Re: haskell parser for java bytecode?
Jerzy Karczmarczuk asked:
Why for goodness' sake, people interested in Haskell *should* worry about parsing of Java bytecode chunks?
Are you asking why I'm interested in bytecode instead of source code, or are you asking why I'm interested in Java at all? One answer to the first question is: because java bytecode is a simpler problem domain than java source. For example, name resolution is trivial when working with bytecode (there's no ambiguity between package/class/variable names, inherited names are already resolved, overloaded functions are already resolved, etc). Also, many of the newer features of java disappear at the bytecode level: inner classes, generics, the new foreach statement, etc. Another answer to the first question is: because java bytecode can be used as a target for several languages these days, and by working with it directly, I can work with any language that compiles to the jvm, rather than restricting myself just to java. The answer to the second question is: because the rest of the world matters to some people. Kimberley Burchett http://www.kimbly.com/
On Thu, Jun 30, 2005 at 11:02:47AM -0400, Kimberley Burchett wrote:
Jerzy Karczmarczuk asked:
Kimberley Burchett writes:
I'm interested in using haskell to do static analysis of java bytecode. To my surprise, I wasn't able to find any existing haskell libraries for parsing java .class files.
Why for goodness' sake, people interested in Haskell *should* worry about parsing of Java bytecode chunks?
Are you asking why I'm interested in bytecode instead of source code, or are you asking why I'm interested in Java at all?
Jerzy is asking why you are surprised. There's nothing wrong in being interested in Java/bytecode, but it's a bit strange to expect every language to have a library for reading Java bytecode. Best regards Tomasz
On Thu, 30 Jun 2005, Tomasz Zielonka wrote:
Jerzy is asking why you are surprised. There's nothing wrong in being interested in Java/bytecode, but it's a bit strange to expect every language to have a library for reading Java bytecode.
My apologies. I didn't mean to imply that there *should* be a library for this, or that the haskell community was in any way failing to live up to my expectations. I was just surprised that nobody had gotten around to doing it before me. Kimberley Burchett http://kimbly.com/
Hello Kimberley, Thursday, June 30, 2005, 9:43:31 PM, you wrote:
library for reading Java bytecode.
btw, it may be much simpler to make binding to existing library, especially if it's written in C/C++. there is a plenty of tools for semi-automated generation of such bindings -- Best regards, Bulat mailto:bulatz@HotPOP.com
Kimberley Burchett writes:
Jerzy Karczmarczuk asked:
Why for goodness' sake, people interested in Haskell *should* worry about parsing of Java bytecode chunks?
Are you asking why I'm interested in bytecode instead of source code, or are you asking why I'm interested in Java at all?
One answer to the first question is: because java bytecode is a simpler problem domain...
The answer to the second question is: because the rest of the world matters to some people.
=== Actually, I didn't ask why *YOU* are interested in Java, sorry for the misunderstanding. You do whatever you wish, and I respect it. All is very fine, and far from any personal doubts. I reacted because you said that you *were surprised* that Haskell milieu did not elaborate libraries for parsing of Java code, as it were natural that everybody, who develops the support for pure functional languages should immediately think about parsing of Java code. I found it a bit preposterous. "People interested in Haskell" did not address you, but the developers of Haskell. Also, I underlined the word *should*. Had your question been formulated more modestly, in a less revindicative way, I wouldn't even open my mouth. That's all to it. The project of writing such a parser may be useful and interesting, but it is far from being the ultimate goal of haskellians... Of course, this is just my private opinion. Now, if YOU are really interested in programming in Haskell, and you need such parser, go ahead. Convince people reading this list - I do speak seriously - that there exists a fabulous application niche for Haskell, that they might become famous and rich doing such thing. All the best. Jerzy Karczmarczuk
participants (4)
-
Bulat Ziganshin -
karczma＠info.unicaen.fr -
Kimberley Burchett -
Tomasz Zielonka