Hi Martin, I developed the current JDEE interface to jdb with the JDK 1.4.1 version of jdb. It's possible that the problems you are encountering are backward compatibility problemw with JDK 1.3. I plan ultimately to enhance the jdb interface to work with older versions of the JDK.
I'm sorry that the recent "improvements" to JDEbug and the jdb interface are proving disruptive for you. I will try to resolve these problems as quickly as possible. BTW, I was gratified to hear that you had had good luck with JDEbug. I have had so many problem reports concerning it (generally on platforms that I don't have) that I had begun to question whether investing any more time it would be worthwhile. Your comments have restored my faith that JDEbug, given the necessary attention, can develop into a serviceable debugger for Emacs users. Paul Martin Dickau writes: > I've used the JDEbug support for several years, but decided to give the jdb > support a try, and I need a little help. > > First of all, JDE 2.3.3beta5, Windows 2000, JDK 1.3.0. > > I work in an environment where everything begins life compiled > with -g:lines. I normally debug by recompiling the particular class(es) I > need with -g and run my program in the debugger, setting breakpoints in the > file(s) I recompiled and examining variables, etc. once the breakpoints are > hit. This works fine with JDEbug in the current beta5. > > If I change my debugger from JDEbug to jdb (leaving my environment otherwise > as it was), however, things do not work for me. If I set a breakpoint on > the same line in the same file where I could stop with JDEbug, jdb > eventually reports no code at that line. If I try to do a stop in > myClass.myMethod( ... ), or com.baa.mypackage.myClass.myMethod( ... ), jdb > eventually tells me it cannot find the method. > > I do have the "use classic VM" option set, but it's not clear to me whether > jdb is paying any attention to that or not. > > Recompiling everything with plain -g isn't a practical option, so I more or > less need to have the debugger function with my current working style > (hence, I haven't tried recompiling everything from scratch). I could use > some suggestions for what might be going on or what to look at to figure out > what might be going on. > > Thanks, > > Martin Dickau, ByAllAccounts >
