date:20071112


How many minutes are used to do a clean checkout?

Emmanuel

[EMAIL PROTECTED] a écrit :

Can you give me a hint what to do to fix this problem?
Or do you need more information?

-Ursprüngliche Nachricht-
Von: Renz, Manuel 
Gesendet: Donnerstag, 8. November 2007 12:59

An: [EMAIL PROTECTED]
Betreff: AW: CVS checkout fails: Terminated with fatal signal 11

It is planned to do a clean checkout once a day  


-Ursprüngliche Nachricht-
Von: Emmanuel Venisse [mailto:[EMAIL PROTECTED]
Gesendet: Mittwoch, 7. November 2007 20:19
An: [EMAIL PROTECTED]
Betreff: Re: CVS checkout fails: Terminated with fatal signal 11

It seems a kill was sent to the cvs client.

How many time is used to do a checkout?

Emmanuel

[EMAIL PROTECTED] a écrit :
 
Hi togehter,
I'm using continuum 1.1 beta 3 at the moment. I have set up 2 Projects 
which work just fine.


Now I'm in the process of setting up my third project, which is quite 
large (clean checked out working copy round about ~260MB ), with which 
I have massive problems.



Here is a short snippet from the logs:

[SocketListener0-1] INFO  FreemarkerManager  - Instantiating
Freemarker ConfigManager!,
com.opensymphony.webwork.views.freemarker.FreemarkerManager
[SocketListener0-0] INFO  Continuum:default  - Enqueuing 'My-App'
(Build definition id=6).
[pool-1-thread-1] INFO  BuildController:default  - Initializing build 
[pool-1-thread-1] INFO  BuildController:default  - Starting build of 
My-App [pool-1-thread-1] INFO  BuildController:default  - Updating 
working dir [pool-1-thread-1] INFO  BuildController:default  - 
Performing action check-working-directory [pool-1-thread-1] INFO 
BuildController:default  - Performing action checkout-project

[pool-1-thread-1] INFO  ContinuumScm:default - Checking out project:
'My-App', id: '63' to 'D:\buildserver\data\working\63'.
[pool-1-thread-1] INFO  ScmManager:default   - Executing: cmd.exe /X
/C 'cvs -z3 -f -d :pserver:[EMAIL PROTECTED]:/CVS -q checkout -d 63 my-app'
[pool-1-thread-1] INFO  ScmManager:default   - Working directory:
D:\buildserver\data\working
...
[pool-1-thread-1] WARN  ContinuumScm:default - Error while checking
out the code for project: 'My-App', id: '63' to 
'D:\buildserver\data\working\63'.

[pool-1-thread-1] WARN  ContinuumScm:default - Command output:
Terminated with fatal signal 11
[pool-1-thread-1] WARN  ContinuumScm:default - Provider message: The
cvs command failed.
[pool-1-thread-1] INFO  BuildController:default  - Merging SCM results 
[pool-1-thread-1] INFO  BuildController:default  - Error updating from 
SCM, not building [pool-1-thread-1] INFO  BuildController:default  - 
Initializing build [pool-1-thread-1] INFO  BuildController:default  - 
Starting build of parent [pool-1-thread-1] INFO 
BuildController:default  - Updating working dir [pool-1-thread-1] INFO 
BuildController:default  - Performing action check-working-directory 
[pool-1-thread-1] INFO  BuildController:default  - Performing action 
update-working-directory-from-scm



In the build logs via the web UI I get just following information and 
nothing more:


Provider message: The cvs command failed.
	Command output: 
	

--
--
---
Terminated with fatal signal 11

--
--
---

When I run cvs -z3 -f -d :pserver:[EMAIL PROTECTED]:/CVS -q checkout -d 63 
my-app manually on the commandline it works like a charme, so I think 
the problem is on some layer between cvs.exe and continuum...


I googled and found this:

Terminated with fatal signal 11

This message usually indicates that CVS (the server, if you're using 
client/server mode) has run out of (virtual) memory. Although CVS 
tries to catch the error and issue a more meaningful message, there 
are many circumstances where that is not possible. If you appear to 
have lots of memory available to the system, the problem is most 
likely that you're running into a system-wide limit on the amount of 
memory a single process can use or a similar process-specific limit. 
The mechanisms for displaying and setting such limits vary from system 
to system, so you'll have to consult an expert for your particular 
system if you don't know how to do that.



But since the command works if it is run outside of continuum I think 
this is not a memory problem on the server.



Hopefully somebody can give me a hint...

Greetings
Manuel Renz
--
Notice: This transmittal and/or attachments may be privileged 
orconfidential. If you are not the intended recipient, you are hereby 
notified that you have received this transmittal in error; any review, 
dissemination, or copying is strictly prohibited. If you received this 
transmittal in error, please notify us immediately by reply and 
immediately delete this

AW: AW: CVS checkout fails: Terminated with fatal signal 11

2007-11-12 Thread Manuel.Renz

Oh sorry, I think I missunderstood you. Thought you asked how often it is used 
to run ;-)
One checkout takes about 3 minutes.

-Ursprüngliche Nachricht-
Von: Emmanuel Venisse [mailto:[EMAIL PROTECTED] 
Gesendet: Montag, 12. November 2007 13:28
An: [EMAIL PROTECTED]
Betreff: Re: AW: CVS checkout fails: Terminated with fatal signal 11

How many minutes are used to do a clean checkout?

Emmanuel

[EMAIL PROTECTED] a écrit :
 Can you give me a hint what to do to fix this problem?
 Or do you need more information?
 
 -Ursprüngliche Nachricht-
 Von: Renz, Manuel
 Gesendet: Donnerstag, 8. November 2007 12:59
 An: [EMAIL PROTECTED]
 Betreff: AW: CVS checkout fails: Terminated with fatal signal 11
 
 It is planned to do a clean checkout once a day
 
 -Ursprüngliche Nachricht-
 Von: Emmanuel Venisse [mailto:[EMAIL PROTECTED]
 Gesendet: Mittwoch, 7. November 2007 20:19
 An: [EMAIL PROTECTED]
 Betreff: Re: CVS checkout fails: Terminated with fatal signal 11
 
 It seems a kill was sent to the cvs client.
 
 How many time is used to do a checkout?
 
 Emmanuel
 
 [EMAIL PROTECTED] a écrit :
  
 Hi togehter,
 I'm using continuum 1.1 beta 3 at the moment. I have set up 2 
 Projects which work just fine.

 Now I'm in the process of setting up my third project, which is quite 
 large (clean checked out working copy round about ~260MB ), with 
 which I have massive problems.


 Here is a short snippet from the logs:

 [SocketListener0-1] INFO  FreemarkerManager  - Instantiating
 Freemarker ConfigManager!,
 com.opensymphony.webwork.views.freemarker.FreemarkerManager
 [SocketListener0-0] INFO  Continuum:default  - Enqueuing 'My-App'
 (Build definition id=6).
 [pool-1-thread-1] INFO  BuildController:default  - Initializing build 
 [pool-1-thread-1] INFO  BuildController:default  - Starting build of 
 My-App [pool-1-thread-1] INFO  BuildController:default  - Updating 
 working dir [pool-1-thread-1] INFO  BuildController:default  - 
 Performing action check-working-directory [pool-1-thread-1] INFO 
 BuildController:default  - Performing action checkout-project
 [pool-1-thread-1] INFO  ContinuumScm:default - Checking out project:
 'My-App', id: '63' to 'D:\buildserver\data\working\63'.
 [pool-1-thread-1] INFO  ScmManager:default   - Executing: cmd.exe /X
 /C 'cvs -z3 -f -d :pserver:[EMAIL PROTECTED]:/CVS -q checkout -d 63 my-app'
 [pool-1-thread-1] INFO  ScmManager:default   - Working directory:
 D:\buildserver\data\working
 ...
 [pool-1-thread-1] WARN  ContinuumScm:default - Error while checking
 out the code for project: 'My-App', id: '63' to 
 'D:\buildserver\data\working\63'.
 [pool-1-thread-1] WARN  ContinuumScm:default - Command output:
 Terminated with fatal signal 11
 [pool-1-thread-1] WARN  ContinuumScm:default - Provider message: The
 cvs command failed.
 [pool-1-thread-1] INFO  BuildController:default  - Merging SCM 
 results [pool-1-thread-1] INFO  BuildController:default  - Error 
 updating from SCM, not building [pool-1-thread-1] INFO  
 BuildController:default  - Initializing build [pool-1-thread-1] INFO  
 BuildController:default  - Starting build of parent [pool-1-thread-1] 
 INFO BuildController:default  - Updating working dir 
 [pool-1-thread-1] INFO BuildController:default  - Performing action 
 check-working-directory [pool-1-thread-1] INFO  
 BuildController:default  - Performing action 
 update-working-directory-from-scm


 In the build logs via the web UI I get just following information and 
 nothing more:

  Provider message: The cvs command failed.
  Command output: 
  
 -
 -
 --
 ---
  Terminated with fatal signal 11
  
 -
 -
 --
 ---

 When I run cvs -z3 -f -d :pserver:[EMAIL PROTECTED]:/CVS -q checkout -d 63 
 my-app manually on the commandline it works like a charme, so I 
 think the problem is on some layer between cvs.exe and continuum...

 I googled and found this:
 
 Terminated with fatal signal 11

 This message usually indicates that CVS (the server, if you're using 
 client/server mode) has run out of (virtual) memory. Although CVS 
 tries to catch the error and issue a more meaningful message, there 
 are many circumstances where that is not possible. If you appear to 
 have lots of memory available to the system, the problem is most 
 likely that you're running into a system-wide limit on the amount of 
 memory a single process can use or a similar process-specific limit.
 The mechanisms for displaying and setting such limits vary from 
 system to system, so you'll have to consult an expert for your 
 particular system if you don't know how to do that.
 

 But since the command works if it is run outside of continuum I think 
 this is not a memory problem on the server.

Re: Cannot add custom resources to EAR's META-INF directory

2007-11-12 Thread Stephen Coy

We are successfully using the Maven ear plugin to build ear files for  
WAS 6.1. I don't believe that your problem is WAS specific however.


The ear plugin looks for its resources in src/main/application, so you  
typically put your META-INF directory there with its extra websphere  
descriptors.


ie.
src/
 main/
  application/
   META-INF/
  ibm-application- 
bnd.xmi

etc
I say typically because if you really need to you can change which  
directory the plugin uses with the earSourceDirectory configuration  
parameter.


Cheers,

Steve Coy


On 12/11/2007, at 11:18 PM, zm wrote:



Hi,

I'm new to maven, and I'm struggling to make a proper EAR build of an
existing project (created and maintained using IBM RAD 6).

The thing is, I'm creating an EAR, and it automatically creates the
application.xml file. But I have other resource files, xml ones,  
some are

created by IBM RAD to be interpreted by WebSphere, and I just want to
include them.

Whenever I include the resources (for testing I'm trying to include  
just the
info.xml file), maven just creates a classes directory with the  
resource
in it (it's META-INF), instead of including it inside the EAR's META- 
INF

file itself ...

Here are the pom.xml and the tree structure after mvn clean package:

Many Thanks in advance.


?xml version=1.0 encoding=UTF-8?
project xmlns=http://maven.apache.org/POM/4.0.0;
xmlns:xsi=http://www.w3.org/2001/XMLSchema-instance;
xsi:schemaLocation=http://maven.apache.org/POM/4.0.0
http://maven.apache.org/maven-v4_0_0.xsd;
modelVersion4.0.0/modelVersion
groupIdcom.mycompany.stock/groupId
artifactIdstock-ear/artifactId
version1.0-SNAPSHOT/version
packagingear/packaging
namestock EAR Project/name
urlhttp://stock.mycompany.com/ear/url
parent
groupIdcom.mycompany.stock/groupId
artifactIdstock-root/artifactId
version1.0-SNAPSHOT/version
/parent
dependencies
dependency
groupIdcom.mycompany.stock/groupId
artifactIdstock-ejb/artifactId
version1.0-SNAPSHOT/version
typejar/type
/dependency
dependency
groupIdcom.mycompany.stock/groupId
artifactIdstock-backoffice/artifactId
version1.0-SNAPSHOT/version
typewar/type
/dependency
dependency
groupIdcom.mycompany.stock/groupId
artifactIdstock-webservices/artifactId
version1.0-SNAPSHOT/version
typewar/type
/dependency
dependency
groupIdcom.mycompany.stock/groupId
artifactIdstock-tests/artifactId
version1.0-SNAPSHOT/version
typewar/type
/dependency
/dependencies
build
resources
resource
directoryMETA-INF/directory
targetPathMETA-INF/targetPath
filteringfalse/filtering
includes
include**/info.xml/include
/includes
/resource
/resources
plugins
plugin
artifactIdmaven-ear-plugin/artifactId
configuration
displayNamestockEAR/displayName
descriptionstock Enterprise Application 
Resource/description
version1.4/version
modules
jarModule

groupIdcom.mycompany.stock/groupId

artifactIdstock-ejb/artifactId

includeInApplicationXmltrue/includeInApplicationXml
/jarModule
webModule

groupIdcom.mycompany.stock/groupId

artifactIdstock-backoffice/artifactId

contextRoot/stockBOWeb/contextRoot
/webModule
webModule

assembly plugin or dependency plugin

2007-11-12 Thread pdelaney


Hello;

I have a problem creating a batch.jar file for a java application that will
run on some UNIX server via a shell script.

