Announcement

Collapse
No announcement yet.

WWJ SDK Alpha 3 - 0.3.0 available

Collapse
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • WWJ SDK Alpha 3 - 0.3.0 available

    The World Wind Java SDK alpha 3 - version 0.3.0 is available.
    Download WWJ SDK 0.3 (Zip 13M) or try the java web start demo 0.3 from NASA Learning Technologies site.

    Here is a list of changes from the release package readme.txt:

    Change Summary for 0.2 to 0.3:
    • - Configuration made consistent and independent
    • - Configuration defaults moved to usage sites
    • - Logging mechanism and semantics cleaned up
    • - Shutdown entry point
    • - Package refactoring
    • - Memory cache refactoring and optimization
    • - Reimplementation of Polylines. Polylines on surface.
    • - Shapes (but the Surface versions will be superseded in next release)
    • - Shape and icon dragging
    • - Math classes re-engineering
    • - View and input handling changes
    • - Examples -- see src/gov/nasa/worldwind/examples
      • - Template for WWJ apps
      • - Template for WWJ panels
      • - WMS Manager
      • - Layer manager
    • - More general icon image specification
    • - WMS capabilities doc parsing
    • - WMS layer specification improved
    • - WMS Layer Manager
    • - Thread-safe icon collections in IconLayer
    • - RPF parsing and display
    • - Applet example and javascript interface
    • - New layers
      • - Country boundaries
      • - USGS Topo
      • - Stars
      • - Fog
      • - Scalebar
      • - World Map
      • - Terrain Profiler
    • - Layer panel utility
    • - Performance statistics utility
    • - Bug fixes
    • - Instrumentation and Run-time statistics
    • - Texture cache
    • - Proxy support
    • - Anaglyph stereo
    • - Many more, but these are the major highlights
    I have also compiled a change list with screenshots here. All previous WWJ SDK updates are listed on the World Wind Central WWJ download page.

    Enjoy
    Last edited by patmurris; 10-11-2007, 07:59 PM.
    My World Wind Java Blog & WW.net Plugins page

  • #2
    The class gov.nasa.worldwind.examples.applet.WWJAp plet has the following import which cannot be resolved:
    import netscape.javascript.JSObject;

    I suspect I'm missing a jar file somewhere.

    Comment


    • #3
      You changed your name, TAG?


      Earth is Square blog

      PUBLIC NOTICE AS REQUIRED BY LAW: Any use of this forum post, in any manner whatsoever, will increase the amount of disorder in the universe. Although no liability is implied herein, the consumer is warned that this process will ultimately lead to the heat death of the universe.

      Comment


      • #4
        This problem has been reported since applets are there... see this thread:
        Originally posted by amnesic View Post
        Since 1.4.2 Java Plug-In ships netscape.javascript in a JAR file called plugin.jar. So I just added this external jar to fixed this problem of compilation. (inside /System/Library/Frameworks/JavaVM.framework/Home/lib/ on my Mac os X)
        Also:
        Originally posted by toolshed View Post
        I had to link to:
        C:\Program Files\Java\jre1.6.0_01\lib\plugin.jar
        to compile w/o errors.
        It obviously fell through the cracks. We will look into it.
        There are a couple other things with the view and the sky gradient too. Expect a bug fix release soon.
        Last edited by patmurris; 10-11-2007, 06:43 PM.
        My World Wind Java Blog & WW.net Plugins page

        Comment


        • #5
          Ok, I found the plugin.jar and that fixed the compile problems. Thanks.

