[jhm]

Software installed

– MyEclipse 5.1 FullStack
– locally installed Java 1.4.2_04-b05
– locally installed Java 1.4.2_05-b04
– locally installed Java 1.6.0-rc-b102
– MS IE 6 SP2
– Firefox 1.5.0.8
– MS Windows XP Pro SP2

Preparation the project for recreation the bug

– create a new Java project
– set the JRE to 1.4.2_05-b04
– import Filesystem: {1.4.2_04-b05}/demo/applets/ArcTest

When running the Applet by “Run As > Applet” it works. Starting the HTML page in IE works. In Firefox also.

But when opening the HTML page in MyEclipse HTMLEditor, the Mozilla Preview shows an error message “Could not load class sun/plugin/JavaRunTime … Bridge was installed under <Java6> but class is missing”.
The debug output is

java.lang.NoClassDefFoundError: sun/plugin/viewer/WNetscapePluginContext
        at org.eclipse.swt.internal.win32.OS.DispatchMessageW(Native Method)
        at org.eclipse.swt.internal.win32.OS.DispatchMessage(OS.java:1925)
        at org.eclipse.swt.widgets.Display.readAndDispatch(Display.java:2966)
        at org.eclipse.ui.internal.Workbench.runEventLoop(Workbench.java:1914)
        at org.eclipse.ui.internal.Workbench.runUI(Workbench.java:1878)
        at org.eclipse.ui.internal.Workbench.createAndRunWorkbench(Workbench.java:419)
        at org.eclipse.ui.PlatformUI.createAndRunWorkbench(PlatformUI.java:149)
        at org.eclipse.ui.internal.ide.IDEApplication.run(IDEApplication.java:95)
        at org.eclipse.core.internal.runtime.PlatformActivator$1.run(PlatformActivator.java:78)
        at org.eclipse.core.runtime.internal.adaptor.EclipseAppLauncher.runApplication(EclipseAppLauncher.java:92)
        at org.eclipse.core.runtime.internal.adaptor.EclipseAppLauncher.start(EclipseAppLauncher.java:68)
        at org.eclipse.core.runtime.adaptor.EclipseStarter.run(EclipseStarter.java:400)
        at org.eclipse.core.runtime.adaptor.EclipseStarter.run(EclipseStarter.java:177)
        at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
        at sun.reflect.NativeMethodAccessorImpl.invoke(Unknown Source)
        at sun.reflect.DelegatingMethodAccessorImpl.invoke(Unknown Source)
        at java.lang.reflect.Method.invoke(Unknown Source)
        at org.eclipse.core.launcher.Main.invokeFramework(Main.java:336)
        at org.eclipse.core.launcher.Main.basicRun(Main.java:280)
        at org.eclipse.core.launcher.Main.run(Main.java:977)
        at org.eclipse.core.launcher.Main.main(Main.java:952)

After that the Mozilla Preview shows a not loaded Applet.

The IE Preview fails also – but much more strictly: it crashes Eclipse.

[Riyad Kalla]

Oh that’s not good. Can you provide this project to me so I can hand it off to the development team so they can see and debug exactly what is going on?

[jhm]

I’ll have a look…

Sending per mail as you suggested in http://www.myeclipseide.com/PNphpBB2+file-viewtopic-t-15206.html ?

[email protected] ATTN Riyad, with a link to this thread so I know why I’m getting the file. I’ll have a look at it ASAP.

[jhm]

Sorry for delay – I prepared a workspace and send it to the support email account.

Let me know if there are files missing …

[Riyad Kalla]

Excellent report, I got all the files and was able to reproduce the problem right away. I filed it as a bug for a developer form the design team to look into ASAP.

[jhm]

Thanks. A Bug report from a developer for other developers 😉

[tom_guyette]

I’m seeing this issue too … is there a bug report somewhere with a workaround I could look at?

The issue seems to be related to running a JVM from within a browser that’s running with a JVM. Sort of like trying to create a ship in a bottle inside a ship in a bottle.