My maven project is a multiproject with the following subprojects
modules
modulecommon/module
moduleserver/module
modulerules/module
modulebatch/module
moduleweb/module
 /modules

the batch module will run on the server and want  is one big JAR file that
contains all of the .class files for the project and its dependencies.   I
would also like to add a Class-Main to the JAR MANIFEST and to have the
ability to remove some MANIFEST files from the other jars.

Should I be looking at the assembly plugin or the dependency plugin and are
there any good examples other than the examples I see in the plugin apache
documentation?

Thanks for the help.
Peter
-- 
View this message in context: 
http://www.nabble.com/assembly-plugin-or-dependency-plugin-tf4791376s177.html#a13706413
Sent from the Maven - Users mailing list archive at Nabble.com.


-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Re: AW: AW: CVS checkout fails: Terminated with fatal signal 11


After a quick search on Google [1], it seems you're not alone with this cvs 
error. But it's sure it isn't a Continuum error but a cvs issue, maybe a memory 
issue on the cvs server [2]

[1] http://www.google.fr/search?q=Terminated+with+fatal+signal+11
[2] http://osdir.com/ml/version-control.cvs.bugs/2002-04/msg00298.html

Emmanuel

[EMAIL PROTECTED] a écrit :

Oh sorry, I think I missunderstood you. Thought you asked how often it is used 
to run ;-)
One checkout takes about 3 minutes.

-Ursprüngliche Nachricht-
Von: Emmanuel Venisse [mailto:[EMAIL PROTECTED] 
Gesendet: Montag, 12. November 2007 13:28

An: [EMAIL PROTECTED]
Betreff: Re: AW: CVS checkout fails: Terminated with fatal signal 11

How many minutes are used to do a clean checkout?

Emmanuel

[EMAIL PROTECTED] a écrit :

Can you give me a hint what to do to fix this problem?
Or do you need more information?

-Ursprüngliche Nachricht-
Von: Renz, Manuel
Gesendet: Donnerstag, 8. November 2007 12:59
An: [EMAIL PROTECTED]
Betreff: AW: CVS checkout fails: Terminated with fatal signal 11

It is planned to do a clean checkout once a day

-Ursprüngliche Nachricht-
Von: Emmanuel Venisse [mailto:[EMAIL PROTECTED]
Gesendet: Mittwoch, 7. November 2007 20:19
An: [EMAIL PROTECTED]
Betreff: Re: CVS checkout fails: Terminated with fatal signal 11

It seems a kill was sent to the cvs client.

How many time is used to do a checkout?

Emmanuel

[EMAIL PROTECTED] a écrit :
 
Hi togehter,
I'm using continuum 1.1 beta 3 at the moment. I have set up 2 
Projects which work just fine.


Now I'm in the process of setting up my third project, which is quite 
large (clean checked out working copy round about ~260MB ), with 
which I have massive problems.



Here is a short snippet from the logs:

[SocketListener0-1] INFO  FreemarkerManager  - Instantiating
Freemarker ConfigManager!,
com.opensymphony.webwork.views.freemarker.FreemarkerManager
[SocketListener0-0] INFO  Continuum:default  - Enqueuing 'My-App'
(Build definition id=6).
[pool-1-thread-1] INFO  BuildController:default  - Initializing build 
[pool-1-thread-1] INFO  BuildController:default  - Starting build of 
My-App [pool-1-thread-1] INFO  BuildController:default  - Updating 
working dir [pool-1-thread-1] INFO  BuildController:default  - 
Performing action check-working-directory [pool-1-thread-1] INFO 
BuildController:default  - Performing action checkout-project

[pool-1-thread-1] INFO  ContinuumScm:default - Checking out project:
'My-App', id: '63' to 'D:\buildserver\data\working\63'.
[pool-1-thread-1] INFO  ScmManager:default   - Executing: cmd.exe /X
/C 'cvs -z3 -f -d :pserver:[EMAIL PROTECTED]:/CVS -q checkout -d 63 my-app'
[pool-1-thread-1] INFO  ScmManager:default   - Working directory:
D:\buildserver\data\working
...
[pool-1-thread-1] WARN  ContinuumScm:default - Error while checking
out the code for project: 'My-App', id: '63' to 
'D:\buildserver\data\working\63'.

[pool-1-thread-1] WARN  ContinuumScm:default - Command output:
Terminated with fatal signal 11
[pool-1-thread-1] WARN  ContinuumScm:default - Provider message: The
cvs command failed.
[pool-1-thread-1] INFO  BuildController:default  - Merging SCM 
results [pool-1-thread-1] INFO  BuildController:default  - Error 
updating from SCM, not building [pool-1-thread-1] INFO  
BuildController:default  - Initializing build [pool-1-thread-1] INFO  
BuildController:default  - Starting build of parent [pool-1-thread-1] 
INFO BuildController:default  - Updating working dir 
[pool-1-thread-1] INFO BuildController:default  - Performing action 
check-working-directory [pool-1-thread-1] INFO  
BuildController:default  - Performing action 
update-working-directory-from-scm



In the build logs via the web UI I get just following information and 
nothing more:


Provider message: The cvs command failed.
	Command output: 
	

-
-
--
---
Terminated with fatal signal 11

-
-
--
---

When I run cvs -z3 -f -d :pserver:[EMAIL PROTECTED]:/CVS -q checkout -d 63 
my-app manually on the commandline it works like a charme, so I 
think the problem is on some layer between cvs.exe and continuum...


I googled and found this:

Terminated with fatal signal 11

This message usually indicates that CVS (the server, if you're using 
client/server mode) has run out of (virtual) memory. Although CVS 
tries to catch the error and issue a more meaningful message, there 
are many circumstances where that is not possible. If you appear to 
have lots of memory available to the system, the problem is most 
likely that you're running into a system-wide limit on the amount of 
memory a single process can use or a similar process-specific limit.
The mechanisms for displaying and setting such limits vary from 
system to

Re: Cleaning up results

2007-11-12 Thread Alex Mayorga Adame


Would this guide
http://maven.apache.org/continuum/documentation/1_1/installation/upgrade.html
work if I'm coming from 1.0.3?

I'll take the plunge and go with 1.1 Beta 4 if only to reclaim some space.
Any bumps along the road I'd need to be aware of?

Thanks in advance,
Alex
-- 
View this message in context: 
http://www.nabble.com/Cleaning-up-results-tf1906827.html#a13707155
Sent from the Continuum - Users mailing list archive at Nabble.com.

Re: Cleaning up results



Alex Mayorga Adame a écrit :

Would this guide
http://maven.apache.org/continuum/documentation/1_1/installation/upgrade.html
work if I'm coming from 1.0.3?


No, it is for 1.1 only.

Emmanuel



I'll take the plunge and go with 1.1 Beta 4 if only to reclaim some space.
Any bumps along the road I'd need to be aware of?

Thanks in advance,
Alex

RE: maven constantly rebuilding everything

2007-11-12 Thread EJ Ciramella

So the fix would be something like:

1 - Make process-resources look at the source of the activated profiles and 
make sure they are newer than the templates and then (and only then) process 
the templates.
2 - Make the install plugin test what's in the repository is out of date with 
the just built jar before installing.
3 - Make the war plugin able to ignore/exclude maven generated files (such as 
pom.properties) - this may/may not solve our war rebuilding issues.

??? 

-Original Message-
From: EJ Ciramella [mailto:[EMAIL PROTECTED] 
Sent: Friday, November 09, 2007 9:33 PM
To: Maven Users List
Subject: RE: maven constantly rebuilding everything

Well, maybe I'm not quite out of the woods.

Process-resources always moves some files into classes so I guess this will 
forever jar stuff up, right? 

-Original Message-
From: EJ Ciramella [mailto:[EMAIL PROTECTED] 
Sent: Friday, November 09, 2007 1:13 PM
To: Maven Users List
Subject: RE: maven constantly rebuilding everything

Yeah, I'm able to prevent rebuilding of all the jars, not re-downloading the 
resources as part of the maven-dependency-plugin - so that much is ok.

I'm still getting everything installed every time I build even if it didn't 
have to re-jar the source.  Is there a solution/jira for that (I couldn't find 
anything with my searches). 

-Original Message-
From: Wayne Fay [mailto:[EMAIL PROTECTED] 
Sent: Friday, November 09, 2007 11:59 AM
To: Maven Users List
Subject: Re: maven constantly rebuilding everything

It certainly sounds like you are dealing with a large project. Since
you mentioned images, I wonder how much of the 200mb is actually code
and how much is everything else? I assume you've run an analysis.

I've looked into this rebuilding every jar issue at some point. In
short, Maven creates a pom.properties file every time it executes in a
given module. Thus you *always* have changes, even when you don't.
Also the filtered resources thing that Daniel mentioned is another
forgotten source of changes.

Damien just sent the link to this bug, he must have been reading my mind... ;-)

Wayne

On 11/9/07, EJ Ciramella [EMAIL PROTECTED] wrote:
 I'd love to know what people are building and the size of things, I think our 
 project is larger than most (I'm not stepping into a pissing contest, simply 
 stating the facts).  We're working on a massive refactoring both in codeline 
 management and actual code refactoring.

 Even with the refactoring done, that will only help to a certain degree.  
 With testing, we're talking about 1 - 1.5 hours.
 The resulting assembly for our main application is 195 - 200 mb (full of 
 jsps, images, classes, jars, etc).  I wish I could better show you the size 
 of the application (we have three ears and 4 - 5 wars depending on what 
 you're building).  There's no real honest to goodness code in the jsps (as it 
 should be), all business logic is handled else where.  As a release engineer, 
 it's not my business to make sweeping changes like that.  I can make 
 suggestions, but in the end, it's up to development to decide what projects 
 have merit/profit/etc associated with them.

 It'd be nice if you could work at the lowest directory all the time (the one 
 responsible for building the ear file), but when your change effects a module 
 that the ear you want depends on (or MULTIPLE subprojects), most developers 
 here build from the top.  Also, when profiles.xml lives at the root of the 
 overall project and contains 90% the configuration for a developers box, 
 sitting at a lower level, aren't these properties unavailable (doesn't maven 
 only look in the working directory for profiles.xml)?

 If maven wouldn't try to rebuild every jar and then every war and then every 
 ear, life would be much better.

 -Original Message-
 From: Wayne Fay [mailto:[EMAIL PROTECTED]
 Sent: Friday, November 09, 2007 8:10 AM
 To: Maven Users List
 Subject: Re: maven constantly rebuilding everything

 Sounds like your projects are entirely too big -- break 'em up a bit.
 Even 10min builds sounds rather long -- is this compilation only or
 including testing? And what portion of that time is consumed by your
 tests -- at this level (compilation), you should be running very fast
 unit tests, generally not integration tests.

 Artifacts/modules are cheap in Maven. War modules in particular
 should not have a lot of code associated with them -- move the code
 into Jars and extract the code out of your JSPs etc into classes that
 go into those Jars.

 Also when you're working, you don't clean install from the top
 parent, but instead run it from the module you're working on. At
 least, this is what we do...

 Wayne

 On 11/9/07, EJ Ciramella [EMAIL PROTECTED] wrote:
  That's incredible, doing that takes 30 - 40 minutes here.
 
  Most developers here are used to a 10 build time.
 
  -Original Message-
  From: Wayne Fay [mailto:[EMAIL PROTECTED]
  Sent: Thursday, November 08, 2007 11:08 PM

Cobertura give me a wrong result and the next execution is right

2007-11-12 Thread Martin Alejandro Villalobos


Hello, I'm having a problem
I'm executing Continuum with a lot of projects including submodules.
CI is executing clean install site site:deploy , including cobertura in 
these reports.
The trouble is that Cobertura give wrong results one time and right in 
the next execution , and so on.

Some knows wich is the problem.

Thanks .