          Comment


          • #6
            Good job.
            After a quick one-minute test of the web start demo I can say that 0.3 seems faster and more responsive than previous releases. However, there is still that issue with tiles loading everywhere but not in the center of the screen...
            There are a couple other things with the view and the sky gradient too.
            Like this? (see the pic)
            Attached Files
            Hot stuff: FAQ, forum search, DirectX, .NET, Video Card Compatibility list

            Comment


            • #7
              Originally posted by m_k View Post
              Like this? (see the pic)
              Yes, the 'big black hole in the sky' bug is always lurking... something like déja vu in the .net version. I thought i tracked it down but the far clipping plane distance seems to change in unpredictable ways at very low altitude... a JIRA issue has been filed.
              My World Wind Java Blog & WW.net Plugins page

              Comment


              • #8
                Great :-) Lib Version information requested, please

                Can you please tell us which are the version of JOGL and GlueGen in this 0.3.0? Thanks

                Comment


                • #9
                  Fix for sky gradient hole

                  Here is a quick fix for the layers.Earth.SkyGradientLayer 'black hole in the sky' at low altitude.
                  In render(), edit this line so that it tests > 10 instead of > 100:
                  Code:
                     if (this.glListId == -1 || Math.abs(this.lastRebuildAltitude - camAlt) > 10)
                  I'd bet it is not the last fix for it
                  My World Wind Java Blog & WW.net Plugins page

                  Comment


                  • #10
                    Weird bug from the JNLP!

                    Hi Pat,

                    Launching the JNLP demo, I got something really weird! I briefly saw *something* and then the GL canvas just flipped. Look at the attached picture and the stack dump from my console... I'm copying a small excerpt here:

                    Code:
                    11-Oct-2007 9:39:02 PM gov.nasa.worldwind.AbstractSceneController pick
                    SEVERE: Exception encountered while picking
                    javax.media.opengl.GLException: array vertex_buffer_object must be disabled to call this method
                    	at com.sun.opengl.impl.GLImpl.checkBufferObject(GLImpl.java:30668)
                    	at com.sun.opengl.impl.GLImpl.checkArrayVBODisabled(GLImpl.java:30716)
                    	at com.sun.opengl.impl.GLImpl.glVertexPointer(GLImpl.java:27937)
                    	at gov.nasa.worldwind.globes.EllipsoidRectangularTessellator$RectTile.render(Unknown Source)
                    	at gov.nasa.worldwind.globes.EllipsoidRectangularTessellator$RectTile.render(Unknown Source)
                    	at gov.nasa.worldwind.globes.SectorGeometryList.pick(Unknown Source)
                    	at gov.nasa.worldwind.AbstractSceneController.pick(Unknown Source)
                    	at gov.nasa.worldwind.BasicSceneController.doRepaint(Unknown Source)
                    	at gov.nasa.worldwind.AbstractSceneController.repaint(Unknown Source)
                    	at gov.nasa.worldwind.WorldWindowGLAutoDrawable.doDisplay(Unknown Source)
                    	at gov.nasa.worldwind.WorldWindowGLAutoDrawable.display(Unknown Source)
                    	at com.sun.opengl.impl.GLDrawableHelper.display(GLDrawableHelper.java:78)
                    	at javax.media.opengl.GLCanvas$DisplayAction.run(GLCanvas.java:324)
                    	at com.sun.opengl.impl.GLDrawableHelper.invokeGL(GLDrawableHelper.java:194)
                    	at javax.media.opengl.GLCanvas.maybeDoSingleThreadedWorkaround(GLCanvas.java:301)
                    	at javax.media.opengl.GLCanvas.display(GLCanvas.java:133)
                    	at javax.media.opengl.GLCanvas.paint(GLCanvas.java:166)
                    	at sun.awt.RepaintArea.paintComponent(Unknown Source)
                    	at sun.awt.RepaintArea.paint(Unknown Source)
                    	at sun.awt.windows.WComponentPeer.handleEvent(Unknown Source)
                    	at java.awt.Component.dispatchEventImpl(Unknown Source)
                    	at java.awt.Component.dispatchEvent(Unknown Source)
                    	at java.awt.EventQueue.dispatchEvent(Unknown Source)
                    	at java.awt.EventDispatchThread.pumpOneEventForFilters(Unknown Source)
                    	at java.awt.EventDispatchThread.pumpEventsForFilter(Unknown Source)
                    	at java.awt.EventDispatchThread.pumpEventsForHierarchy(Unknown Source)
                    	at java.awt.EventDispatchThread.pumpEvents(Unknown Source)
                    	at java.awt.EventDispatchThread.pumpEvents(Unknown Source)
                    	at java.awt.EventDispatchThread.run(Unknown Source)
                    I haven't had the time yet to check if I can reproduce it with the SDK itself.

