Thanks, we'd certainly be interested in bug reports against bnd. Our issue
tracker is on Github: https://github.com/bndtools/bnd/issues

If ASM fails on default methods then I would expect we would too. Bnd does
not use ASM but it does some similar things.

Regards,
Neil



On Tue, Mar 18, 2014 at 9:21 PM, Christopher BROWN <[email protected]> wrote:

> I got Felix running on the developer preview last autumn following advice
> on this list:
>
> http://mail-archives.apache.org/mod_mbox/felix-users/201309.mbox/%3CCAHL_zcOO23xUaPc_gCFKaw=bvbcecg_hhh2pv7ty5yld-xd...@mail.gmail.com%3E
>
> I don't know if this sort of patching is still required with the most
> recent releases, but it wasn't hard to apply.
>
> As for Java 8 features with BND, I've used Java 8 outside of OSGi
> applications and it's proving very productive.  I doubt I could help in any
> way with modifying BND but I'd certainly be interested in beta testing an
> updated version.  Of note, I've been able to examine Java 8 bytecode with
> ASM versions that were designed for Java 7, and it works fine until it hits
> code using default interface methods.
>
> I'm guessing therefore that in the short term, I'll try using Java 8 as the
> build JDK with "source" and "target" options for "javac" set to Java 7...
> at least that way I can benefit from the new APIs such as the excellent
> "java.time" packages.
>
> Let me know if I can help out.
>
> Thanks,
> Christopher
>
>
> On 18 March 2014 21:59, Neil Bartlett <[email protected]> wrote:
>
> > I would expect Felix to "Just Work" on the Java 8 runtime; after all, J8
> is
> > meant to be backwards compatible.
> >
> > Bnd will almost certainly NOT work on classes compiled for Java 8 if they
> > use the new bytecode features. We have not done any testing with this yet
> > but I'm sure there will be issues for us to work on. For example,
> interface
> > default methods will have a large impact on compatibility calculations
> and
> > baselining.
> >
> > Regards,
> > Neil
> >
>

Reply via email to