You'll probably want to put your main class in a separate jar and then use
dependency:unpack-dependencies to unpack the main class
2009/6/2 nambi sankaran
> Hi All
>
> I am trying to embed jetty into a war file, so that the war file becomes
> executable.
>java -jar mywar.war
> will start jet
Hi All
I am trying to embed jetty into a war file, so that the war file becomes
executable.
java -jar mywar.war
will start jetty on port 8080. ie, Main-Class will start Jetty programmatically.
To achieve this I have added configuration section for war plugin,
testwar
goovy Startable interface allows you to determine when an object is full
constructed
http://groovy.codehaus.org/gapi/groovy/swing/impl/Startable.html
does this help?
Martin
__
Verzicht und Vertraulichkeitanmerkung/Note de déni et de confidentialité
I don't think Maven really over-writes to the central repository. I
understand it only writes in the local repository. If you want the
central repository updated, you have to do that specifically.
In fact, as I understand it, the files in the central repository are
supposed to be immutable onc
Yes, this could be a solution. But I was wondering if there will be syncing
problems when Maven is trying to overwrite read-only VCSed dependencies.
Marilyn Sander -X (marilysa - Digital-X, Inc. at Cisco) wrote:
>
> It ought to be simple. You want to have a Maven central repository, and
> hav
It ought to be simple. You want to have a Maven central repository, and
have it under version
control, presumably because your IT department will take responsibility
for backups of the VCS system, while it will not be responsible for just
any old file system.
So, make the central repository refer
The reason why I mentioned VCS is because it is the only company supported IT
service we have. To ensure absolute uptime for your build system, this is
the only way to go. Of course we could setup ftp/http to store dependencies.
But that gives rise to the same problem for maintenance.
nhajrat
If your org is set on not using a repo manager, then I'm not sure that
a VCS is the way to go anyway.
Since maven doesn't overwrite it's artifacts with a new revision, but
rather keeps distinct copies of each deployed artifact (excepting
SNAPSHOTS) -- I think a simple filesystem layout corr
We have done that already. The internal repo is up and running like a charm.
The reason why I asked for VCS support is that nobody wants to maintain this
repo (although it probably does not require any maintenance). The IT service
of my org does not either. We need a backup plan.
nhajratw wrote
I'd strongly suggest revisiting the internal proxy. In the time it
takes you to come up with a solution to store your artifacts in a VCS,
you could have set up Nexus and left early for the day :-)
It's really simple, and once up, requires very little maintenance
effort.
http://nexus.sonat
I need some suggestions for Enterprise build process using Maven.
We are currently developing a Java project using Maven. We version controll
our project source. Our Continuous integration system syncs the source and
uses maven assembly to build a jar with dependencies. This work flow works
fine
Hi,
Do we have lifecycle defined for javascript?
Like:
concatenate
static-check (run jslint)
test
minify
install
deploy
Regards,
Nitin Verma
-
To unsubscribe, e-mail: users-unsubscr...@maven.apache.org
For additional commands
since B already depend on the zip file, so you dont have to make A to
depend on the Zip, but use maven-dependency-plugin to unpack for both
A and B.
-D
On Mon, Jun 1, 2009 at 12:46 PM, dkowis wrote:
> On Mon, 1 Jun 2009 12:43:35 -0700, Dan Tran wrote:
>> you must do the same for project A.
>
>
Hey,
I have different streams for the same projects.
Each task effort has its own stream (all streams are child streams of the
integration stream).
I want to be able to run builds for each stream on the integration machine.
The problem is that the artifact names are the same , so the only possib
in maven everything is a plugin so i
was trying to convert your C binary to plugin so it will install as a
configured plugin
(btw: still looking for a solution to convert a C binary to a maven plugin)
thanks,
Martin
__
Note de déni et de confidential
On Mon, 1 Jun 2009 12:43:35 -0700, Dan Tran wrote:
> you must do the same for project A.
So project A must explicitly depend on the dlls for runtime?
Looks like a good case for a superpom.
It'd be nice if there was a way to be able to add runtime dependencies that
aren't explicitly jar files an
you must do the same for project A.
-
On Mon, Jun 1, 2009 at 12:40 PM, dkowis wrote:
> On Mon, 1 Jun 2009 12:08:48 -0700, Dan Tran wrote:
>> use maven-depdendency-plugin to unpack your zip file and place some
>> where your test can understand and load
>>
>> -D
>
>
> Yep, already did that in Pro
On Mon, 1 Jun 2009 12:08:48 -0700, Dan Tran wrote:
> use maven-depdendency-plugin to unpack your zip file and place some
> where your test can understand and load
>
> -D
Yep, already did that in Project B and found about that on the mailing list
archives. What I'm asking about is how do I get i
I did do a quick test some time ago and my findings agree with Wayne's
assertion here. ie: Maven treats .SNAPSHOT like a regular release version and
not a snapshot version. Deploying a .SNAPSHOT version goes to the release repo.
