-----Original Message-----
From: Sieger, Nicholas <[EMAIL PROTECTED]>
To: '[EMAIL PROTECTED]' <[EMAIL PROTECTED]>
Date: Tuesday, December 05, 2000 10:48 AM
Subject: Debugging JDEbug (was RE: Problems with compilation of the debugger
source ...)
>I've experienced this problem too, and may have some time to contribute to
>fixing it. JDEbug used to work for me, but then it stopped working
>somewhere around the time that I upgraded to 2.2.5, or upgraded NT to SP6 -
>not sure which, unfortunately. I wouldn't be surprised if it was the
latter.
>I should probably spend some time experimenting with a minimal .emacs and
>possibly downgrading to a previous version to see if the problem goes away.
>
>Anyhow, I was wondering if any of the JDE developers has any notes or tips
>for getting an environment up in which one can test and debug JDEbug.
1. Set the variable jde-bug-debug on.
This causes the JDE to set the JDEbug classpath to the java/classes
directory
instead of to the java/lib/jde.jar file.
2. Use the JDEbug message method to instrument the JDEbug code. The message
method displays the specified message in the *JDEbug* buffer.
3. Recompile and run JDEbug.
If you can get JDEbug to launch, you can use JDEbug to debug itself. The way
to do
this is to use two instances of Emacs/JDEbug. Start JDEbug in attach mode in
one instance.
Use the other instance to attach JDEbug to the first instance.
- Paul
>
>/Nick
>
>-----Original Message-----
>From: Caryn [mailto:[EMAIL PROTECTED]]
>Sent: Monday, December 04, 2000 8:22 PM
>To: Richard den Adel; [EMAIL PROTECTED]
>Subject: Re: Problems with compilation of the debugger source
>Application.java in both 2.2.5 and 2.2.6 beta7
>
>
>
>-----Original Message-----
>From: Richard den Adel <[EMAIL PROTECTED]>
>To: [EMAIL PROTECTED] <[EMAIL PROTECTED]>
>Date: Monday, December 04, 2000 1:50 AM
>Subject: Problems with compilation of the debugger source Application.java
>in both 2.2.5 and 2.2.6 beta7
>
>
>>Hi all,
>>
>>Since 2.2.5 I've had problems using the debugger. Sometimes the debugger
>>won't start correctly and times out while waiting for a connection to
>>the application process.
>>
>>So I thought that there might be a problem with compatibility. We're
>>still using jdk1.2.2, the jde sources are compiled (I concluded that
>>from the makefile) with jdk1.3.
>>
>>Compilation of Application.java in jde/debugger gives the following result
>:
>>
>>
>>
>>/home/richard/jde/java/src/jde/debugger/Application.java:259: Can't
>>reference appID before the superclass constructor has been called.
>> super("Input Processor for App #"+appID);
>> ^
>>/home/richard/jde/java/src/jde/debugger/Application.java:427: Can't
>>reference appID before the superclass constructor has been called.
>> super("Standard Error Processor for App #" + appID);
>> ^
>>/home/richard/jde/java/src/jde/debugger/Application.java:720: Blank
>>final variable 'appID' may not have been initialized. It must be
>>assigned a value in an initializer, or in every constructor.
>> final Integer appID;
>> ^
>>/home/richard/jde/java/src/jde/debugger/Application.java:723: Blank
>>final variable 'jdebug' may not have been initialized. It must be
>>assigned a value in an initializer, or in every constructor.
>> final Jdebug jdebug;
>> ^
>>4 errors
>>
>>
>>
>>I've tried the sources from jde2.2.5 and jde2.2.6beta7. I have tried the
>>jdk1.2.2 compiler and the jdk1.3 compiler from sun on the solaris
platform.
>>Anyone else had this problem?
>>
>>Am I doing something wrong here?
>
>
>A lot of people have had this problem, mainly on Windows. Unfortunately I
>have never been able to reproduce it on any of my Windows or Solaris
>systems. It seems to occur when Emacs tries to connect to a socket created
>by JDEbug to connect the standard I/O of the application to Emacs. The
>socket connection never gets made. Some people have found that removing all
>versions of the JDK and reinstalling a single version clears up the
problem.
>I suspect the problem may be a thread deadlock somewhere. The JDEbug Java
>code is heavily threaded. However, I didn't write the code and I haven't
had
>time to try to analyze it to see where deadlocks might occur. I put in some
>messages designed to instrument each stage in the standard I/O socket
>creation and connection process. It is my hope that somebody, hopefully a
>multithreading expert, who has this problem will take a cue from this and
>try to pinpoint exactly where the hangup occurs.
>
>- Paul
>
