1.1 incontainer test failures
by Marko Lukša
Caused by commit
https://github.com/weld/core/commit/add07e82562c8da314556a96678aa5840dc037ec
How come CI didn't catch this?
----------------------------------
Failed tests:
testBeginAlreadyLongRunningConversationThrowsException(org.jboss.jsr299.tck.tests.context.conversation.ClientConversationContextTest):
Found class com.gargoylesoftware.htmlunit.WebResponse, but interface was
expected
testConversationBeginMakesConversationLongRunning(org.jboss.jsr299.tck.tests.context.conversation.ClientConversationContextTest):
Found class com.gargoylesoftware.htmlunit.WebResponse, but interface was
expected
...
Tests run: 821, Failures: 22, Errors: 0, Skipped: 0
[INFO]
------------------------------------------------------------------------
[INFO] Reactor Summary:
[INFO]
[INFO] Weld: Build Configuration ......................... SUCCESS [1.699s]
[INFO] Weld Parent ....................................... SUCCESS [1.052s]
[INFO] Weld Implementation ............................... SUCCESS [3.503s]
[INFO] Weld Integration Tests - JBoss Test Harness (JBoss AS) SUCCESS
[6.885s]
[INFO] Weld Porting Package for the CDI TCK .............. SUCCESS [1.523s]
[INFO] Weld Integration Tests (JBoss AS) ................. SUCCESS
[1:13.289s]
[INFO] AtInject TCK runner for Weld ...................... SUCCESS [2.675s]
[INFO] CDI TCK runner for Weld (JBoss AS) ................ FAILURE
[5:18.466s]
[INFO] Weld SE (Core) .................................... SKIPPED
12 years
Fwd: [jboss-as7-dev] FYI, a new version of Javassist will soon be available which will have a fix for JASSIST-163 (impacts Weld, jbossws-native-core, scannotation, slf4j, Hibernate)...
by Scott Marlow
The Javassist project will soon release version 3.1.17 (including the
fix for JASSIST-163). Will Weld want to upgrade to 3.1.17 also?
Hibernate will probably upgrade and AS also (if everything works).
I have a local copy of weld-core but was not sure if that is enough to
test with the javassist trunk. Is there more that we would want to do
other than just bumping the javassist version in the weld-core pom.xml
and "mvn clean install test"?
Scott
-------- Original Message --------
Subject: [jboss-as7-dev] FYI, a new version of Javassist will soon be
available which will have a fix for JASSIST-163 (impacts Weld,
jbossws-native-core, scannotation, slf4j, Hibernate)...
Date: Fri, 26 Oct 2012 10:51:33 -0400
From: Scott Marlow <smarlow(a)redhat.com>
To: jboss-as7-dev(a)lists.jboss.org <jboss-as7-dev(a)lists.jboss.org>
After 3.1.17 of Javassist is released and testing looks good, we can
look at AS7-5127 which is about upgrading javassist in as7 to improve
performance (as long as the new javassist doesn't break anything).
Obviously, if Javassist 3.1.17 breaks something, we will not be able to
upgrade.
I took a look at which AS7 modules (statically) depend on javassist and
the list is:
1. Weld
2. jbossws-native-core
3. scannotation
4. slf4j
5. Hibernate
I'm testing Hibernate with the new javassist but don't know how to test
(1), (2), (3), (4) above. Is there anything else that we need to try
besides the AS7 testsuite to gain confidence that we can upgrade to
javassist 3.1.17 (after its released soon)?
I built trunk of javassist locally and caught one minor failure with the
Hibernate test suite (see yesterdays comments in JASSIST-163). I'm
built AS7 with the updated hibernate/javassist jars and passed the AS7
testsuite.
If anyone wants to do additional testing with the javassist trunk, let
me know and I'll ask for more time to do testing before 3.1.17 is
released. Or others could also coordinate through JASSIST-163.
Scott
_______________________________________________
jboss-as7-dev mailing list
jboss-as7-dev(a)lists.jboss.org
https://lists.jboss.org/mailman/listinfo/jboss-as7-dev
12 years
Re: [weld-dev] [cdi-dev] Problem with EJB´s, Stereotype with an @Alternative and Specialization
by Pete Muir
This looks like purely an impl issue, moving to weld-dev.
On 9 Oct 2012, at 12:56, Daniel Sachse wrote:
> Hey guys,
>
> I ran into a problem today and I am unsure if the behavior I encountered is the way it should be.
>
> My Environment:
> JBoss 7.1.1.Final with Weld 1.1.5
>
> My Scenario:
>
> ------------------------------------------------------------------------------------------------------
> public abstract class AbstractClass implements DemoInterface {…}
>
> @Singleton
> @ConcurrencyManagement(ConcurrencyManagementType.BEAN)
> public class Real extends AbstractClass {…}
>
> @Singleton
> @ConcurrencyManagement(ConcurrencyManagementType.BEAN)
> @Alternative
> @Specializes
> public class Mock extends Real {…}
>
> ------------------------------------------------------------------------------------------------------
>
> If I register class Mock as an @Alternative in beans.xml, everything works fine.
>
> <alternatives>
> <class>at.w0mb.example.Mock</class>
> </alternatives>
>
> As soon as I try:
>
> ------------------------------------------------------------------------------------------------------
>
> @Alternative
> @Stereotype
> @Retention(RUNTIME)
> @Target(TYPE)
> public @interface Mock {}
>
>
> @Singleton
> @ConcurrencyManagement(ConcurrencyManagementType.BEAN)
> @Mock
> //@Alternative
> @Specializes
> public class Mock extends Real {…}
>
> ------------------------------------------------------------------------------------------------------
>
> If I now register the @Stereotype Mock as an @Alternative in beans.xml, the application won´t deploy.
> <alternatives>
> <stereotype>at.w0mb.example.stereotype.Mock</stereotype>
> </alternatives>
>
> The thrown error message is:
> WELD-001408 Unsatisfied dependencies for type [DemoInterface] …….
>
>
> Do you guys think that this is normal? Is it a specification Issue? Or Weld?
>
> Regards,
>
> Daniel
> _______________________________________________
> cdi-dev mailing list
> cdi-dev(a)lists.jboss.org
> https://lists.jboss.org/mailman/listinfo/cdi-dev
12 years, 1 month
