding filing compiler bug reports...
>
> Compiler questions and possible bug reports are often posted to this list,
> but this list is obviously not meant to be a general triage area for bug
> reports from the entire Java community, so I'm hesitant to post occasional
> issues that crop up here.
>
> I've also noticed that the JDK bug submission form on
> http://bugreport.java.com/ doesn't seem to really work any more... I've
> submitted a couple of bug reports there lately but never heard anything
> back (example
> <http://stackoverflow.com/questions/33023332/why-doesnt-this-java-program-compile>).
> I have another one now but don't want it to be lost like the others.
>
> It also doesn't seem possible for mere mortals to create accounts on
> https://bugs.openjdk.java.net/
>
> So what is the correct procedure for non-compiler-developers to
> productively report suspected compiler issues?
>
> Thanks,
> -Archie
>
> --
> Archie L. Cobbs
>
>
> Archie,
>
> Thanks for asking.
>
> You are correct that this is not an area to help debug users programs.
> The more you have reasonable cause to suspect that javac may not be
> behaving correctly, (or could reasonably be improved) and the more you can
> produce evidence, such as reproducible test cases, to back up your claim,
> the more interested folk will be in general.  Ideally, test cases should be
> small, and should be "just pure javac" -- meaning, so simple there's no
> need to involve any build tools, unless you believe they are an inherent
> part of the problem.
>
> If you file an issue on bugreport.java.com, and you don't hear back in a
> timely manner, you can try posting a reference here and we can help ensure
> the issue is migrated to bugs.openjdk.java.net.
>
> -- Jon
>



-- 
Archie L. Cobbs
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.openjdk.java.net/pipermail/compiler-dev/attachments/20160217/2138d724/attachment.html>