I’m not sure if it’s possible, but I think what we need is a way to tell the browser plugin’s applet loader to use the JVM that Eclipse is currently using, rather than bootstrapping its own from scratch.

[Riyad Kalla]

The bug tracker following this issue is private at the moment (because we track other non-MyEclipse things with it) but I am adding your comments to the report to see if some headway can get made for our next release (lower priority behind some other important tasks that management is requiring get done).

[tom_guyette]

This bug is keeping me from using MyEclipse on the entire application my company develops, so if your developers even have suspicions of a diagnosis / things to experiment with, I’m glad to devote some time to experimenting. That’s why I wanted to get a look at the bug’s developer notes. If you could share just those here (if there are any), that would help.

The most important result isn’t necessarily a fix, but a workaround if one is possible.

Is it known whether the MyEclipse Web 2.0 browser nominally works with web pages that have applets embedded? If it does, then obviously there’s something special about my case. If it doesn’t, there’s some sort of architectural issue. Knowing which of these is the case chops the issue into smaller pieces.

[Riyad Kalla]

This bug is keeping me from using MyEclipse on the entire application my company develops, so if your developers even have suspicions of a diagnosis / things to experiment with, I’m glad to devote some time to experimenting.

I was under the impression that this bug only effected people using the IE Preview… is it effecting you in some other day, like in design mode?

The most important result isn’t necessarily a fix, but a workaround if one is possible.

I agree, and moving forward Java 6 is going to become more common since it’s final now, so we need to address this.

Is it known whether the MyEclipse Web 2.0 browser nominally works with web pages that have applets embedded? If it does, then obviously there’s something special about my case. If it doesn’t, there’s some sort of architectural issue. Knowing which of these is the case chops the issue into smaller pieces.

I don’t know the guts of the implementation enough to answer this. I did post your comments for the dev working on this to see, he should follow up if he gets a chance.

[tom_guyette]

I’m sorry, I think I inadvertantly withheld useful information. Thanks twice for the quick response.

I’m getting essentially the same “fatal error” message, “Could not load class sun/plugin/JavaRunTime … Bridge was installed … but class is missing” but under different circumstances. I haven’t touched Java 6, it’s all Java 5, and it’s happening whenever I turn on just-in-time Javascript debugging on an html page that contains an applet.

My understanding is that the html page is being opened in the internal browser, which appears to be based on Mozilla. The internal browser doesn’t seem to have any way to tell it to use the currently running Eclipse JVM for the Java plug-in, nor does it seem to have an explicit preference for what Java plugin path to use. It seems to simply be pointing to the wrong JVM, possibly to a JVM that doesn’t even exist. I’m wondering if I need to re-install the Sun VM package so it will detect the Eclipse internal browser and install the plugin in the correct location, but I understand little enough about how the internal browser works that I’m not sure that would be useful. I cached a screenshot at my work account and can post it on Monday if that would be useful. I’m also happy to run through troubleshooting / investigative steps your dev team may prescribe. If you tell me what info you need and how to gather it, I’ll gladly provide it.

Applets work fine in my regular installation of Firefox, and the debugger works fine on any code that doesn’t contain applets.

It may also be useful to note that after the fatal error occurs, the entire Eclipse IDE becomes dog slow, takes up a consistent 50% of CPU cycles, and quickly grows an unusually large (400MB+) memory footprint.

[Riyad Kalla]

Tom,
Thank you for the detail, I just passed that off for an evaluation to our designer team. They are on a 2-day crunch timeline for Monday code freeze though, so I may not hear anything back. Please ping me Wednesday if I haven’t replied with an update.

[tom_guyette]

Just refreshing this thread — I still can’t debug JavaScript when there’s Java embedded.

[Riyad Kalla]

Tom,
Our developers identified the issue, but it hasn’t been targetted by management yet for a specific release.

[Author]

Posts