Re: [DotGNU]Fw: javaByteCode -> IL Compiler

[Gopal V]

Hi Gaurav,

> Right now, everything's sounding totally wierd to me. As far as the problem
> of local variables go, Rhys has explained me in detail. I am worried about 
> when you say "java compiler-plugin".
        I tried writing the stack type prediction algorithm from a bit of
the Kaffe verifier. But ended up getting dumped. Also I'm alone in
this stuff and the exams came up at that time etc.....
> 
>  Don't know. Is he Abhay? He also approached me regarding the "heaviness" of
> the topic if that's taken up as a part of course term-project, but then being
> alone he left it. Now that I also joined the course (auditing it), along with 
> a third guy we are thinking of taking up this challenge.
        If you need another developer, I am quite ready to work on this stuff.
I'll help you out with code/JVM reference. I have some useful experience
with the JVM. If you are not using Pnet's library for reading class files,
I have my mini-project (java assembler/disassembler) which I can hack into
sort of an API for reading JVM bytecode. But I think Rhys will prefer that
you use getMethodCode for getting method info. (I dunno how you're going
to resolve constant pool entries from this !, ask Rhys...)

        I think we could use one more "Meet Rhys" irc chat. Can you use
IRC (firewall ?). The first chat resulted in what you now see as the
pnetlib class-status page.

Gopal.V
-- 
 The difference between insanity and genius is only measured by success
 //===<=>===\\
|| GNU RULEZ ||
 \\===<=>===//