Regards,

Martin.

from 1.1-beta3 to beta4

Hi all,

I'm trying to upgrade from beta3 to beta4 but no succes.
I have been following the migration page instructions:
http://maven.apache.org/continuum/documentation/1_1/installation/upgrade.html

but I get the following exception when importing the builds database:

9660 [main] INFO JPOX.RDBMS  - Validated 2 foreign key(s) for table
PROJECTGROUP_NOTIFIERS
9698 [main] INFO JPOX.RDBMS  - Validated 3 index(es) for table
PROJECTGROUP_NOTIFIERS
Exception in thread main javax.jdo.JDOUserException: One or more
instances could not be deleted
at 
org.jpox.AbstractPersistenceManager.deletePersistentAll(AbstractPersistenceManager.java:1525)
at 
org.jpox.store.rdbms.query.JDOQLQuery.performDeletePersistentAll(JDOQLQuery.java:687)
at org.jpox.store.query.Query.deletePersistentAll(Query.java:1120)
at org.jpox.store.query.Query.deletePersistentAll(Query.java:1075)
at org.jpox.store.query.Query.deletePersistentAll(Query.java:1049)
at 
org.codehaus.plexus.jdo.PlexusJdoUtils.removeAll(PlexusJdoUtils.java:444)
at 
org.apache.maven.continuum.store.JdoContinuumStore.eraseDatabase(JdoContinuumStore.java:1961)
at 
org.apache.maven.continuum.management.JdoDataManagementTool.eraseDatabase(JdoDataManagementTool.java:127)
at 
org.apache.maven.continuum.management.DataManagementCli.processDatabase(DataManagementCli.java:254)
at 
org.apache.maven.continuum.management.DataManagementCli.main(DataManagementCli.java:158)
NestedThrowablesStackTrace:
javax.jdo.JDOUserException: One or more instances could not be deleted
at 
org.jpox.AbstractPersistenceManager.deletePersistentAll(AbstractPersistenceManager.java:1525)
at 
org.jpox.store.rdbms.scostore.ElementContainerStore.clear(ElementContainerStore.java:659)
at 
org.jpox.store.mapping.CollectionMapping.deleteDependent(CollectionMapping.java:334)
at 
org.jpox.store.rdbms.table.ClassTable.deleteDependent(ClassTable.java:2651)
at org.jpox.store.StoreManager.deleteDependent(StoreManager.java:1019)
at 
org.jpox.state.StateManagerImpl.deletePersistent(StateManagerImpl.java:4332)
at 
org.jpox.AbstractPersistenceManager.internalDeletePersistent(AbstractPersistenceManager.java:1473)
at 
org.jpox.AbstractPersistenceManager.deletePersistentAll(AbstractPersistenceManager.java:1516)
at 
org.jpox.store.rdbms.query.JDOQLQuery.performDeletePersistentAll(JDOQLQuery.java:687)
at org.jpox.store.query.Query.deletePersistentAll(Query.java:1120)
at org.jpox.store.query.Query.deletePersistentAll(Query.java:1075)
at org.jpox.store.query.Query.deletePersistentAll(Query.java:1049)
at 
org.codehaus.plexus.jdo.PlexusJdoUtils.removeAll(PlexusJdoUtils.java:444)
at 
org.apache.maven.continuum.store.JdoContinuumStore.eraseDatabase(JdoContinuumStore.java:1961)
at 
org.apache.maven.continuum.management.JdoDataManagementTool.eraseDatabase(JdoDataManagementTool.java:127)
at 
org.apache.maven.continuum.management.DataManagementCli.processDatabase(DataManagementCli.java:254)
at 
org.apache.maven.continuum.management.DataManagementCli.main(DataManagementCli.java:158)
NestedThrowablesStackTrace:
javax.jdo.JDODataStoreException: Delete request failed: DELETE FROM
BUILDDEFINITION WHERE ID = ?
at 
org.jpox.store.rdbms.request.DeleteRequest.execute(DeleteRequest.java:288)
at org.jpox.store.rdbms.table.ClassTable.delete(ClassTable.java:2611)
at org.jpox.store.StoreManager.delete(StoreManager.java:981)
at 
org.jpox.state.StateManagerImpl.internalDeletePersistent(StateManagerImpl.java:4395)
at 
org.jpox.state.StateManagerImpl.deletePersistent(StateManagerImpl.java:4350)
at 
org.jpox.AbstractPersistenceManager.internalDeletePersistent(AbstractPersistenceManager.java:1473)
at 
org.jpox.AbstractPersistenceManager.deletePersistentAll(AbstractPersistenceManager.java:1516)
at 
org.jpox.store.rdbms.scostore.ElementContainerStore.clear(ElementContainerStore.java:659)
at 
org.jpox.store.mapping.CollectionMapping.deleteDependent(CollectionMapping.java:334)
at 
org.jpox.store.rdbms.table.ClassTable.deleteDependent(ClassTable.java:2651)
at org.jpox.store.StoreManager.deleteDependent(StoreManager.java:1019)
at 
org.jpox.state.StateManagerImpl.deletePersistent(StateManagerImpl.java:4332)
at 
org.jpox.AbstractPersistenceManager.internalDeletePersistent(AbstractPersistenceManager.java:1473)
at 
org.jpox.AbstractPersistenceManager.deletePersistentAll(AbstractPersistenceManager.java:1516)
at 
org.jpox.store.rdbms.query.JDOQLQuery.performDeletePersistentAll(JDOQLQuery.java:687)
at org.jpox.store.query.Query.deletePersistentAll(Query.java:1120)
at org.jpox.store.query.Query.deletePersistentAll(Query.java:1075)
at org.jpox.store.query.Query.deletePersistentAll(Query.java:1049)
at

Re: Cobertura give me a wrong result and the next execution is right

could you give some more details about 'wrong results'?

On Nov 12, 2007 4:45 PM, Martin Alejandro Villalobos
[EMAIL PROTECTED] wrote:
 Hello, I'm having a problem
 I'm executing Continuum with a lot of projects including submodules.
 CI is executing clean install site site:deploy , including cobertura in
 these reports.
 The trouble is that Cobertura give wrong results one time and right in
 the next execution , and so on.
 Some knows wich is the problem.

 Thanks .

 Regards,

 Martin.





-- 
http://www.somatik.be
Microsoft gives you windows, Linux gives you the whole house.

Re: Properties files exclude from jar artifact

2007-11-12 Thread Saloucious


Sure assembly will unpack properites in conf dir.

But in my assembly, i would like to define a dependency to this module
So finally, i will have my jar artifact with my properties files embded (for
eg in a lib dir)
and my properties files in conf dir (double)



Wayne Fay wrote:
 
 Use dependencies plugin to unpack the properties files into your
 current module so they will be packaged by assembly into the conf dir
 you require.
 
 Wayne
 
 On 11/9/07, Saloucious [EMAIL PROTECTED] wrote:

 So if  i package properties file into my jars, how can I create finally
 an
 assembly which create a deploy/conf dir with all properties from jar.

 We put all properties into this dir because it is easiest to patch it.





 Wayne Fay wrote:
 
  No such all classifier exists. You must specify each dependency.
 
  Wayne
 
  On 11/6/07, Saloucious [EMAIL PROTECTED] wrote:
 
  Ok, but is it possible to get jar artifact and properties artifact in
 one
  dependency decalaration:
 
  For example, I deploy module1 with jar artifact and properties
 artifact
 
  and in a module2, i declare a dependency  with module1
  Is Maven able to retrieve 2 artifacts (for example with a
  classifierALL/classifier or something like that)
 
  dependency
  groupidmygroup/groupid
  artifactidmodule1/artifactid
  /dependency
 
  Otherwise i must declare 2 dependencies by module, one for binaries
 files
  and one for properties files...
 
 
  Wayne Fay wrote:
  
   You could make another artifact (jar) that consists only of
 properties
   files that are jar'ed together and deploy it into your repo.
  
   Wayne
  
   On 10/26/07, Saloucious [EMAIL PROTECTED] wrote:
  
   Hi,
  
   Acutally, i'm working on a migration from Ant to Maven.
   The old Ant script creates for each project a  jar and  copy
  properties
   files in a deploy dir.
   So these properties file are not embeded in jar.
  
   Now i'm on Maven,
   If I excludes files from jar, they will not be in repository
   I don't find the way to be able to not embed properties file from
 jar
  and
   install them in repo.
   --
   View this message in context:
  
 
 http://www.nabble.com/Properties-files-exclude-from-jar-artifact-tf4698085s177.html#a13429857
   Sent from the Maven - Users mailing list archive at Nabble.com.
  
  
  
 -
   To unsubscribe, e-mail: [EMAIL PROTECTED]
   For additional commands, e-mail: [EMAIL PROTECTED]
  
  
  
  
 -
   To unsubscribe, e-mail: [EMAIL PROTECTED]
   For additional commands, e-mail: [EMAIL PROTECTED]
  
  
  
 
  --
  View this message in context:
 
 http://www.nabble.com/Properties-files-exclude-from-jar-artifact-tf4698085s177.html#a13616078
  Sent from the Maven - Users mailing list archive at Nabble.com.
 
 
  -
  To unsubscribe, e-mail: [EMAIL PROTECTED]
  For additional commands, e-mail: [EMAIL PROTECTED]
 
 
 
  -
  To unsubscribe, e-mail: [EMAIL PROTECTED]
  For additional commands, e-mail: [EMAIL PROTECTED]
 
 
 

 --
 View this message in context:
 http://www.nabble.com/Properties-files-exclude-from-jar-artifact-tf4698085s177.html#a13669183
 Sent from the Maven - Users mailing list archive at Nabble.com.


 -
 To unsubscribe, e-mail: [EMAIL PROTECTED]
 For additional commands, e-mail: [EMAIL PROTECTED]


 
 -
 To unsubscribe, e-mail: [EMAIL PROTECTED]
 For additional commands, e-mail: [EMAIL PROTECTED]
 
 
 

-- 
View this message in context: 
http://www.nabble.com/Properties-files-exclude-from-jar-artifact-tf4698085s177.html#a13708126
Sent from the Maven - Users mailing list archive at Nabble.com.


-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Re: Cobertura give me a wrong result and the next execution is right

2007-11-12 Thread mav007ar


Yes, by example, there are some classes that have 98% of line coverage and
cobertura switch between this results and 2%
Thanks 


Francis De Brabandere-2 wrote:
 
 could you give some more details about 'wrong results'?
 
 On Nov 12, 2007 4:45 PM, Martin Alejandro Villalobos
 [EMAIL PROTECTED] wrote:
 Hello, I'm having a problem
 I'm executing Continuum with a lot of projects including submodules.
 CI is executing clean install site site:deploy , including cobertura in
 these reports.
 The trouble is that Cobertura give wrong results one time and right in
 the next execution , and so on.
 Some knows wich is the problem.

 Thanks .

 Regards,

 Martin.


 
 
 
 -- 
 http://www.somatik.be
 Microsoft gives you windows, Linux gives you the whole house.
 
 

-- 
View this message in context: 
http://www.nabble.com/Cobertura-give-me-a-wrong-result-and-the-next-execution-is-right-tf4791862.html#a13708174
Sent from the Continuum - Users mailing list archive at Nabble.com.

Re: Cobertura give me a wrong result and the next execution is right

I think this might be something to ask on the cobertura mailing list
rather than here. I used to have cobertura problems but they were gone
after switching back to version 2.0 instead of 2.1

add this to your buildplugins

plugin
groupIdorg.codehaus.mojo/groupId
artifactIdcobertura-maven-plugin/artifactId
version2.0/version!-- version 2.1 is
broken? --
executions
execution
goals
goalclean/goal
/goals
/execution
/executions
/plugin

On Nov 12, 2007 5:02 PM, mav007ar [EMAIL PROTECTED] wrote:

Yes, by example, there are some classes that have 98% of line coverage and
cobertura switch between this results and 2%
Thanks

Francis De Brabandere-2 wrote:

could you give some more details about 'wrong results'?

On Nov 12, 2007 4:45 PM, Martin Alejandro Villalobos
[EMAIL PROTECTED] wrote:
Hello, I'm having a problem
I'm executing Continuum with a lot of projects including submodules.
CI is executing clean install site site:deploy , including cobertura in
these reports.
The trouble is that Cobertura give wrong results one time and right in
the next execution , and so on.
Some knows wich is the problem.

Thanks .

Regards,

Martin.

--
http://www.somatik.be
Microsoft gives you windows, Linux gives you the whole house.

--
View this message in context:
http://www.nabble.com/Cobertura-give-me-a-wrong-result-and-the-next-execution-is-right-tf4791862.html#a13708174
Sent from the Continuum - Users mailing list archive at Nabble.com.

--
http://www.somatik.be
Microsoft gives you windows, Linux gives you the whole house.

