Uploaded image for project: 'JDK'
  1. JDK
  2. JDK-4035778

sun.applet.AppletPanel bug: applets can disappear if new browser window is creat

    XMLWordPrintable

    Details

    • Type: Bug
    • Status: Closed
    • Priority: P1
    • Resolution: Fixed
    • Affects Version/s: 1.1.1
    • Fix Version/s: 1.1.1
    • Component/s: core-libs
    • Labels:
    • Subcomponent:
    • Resolved In Build:
      1.1.1
    • CPU:
      generic
    • OS:
      solaris_2.4
    • Verification:
      Verified

      Description

      (bill.foote@Eng 1997-03-02: Reclassified under java.classes_util, even though
       this is a sun.applet bug. What _is_ the right category of a sun.applet
       bug?)

        Sometimes applets on a page will disappear if the browser
      window is cloned and the user 'backs' to the page with the applet.

      Here are the steps to reproduce this using the attached applets
      and test html. The two applets (testApplet1 and testApplet2)
      simply paint some text. The two html files have an applet and
      a link to the other html page. So first save the attachments
      in some directory.

      1. Open test1.html.
      The page should have "Test Applet 1" applet text.

      2. Click on "Click here".
      This will take you to the test2.html page.
      The page should have "Test Applet 2" applet text.

      3. Shift-Click on the Back button to create a new browser window
      with test1.html page.

      4. Close the new browser window.
      Note the "Interrupted: bailing" message in the original
      browser windows message line.

      5. Click on the "Go back" link to go to first page.
      The page should have "Test Applet 1" applet text but doesn't!
      Clicking on Reload brings the applet back.

      This is reproducible with hotjava1.0j and JDK1.1.1A as
      well as a current build of hotjava (again on JDK1.1.1).

      I tracked down the "Interrupted: bailing" message to the
      sun.applet.AppletPanel.run() method. If it cannot get
      the next event, it returns. This method handles the various
      APPLET_* events (e.g. APPLET_LOAD, APPLET_START, APPLET_STOP).
      I'm guessing that one of these events was coming in (perhaps
      APPLET_STOP?) and was not fully handled. This may have left
      the applet in an indeterminate state so that when it is activated
      again it fails to paint properly.

      This test case may seem somewhat odd but it was created by
      distilling the Genie stuff down. They found that their applets
      were not always functioning if the users went back and forth
      in various ways. For what its worth, this test case work
      fine in Netscape.

        Attachments

          Activity

            People

            Assignee:
            rfield Robert Field
            Reporter:
            duke J. Duke (Inactive)
            Votes:
            0 Vote for this issue
            Watchers:
            0 Start watching this issue

              Dates

              Created:
              Updated:
              Resolved:
              Imported:
              Indexed: