Re: Interpreted debugging with gcj/gij

[Keith Seitz <keiths at redhat dot com>]

On Thu, 2005-05-26 at 10:51 -0600, Tom Tromey wrote:
> >>>>> "Keith" == Keith Seitz <keiths@redhat.com> writes:
> Keith> I can certainly send a follow-on message, explaining the
> Keith> overall design of the back-end which was designed with an eye
> Keith> on minimal impact on non- debugging performance and
> Keith> VM-neutrality.
> 
> Yeah, this would be very helpful, so that folks can get an overview of
> what is going on -- a context in which to review the patches.

Coming to an inbox near you very, very soon.

> As far as administrative stuff, the URLs Mark posted are a good start.

I believe that I have followed most of the standard GNU (C/C++) coding
conventions within the code. Okay, truthfully, I just let emacs do its
thing. ;-) The only major departure from standard GNU (C/C++)
conventions is my avoidance of the underscore. I prefer the more common
(to me at least) "fooBarBaz" to "foo_bar_baz" when coding in java.

Classpath appears to depart from GNU in one major aspect: putting spaces
between function names, e.g., "myfunction()" instead of the (more
standard) "myfunction ()". Call me old-fashioned. Working on gdb for a
long time has really hammered that space into my brain.

None of it's really a big deal, IMO. I've looked at a lot of the gcj
code, and there appears to be significant departures from either of
these standards all over the place.

I'll presume that by the first few patches I'll understand better what
people find offensive and fix such offenses in subsequent patches.

> If there is code that could go in Classpath, it would be best to start
> talking about that too.  What does it do, what should we name it,
> those kinds of things.

Yes, nearly the whole thing could go into classpath, I think. Right now,
I've got everything arranged under gnu.gcj.jdwp. If there is a better
place to put this for Classpath's sake, I could refactor the code easily
enough. Just tell me where.

Keith