It was what I wanted to do. I put a comment in an issue but it's not yet
implemented.
My idea it's just to move the cache from the target dir of each module in
the root directory of the localrepo. We have also to improve caching support
for multiprojects build (it seems that cache isn't shared betw
I can't find anything searching nabble about whether it was considered
a bad thing to write the cache files into ~/.m2/repository.
The problem I am finding is that you want to run "mvn eclipse:eclipse"
when you have a new baseline and all your version numbers have
changed. And often in this scenar
IMO we need to give people a place to work that's ahead of the current
work, but we can definitely keep it on a feature branch. Doesn't matter
to me.
Brett Porter wrote:
On 19/09/2008, at 1:47 AM, [EMAIL PROTECTED] wrote:
Author: jdcasey
Date: Thu Sep 18 08:47:01 2008
New Revision: 696678
Sure, I can re-update the site.
Brett Porter wrote:
Shouldn't this still be listed separately? 2.0.9 is the current release
and 2.1.0-M1 is the current preview release?
- Brett
On 19/09/2008, at 10:14 AM, [EMAIL PROTECTED] wrote:
Author: jdcasey
Date: Thu Sep 18 17:14:07 2008
New Revision:
The Maven team is pleased to announce the release of the Maven 2.1.0-M1.
This release signals the beginning of a new direction in the development
of Maven. Recently, Maven's development efforts were split into three
major goals. First, the Maven 2.0.x code line will be minimally
maintained for
Issue Subscription
Filter: Design & Best Practices (28 issues)
Subscriber: mavendevlist
Key Summary
MNG-2184Possible problem with @aggregator and forked lifecycles
http://jira.codehaus.org/browse/MNG-2184
MNG-612 implement conflict resolution techniques
htt
Mercury wagon provider can now upload to [?any] webdav server, tested on
Apache WebDAV and Nexus. Don't miss the PGP signature generation feature
for uploads - configuration details in
http://jira.codehaus.org/browse/MERCURY-8. You can use it in any Maven
2.0.9+, maybe even earlier.
Now it's
I apologize if you get this twice. My ISP has been doing some work on their
email and it isn't quite working properly.
I'm confused how the build can continue without the parent pom. I wouldn't
call the behavior "inconsistent", I would call it completely indeterminate.
The parent plugin could have
+1
On 15/09/2008, at 10:07 PM, Dennis Lundberg wrote:
Hi
Now that maven-parent 9 is out, it's time to follow up with the other
parents as well.
Diff for the pom.xml:
http://svn.eu.apache.org/viewvc/maven/shared/trunk/pom.xml?r1=637254&r2=695193&diff_format=h
Diff for the site.xml:
http://svn
+1
On 15/09/2008, at 10:29 PM, Dennis Lundberg wrote:
Hi
Here's another parent in need of a release.
Diff for the pom.xml:
http://svn.eu.apache.org/viewvc/maven/plugins/trunk/pom.xml?r1=646036&r2=695450&diff_format=h
Diff for the site.xml:
http://svn.eu.apache.org/viewvc/maven/plugins/trunk/
Shouldn't this still be listed separately? 2.0.9 is the current
release and 2.1.0-M1 is the current preview release?
- Brett
On 19/09/2008, at 10:14 AM, [EMAIL PROTECTED] wrote:
Author: jdcasey
Date: Thu Sep 18 17:14:07 2008
New Revision: 696865
URL: http://svn.apache.org/viewvc?rev=696865&
For the record...
A late +1 from me. I was using it for all my builds for the last few
days and RC17 before that.
BTW, Deng is also on the PMC :)
Cheers,
Brett
On 19/09/2008, at 9:46 AM, John Casey wrote:
Vote Result:
Binding:
John Casey
Lukas Theussl
Brian Fox
Vincent Siveton
Olivier L
On 19/09/2008, at 4:16 AM, Benjamin Bentmann wrote:
Brett Porter wrote:
(see r682889)
Ah, and I was already wondering why the Hudson bundle ever (almost)
worked for me back in July.
it then doesn't honour the installation settings at all, since the
value is set by the ITs by reading s
It's tied more to the decision of what role the Maven reactor session
plays in the build. The same question could be asked of JARs that were
just built but not installed.
I think as long as the POM is queued to be *built* in the current
reactor session it makes sense to use that, but pullin
I agree - I ran into this last night trying to test some internal
tooling...
James
On Fri, 2008-09-19 at 10:19 +1000, Brett Porter wrote:
>
> On 19/09/2008, at 1:31 AM, John Casey wrote:
>
> > What are you trying to accomplish? Maybe I can help you out more if
> > I know that?
>
> I was just
On 19/09/2008, at 1:31 AM, John Casey wrote:
What are you trying to accomplish? Maybe I can help you out more if
I know that?
I was just wondering if we were only voting on the release of the
tarballs, or if we were voting on the stage repository contents as well.
On 19/09/2008, at 1:30
On 19/09/2008, at 1:47 AM, [EMAIL PROTECTED] wrote:
Author: jdcasey
Date: Thu Sep 18 08:47:01 2008
New Revision: 696678
URL: http://svn.apache.org/viewvc?rev=696678&view=rev
Log:
Creating 2.2.x branch for the beginnings of some planning/work on
that release.
Is this getting ahead of oursel
Vote Result:
Binding:
John Casey
Lukas Theussl
Brian Fox
Vincent Siveton
Olivier Lamy
Jesse McConnell
Emmanuel Venisse
Non-Binding:
Paul Benedict
Fabrice Bellingard
Petar Tahchiev
Jason Dillon
Nicolas de Loof
Mauro Talevi
Maria Odea Ching
Thanks everyone,
-john
---
John Casey
Developer and
I'm confused how the build can continue without the parent pom. I
wouldn't call the behavior "inconsistent", I would call it completely
indeterminate. The parent plugin could have managed dependencies, plugin
definitions, variable definitions, etc. that will simply make the build
not work corre
I'm wondering what people's take on
http://jira.codehaus.org/browse/MNG-2196for Maven 3.0 is. Continuing
with a build if a parent pom is not found on
the local file-system or in the repository is something that could cause
inconsistent behavior between builds.
Thanks,
Shane
Brett Porter wrote:
(see r682889)
Ah, and I was already wondering why the Hudson bundle ever (almost)
worked for me back in July.
it then
doesn't honour the installation settings at all, since the value is set
by the ITs by reading settings.xml itself.
What use case had you in mind abou
I managed to have it with 'svn co .../archetype/trunk arch'
Regards
Raphaël
2008/9/17, Dennis Lundberg <[EMAIL PROTECTED]>:
> To be able to check out all of archetype directly to a Windows drive,
> like to C:\archetype, the path would need to be 20 characters shorter.
>
> The maximum path le
Jason van Zyl wrote:
Though if you want to work on the ITs, I think the first thing that
would be useful is to get the core Maven plugins out of the ITs which we
are using for testing. Integration tests for plugins themselves should
be moved to their respective plugins and stay there.
O
What are you trying to accomplish? Maybe I can help you out more if I
know that?
Brett Porter wrote:
is there a staged repo for this too?
On 16/09/2008, at 7:12 AM, John Casey wrote:
Hi everyone,
After fixing 70 issues and spending about 2 months going through
release candidate after relea
I've only been running 'mvn clean release:prepare release:perform', and
using a staging location that separates artifacts by artifactId and
version directories, to keep from messing up the stage:copy plugin
eventually. So, no, I don't think that will produce a staging repository
where all of th
25 matches
Mail list logo
