|
From: | Elliott Hughes |
Subject: | Re: [Cp-tools-discuss] gnu.bytecode |
Date: | Sun, 30 Jan 2005 13:14:18 -0800 |
On Jan 29, 2005, at 20:34, C. Brian Jones wrote:
On Thu, 2004-12-30 at 01:36, C. Brian Jones wrote:On Wed, 2004-12-29 at 22:57, Elliott Hughes wrote:is this the right/a sensible place to talk about gnu.bytecode? i know there's *a* copy of gnu.bytecode in the repository, but i know it's not *the* copy. for some reason, it was easier to find this list than any equivalent kawa one. i've been writing a tool similar to javap/jcf-dump, but which outputssomething in a form suitable for re-assembly into a .class file. asidefrom the lack of support for 1.5 changes, i've found that the LocalVariableTable support doesn't work (trygnu.classpath.tools.JavapMain or gnu.bytecode.dump on a class compiled with "javac -g" to see). there's a really simple fix for JavapMain and my disassembler, but i don't know if it'll break kawa or anything elsethat uses gnu.bytecode. is this where i'd find people who're interested?I wanted to reply again because Tom has mentioned adding an option to javap to output in a format recognizable by Jasmin, see http://jasmin.sourceforge.net/. I've not yet dug into it to determine if this project is up to date enough to support 1.5 features.
it's not. it also doesn't support the generation of invalid class files, which would seem like one of the uses of a Java assembler.
I've removed gnu.bytecode from the cp-tools CVS in favor of the version in Kawa CVS. Here are a few of the things I think need fixing in gnu.bytecode... a) Remove use of Class.forName(), if it exists, and substitute in something that can search a specified path instead. b) Support 1.5 c) Ensure API provides enough information to output all details of the class file without using a Writer which may not output in a format you desire and tends to have special access to internal variables...
i think a better choice is to make the writer methods public and non-final. i fear the other way duplication lies.
Do you want to work on something to output in jasmin format together?
can do. i think that the class i already posted shows there's some mileage in this idea; its output is way easier to read than javap's.
i'm not sure we should worry too much about using exactly jasmin format, though. for one thing, it doesn't support everything we need, and for another, it seems offensively over-complicated for such a simple job. (i have a semi-irrational dislike of JavaCC.)
-- http://www.jessies.org/~enh/
[Prev in Thread] | Current Thread | [Next in Thread] |