                    Bye bye!

                    Chiss!
                    Attached Files

                    Comment


                    • #11
                      Originally posted by Chiss View Post
                      Launching the JNLP demo, I got something really weird! I briefly saw *something* and then the GL canvas just flipped.

                      11-Oct-2007 9:39:02 PM gov.nasa.worldwind.AbstractSceneControll er pick
                      SEVERE: Exception encountered while picking
                      javax.media.opengl.GLException: array vertex_buffer_object must be disabled to call this method
                      Quite surprising! It looks like a gl state leak - some GPU state not restored properly after a rendering operation and contaminating the following ones.
                      Did it happen just once or can you reproduce it at will ?
                      The partial upside down flip is a weird artefact though...
                      My World Wind Java Blog & WW.net Plugins page

                      Comment


                      • #12
                        Rats. This is the identical problem we encountered on Wednesday prior to the 0.3.0 release after Sun updated the jogl libs at their jws location. Our Java Web Start files point directly to those, and if they change and the new libs are broken then this happens. I've just contacted the JOGL lead. I hope to hear back soon, and that he's reverted to the previous known-good jogl libs.

                        In the meantime, would you please bring up the Java Web Start Cache Viewer and clear any existing jogl or gluegen -- or worldwind, for that matter -- files from the cache. Then try to run the wwj jws app again. There's a chance of the cache having picked up a broken jogl version between releases.

                        This is the problem with the jws model. We test code on a specific instance of support libs. If those change and introduce bugs, then all the apps depending on them break. This has happened several times now, and I'm ready to throw in the towel and just include known-good instances of support libs in the release itself.

                        Comment


                        • #13
                          I just cleaned up two jogl related objects updated most recently from the java resource cache... and voila, it works again

                          The JNLPAppletLaucher may suffer similar setbacks for the same reasons.
                          Last edited by patmurris; 10-13-2007, 09:11 PM.
                          My World Wind Java Blog & WW.net Plugins page

                          Comment


                          • #14
                            Originally posted by tag View Post
                            Rats. This is the identical problem we encountered on Wednesday prior to the 0.3.0 release after Sun updated the jogl libs at their jws location. Our Java Web Start files point directly to those, and if they change and the new libs are broken then this happens. I've just contacted the JOGL lead. I hope to hear back soon, and that he's reverted to the previous known-good jogl libs.
                            I'm the JOGL project lead. First, I apologize for the breakage. We failed to test an update of JOGL's TextRenderer with World Wind Java. This won't happen again.

                            Judging from the reply above, it sounds like clearing out the Java Web Start cache has solved the problem you were seeing, Chiss; can you confirm?

                            Originally posted by tag View Post
                            This is the problem with the jws model. We test code on a specific instance of support libs. If those change and introduce bugs, then all the apps depending on them break. This has happened several times now, and I'm ready to throw in the towel and just include known-good instances of support libs in the release itself.
                            Again, I apologize for the breakage, but there is a less drastic solution: point the World Wind Java JNLP file at one of our stable JNLP files rather than the most current one. We document where this is (approximately) in the JOGL Users' Guide, but to be clear, here's the JNLP extension link for the release of JOGL prior to the TextRenderer changes:

                            Code:
                            http://download.java.net/media/jogl/builds/archive/jsr-231-1.1.1-rc3/webstart/jogl.jnlp
                            As the premier JOGL based application to date, World Wind Java's use of the latest JOGL builds is definitely appreciated, but I understand the need for stability. The stable JNLP links should address this problem.

                            Comment


                            • #15
                              Thanks for the quick response Ken. One of these days I'd like to understand what causes cached contents to be replaced and what does not.

                              I don't think we've seen Ken here before Everyone should know that if it were not for him we would not have the quality OpenGL interface we have. While I'm sure there were others involved, Ken has tirelessly spearheaded the effort within Sun and produced most of the implementation. WWJ owes a tremendous amount to his and Sun's exceptionally strong support of our effort.

                              Comment

                              Working...
                              X