Community
Participate
Working Groups
Created attachment 139065 [details] Screendump Build ID: Build id: I20090430-2300 Steps To Reproduce: I was (happily) typing some code when an error appeared...see attachment. More information: Microsoft Windows [Version 6.0.6001] Copyright (c) 2006 Microsoft Corporation. All rights reserved. C:\Users\Administrator>java -version java version "1.6.0_14" Java(TM) SE Runtime Environment (build 1.6.0_14-b08) Java HotSpot(TM) 64-Bit Server VM (build 14.0-b16, mixed mode)
Please attach the .log (or the relevant entry copied from the Error Log view).
I have attached the screenshot of the error, and, said what I was doing. That's all I know.
As you can imagine, the screenshot does not help much if we cannot reproduce the problem and have no additional info on the circumstances. That's why I asked you to attach your log file. The log file is at <workspace location>/.metadata/.log If you don't want to attach the whole log, you can also select Window > Show View > Error Log and then copy the entry with the same title as given in the screendump (including the stack trace, of course).
Created attachment 139194 [details] log file Attached log file.
Created attachment 139196 [details] Another log file
There are two JDT related entries in the .log: >java.lang.ClassCastException: org.eclipse.jdt.core.dom.EnumDeclaration cannot be >cast to org.eclipse.jdt.core.dom.TypeDeclaration This is fixed, see bug 239887. !ENTRY org.eclipse.core.jobs 4 2 2009-06-12 21:40:12.984 !MESSAGE An internal error occurred during: "Requesting Java AST from selection". !STACK 0 java.lang.NullPointerException at org.eclipse.jdt.internal.compiler.lookup.WildcardBinding.erasure(WildcardBinding.java:389) Jonathan, it would be good to have a test case. Moving to JDT Core for investigation.
Could you please provide the compilation unit that caused the problem? Thanks.
I don't remember sorry.
Kent, it looks like typeVariable() can return null, but this is not handled everywhere.
Created attachment 140015 [details] Proposed patch
No reproduceable case to verify - just check code Released in HEAD for 3.6M1
Without a reproducable test case and/or steps to reproduce to assess how serious this problem is, this won't be backported to 3.5.1.
Verfied for 3.6M1 (by code inspection)