Mailing-List: contact dev-help@maven.apache.org; run by ezmlm
Precedence: bulk
Reply-To: "Maven Developers List" <dev@maven.apache.org>
Received-SPF: pass (athena.apache.org: domain of jason@maven.org designates
 63.246.20.110 as permitted sender)
Message-Id: <57A38C71-02C2-4D5F-819D-EAE988AC4BE6@maven.org>
From: Jason van Zyl <jason@maven.org>
To: "Maven Developers List" <dev@maven.apache.org>
In-Reply-To: <f84708e90806221213k3146ff33kb61b70b50750cb5a@mail.gmail.com>
Content-Type: text/plain; charset=WINDOWS-1252; format=flowed; delsp=yes
Content-Transfer-Encoding: quoted-printable
Mime-Version: 1.0 (Apple Message framework v926)
Subject: Re: Testing Maven 2.x Embedding
Date: Sat, 5 Jul 2008 12:00:59 -0400
References: <619DA980-0DEE-4743-96D0-BF8FE2F5100B@maven.org>
 <f84708e90806221137p24c02adw92e127b1075eca8a@mail.gmail.com>
 <B9597CAB-BA42-476B-9C68-438CE46B9604@maven.org>
 <f84708e90806221213k3146ff33kb61b70b50750cb5a@mail.gmail.com>

Milos, can you help me copy the Mevenide job from deadlock?

I can see from the output here what the CLI looks like but there are a =20=

couple options I don't really understand:

http://deadlock.netbeans.org/hudson/job/mevenide/423/console

On 22-Jun-08, at 3:13 PM, Milos Kleint wrote:

> On Sun, Jun 22, 2008 at 8:44 PM, Jason van Zyl <jason@maven.org> =20
> wrote:
>>
>> On 22-Jun-08, at 11:37 AM, Milos Kleint wrote:
>>
>>> On Sun, Jun 22, 2008 at 8:18 PM, Jason van Zyl <jason@maven.org> =20
>>> wrote:
>>>>
>>>> Milos,
>>>>
>>>> Do you need anything other then the standard embedder that's =20
>>>> produced by
>>>> the
>>>> build to run your automated tests?
>>>
>>> I don't really have any automated tests. The wide variety of =20
>>> problems
>>> I've encountered so far sort of convince me that I would just spend
>>> time on them without much effect. I rely on the maven's own
>>> integration tests instead. That's what people working on maven care
>>> about anyway primarily.
>>>
>>> I'm using the -Pnetbeans profile for building the embedder bits, =20
>>> as I
>>> think shading is still necessary in embedded env.
>>>
>>>
>>>>
>>>> I'm trying to hook the standard output to m2e, and I wanted to =20
>>>> try and do
>>>> the same with the Netbeans integration so that we can get more =20
>>>> feedback
>>>> on
>>>> changes in the trunk as they happen over the course of the =20
>>>> releases that
>>>> will start shortly.
>>>
>>> what does "hook the standard output to m2e" mean?
>>>
>>
>> We have quite a few automated tests, so to feed the embedder that =20
>> produced
>> and wire it into m2e's automated tests as a smoke test.
>>
>>>
>>>>
>>>> I know you have your builds in the Sun instance of Hudson but can =20=

>>>> you
>>>> help
>>>> me setup a Netbeans build in Hudson so we can run whatever =20
>>>> automated
>>>> tests
>>>> you have so that we can catch problems with embedding faster?
>>>
>>> it's basically a "mvn install nbm:nbm nbm:cluster" but mvn install
>>> shall be enough as well. only a few unit tests there (as I don't
>>> believe in UI testing :) - or rather I don't believe that the value
>>> added to time spent ratio is good in the long run)
>>>
>>> My testing usually involves opening and working with multiple =20
>>> complex
>>> projects like mevenide, glassfish or hudson to see if things keep
>>> working..
>>>
>>> If writing tests is reasonably easy at maven, I could write some =20
>>> here
>>> eventually, but generally I'm turned down by any complex test setup.
>>> That's where the tests belong anyway IMHO.
>>>
>>
>> We really just have tests for the actions in m2e. Testing the UI is =20=

>> a test
>> plan and human, no way around that but where I can determine =20
>> something done
>> in the embedder causes easily detectable problems I'd like to know =20=

>> about
>> them sooner rather then later.
>>
>> Given m2e and netbeans are the primary consumers at this point it =20
>> would be
>> good to at least keep you guys in Hudson and make sure everything is
>> working. At least at a high level.
>>
>
> Fine with me. However please note that without the manual UI testing
> you probably only figure netbeans integration doesn't compile (which
> actually can be useful as I do override implementations of a few core
> components.)
>
> =46rom top of my head the 3.1.1 release of mevenide had a problem with
> RELEASE version resolution, I've fixed it later, updated to latest
> version just to find out that now the 3.1.3 version has a problem with
> long dependency versions (fixed now in trunk but still I need to
> perform a re-release to push it to users)
> These should have been cought by the integration or unit tests in
> maven itself. I usually only figure when people report the problem to
> me or encounter the issue myself.
>
> Milos
>
>>> greetings.
>>>
>>> Milos
>>>
>>>>
>>>> Thanks,
>>>>
>>>> Jason
>>>>
>>>> ----------------------------------------------------------
>>>> Jason van Zyl
>>>> Founder,  Apache Maven
>>>> jason at sonatype dot com
>>>> ----------------------------------------------------------
>>>>
>>>> A party which is not afraid of letting culture,
>>>> business, and welfare go to ruin completely can
>>>> be omnipotent for a while.
>>>>
>>>> -- Jakob Burckhardt
>>>>
>>>>
>>>> =
---------------------------------------------------------------------
>>>> To unsubscribe, e-mail: dev-unsubscribe@maven.apache.org
>>>> For additional commands, e-mail: dev-help@maven.apache.org
>>>>
>>>>
>>>
>>> =
---------------------------------------------------------------------
>>> To unsubscribe, e-mail: dev-unsubscribe@maven.apache.org
>>> For additional commands, e-mail: dev-help@maven.apache.org
>>>
>>
>> Thanks,
>>
>> Jason
>>
>> ----------------------------------------------------------
>> Jason van Zyl
>> Founder,  Apache Maven
>> jason at sonatype dot com
>> ----------------------------------------------------------
>>
>> In short, man creates for himself a new religion of a rational
>> and technical order to justify his work and to be justified in it.
>>
>> -- Jacques Ellul, The Technological Society
>>
>>
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail: dev-unsubscribe@maven.apache.org
>> For additional commands, e-mail: dev-help@maven.apache.org
>>
>>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@maven.apache.org
> For additional commands, e-mail: dev-help@maven.apache.org
>

Thanks,

Jason

----------------------------------------------------------
Jason van Zyl
Founder,  Apache Maven
jason at sonatype dot com
----------------------------------------------------------

To do two things at once is to do neither.

  -=97Publilius Syrus, Roman slave, first century B.C.


---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@maven.apache.org
For additional commands, e-mail: dev-help@maven.apache.org