AW: AW: AW: CVS checkout fails: Terminated with fatal signal 11

2007-11-12 Thread Manuel.Renz

 
As I wrote in my initial request I'm aware of the virtual memory thingy... (I 
even quoted that)
The thing is that the same cvs command runs fine via shell on the same machine 
that continuum runs on. 
So I'm pretty sure that this is not [2] a memory issue on the cvs server. 
About 40 people are using the cvs server and memory never was a problem. 

Continuum itself runs on a dedicated machine with 3GB RAM and an intel xeon 
2GHz multicore cpu, this is rather oversized than insufficient hardware.

Is it possible that the local JVM is out of memory? Although, wouldn't I got an 
OutOfMemory Exception if this was the case?
Are you aware of situations when continuum's process  needs a lot of memory?

Is it possible to write the output of the cvs command to the continuum logs?

-Ursprüngliche Nachricht-
Von: Emmanuel Venisse [mailto:[EMAIL PROTECTED] 
Gesendet: Montag, 12. November 2007 15:38
An: [EMAIL PROTECTED]
Betreff: Re: AW: AW: CVS checkout fails: Terminated with fatal signal 11

After a quick search on Google [1], it seems you're not alone with this cvs 
error. But it's sure it isn't a Continuum error but a cvs issue, maybe a memory 
issue on the cvs server [2]

[1] http://www.google.fr/search?q=Terminated+with+fatal+signal+11
[2] http://osdir.com/ml/version-control.cvs.bugs/2002-04/msg00298.html

Emmanuel

[EMAIL PROTECTED] a écrit :
 Oh sorry, I think I missunderstood you. Thought you asked how often it 
 is used to run ;-) One checkout takes about 3 minutes.
 
 -Ursprüngliche Nachricht-
 Von: Emmanuel Venisse [mailto:[EMAIL PROTECTED]
 Gesendet: Montag, 12. November 2007 13:28
 An: [EMAIL PROTECTED]
 Betreff: Re: AW: CVS checkout fails: Terminated with fatal signal 11
 
 How many minutes are used to do a clean checkout?
 
 Emmanuel
 
 [EMAIL PROTECTED] a écrit :
 Can you give me a hint what to do to fix this problem?
 Or do you need more information?

 -Ursprüngliche Nachricht-
 Von: Renz, Manuel
 Gesendet: Donnerstag, 8. November 2007 12:59
 An: [EMAIL PROTECTED]
 Betreff: AW: CVS checkout fails: Terminated with fatal signal 11

 It is planned to do a clean checkout once a day

 -Ursprüngliche Nachricht-
 Von: Emmanuel Venisse [mailto:[EMAIL PROTECTED]
 Gesendet: Mittwoch, 7. November 2007 20:19
 An: [EMAIL PROTECTED]
 Betreff: Re: CVS checkout fails: Terminated with fatal signal 11

 It seems a kill was sent to the cvs client.

 How many time is used to do a checkout?

 Emmanuel

 [EMAIL PROTECTED] a écrit :
  
 Hi togehter,
 I'm using continuum 1.1 beta 3 at the moment. I have set up 2 
 Projects which work just fine.

 Now I'm in the process of setting up my third project, which is 
 quite large (clean checked out working copy round about ~260MB ), 
 with which I have massive problems.


 Here is a short snippet from the logs:

 [SocketListener0-1] INFO  FreemarkerManager  - Instantiating
 Freemarker ConfigManager!,
 com.opensymphony.webwork.views.freemarker.FreemarkerManager
 [SocketListener0-0] INFO  Continuum:default  - Enqueuing 'My-App'
 (Build definition id=6).
 [pool-1-thread-1] INFO  BuildController:default  - Initializing 
 build [pool-1-thread-1] INFO  BuildController:default  - Starting 
 build of My-App [pool-1-thread-1] INFO  BuildController:default  - 
 Updating working dir [pool-1-thread-1] INFO  BuildController:default  
 - Performing action check-working-directory [pool-1-thread-1] INFO 
 BuildController:default  - Performing action checkout-project
 [pool-1-thread-1] INFO  ContinuumScm:default - Checking out project:
 'My-App', id: '63' to 'D:\buildserver\data\working\63'.
 [pool-1-thread-1] INFO  ScmManager:default   - Executing: cmd.exe /X
 /C 'cvs -z3 -f -d :pserver:[EMAIL PROTECTED]:/CVS -q checkout -d 63 
 my-app'
 [pool-1-thread-1] INFO  ScmManager:default   - Working directory:
 D:\buildserver\data\working
 ...
 [pool-1-thread-1] WARN  ContinuumScm:default - Error while checking
 out the code for project: 'My-App', id: '63' to 
 'D:\buildserver\data\working\63'.
 [pool-1-thread-1] WARN  ContinuumScm:default - Command output:
 Terminated with fatal signal 11
 [pool-1-thread-1] WARN  ContinuumScm:default - Provider message: The
 cvs command failed.
 [pool-1-thread-1] INFO  BuildController:default  - Merging SCM 
 results [pool-1-thread-1] INFO  BuildController:default  - Error 
 updating from SCM, not building [pool-1-thread-1] INFO 
 BuildController:default  - Initializing build [pool-1-thread-1] INFO 
 BuildController:default  - Starting build of parent 
 [pool-1-thread-1] INFO BuildController:default  - Updating working 
 dir [pool-1-thread-1] INFO BuildController:default  - Performing 
 action check-working-directory [pool-1-thread-1] INFO 
 BuildController:default  - Performing action 
 update-working-directory-from-scm


 In the build logs via the web UI I get just following information 
 and nothing more:

 Provider message: The cvs command failed.
 Command output:

Re: How to create a Mojo that controls an ANT task

2007-11-12 Thread Siegfried Goeschl


Hi Pedro,

I faced a similar task and it took a while to find a plugin invoking ANT 
within a Maven build - unfortunately I can't remember what plugin I 
looked at (oehaus?!) but I copywasted to code for my Canoo WebTest plugin


http://people.apache.org/~sgoeschl/download/maven-plugins/

The code uses an AntExecutor which does the trick

Cheers,

Siegfried Goeschl


Pedro Viegas wrote:

Hi there guys,

I am trying to create a Maven plugin that executes a given ANT task inside
it.
My real case is to call the Hibernate Tools task to generate a bunch of ORM
classes and Hibernate mappings.
But I need to make some decisions and have to pass some parameters to the
several Hibernate Tools tasks I need to call depending on some conditions
that I need to analize in a Java Mojo.

Basically I need to:

   1. Create a Mojo with some parameters and do some processing and file
   generation of my own
   2. Call the ANT Hibernate Tools tasks a bunch of times to generate all
   my classes/mapings on several packages for several database schemas
   3. Try to keep it as simple as possible to the end programmer user,
   simply call a Maven plugin and state some configuration files and
   destination package and let the conventioned behavior take charge.

I have created the first part easily by creating a Java Maven Plugin.
The second part I have configured like a maven-ant-plugin addition to the
POM and got it working, but this has to be copied into each of the projects
that need this goal and all the configuration I could pragmatically do have
to be typed repeatedly.
So 1 and 2 are go, 3 is what I can't find any documentation or sample to
help me.

I have seen the example from:
http://maven.apache.org/guides/plugin/guide-ant-plugin-development.html

But this is how to build a maven plugin with ANT. What I need is to build a
maven plugin in Java that can call an ant task, like antrun plugin does but
adding a few more java actions and decisions before and after.

So, how can I use the already running maven java plugin to call the ANT
tasks?
Can I call a plugin inside another plugin? How?
Should I try to extend the maven-antrun-plugin to try to accomplish this?

Thanks in advance,

  


-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Re: How to create a Mojo that controls an ANT task

2007-11-12 Thread Manos Batsis



maven-antrun-plugin?

Siegfried Goeschl wrote:

Hi Pedro,

I faced a similar task and it took a while to find a plugin invoking ANT 
within a Maven build - unfortunately I can't remember what plugin I 
looked at (oehaus?!) but I copywasted to code for my Canoo WebTest plugin


http://people.apache.org/~sgoeschl/download/maven-plugins/

The code uses an AntExecutor which does the trick

Cheers,

Siegfried Goeschl


Pedro Viegas wrote:

Hi there guys,

I am trying to create a Maven plugin that executes a given ANT task 
inside

it.
My real case is to call the Hibernate Tools task to generate a bunch 
of ORM

classes and Hibernate mappings.
But I need to make some decisions and have to pass some parameters to the
several Hibernate Tools tasks I need to call depending on some conditions
that I need to analize in a Java Mojo.

Basically I need to:

   1. Create a Mojo with some parameters and do some processing and file
   generation of my own
   2. Call the ANT Hibernate Tools tasks a bunch of times to generate all
   my classes/mapings on several packages for several database schemas
   3. Try to keep it as simple as possible to the end programmer user,
   simply call a Maven plugin and state some configuration files and
   destination package and let the conventioned behavior take charge.

I have created the first part easily by creating a Java Maven Plugin.
The second part I have configured like a maven-ant-plugin addition to the
POM and got it working, but this has to be copied into each of the 
projects
that need this goal and all the configuration I could pragmatically do 
have

to be typed repeatedly.
So 1 and 2 are go, 3 is what I can't find any documentation or sample to
help me.

I have seen the example from:
http://maven.apache.org/guides/plugin/guide-ant-plugin-development.html

But this is how to build a maven plugin with ANT. What I need is to 
build a
maven plugin in Java that can call an ant task, like antrun plugin 
does but

adding a few more java actions and decisions before and after.

So, how can I use the already running maven java plugin to call the ANT
tasks?
Can I call a plugin inside another plugin? How?
Should I try to extend the maven-antrun-plugin to try to accomplish this?

Thanks in advance,

  


-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]




-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Re: Cobertura give me a wrong result and the next execution is right

2007-11-12 Thread mav007ar

Thanks, Francis, I'll try in the cobertura mailing list.
I'll tell you when I fix it.
Cheers.
Martin.

Francis De Brabandere-2 wrote:

I think this might be something to ask on the cobertura mailing list
rather than here. I used to have cobertura problems but they were gone
after switching back to version 2.0 instead of 2.1

add this to your buildplugins

On Nov 12, 2007 5:02 PM, mav007ar [EMAIL PROTECTED] wrote:

Yes, by example, there are some classes that have 98% of line coverage
and
cobertura switch between this results and 2%
Thanks

Francis De Brabandere-2 wrote:

could you give some more details about 'wrong results'?

On Nov 12, 2007 4:45 PM, Martin Alejandro Villalobos
[EMAIL PROTECTED] wrote:
Hello, I'm having a problem
I'm executing Continuum with a lot of projects including submodules.
CI is executing clean install site site:deploy , including cobertura
in
these reports.
The trouble is that Cobertura give wrong results one time and right in
the next execution , and so on.
Some knows wich is the problem.

Thanks .

Regards,

Martin.

--
http://www.somatik.be
Microsoft gives you windows, Linux gives you the whole house.

--
View this message in context:
http://www.nabble.com/Cobertura-give-me-a-wrong-result-and-the-next-execution-is-right-tf4791862.html#a13708611
Sent from the Continuum - Users mailing list archive at Nabble.com.

Run Archiva on Aix

2007-11-12 Thread MATHUS Baptiste

Hi all,

Thanks to cb, I solved my repo problem. I finally decided to separate my 
corporate and publicly retrieved artifact under two distinct urls.
Now, my tests are finished, I want to install my archiva config on an Aix. My 
tests were done on a Debian Linux.

Do you think it's a lost cause ? :-).
Which script in the archiva/bin directory would be the closest to the Aix needs 
if someone has an idea?

Thanks a lot.

-- 
B. MATHUS

Re: AW: AW: AW: CVS checkout fails: Terminated with fatal signal 11


The message is Terminated with fatal signal 11 and it is sent by cvs directly.

Do you use the embedded pure java cvs client (it is the default) or the native 
cvs client?

With 1.1-beta-3, logs are configured in 
WEB-INF/classes/META-INF/plexus/application.xml, you can set the maven-scm log 
level to debug.

Emmanuel

[EMAIL PROTECTED] a écrit :
 
As I wrote in my initial request I'm aware of the virtual memory thingy... (I even quoted that)
The thing is that the same cvs command runs fine via shell on the same machine that continuum runs on. 
So I'm pretty sure that this is not [2] a memory issue on the cvs server. 
About 40 people are using the cvs server and memory never was a problem. 


Continuum itself runs on a dedicated machine with 3GB RAM and an intel xeon 
2GHz multicore cpu, this is rather oversized than insufficient hardware.

Is it possible that the local JVM is out of memory? Although, wouldn't I got an 
OutOfMemory Exception if this was the case?
Are you aware of situations when continuum's process  needs a lot of memory?

Is it possible to write the output of the cvs command to the continuum logs?

