Re: OWB-2.0.0 to be released soon

[Mark Struberg <st...@yahoo.de.INVALID>]

Hi John!

I've now found a bit time to investigate OWB-2 behaviour in DeltaSpike.
I'm now down to just a few tests and they all seems to be caused by different behaviour of arquillian.

There is a test which checks that a resource must only be served a single time from the ClassPath.
In this test we package the resource into the ShrinkWrap war. And since the OWB-Arquillian adapter does it actually correct, we see this resource twice. The first time from the @Deployment of the test, the other time from the test classpath.
If i move the original test resource to e.g. testonly/myresource.properties then the test passes with OWB. But then it fails with older Weld and OWB arquillian versions as they do not properly serve resources from deployment archives :(

To sum it up: Yes, it's something we need to work on, but it's not an OWB core bug but simply an edge case and portability issue with the various Arquillian containers.

Anyway, thanks for catching it and bringing it up!

txs and LieGrue,
strub

> Am 19.06.2017 um 04:47 schrieb John D. Ament <jo...@apache.org>:
> 
> Hmm so DeltaSpike is seeing 11 test failures with OWB2.
> 
> https://builds.apache.org/view/A-D/view/DeltaSpike/job/DeltaSpike%20OWB%202.0.0/lastCompletedBuild/testReport/
> 
> CDI Ctrl Servlet - it looks like a weird error, the impl of this bean
> hasn't been changed in 2 years.  Not sure if this is a new validation in
> place.
> 
> ClasspathResourceTest - I'm not sure if the path is getting skewed due to
> duplicate resources, or the same resource being added multiple times.
> 
> JSF & Scheduler look like changes in the dependency structure.  If on
> purpose, I'll push up a fix sometime tomorrow.
> 
> On Sun, Jun 18, 2017 at 10:04 AM Romain Manni-Bucau <rm...@gmail.com>
> wrote:
> 
>> 2 things to check before the release:
>> 
>> 1. se API (not covered by tck and poorly covered by us ATM)
>> 2. the fastMatching flag should get removed if we can (created due to 1
>> tck)
>> 
>> 
>> Romain Manni-Bucau
>> @rmannibucau <https://twitter.com/rmannibucau> |  Blog
>> <https://blog-rmannibucau.rhcloud.com> | Old Blog
>> <http://rmannibucau.wordpress.com> | Github
>> <https://github.com/rmannibucau> | LinkedIn
>> <https://www.linkedin.com/in/rmannibucau> | JavaEE Factory
>> <https://javaeefactory-rmannibucau.rhcloud.com>
>> 
>> 2017-06-18 15:41 GMT+02:00 Mark Struberg <st...@yahoo.de>:
>> 
>>> OWB itself did always ignore it and log a warning. This code did not get
>>> changed for a while.
>>> 
>>> But we have/had a bug in the arquillian connector which lead to blowing
>>> up on NoClassDefFound.
>>> I need to check whether we fixed this already with another commit.
>>> 
>>> The ticket is https://issues.apache.org/jira/browse/OWB-1179
>>> 
>>> LieGrue,
>>> strub
>>> 
>>> 
>>> 
>>>> Am 18.06.2017 um 13:45 schrieb John D. Ament <jo...@apache.org>:
>>>> 
>>>> Just wondering, did you change the behavior when a class isn't found to
>>>> ignore the bean? Prior versions of OWB would throw an exception.
>>>> 
>>>> John
>>>> 
>>>> On Sat, Jun 17, 2017 at 4:43 PM Mark Struberg <struberg@yahoo.de.invalid
>>>> 
>>>> wrote:
>>>> 
>>>>> Yes, it should 'just work' afaict.
>>>>> We even kept the SPI the same.
>>>>> We most likely will add an async event related API for better
>>> integration
>>>>> within TomEE in the future.
>>>>> But this should still be perfectly backward compatible as OWB will
>>> provide
>>>>> a default implementation anyway!
>>>>> 
>>>>> If you catch any (unexpected) problems then just ping us.
>>>>> 
>>>>> txs and LieGrue,
>>>>> strub
>>>>> 
>>>>> 
>>>>>> Am 17.06.2017 um 22:36 schrieb John D. Ament <jo...@apache.org>:
>>>>>> 
>>>>>> So... just to confirm.  Outside of changing the geronimo specs,
>>> taking a
>>>>> OWB 1.x profile that I may have will just work with OWB 2?
>>>>>> 
>>>>>> John
>>>>>> 
>>>>>> On Sat, Jun 17, 2017 at 1:53 PM Mark Struberg
>>> <st...@yahoo.de.invalid>
>>>>> wrote:
>>>>>> Yes, it is 1:1 backward compatible.
>>>>>> 
>>>>>> The only thing you need to update is the jcdi and common-annotations
>>> API:
>>>>>> 
>>>>>> <dependency>
>>>>>>   <groupId>org.apache.geronimo.specs</groupId>
>>>>>>   <artifactId>geronimo-annotation_1.3_spec</artifactId>
>>>>>>   <version>1.0-SNAPSHOT</version>
>>>>>> </dependency>
>>>>>> 
>>>>>> <dependency>
>>>>>>   <groupId>org.apache.geronimo.specs</groupId>
>>>>>>   <artifactId>geronimo-jcdi_2.0_spec</artifactId>
>>>>>>   <version>1.0-SNAPSHOT</version>
>>>>>> </dependency>
>>>>>> 
>>>>>> They will be released this week as well.
>>>>>> 
>>>>>> LieGrue,
>>>>>> strub
>>>>>> 
>>>>>> 
>>>>>>> Am 17.06.2017 um 19:41 schrieb Ludovic Pénet <l....@senat.fr>:
>>>>>>> 
>>>>>>> Is it a drop i' remplacement ?
>>>>>>> If yes, I would gladky test this works, before test driving the new
>>>>> features.
>>>>>>> 
>>>>>>> Ludovic
>>>>>>> 
>>>>>>> Le 17 juin 2017 19:38:01 GMT+02:00, Mark Struberg <struberg@yahoo.de
>>>> 
>>>>> a écrit :
>>>>>>> Hi folks!
>>>>>>> 
>>>>>>> We are finished with implementing all CDI-2.0 features and now
>>>>> successfully pass the standalone TCK!
>>>>>>> A recent owb-2.0.0-SNAPSHOT is deployed to the Apache Snapshots
>>>>> repository [1].
>>>>>>> This get's deployed via Jenkins each night.
>>>>>>> 
>>>>>>> It would be great if you could try it out and give us some feedback!
>>>>>>> We gonna release it somewhen next week.
>>>>>>> 
>>>>>>> txs and LieGrue,
>>>>>>> strub
>>>>>>> 
>>>>>>> 
>>>>>>> [1] https://repository.apache.org/content/groups/snapshots/
>>>>>>> 
>>>>>>> --
>>>>>>> Envoyé de mon appareil Android avec K-9 Mail. Veuillez excuser ma
>>>>> brièveté.
>>>>>> 
>>>>> 
>>>>> 
>>> 
>>> 
>>