Yes, it is one of the things that have been
replaced by a Java
>> Could you investigate the OO.org
Java requirements, and then take a
>> look at the state of JAmiga (speak to the devs if necessary), if its
>> not far enough advanced, nor likely to be soon, please take a look at
>> Kaffe/cacao/Jikes/etc and make an evaluation as to the best solution
>> for our needs and report back, we can then decide what to do.
> OK, will do.
Don't want to stop anyone, but maybe... as
there "is" already JAmiga...
why not try to get gcj to work?
I think that was the point of my request no? basically he needs to go and
find out exactly what OO.org's Java requirements are, if JAmiga can meet
those requirements, job done, otherwise find the best solution that will
take the least amount of time to port/bring up to scratch.
yes, it was just a surgestion, that the Java Compiler could be usefull
instead of an JVM (should also be faster, if I am not totally wrong).
As I understand it we currently have some (68k) based
Java compilers and
from Thomas' message we could have a PPC one, and maybe a library fairly
easily (gcj based), however I dont know if its that simple, do we need the
GUI classes, do we need the JVM etc?
IIRC libjava uses the same classes library that JAmiga, Kaffe and almost
all other opensource JVMs use... GNU Classpath.
I guess if someting is needed in the Classpath port that JAmiga is using,
would they not mind to get some help (but I don't know).
if things get compiled using gcj, would it not need a JVM.
and yes, gcj is not the big problem by itself (as the work is done by
So Tony gets to investigate first, report back, then
we decide the best
next step Java wise.
fine with me... it was just an idea, to get the java compiler working
Worst case scenario is we need to port a full Java
hnl_dk - Henning Nielsen Lund
µA1-C (IBM PowerPC 750 FX V2.2 @ 800MHz), 256MB RAM, NEC ND-3500A DVD/CD
RW, Samsung SP1604N 160GB PATA Harddisk drive, Acer AL1721 17" TFT,
modified Tyan M2042 1->3 PCI Riser...