-Ursprüngliche Nachricht-
Von: Emmanuel Venisse [mailto:[EMAIL PROTECTED] 
Gesendet: Montag, 12. November 2007 15:38

An: [EMAIL PROTECTED]
Betreff: Re: AW: AW: CVS checkout fails: Terminated with fatal signal 11

After a quick search on Google [1], it seems you're not alone with this cvs 
error. But it's sure it isn't a Continuum error but a cvs issue, maybe a memory 
issue on the cvs server [2]

[1] http://www.google.fr/search?q=Terminated+with+fatal+signal+11
[2] http://osdir.com/ml/version-control.cvs.bugs/2002-04/msg00298.html

Emmanuel

[EMAIL PROTECTED] a écrit :
Oh sorry, I think I missunderstood you. Thought you asked how often it 
is used to run ;-) One checkout takes about 3 minutes.


-Ursprüngliche Nachricht-
Von: Emmanuel Venisse [mailto:[EMAIL PROTECTED]
Gesendet: Montag, 12. November 2007 13:28
An: [EMAIL PROTECTED]
Betreff: Re: AW: CVS checkout fails: Terminated with fatal signal 11

How many minutes are used to do a clean checkout?

Emmanuel

[EMAIL PROTECTED] a écrit :

Can you give me a hint what to do to fix this problem?
Or do you need more information?

-Ursprüngliche Nachricht-
Von: Renz, Manuel
Gesendet: Donnerstag, 8. November 2007 12:59
An: [EMAIL PROTECTED]
Betreff: AW: CVS checkout fails: Terminated with fatal signal 11

It is planned to do a clean checkout once a day

-Ursprüngliche Nachricht-
Von: Emmanuel Venisse [mailto:[EMAIL PROTECTED]
Gesendet: Mittwoch, 7. November 2007 20:19
An: [EMAIL PROTECTED]
Betreff: Re: CVS checkout fails: Terminated with fatal signal 11

It seems a kill was sent to the cvs client.

How many time is used to do a checkout?

Emmanuel

[EMAIL PROTECTED] a écrit :
 
Hi togehter,
I'm using continuum 1.1 beta 3 at the moment. I have set up 2 
Projects which work just fine.


Now I'm in the process of setting up my third project, which is 
quite large (clean checked out working copy round about ~260MB ), 
with which I have massive problems.



Here is a short snippet from the logs:

[SocketListener0-1] INFO  FreemarkerManager  - Instantiating
Freemarker ConfigManager!,
com.opensymphony.webwork.views.freemarker.FreemarkerManager
[SocketListener0-0] INFO  Continuum:default  - Enqueuing 'My-App'
(Build definition id=6).
[pool-1-thread-1] INFO  BuildController:default  - Initializing 
build [pool-1-thread-1] INFO  BuildController:default  - Starting 
build of My-App [pool-1-thread-1] INFO  BuildController:default  - 
Updating working dir [pool-1-thread-1] INFO  BuildController:default  
- Performing action check-working-directory [pool-1-thread-1] INFO 
BuildController:default  - Performing action checkout-project

[pool-1-thread-1] INFO  ContinuumScm:default - Checking out project:
'My-App', id: '63' to 'D:\buildserver\data\working\63'.
[pool-1-thread-1] INFO  ScmManager:default   - Executing: cmd.exe /X
/C 'cvs -z3 -f -d :pserver:[EMAIL PROTECTED]:/CVS -q checkout -d 63 my-app'
[pool-1-thread-1] INFO  ScmManager:default   - Working directory:
D:\buildserver\data\working
...
[pool-1-thread-1] WARN  ContinuumScm:default - Error while checking
out the code for project: 'My-App', id: '63' to 
'D:\buildserver\data\working\63'.

[pool-1-thread-1] WARN  ContinuumScm:default - Command output:
Terminated with fatal signal 11
[pool-1-thread-1] WARN  ContinuumScm:default - Provider message: The
cvs command failed.
[pool-1-thread-1] INFO  BuildController:default  - Merging SCM 
results [pool-1-thread-1] INFO  BuildController:default  - Error 
updating from SCM, not building [pool-1-thread-1] INFO 
BuildController:default  - Initializing build [pool-1-thread-1] INFO 
BuildController:default  - Starting build of parent 
[pool-1-thread-1] INFO BuildController:default  - Updating working 
dir [pool-1-thread-1] INFO BuildController:default  - Performing 
action check-working-directory

Re: Access is denied when running mvn release:prepare (Help!)

2007-11-12 Thread Heinrich Nirschl

On Nov 12, 2007 3:19 AM, dev dev [EMAIL PROTECTED] wrote:
 Here is the stack trace.

 [INFO] Executing goals 'clean verify'...
 [INFO] Executing: mvn clean verify --no-plugin-updates -P devel
 Access is denied.


If this happened on Windows, there was maybe a file in the target
directory opened by another application. In this case, the clean
plugin cannot remove the target directory.

- Henry

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Artifact names truncated in HTML after deployment

2007-11-12 Thread ecc11718


My searches in this forum and Apache web server were fruitless, apologies if
this has been answered.

Projects I'm working on have very long artifact names
[(groupId:artifactId:version) + type (-sources, -javadoc, etc)]. The
generated HTML from a deploy shows that the names of the artifacts are being
truncated with ellipses. Couple this with uniqueVersion set to true and
you can imagine the browsing difficulties. What is generating the HTML?

Is this a maven-deploy-plugin issue?

Web server configuration (using Apache 2.0.x)?

What can be done to guarantee names aren't truncated?
-- 
View this message in context: 
http://www.nabble.com/Artifact-names-truncated-in-HTML-after-deployment-tf4792329s177.html#a13709289
Sent from the Maven - Users mailing list archive at Nabble.com.


-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Re: from 1.1-beta3 to beta4


It's a known issue (CONTINUUM-1540).

the workaround is to remove the continuum database and add ;create=true on 
the jdbc url you use for the import

Emmanuel

Francis De Brabandere a écrit :

Hi all,

I'm trying to upgrade from beta3 to beta4 but no succes.
I have been following the migration page instructions:
http://maven.apache.org/continuum/documentation/1_1/installation/upgrade.html

but I get the following exception when importing the builds database:

9660 [main] INFO JPOX.RDBMS  - Validated 2 foreign key(s) for table
PROJECTGROUP_NOTIFIERS
9698 [main] INFO JPOX.RDBMS  - Validated 3 index(es) for table
PROJECTGROUP_NOTIFIERS
Exception in thread main javax.jdo.JDOUserException: One or more
instances could not be deleted
at 
org.jpox.AbstractPersistenceManager.deletePersistentAll(AbstractPersistenceManager.java:1525)
at 
org.jpox.store.rdbms.query.JDOQLQuery.performDeletePersistentAll(JDOQLQuery.java:687)
at org.jpox.store.query.Query.deletePersistentAll(Query.java:1120)
at org.jpox.store.query.Query.deletePersistentAll(Query.java:1075)
at org.jpox.store.query.Query.deletePersistentAll(Query.java:1049)
at 
org.codehaus.plexus.jdo.PlexusJdoUtils.removeAll(PlexusJdoUtils.java:444)
at 
org.apache.maven.continuum.store.JdoContinuumStore.eraseDatabase(JdoContinuumStore.java:1961)
at 
org.apache.maven.continuum.management.JdoDataManagementTool.eraseDatabase(JdoDataManagementTool.java:127)
at 
org.apache.maven.continuum.management.DataManagementCli.processDatabase(DataManagementCli.java:254)
at 
org.apache.maven.continuum.management.DataManagementCli.main(DataManagementCli.java:158)
NestedThrowablesStackTrace:
javax.jdo.JDOUserException: One or more instances could not be deleted
at 
org.jpox.AbstractPersistenceManager.deletePersistentAll(AbstractPersistenceManager.java:1525)
at 
org.jpox.store.rdbms.scostore.ElementContainerStore.clear(ElementContainerStore.java:659)
at 
org.jpox.store.mapping.CollectionMapping.deleteDependent(CollectionMapping.java:334)
at 
org.jpox.store.rdbms.table.ClassTable.deleteDependent(ClassTable.java:2651)
at org.jpox.store.StoreManager.deleteDependent(StoreManager.java:1019)
at 
org.jpox.state.StateManagerImpl.deletePersistent(StateManagerImpl.java:4332)
at 
org.jpox.AbstractPersistenceManager.internalDeletePersistent(AbstractPersistenceManager.java:1473)
at 
org.jpox.AbstractPersistenceManager.deletePersistentAll(AbstractPersistenceManager.java:1516)
at 
org.jpox.store.rdbms.query.JDOQLQuery.performDeletePersistentAll(JDOQLQuery.java:687)
at org.jpox.store.query.Query.deletePersistentAll(Query.java:1120)
at org.jpox.store.query.Query.deletePersistentAll(Query.java:1075)
at org.jpox.store.query.Query.deletePersistentAll(Query.java:1049)
at 
org.codehaus.plexus.jdo.PlexusJdoUtils.removeAll(PlexusJdoUtils.java:444)
at 
org.apache.maven.continuum.store.JdoContinuumStore.eraseDatabase(JdoContinuumStore.java:1961)
at 
org.apache.maven.continuum.management.JdoDataManagementTool.eraseDatabase(JdoDataManagementTool.java:127)
at 
org.apache.maven.continuum.management.DataManagementCli.processDatabase(DataManagementCli.java:254)
at 
org.apache.maven.continuum.management.DataManagementCli.main(DataManagementCli.java:158)
NestedThrowablesStackTrace:
javax.jdo.JDODataStoreException: Delete request failed: DELETE FROM
BUILDDEFINITION WHERE ID = ?
at 
org.jpox.store.rdbms.request.DeleteRequest.execute(DeleteRequest.java:288)
at org.jpox.store.rdbms.table.ClassTable.delete(ClassTable.java:2611)
at org.jpox.store.StoreManager.delete(StoreManager.java:981)
at 
org.jpox.state.StateManagerImpl.internalDeletePersistent(StateManagerImpl.java:4395)
at 
org.jpox.state.StateManagerImpl.deletePersistent(StateManagerImpl.java:4350)
at 
org.jpox.AbstractPersistenceManager.internalDeletePersistent(AbstractPersistenceManager.java:1473)
at 
org.jpox.AbstractPersistenceManager.deletePersistentAll(AbstractPersistenceManager.java:1516)
at 
org.jpox.store.rdbms.scostore.ElementContainerStore.clear(ElementContainerStore.java:659)
at 
org.jpox.store.mapping.CollectionMapping.deleteDependent(CollectionMapping.java:334)
at 
org.jpox.store.rdbms.table.ClassTable.deleteDependent(ClassTable.java:2651)
at org.jpox.store.StoreManager.deleteDependent(StoreManager.java:1019)
at 
org.jpox.state.StateManagerImpl.deletePersistent(StateManagerImpl.java:4332)
at 
org.jpox.AbstractPersistenceManager.internalDeletePersistent(AbstractPersistenceManager.java:1473)
at 
org.jpox.AbstractPersistenceManager.deletePersistentAll(AbstractPersistenceManager.java:1516)
at 
org.jpox.store.rdbms.query.JDOQLQuery.performDeletePersistentAll(JDOQLQuery.java:687)
at

Re: from 1.1-beta3 to beta4

Thanks!