It must say -SNAPSHOT to deploy to a SNAPSHOT repo.
---
Todd Thies
use maven-depdendency-plugin to unpack your zip file and place some
where your test can understand and load
-D
On Mon, Jun 1, 2009 at 11:56 AM, dkowis wrote:
> On Mon, 1 Jun 2009 13:58:25 -0400, Martin Gainty
> wrote:
>> you would need to build in some manner of intelligently defining the
>> (g
> Need your expertise to tell me if anyone sees an issue in using .SNAPSHOT
> instead of -SNAPSHOT in an artifact version. (The reason is because the
> artifact name has to have a minimum amount of '-').
I believe Maven (at this point) is hard-coded to use -SNAPSHOT, so I
don't think you will enj
On Mon, 1 Jun 2009 13:58:25 -0400, Martin Gainty
wrote:
> you would need to build in some manner of intelligently defining the
> (groupId,artifactId,version) for your DLL e.g.
>
> i believe the analog for groupId could be namespace //here is an example
>
> #include
>
>
> using namespace
> st
This question was asked/answered on Apr30.
The person who answered said "Use -SNAPSHOT"
Some plugins are more picky than others, so use the standard and spare
yourself some pain later.
The artifact, when it leaves your premises should be a released version,
so it should not have the SNAPSHOT suf
you would need to build in some manner of intelligently defining the
(groupId,artifactId,version) for your DLL e.g.
i believe the analog for groupId could be namespace //here is an example
#include
using namespace
std;
namespace
SampleOne
{
float p = 10.34;
}http://www.tenouk
I've searched the mailing lists and they don't quite get far enough on this
subject for me.
I've got a project, incidentally it's the rxtx.org package, that uses the
rxtxSerial.dll. I then have another project that uses that project:
Project A
|
| - Project b
|
|- zipfile containing r
I am testing out the site plugin on a multi module project and the
module project's site is not getting a link back to the parent site.
I am using maven version...
Apache Maven 2.1.0 (r755702; 2009-03-18 14:10:27-0500)
Java version: 1.5.0_16
OS name: "mac os x" version: "10.5.7" arch: "i386" Fa
Hello All,
Need your expertise to tell me if anyone sees an issue in using .SNAPSHOT
instead of -SNAPSHOT in an artifact version. (The reason is because the
artifact name has to have a minimum amount of '-').
Example:
1.0.0.SNAPSHOT
I'd appreciate feedback
Sonia
--
View this message in cont
> Currently when I ctrl+c during a "mvn test" in windows, it displays
>
> Destroying 1 processes
> Destroying process..
> Destroyed 1 processes
>
> however the process seems to still be alive.
Guess I'll just report this as a bug for now.
Thanks!
-=r
--
Thank you very much.
For now I switched to "excalibur-datasource:1.2.0". If I notice some
weird behavior, I will follow your advice.
This is no big deal. However, the dependencies should be in the same
repository and it is also not the only package with missing
dependencies. I think this is somet
On Fri, May 22, 2009 at 4:41 PM, fachhoch wrote:
>
> my mvn version
>
> Apache Maven 2.1.0 (r755702; 2009-03-18 15:10:27-0400)
> Java version: 1.6.0_11
> Java home: C:\Program Files\Java\jdk1.6.0_11\jre
> Default locale: en_US, platform encoding: Cp1252
> OS name: "windows xp" version: "5.1" arch
It's available in an m1 repo here
https://svn.apache.org/repos/asf/excalibur/repository/d-haven-managed-pool/jars/
Jorg
On Fri, May 29, 2009 at 4:56 PM, Walter B. Rasmann
wrote:
> I hope this is the right place to report this issue.
>
>
> Problem description:
>
> Excalibur-datasource depends on
if you only run your tests with code coverage enabled, you cannot
trust a passing result.
I have seen quite a number of cases where tests fail if ran standalone
and pass with coverage. usually due to misunderstanding the java
memory model and how the jvm spec allows optimisations
run the
Can anyone tell me how to prevent unittest run three times when using mvn
site command??
Thanks all.
struberg wrote:
>
>
> Old story, and as far as I remember (please correct me if I'm wrong) the
> reason is:
>
> The site lifecycle and the default build lifecycle are isolated things.
> So if
33 matches
Mail list logo
