Summary: | Internal compiler error | ||
---|---|---|---|
Product: | [Eclipse Project] JDT | Reporter: | Mark Brodhun <brodhun> |
Component: | Core | Assignee: | JDT-Core-Inbox <jdt-core-inbox> |
Status: | RESOLVED DUPLICATE | QA Contact: | |
Severity: | normal | ||
Priority: | P3 | ||
Version: | 2.1 | ||
Target Milestone: | 2.1 RC4 | ||
Hardware: | PC | ||
OS: | Windows 2000 | ||
Whiteboard: |
Description
Mark Brodhun
2003-02-10 17:12:18 EST
An internal compiler error is always reported against position 0 of the compilation unit (likely package statement line). However it has nothing to do with the package statement per se. Could you provide a source sample which reproduces it ? The stack trace doesn't provide line numbers, and the String constructor seems suspicious since the code should not perform any String manipulation (only char []s). I doubt this will help but the below is the complete contents of the last file that had the internal compiler error. The error moved from another file to this file after I shut down and restarted Eclipse (W2k IBM JRE 1.3.1). As I said this error comes in at random times and I don't think it has anything to do with one particular file. I have seen related error traces that say error in the org.eclipse.core.resources package but it isn't currently failing so I can't get the exact msg for you. In all cases it is due to a save operation generating the null pointer exception in the string constructor. package com.ibm.edts.tcrepos; /** * */ public interface ESDTConstants { public static final String ATTR_COM_IBM_EDTS_PRODUCT = "com.ibm.edts.product"; public static final String ATTR_COM_IBM_EDTS_RELEASE = "com.ibm.edts.release"; public static final String REPOS_NAME = "com.ibm.edts.directoryartifacts"; public static final String ATTR_COM_IBM_ESDT_CRTDATE = "com.ibm.edts.crtDate"; } It feels highly unlikely that this code would cause such a grief. I am wondering if you aren't rather seeing a VM bug. Could you try to use a different JRE and see if it occurs again ? There have been numerous unexpected NPEs using certain IBM 1.3.1 builds. Cannot reproduce on JRE 1.4.1_01 Closing, please reopen if you have more evidence than the compiler is guilty. Reopen to workaround |