On Nov 12, 2007 6:00 PM, Emmanuel Venisse [EMAIL PROTECTED] wrote:
 It's a known issue (CONTINUUM-1540).

 the workaround is to remove the continuum database and add ;create=true on 
 the jdbc url you use for the import

 Emmanuel

 Francis De Brabandere a écrit :

  Hi all,
 
  I'm trying to upgrade from beta3 to beta4 but no succes.
  I have been following the migration page instructions:
  http://maven.apache.org/continuum/documentation/1_1/installation/upgrade.html
 
  but I get the following exception when importing the builds database:
 
  9660 [main] INFO JPOX.RDBMS  - Validated 2 foreign key(s) for table
  PROJECTGROUP_NOTIFIERS
  9698 [main] INFO JPOX.RDBMS  - Validated 3 index(es) for table
  PROJECTGROUP_NOTIFIERS
  Exception in thread main javax.jdo.JDOUserException: One or more
  instances could not be deleted
  at 
  org.jpox.AbstractPersistenceManager.deletePersistentAll(AbstractPersistenceManager.java:1525)
  at 
  org.jpox.store.rdbms.query.JDOQLQuery.performDeletePersistentAll(JDOQLQuery.java:687)
  at org.jpox.store.query.Query.deletePersistentAll(Query.java:1120)
  at org.jpox.store.query.Query.deletePersistentAll(Query.java:1075)
  at org.jpox.store.query.Query.deletePersistentAll(Query.java:1049)
  at 
  org.codehaus.plexus.jdo.PlexusJdoUtils.removeAll(PlexusJdoUtils.java:444)
  at 
  org.apache.maven.continuum.store.JdoContinuumStore.eraseDatabase(JdoContinuumStore.java:1961)
  at 
  org.apache.maven.continuum.management.JdoDataManagementTool.eraseDatabase(JdoDataManagementTool.java:127)
  at 
  org.apache.maven.continuum.management.DataManagementCli.processDatabase(DataManagementCli.java:254)
  at 
  org.apache.maven.continuum.management.DataManagementCli.main(DataManagementCli.java:158)
  NestedThrowablesStackTrace:
  javax.jdo.JDOUserException: One or more instances could not be deleted
  at 
  org.jpox.AbstractPersistenceManager.deletePersistentAll(AbstractPersistenceManager.java:1525)
  at 
  org.jpox.store.rdbms.scostore.ElementContainerStore.clear(ElementContainerStore.java:659)
  at 
  org.jpox.store.mapping.CollectionMapping.deleteDependent(CollectionMapping.java:334)
  at 
  org.jpox.store.rdbms.table.ClassTable.deleteDependent(ClassTable.java:2651)
  at 
  org.jpox.store.StoreManager.deleteDependent(StoreManager.java:1019)
  at 
  org.jpox.state.StateManagerImpl.deletePersistent(StateManagerImpl.java:4332)
  at 
  org.jpox.AbstractPersistenceManager.internalDeletePersistent(AbstractPersistenceManager.java:1473)
  at 
  org.jpox.AbstractPersistenceManager.deletePersistentAll(AbstractPersistenceManager.java:1516)
  at 
  org.jpox.store.rdbms.query.JDOQLQuery.performDeletePersistentAll(JDOQLQuery.java:687)
  at org.jpox.store.query.Query.deletePersistentAll(Query.java:1120)
  at org.jpox.store.query.Query.deletePersistentAll(Query.java:1075)
  at org.jpox.store.query.Query.deletePersistentAll(Query.java:1049)
  at 
  org.codehaus.plexus.jdo.PlexusJdoUtils.removeAll(PlexusJdoUtils.java:444)
  at 
  org.apache.maven.continuum.store.JdoContinuumStore.eraseDatabase(JdoContinuumStore.java:1961)
  at 
  org.apache.maven.continuum.management.JdoDataManagementTool.eraseDatabase(JdoDataManagementTool.java:127)
  at 
  org.apache.maven.continuum.management.DataManagementCli.processDatabase(DataManagementCli.java:254)
  at 
  org.apache.maven.continuum.management.DataManagementCli.main(DataManagementCli.java:158)
  NestedThrowablesStackTrace:
  javax.jdo.JDODataStoreException: Delete request failed: DELETE FROM
  BUILDDEFINITION WHERE ID = ?
  at 
  org.jpox.store.rdbms.request.DeleteRequest.execute(DeleteRequest.java:288)
  at 
  org.jpox.store.rdbms.table.ClassTable.delete(ClassTable.java:2611)
  at org.jpox.store.StoreManager.delete(StoreManager.java:981)
  at 
  org.jpox.state.StateManagerImpl.internalDeletePersistent(StateManagerImpl.java:4395)
  at 
  org.jpox.state.StateManagerImpl.deletePersistent(StateManagerImpl.java:4350)
  at 
  org.jpox.AbstractPersistenceManager.internalDeletePersistent(AbstractPersistenceManager.java:1473)
  at 
  org.jpox.AbstractPersistenceManager.deletePersistentAll(AbstractPersistenceManager.java:1516)
  at 
  org.jpox.store.rdbms.scostore.ElementContainerStore.clear(ElementContainerStore.java:659)
  at 
  org.jpox.store.mapping.CollectionMapping.deleteDependent(CollectionMapping.java:334)
  at 
  org.jpox.store.rdbms.table.ClassTable.deleteDependent(ClassTable.java:2651)
  at 
  org.jpox.store.StoreManager.deleteDependent(StoreManager.java:1019)
  at 
  org.jpox.state.StateManagerImpl.deletePersistent(StateManagerImpl.java:4332)
  at

Re: maven constantly rebuilding everything

Sounds reasonable, but how would you do #2? One way I was thinking was
grab latest jar from repo, compare contents to just built jar (hash
both and compare?). Or did you have another plan?

Wayne

On 11/12/07, EJ Ciramella [EMAIL PROTECTED] wrote:
 So the fix would be something like:

 1 - Make process-resources look at the source of the activated profiles and
 make sure they are newer than the templates and then (and only then) process
 the templates.
 2 - Make the install plugin test what's in the repository is out of date
 with the just built jar before installing.
 3 - Make the war plugin able to ignore/exclude maven generated files (such
 as pom.properties) - this may/may not solve our war rebuilding issues.

 ???


-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Re: Running a Test Conditionally -- Including/Excluding Tests with Profiles

Without looking at the code I can't be sure, but I would assume that
you can't include -and- exclude the same file. Probably the exclude
list is processed last' in the code which means it won't run.

Wayne

On 11/11/07, Dan Kigelman [EMAIL PROTECTED] wrote:
 Thank you Wayne,

 I will be careful about that.  It seems it was working on my linux laptop,
 but I'm sure you've averted some of my future headaches for executing this
 on all kinds of shells.  I renamed the profile and activation to the
 following:

 profile
   idconditionProfile/id
   activation
 property
   namesome_condition/name
   valuetrue/value
 /property
   /activation
   build

 I execute now with mvn test -Dsome_condition=true but still with the same
 result:

 ---
  T E S T S
 ---
 Running sample.test.ExecuteAlwaysTest
 Tests run: 1, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 0.04 sec

 Results :

 Tests run: 1, Failures: 0, Errors: 0, Skipped: 0

 [INFO] [surefire:test {execution: rareTest}]
 [INFO] Surefire report directory:
 /home/danik/workspace/test2/execute-test-conditionally/target/surefire-reports

 ---
  T E S T S
 ---
 There are no tests to run.

 Results :

 Tests run: 0, Failures: 0, Errors: 0, Skipped: 0

 [INFO]
 
 [INFO] BUILD SUCCESSFUL
 [INFO]
 

 Somehow the second surefire execution includes no tests...

 -- Dan


 On Nov 11, 2007 5:41 PM, Wayne Fay [EMAIL PROTECTED] wrote:

  Try renaming the profile to some_condition, and then using
  -Dsome_condition=true. I've never tried to use a property named a-b-c
  and I'm not entirely sure how different operating systems pass that to
  Java and how it is interpreted prior to handing it to Maven (eg -D, -h
  etc). So I would just avoid -'s in profile names, personally.
 
  Wayne
 
  On 11/11/07, Dan Kigelman [EMAIL PROTECTED] wrote:
   Hello!
  
   I am trying to execute a test if and only if a certain condition is set
   (such as a profile is activated).  The test would fail if it is run on
  some
   systems, so I am trying to disable it by default, and run it only when
  run
   on the relevant system.
  
   I made a sample project to simplify the problem.  Below is the relevant
  part
   my pom:
  
build
  plugins
plugin
  groupIdorg.apache.maven.plugins/groupId
  artifactIdmaven-surefire-plugin/artifactId
  configuration
excludes
  exclude**/ExecuteSometimesTest.java/exclude
/excludes
  /configuration
/plugin
  /plugins
/build
  
profiles
  profile
idcondition-profile/id
activation
  property
namesome-condition/name
  /property
/activation
build
  plugins
plugin
  groupIdorg.apache.maven.plugins/groupId
  artifactIdmaven-surefire-plugin/artifactId
  executions
execution
  idrare-test/id
  goals
goaltest/goal
  /goals
  phasetest/phase
  configuration
includes
  include**/ExecuteSometimesTest.java/include
/includes
excludes /
  /configuration
/execution
  /executions
/plugin
  /plugins
/build
  /profile
/profiles
  
   When I execute mvn test it works as expected:
   ---
T E S T S
   ---
   Running sample.test.ExecuteAlwaysTest
   Tests run: 1, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 0.085sec
  
   Results :
  
   Tests run: 1, Failures: 0, Errors: 0, Skipped: 0
  
   [INFO]
   
   [INFO] BUILD SUCCESSFUL
   [INFO]
   
  
  
   However, when I execute mvn test -Dsome-condition the
   ExecuteSometimesTest.java is not executed:
  
   ---
T E S T S
   ---
   Running sample.test.ExecuteAlwaysTest
   Tests run: 1, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 0.04 sec
  
   Results :
  
   Tests run: 1, Failures: 0, Errors: 0, Skipped: 0
  
   [INFO] [surefire:test {execution: rare-test}]
   [INFO] Surefire report directory:
  
 
 /home/danik/workspace/test2/execute-test-conditionally/target/surefire-reports

Re: Compiler fails to include generated sources by axistools-maven-plugin

You may need to bind the execution of the plugin to one of Maven's
standard phases. I'm not using axistools so I'm not entirely sure --
many plugins are bound to phases in the plugin code directly. Perhaps
someone else using axistools (successfully) will reply.

Wayne

On 11/11/07, urir [EMAIL PROTECTED] wrote:

 plugin
   groupIdorg.codehaus.mojo/groupId
   artifactIdaxistools-maven-plugin/artifactId
   version1.1/version
   configuration
   
 sourceDirectorysrc/main/resources/META-INF/wsdl/sourceDirectory
   /configuration
   executions
   execution
   goals
   goalwsdl2java/goal
   /goals
   /execution
   /executions
   /plugin


 Wayne Fay wrote:
 
  Show us the section of your pom where you are declaring the
  axistools-m-p execution.
 
  Wayne
 
  On 11/11/07, urir [EMAIL PROTECTED] wrote:
 
  Hi,
 
  I am using the axistools-maven-plugin to generate sources from wsdl.
  The sources are generated into:
  target\generated-sources\axistools\wsdl2java
 
  but in compile stage, those sources are not included if I run 'mvn clean
  install'.
 
  If I run 'mvn clean install' and then 'mvn install' - then the files are
  included.
 
  I am working over daytrader example application from Better Builds with
  Maven - July 2007.
 
  Regards,
  Uri Ravzin.
  --
  View this message in context:
 
 http://www.nabble.com/Compiler-fails-to-include-generated-sources-by-axistools-maven-plugin-tf4785031s177.html#a13689179
  Sent from the Maven - Users mailing list archive at Nabble.com.
 
 
  -
  To unsubscribe, e-mail: [EMAIL PROTECTED]
  For additional commands, e-mail: [EMAIL PROTECTED]
 
 
 
  -
  To unsubscribe, e-mail: [EMAIL PROTECTED]
  For additional commands, e-mail: [EMAIL PROTECTED]
 
 
 

 --
 View this message in context:
 http://www.nabble.com/Compiler-fails-to-include-generated-sources-by-axistools-maven-plugin-tf4785031s177.html#a13699871
 Sent from the Maven - Users mailing list archive at Nabble.com.


 -
 To unsubscribe, e-mail: [EMAIL PROTECTED]
 For additional commands, e-mail: [EMAIL PROTECTED]



-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Re: Cleaning up results

2007-11-12 Thread Alex Mayorga Adame



Emmanuel Venisse wrote:
 
 
 Alex Mayorga Adame a écrit :
 Would this guide
 http://maven.apache.org/continuum/documentation/1_1/installation/upgrade.html
 work if I'm coming from 1.0.3?
 
 No, it is for 1.1 only.
 
 Emmanuel
 

Is there a way from 1.0.3 to the latest beta?
-- 
View this message in context: 
http://www.nabble.com/Cleaning-up-results-tf1906827.html#a13709788
Sent from the Continuum - Users mailing list archive at Nabble.com.

RE: maven constantly rebuilding everything

2007-11-12 Thread EJ Ciramella

Here's the pseudo code:

1 - Don't have to build new jar because everything is uptodate?  Don't
install - maybe this means install takes a new param true/false style on
whether it needs to be installed or not.
2 - If local jar has newer timestamp - install.

Install only cares about local repository, right?  It's deploy that
looks across to the internal repository - and the only machine to do
that is the build machine.

About the process-resources solution - is that kind of info available?
Can I track back via the profiles on the commandline to the files in
question (where the profiles came from)?

Am I missing anything?

-Original Message-
From: Wayne Fay [mailto:[EMAIL PROTECTED] 
Sent: Monday, November 12, 2007 12:06 PM
To: Maven Users List
Subject: Re: maven constantly rebuilding everything

Sounds reasonable, but how would you do #2? One way I was thinking was
grab latest jar from repo, compare contents to just built jar (hash
both and compare?). Or did you have another plan?

