First release of JSR292 backport

[Rémi Forax]

I've just released the first (in fact, the second) version of the JSR
292 backport.
See my blog:
http://weblogs.java.net/blog/forax/

Dear language runtime developers,
I hope you will have fun with it.

cheers,
Rémi

[Chanwit Kaewkasi]

Thanks, Rémi !
It will be useful for me.

Cheers,

Chanwit

--
Chanwit Kaewkasi
PhD Candidate,
Centre for Novel Computing
School of Computer Science
The University of Manchester
Oxford Road
Manchester
M13 9PL, UK

[Frank Wierzbicki]

Thanks Remi! I know I will have some fun with it :)

-Frank

[Charles Oliver Nutter]

Spectacular effort, Remi...I hope to give it a try soon. I'm also
hoping I'll have opportunity to help improve it given what we have
learned about making dispatch fast in JRuby.

[Rémi Forax]

Charles Oliver Nutter a écrit :

> Spectacular effort, Remi...I hope to give it a try soon. I'm also
> hoping I'll have opportunity to help improve it given what we have
> learned about making dispatch fast in JRuby.
>

All contributions are welcome :)
I haven't your experience.

Rémi

[Rémi Forax]

Frank Wierzbicki a écrit :

I can't wait to see an invoke-jython-dynamic prototype.
I feel like I had put a stone in the middle of the great pyramid
and can't wait the others to finish the dream.

> -Frank
>

Rémi

[Rémi Forax]

Chanwit Kaewkasi a écrit :

> Thanks, Rémi !
> It will be useful for me.
>

Good luck for G7 !
> Cheers,
>
> Chanwit
>

Rémi

[Charles Oliver Nutter]

Remi: I have not had any look at your stuff yet, but would it be an
appropriate replacement for the "mock" indy jar I've been using? i.e.
even if we don't run against the backport, would it enable compiling
JRuby's invokedynamic stuff on non-MLVM JVMs?

http://github.com/jruby/indy_mock/tree/master

- Charlie

[Rémi Forax]

Charles Oliver Nutter a écrit :

> Remi: I have not had any look at your stuff yet, but would it be an
> appropriate replacement for the "mock" indy jar I've been using? i.e.
> even if we don't run against the backport, would it enable compiling
> JRuby's invokedynamic stuff on non-MLVM JVMs?
>
> http://github.com/jruby/indy_mock/tree/master
>
> - Charlie
>

Yes, it should be a drop in replacement.

Rémi

[Attila Szegedi]

Hi Rémi,

your backport only compiles with and runs on Java 6, as it relies on
several 6-only methods, notably:

java.util.Arrays.copyOf()
java.util.Arrays.copyOfRange()
java.lang.instrument.Instrumentation.redefineClasses(ClassDefinition)
(Java 5 only has Instrumentation.redefineClasses(ClassDefinition[]))

Just saying, as I thought your intent was to support Java 5 too.

Attila.

[Rémi Forax]

Attila Szegedi a écrit :

> Hi Rémi,
>
> your backport only compiles with and runs on Java 6, as it relies on
> several 6-only methods, notably:
>
> java.util.Arrays.copyOf()
> java.util.Arrays.copyOfRange()
> java.lang.instrument.Instrumentation.redefineClasses(ClassDefinition)
> (Java 5 only has Instrumentation.redefineClasses(ClassDefinition[]))
>
> Just saying, as I thought your intent was to support Java 5 too.
>
> Attila.
>

Oups.
will fix ASAP.

Rémi

[Rémi Forax]

Attila Szegedi a écrit :

> Hi Rémi,
>
> your backport only compiles with and runs on Java 6, as it relies on
> several 6-only methods, notably:
>
> java.util.Arrays.copyOf()
> java.util.Arrays.copyOfRange()
> java.lang.instrument.Instrumentation.redefineClasses(ClassDefinition)
> (Java 5 only has Instrumentation.redefineClasses(ClassDefinition[]))
>
> Just saying, as I thought your intent was to support Java 5 too.
>
> Attila.
>

It should be fixed in the trunk and I've put a new zip in download
section (1.0.2)

Rémi