Hello,
I have identified a problem with the pax-web-jsp project and I would like to
contribute a fix. What is the best approach ? Opening a JIRA ticket, asking for
commitership ?
Here are some details about the issue. I have been testing the pax-web-jsp
under the IBM Websphere 7 platform, and this server comes with it's own JVM
that has a different behavior than Oracle's one. Basically I had a problem with
the JasperClassLoader class that was raising an exception inside the
constructor. Here is the stack trace:
java.lang.NullPointerException
at
org.ops4j.pax.web.jsp.internal.JasperClassLoader.hashCode(JasperClassLoader.java:197)
at
com.ibm.oti.shared.SharedAbstractHelperFactory.objectToString(SharedAbstractHelperFactory.java:92)
at
com.ibm.oti.shared.SharedAbstractHelperFactory.findSharedHelperForClassLoader(SharedAbstractHelperFactory.java:126)
at
com.ibm.oti.shared.SharedClassHelperFactoryImpl.findHelperForClassLoader(SharedClassHelperFactoryImpl.java:51)
at
com.ibm.oti.shared.SharedClassHelperFactoryImpl.getURLClasspathHelper(SharedClassHelperFactoryImpl.java:145)
at
java.net.URLClassLoader.initializeSharedClassesSupport(URLClassLoader.java:180)
at java.net.URLClassLoader.<init>(URLClassLoader.java:249)
at
org.ops4j.pax.web.jsp.internal.JasperClassLoader.<init>(JasperClassLoader.java:70)
at
org.ops4j.pax.web.jsp.JspServletWrapper.<init>(JspServletWrapper.java:62)
at
org.jahia.bundles.extender.jahiamodules.JspBundleObserver.addingEntries(JspBundleObserver.java:30)
at
org.ops4j.pax.swissbox.extender.BundleWatcher$3.run(BundleWatcher.java:226)
at
java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:452)
at java.util.concurrent.FutureTask$Sync.innerRun(FutureTask.java:314)
at java.util.concurrent.FutureTask.run(FutureTask.java:149)
at
java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.access$301(ScheduledThreadPoolExecutor.java:109)
at
java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.run(ScheduledThreadPoolExecutor.java:217)
at
java.util.concurrent.ThreadPoolExecutor$Worker.runTask(ThreadPoolExecutor.java:897)
at
java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:919)
at java.lang.Thread.run(Thread.java:738)
Looking at the details it seems that IBM's JVM is calling the hashCode method
before the object has finished it's constructor, and therefore the
m_bundleClassLoader field is still null. I fixed this issue with the following
patch :
===================================================================
---
pax-web-jsp/src/main/java/org/ops4j/pax/web/jsp/internal/JasperClassLoader.java
(revision 2b46fb08893ed5cc905ae1e5d920a5afdf82d29b)
+++
pax-web-jsp/src/main/java/org/ops4j/pax/web/jsp/internal/JasperClassLoader.java
(revision )
@@ -189,6 +189,11 @@
*/
LOG.trace("Using m_bundleClassloader.hashCode()");
+
+ if (m_bundleClassLoader == null) {
+ return super.hashCode();
+ }
+
final Bundle bundle = m_bundleClassLoader.getBundle();
// This operation guarantees that the JasperClassLoader will fall in
the same bucket as the BundleClassLoader it is wrapping
Let me know what is the best way to contribute this, I'd be more than happy to
contribute this :)
Best regards,
Serge Huber.
- -- --- -----=[ shuber at jahia dot com ]=---- --- -- -
CTO & Co-founder - Jahia Solutions Group, 9 Routes de Jeunes, 1227 Acacias,
Switzerland
Jahia’s next-generation, open source CMS stems from a widely acknowledged
vision
of enterprise application convergence – web, search, document, social and
portal –
unified by the simplicity of web content management.
http://www.jahia.com
_______________________________________________
general mailing list
[email protected]
http://lists.ops4j.org/mailman/listinfo/general