Wayne

On 11/12/07, EJ Ciramella [EMAIL PROTECTED] wrote:
 So the fix would be something like:

 1 - Make process-resources look at the source of the activated
profiles and
 make sure they are newer than the templates and then (and only then)
process
 the templates.
 2 - Make the install plugin test what's in the repository is out of
date
 with the just built jar before installing.
 3 - Make the war plugin able to ignore/exclude maven generated files
(such
 as pom.properties) - this may/may not solve our war rebuilding issues.

 ???


-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]


-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Re: How to create a Mojo that controls an ANT task

2007-11-12 Thread Saloucious


Just an idea

May be you can fork maven-antrun-plugin, and instead of let plexus bind
tasks attribute, instanciate tasks attribute with an xml file embded in you
plugin artifact

I was just having a look in AntRunMojo class but i think you have to use
AntTargetConverter 

This is just assumptions ...

Please keep me in touch


Pedro Viegas-2 wrote:
 
 Hi there guys,
 
 I am trying to create a Maven plugin that executes a given ANT task inside
 it.
 My real case is to call the Hibernate Tools task to generate a bunch of
 ORM
 classes and Hibernate mappings.
 But I need to make some decisions and have to pass some parameters to the
 several Hibernate Tools tasks I need to call depending on some conditions
 that I need to analize in a Java Mojo.
 
 Basically I need to:
 
1. Create a Mojo with some parameters and do some processing and file
generation of my own
2. Call the ANT Hibernate Tools tasks a bunch of times to generate all
my classes/mapings on several packages for several database schemas
3. Try to keep it as simple as possible to the end programmer user,
simply call a Maven plugin and state some configuration files and
destination package and let the conventioned behavior take charge.
 
 I have created the first part easily by creating a Java Maven Plugin.
 The second part I have configured like a maven-ant-plugin addition to the
 POM and got it working, but this has to be copied into each of the
 projects
 that need this goal and all the configuration I could pragmatically do
 have
 to be typed repeatedly.
 So 1 and 2 are go, 3 is what I can't find any documentation or sample to
 help me.
 
 I have seen the example from:
 http://maven.apache.org/guides/plugin/guide-ant-plugin-development.html
 
 But this is how to build a maven plugin with ANT. What I need is to build
 a
 maven plugin in Java that can call an ant task, like antrun plugin does
 but
 adding a few more java actions and decisions before and after.
 
 So, how can I use the already running maven java plugin to call the ANT
 tasks?
 Can I call a plugin inside another plugin? How?
 Should I try to extend the maven-antrun-plugin to try to accomplish this?
 
 Thanks in advance,
 
 -- 
 Pedro Viegas
 
 
 Walking on water and developing software
 from a specification are easy if both are
 frozen.
 - Edward V. Berard
 
 

-- 
View this message in context: 
http://www.nabble.com/How-to-create-a-Mojo-that-controls-an-ANT-task-tf4788523s177.html#a13710554
Sent from the Maven - Users mailing list archive at Nabble.com.


-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Re: How to create a Mojo that controls an ANT task

2007-11-12 Thread Pedro Viegas

Hi there,

Thanks a bunch for your responses.
You pointed me in the right direction. I was wishing I could use the antrun
plugin resources but without having to copy-paste or in other way break
future updates of it.
I'm drilling into the plugin's source right now and will post my result as
soon as I have new information.

This plugin we're working on at my company will be open-sourced so if there
is any interest I can post here the link to the project when we finish it.
We have yet to crerate a Maven repository in our server, but even it that is
not available the Maven Generated site will have all needed information
including the source and package, until we publish the repository.

Regards,

On Nov 12, 2007 5:53 PM, Saloucious [EMAIL PROTECTED] wrote:


 Just an idea

 May be you can fork maven-antrun-plugin, and instead of let plexus bind
 tasks attribute, instanciate tasks attribute with an xml file embded in
 you
 plugin artifact

 I was just having a look in AntRunMojo class but i think you have to use
 AntTargetConverter

 This is just assumptions ...

 Please keep me in touch


 Pedro Viegas-2 wrote:
 
  Hi there guys,
 
  I am trying to create a Maven plugin that executes a given ANT task
 inside
  it.
  My real case is to call the Hibernate Tools task to generate a bunch of
  ORM
  classes and Hibernate mappings.
  But I need to make some decisions and have to pass some parameters to
 the
  several Hibernate Tools tasks I need to call depending on some
 conditions
  that I need to analize in a Java Mojo.
 
  Basically I need to:
 
 1. Create a Mojo with some parameters and do some processing and file
 generation of my own
 2. Call the ANT Hibernate Tools tasks a bunch of times to generate
 all
 my classes/mapings on several packages for several database schemas
 3. Try to keep it as simple as possible to the end programmer user,
 simply call a Maven plugin and state some configuration files and
 destination package and let the conventioned behavior take charge.
 
  I have created the first part easily by creating a Java Maven Plugin.
  The second part I have configured like a maven-ant-plugin addition to
 the
  POM and got it working, but this has to be copied into each of the
  projects
  that need this goal and all the configuration I could pragmatically do
  have
  to be typed repeatedly.
  So 1 and 2 are go, 3 is what I can't find any documentation or sample to
  help me.
 
  I have seen the example from:
  http://maven.apache.org/guides/plugin/guide-ant-plugin-development.html
 
  But this is how to build a maven plugin with ANT. What I need is to
 build
  a
  maven plugin in Java that can call an ant task, like antrun plugin does
  but
  adding a few more java actions and decisions before and after.
 
  So, how can I use the already running maven java plugin to call the ANT
  tasks?
  Can I call a plugin inside another plugin? How?
  Should I try to extend the maven-antrun-plugin to try to accomplish
 this?
 
  Thanks in advance,
 
  --
  Pedro Viegas
 
  
  Walking on water and developing software
  from a specification are easy if both are
  frozen.
  - Edward V. Berard
 
 

 --
 View this message in context:
 http://www.nabble.com/How-to-create-a-Mojo-that-controls-an-ANT-task-tf4788523s177.html#a13710554
 Sent from the Maven - Users mailing list archive at Nabble.com.


 -
 To unsubscribe, e-mail: [EMAIL PROTECTED]
 For additional commands, e-mail: [EMAIL PROTECTED]




-- 
Pedro Viegas


Walking on water and developing software
from a specification are easy if both are
frozen.
- Edward V. Berard

scm:bootstrap checks out twice

2007-11-12 Thread Ilyevsky, Leonid (Equity Trading)

I am using the maven-scm-plugin, the scm:bootstrap goal.
Everything works OK, except that it does the checkout from svn twice: it
cleans the .../target/checkout directory, performs the checkout, and
then cleans again and checks out again. Only after that it does the
specified goals.
This is annoying because the checkout may take essential time.

What is the problem? Is anybody else experiencing the same ?


This message w/attachments (message) may be privileged, confidential or 
proprietary, and if you are not an intended recipient, please notify the 
sender, do not use or share it and delete it. Unless specifically indicated, 
this message is not an offer to sell or a solicitation of any investment 
products or other financial product or service, an official confirmation of any 
transaction, or an official statement of Merrill Lynch. Subject to applicable 
law, Merrill Lynch may monitor, review and retain e-communications (EC) 
traveling through its networks/systems. The laws of the country of each 
sender/recipient may impact the handling of EC, and EC may be archived, 
supervised and produced in countries other than the country in which you are 
located. This message cannot be guaranteed to be secure or error-free. This 
message is subject to terms available at the following link: 
http://www.ml.com/e-communications_terms/. By messaging with Merrill Lynch you 
consent to the foregoing.

Re: Accessing java.class.path inside a test execution

2007-11-12 Thread Rémy Sanlaville

Hi,

This a common request and unfortunately the ${maven.test.classpath} variable
is not valuated in the surefire configuration plugin.

For the moment, I found a workaround but it is unsatisfactory

 build
plugins
plugin
artifactIdmaven-antrun-plugin/artifactId
executions
  execution
phasegenerate-test-resources/phase
configuration
  tasks
   property
   refid=maven.test.classpath name=
maven.test.classpath
   /property
 mkdir dir=${project.build.directory}/test-classes/
 echo
file=${project.build.directory}/test-classes/classpath.txt
append=false${maven.test.classpath}/echo
/tasks
/configuration
goals
  goalrun/goal
/goals
  /execution
/executions
/plugin
/plugins
  /build

Then, in your testcase, you just have to read the generated classpath file.

HTH,

Rémy

Re: maven constantly rebuilding everything

2007-11-12 Thread Lee Meador

When a child pom indicates there is a parent pom, the parent pom is
always processed. It doesn't depend on where in the hierarchy the
child pom is sitting. Notice that the child pom will have a parent
section that completely identifies the parent OR defaults to the
parent being in the parent folder.

Notice that with all the things you can inject into the generated
artifacts, it is very difficult to tell if the current generated
artifact is going to be different than the last version. The problem
is that the output file doesn't just depend on the input files. It
depends on various properties and such, as well. (Think build number.)
It probably isn't any faster to build the new version (say, in memory)
and then read the old version and compare than to just build the new
version and store it. (You do have the down side of having a new file
date/time on the output.)

Once you change the one output file, then all the downstream
dependencies have to be fired off as well. (That may be a reason to
try to keep from changing the date/time on what was the output file
but has become the input for the next phase.) But, if that didn't
happen, you would still have to build the next phase because you don't
know if it depends on some changed property. Sorting out which
properties can change and NOT cause the next phase to be built is a
big problem too. You would have to say stuff like, When the date/time
changes, you don't do a build. But when the deployment environment
changes you do need a build. And, you can define properties on the
command line and they can come from the system environment. Do you
store all the properties values on each build and then compare on the
next build to see if anything changed?

As you can see, there might be some things that could be done to make
less stuff get built. But there may not ever be a way to make it work
100%. This may take a lot of thought and testing and defining to even
know.

-- Lee

On Nov 9, 2007 7:57 AM, EJ Ciramella [EMAIL PROTECTED] wrote:
 I'd love to know what people are building and the size of things, I think our 
 project is larger than most (I'm not stepping into a pissing contest, simply 
 stating the facts).  We're working on a massive refactoring both in codeline 
 management and actual code refactoring.

 Even with the refactoring done, that will only help to a certain degree.  
 With testing, we're talking about 1 - 1.5 hours.
 The resulting assembly for our main application is 195 - 200 mb (full of 
 jsps, images, classes, jars, etc).  I wish I could better show you the size 
 of the application (we have three ears and 4 - 5 wars depending on what 
 you're building).  There's no real honest to goodness code in the jsps (as it 
 should be), all business logic is handled else where.  As a release engineer, 
 it's not my business to make sweeping changes like that.  I can make 
 suggestions, but in the end, it's up to development to decide what projects 
 have merit/profit/etc associated with them.

 It'd be nice if you could work at the lowest directory all the time (the one 
 responsible for building the ear file), but when your change effects a module 
 that the ear you want depends on (or MULTIPLE subprojects), most developers 
 here build from the top.  Also, when profiles.xml lives at the root of the 
 overall project and contains 90% the configuration for a developers box, 
 sitting at a lower level, aren't these properties unavailable (doesn't maven 
 only look in the working directory for profiles.xml)?

 If maven wouldn't try to rebuild every jar and then every war and then every 
 ear, life would be much better.

 -Original Message-
 From: Wayne Fay [mailto:[EMAIL PROTECTED]

 Sent: Friday, November 09, 2007 8:10 AM
 To: Maven Users List
 Subject: Re: maven constantly rebuilding everything

 Sounds like your projects are entirely too big -- break 'em up a bit.
 Even 10min builds sounds rather long -- is this compilation only or
 including testing? And what portion of that time is consumed by your
 tests -- at this level (compilation), you should be running very fast
 unit tests, generally not integration tests.

 Artifacts/modules are cheap in Maven. War modules in particular
 should not have a lot of code associated with them -- move the code
 into Jars and extract the code out of your JSPs etc into classes that
 go into those Jars.

 Also when you're working, you don't clean install from the top
 parent, but instead run it from the module you're working on. At
 least, this is what we do...

 Wayne

 On 11/9/07, EJ Ciramella [EMAIL PROTECTED] wrote:
  That's incredible, doing that takes 30 - 40 minutes here.
 
  Most developers here are used to a 10 build time.
 
  -Original Message-
  From: Wayne Fay [mailto:[EMAIL PROTECTED]
  Sent: Thursday, November 08, 2007 11:08 PM
  To: Maven Users List
  Subject: Re: maven constantly rebuilding everything
 
  I nearly always build clean package or clean install so I expect
  them to rebuild

Re: How to create a Mojo that controls an ANT task

2007-11-12 Thread Saloucious


If you have multi-projects and a parent, a easier solution, should to add
your maven-antrun-plugin to pluginsMagement/ tag in parent pom.xml.

As far as I know you can refer in your children pom to your 
maven-antrun-plugin without redefine tasks.





