Bug 324804 - NullPointerException at AspectJElementHierarchy.java:677
Summary: NullPointerException at AspectJElementHierarchy.java:677
Status: RESOLVED FIXED
Alias: None
Product: AspectJ
Classification: Tools
Component: Compiler (show other bugs)
Version: unspecified   Edit
Hardware: PC Windows Vista
: P3 normal (vote)
Target Milestone: 1.6.10   Edit
Assignee: aspectj inbox CLA
QA Contact:
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2010-09-08 18:30 EDT by Jesse Englert CLA
Modified: 2010-09-09 12:24 EDT (History)
1 user (show)

See Also:

Attachments
JVM Error Log (19.70 KB, application/octet-stream)
2010-09-08 18:41 EDT, Jesse Englert CLA 		no flags Details
View All Add an attachment (proposed patch, testcase, etc.)

Note You need to log in before you can comment on or make changes to this bug.
Description Jesse Englert CLA 2010-09-08 18:30:31 EDT 
Build Identifier: 20100617-1415

After removing a "throws" clause from the constructor of a class and then saving I got the following error:

java.lang.NullPointerException
at org.aspectj.asm.internal.AspectJElementHierarchy.getCanonicalFilePath(AspectJElementHierarchy.java:677)
at org.aspectj.asm.internal.AspectJElementHierarchy.updateHandleMap(AspectJElementHierarchy.java:641)
at org.aspectj.asm.AsmManager.removeStructureModelForFiles(AsmManager.java:572)
at org.aspectj.asm.AsmManager.processDelta(AsmManager.java:604)
at org.aspectj.ajdt.internal.core.builder.AjBuildManager ... oBuildJob.run(AutoBuildJob.java:242)
at org.eclipse.core.internal.jobs.Worker.run(Worker.java:54)

Compile error: NullPointerException thrown: null

Reproducible: Didn't try
Comment 1 Jesse Englert CLA 2010-09-08 18:41:43 EDT 
Created attachment 178463 [details]
JVM Error Log

I was able to reproduce this error when adding or removing the throws clause from the constructor...until Eclipse eventually crashed. Attaching the JVM hs_err log. The JVM identified the following: 

# Problematic frame:
# J  org.aspectj.asm.internal.AspectJElementHierarchy.updateHandleMap(Ljava/util/Set;)V
Comment 2 Andrew Clement CLA 2010-09-09 12:24:56 EDT 
Possibly a threading issue with an AJDT thread affecting the handleMap which the compiler is in the middle of processing.  A guard would seem to suffice here as null effectively means the same as the check the code was going to perform.

Not sure why a simple NPE would lead to a vm crash.

I've committed a guard on null to stop this occurring.