Pedro Viegas-2 wrote:
 
 Hi there,
 
 Thanks a bunch for your responses.
 You pointed me in the right direction. I was wishing I could use the
 antrun
 plugin resources but without having to copy-paste or in other way break
 future updates of it.
 I'm drilling into the plugin's source right now and will post my result as
 soon as I have new information.
 
 This plugin we're working on at my company will be open-sourced so if
 there
 is any interest I can post here the link to the project when we finish it.
 We have yet to crerate a Maven repository in our server, but even it that
 is
 not available the Maven Generated site will have all needed information
 including the source and package, until we publish the repository.
 
 Regards,
 
 On Nov 12, 2007 5:53 PM, Saloucious [EMAIL PROTECTED] wrote:
 

 Just an idea

 May be you can fork maven-antrun-plugin, and instead of let plexus bind
 tasks attribute, instanciate tasks attribute with an xml file embded in
 you
 plugin artifact

 I was just having a look in AntRunMojo class but i think you have to use
 AntTargetConverter

 This is just assumptions ...

 Please keep me in touch


 Pedro Viegas-2 wrote:
 
  Hi there guys,
 
  I am trying to create a Maven plugin that executes a given ANT task
 inside
  it.
  My real case is to call the Hibernate Tools task to generate a bunch of
  ORM
  classes and Hibernate mappings.
  But I need to make some decisions and have to pass some parameters to
 the
  several Hibernate Tools tasks I need to call depending on some
 conditions
  that I need to analize in a Java Mojo.
 
  Basically I need to:
 
 1. Create a Mojo with some parameters and do some processing and
 file
 generation of my own
 2. Call the ANT Hibernate Tools tasks a bunch of times to generate
 all
 my classes/mapings on several packages for several database schemas
 3. Try to keep it as simple as possible to the end programmer user,
 simply call a Maven plugin and state some configuration files and
 destination package and let the conventioned behavior take charge.
 
  I have created the first part easily by creating a Java Maven Plugin.
  The second part I have configured like a maven-ant-plugin addition to
 the
  POM and got it working, but this has to be copied into each of the
  projects
  that need this goal and all the configuration I could pragmatically do
  have
  to be typed repeatedly.
  So 1 and 2 are go, 3 is what I can't find any documentation or sample
 to
  help me.
 
  I have seen the example from:
  http://maven.apache.org/guides/plugin/guide-ant-plugin-development.html
 
  But this is how to build a maven plugin with ANT. What I need is to
 build
  a
  maven plugin in Java that can call an ant task, like antrun plugin does
  but
  adding a few more java actions and decisions before and after.
 
  So, how can I use the already running maven java plugin to call the ANT
  tasks?
  Can I call a plugin inside another plugin? How?
  Should I try to extend the maven-antrun-plugin to try to accomplish
 this?
 
  Thanks in advance,
 
  --
  Pedro Viegas
 
  
  Walking on water and developing software
  from a specification are easy if both are
  frozen.
  - Edward V. Berard
 
 

 --
 View this message in context:
 http://www.nabble.com/How-to-create-a-Mojo-that-controls-an-ANT-task-tf4788523s177.html#a13710554
 Sent from the Maven - Users mailing list archive at Nabble.com.


 -
 To unsubscribe, e-mail: [EMAIL PROTECTED]
 For additional commands, e-mail: [EMAIL PROTECTED]


 
 
 -- 
 Pedro Viegas
 
 
 Walking on water and developing software
 from a specification are easy if both are
 frozen.
 - Edward V. Berard
 
 

-- 
View this message in context: 
http://www.nabble.com/How-to-create-a-Mojo-that-controls-an-ANT-task-tf4788523s177.html#a13712774
Sent from the Maven - Users mailing list archive at Nabble.com.


-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Is there already a Module-Overview-Plugin?

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Hi there,

I just wonder if there is already a plugin for maven2 that can generate a table
with the links, names, and descriptions of the child-modules of a project.
I remember that this was produced by the multi-site in maven1.
I might write such plugin at the mojo sandbox but want to get sure, if
this does NOT already exist.

Does anybody know???

Thanks a lot.
  Jörg
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.5 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFHOK/OmPuec2Dcv/8RAsjCAJ4gnAFtg+ALUXHehc8S+v1MqEFUWACgiXyU
PmDhDAWIuVBopL1/FbDFPtE=
=GiXl
-END PGP SIGNATURE-

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Re: maven offline

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Hi there,

you might also be interested in this one:
http://maven.apache.org/plugins/maven-dependency-plugin/go-offline-mojo.html

However there are still some bugs when using snapshots and beeing offline.

Regards
  J?rg

 mvn -o will automatically use your local repo cache. If you are using
 mvn -o, it will not use any other repos, so the only files available
 to Maven are those already downloaded to your local machine.
 
 Wayne
 
 On 10/15/07, MavenMan [EMAIL PROTECTED] wrote:


 Wayne Fay wrote:
 Try mvn -h and then read the help text which is output. One of the
 printed options is what you're looking for.

 Wayne
 If I select -o ，need I to set local Repositories ?how to set ?
 Best Regards !

 On 10/15/07, MavenMan [EMAIL PROTECTED] wrote:
 I am a beginer of maven .but I think it is a great artifact.  but now I
 can
 not connect to Internet.
 I need a detail step to deel with this problem .By the way ,I use
 m2eclipse
 pulgin to control maven  . who can help me .
 thank you in advance!
 --
 View this message in context:
 http://www.nabble.com/maven-offline-tf4631310s177.html#a13224819
 Sent from the Maven - Users mailing list archive at Nabble.com.


 -
 To unsubscribe, e-mail: [EMAIL PROTECTED]
 For additional commands, e-mail: [EMAIL PROTECTED]


 -
 To unsubscribe, e-mail: [EMAIL PROTECTED]
 For additional commands, e-mail: [EMAIL PROTECTED]



 --
 View this message in context: 
 http://www.nabble.com/maven-offline-tf4631310s177.html#a13225985
 Sent from the Maven - Users mailing list archive at Nabble.com.


 -
 To unsubscribe, e-mail: [EMAIL PROTECTED]
 For additional commands, e-mail: [EMAIL PROTECTED]


 
 -
 To unsubscribe, e-mail: [EMAIL PROTECTED]
 For additional commands, e-mail: [EMAIL PROTECTED]
 
 

-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.5 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFHOLB6mPuec2Dcv/8RAoc5AJ9/cQPOJmpHUYAEU/R4L3IuABdiVACfcDJy
cOEgpanpoaqYNYFtG61nVeQ=
=6L9H
-END PGP SIGNATURE-

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Re: solaris package (pkg) support

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Hi there,

sorry for catching up so late...


 Thanks for responding.  I already know how to make Solaris packages,
 but I don't know how to integrate the process into my pom.xml.
 Someone posted a reference to a maven plugin that could create Solaris
 packages, but I couldn't find it when I searched.
 
 Dan Tran replied with this link which looks interesting:
 
 http://mojo.codehaus.org/solaris-maven-plugin/
If you look at the examples in src/it/*
you can get an idea of how it should work.

 
 If you don't find the problem already solved, then the two approaches to
 take are to write a maven plugin for yourself, or to implement the calling
 of pkg* using ant, and then wrap the build file in the
 maven-antrun-plugin.

 We use the antrun plugin to build Matlab code for us, it works pretty well.
 
 Ah, this looks promising as well, thanks.
Actually the solaris-maven-plugin uses an ant-skript to do its job.

Regards
  Jörg
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.5 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFHOLGJmPuec2Dcv/8RApGSAJ9WMAfOxn2K7HteuAitNsZbwlCFYwCdFTXf
/OwNgm2L9jvG39+GZVZ9HAs=
=b8FS
-END PGP SIGNATURE-

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Re: how to change version numbers in multiple pom.xml?

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

 Hi,
Hi there,
 
 What is the recommended methods to change contents (version numbers)
 version within multiple pom.xml? Is there anything like REXML in RUBY?
You can find an experimental approach that might or might not help you:
http://maven.apache.org/plugins/maven-dependency-plugin/go-offline-mojo.html
 
 Thanks.
 
 A.
 
Regards
  Jörg
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.5 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFHOLYAmPuec2Dcv/8RAm9yAKCSDKnTYiY7erD5XGhRjiYlBYgz3ACfaXpT
E2yDhyEM2MorwAO3Z5AaAT4=
=BjE6
-END PGP SIGNATURE-

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Re: Cleaning up results



Alex Mayorga Adame a écrit :


Emmanuel Venisse wrote:


Alex Mayorga Adame a écrit :

Would this guide
http://maven.apache.org/continuum/documentation/1_1/installation/upgrade.html
work if I'm coming from 1.0.3?

No, it is for 1.1 only.

Emmanuel



Is there a way from 1.0.3 to the latest beta?


We don't have a real 1.0.3 instance to test the migration.

You can test the datamangement tool described in the upgrade guide with this 
parameter (-databaseFormat CONTINUUM_103) added to the export command in this 
page.

Let us know if it work.

Emmanuel

Receiving email notification of successful build

2007-11-12 Thread Daniel Marques

Hello.

 

I am very new to Continuum, and so I'm sorry if this question has a
simple answer -- I searched the web and could not find it.

 

I'm attempting to setup Continuum (Version: 1.1-beta-3) so that I can
receive notification of successful build.  Build failures get reported
correctly, but successful build notices do not get sent.

 

I add a new email notifier and selected
Success/Failures/Warnings/Errors, but am still without success. Can
someone please explain the missing steps?

 

Thanks very much.

 

Sincerely, 


Dan

 

 

 

Daniel Marques

Chief Technical Officer

Optifreeze, LLC

646.307.4703

Re: scm:bootstrap checks out twice


How do you use it?

Emmanuel

Ilyevsky, Leonid (Equity Trading) a écrit :

I am using the maven-scm-plugin, the scm:bootstrap goal.
Everything works OK, except that it does the checkout from svn twice: it
cleans the .../target/checkout directory, performs the checkout, and
then cleans again and checks out again. Only after that it does the
specified goals.
This is annoying because the checkout may take essential time.

What is the problem? Is anybody else experiencing the same ?


This message w/attachments (message) may be privileged, confidential or 
proprietary, and if you are not an intended recipient, please notify the 
sender, do not use or share it and delete it. Unless specifically indicated, 
this message is not an offer to sell or a solicitation of any investment 
products or other financial product or service, an official confirmation of any 
transaction, or an official statement of Merrill Lynch. Subject to applicable 
law, Merrill Lynch may monitor, review and retain e-communications (EC) 
traveling through its networks/systems. The laws of the country of each 
sender/recipient may impact the handling of EC, and EC may be archived, 
supervised and produced in countries other than the country in which you are 
located. This message cannot be guaranteed to be secure or error-free. This 
message is subject to terms available at the following link: 
http://www.ml.com/e-communications_terms/. By messaging with Merrill Lynch you 
consent to the foregoing.








-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

RE: maven constantly rebuilding everything

2007-11-12 Thread EJ Ciramella

I'm not sure what you're trying to tell me to do, nothing?

-Original Message-
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Lee Meador
Sent: Monday, November 12, 2007 2:22 PM
To: Maven Users List
Subject: Re: maven constantly rebuilding everything

When a child pom indicates there is a parent pom, the parent pom is
always processed. It doesn't depend on where in the hierarchy the
child pom is sitting. Notice that the child pom will have a parent
section that completely identifies the parent OR defaults to the
parent being in the parent folder.

Notice that with all the things you can inject into the generated
artifacts, it is very difficult to tell if the current generated
artifact is going to be different than the last version. The problem
is that the output file doesn't just depend on the input files. It
depends on various properties and such, as well. (Think build number.)
It probably isn't any faster to build the new version (say, in memory)
and then read the old version and compare than to just build the new
version and store it. (You do have the down side of having a new file
date/time on the output.)

Once you change the one output file, then all the downstream
dependencies have to be fired off as well. (That may be a reason to
try to keep from changing the date/time on what was the output file
but has become the input for the next phase.) But, if that didn't
happen, you would still have to build the next phase because you don't
know if it depends on some changed property. Sorting out which
properties can change and NOT cause the next phase to be built is a
big problem too. You would have to say stuff like, When the date/time
changes, you don't do a build. But when the deployment environment
changes you do need a build. And, you can define properties on the
command line and they can come from the system environment. Do you
store all the properties values on each build and then compare on the
next build to see if anything changed?

As you can see, there might be some things that could be done to make
less stuff get built. But there may not ever be a way to make it work
100%. This may take a lot of thought and testing and defining to even
know.

-- Lee

On Nov 9, 2007 7:57 AM, EJ Ciramella [EMAIL PROTECTED] wrote:
 I'd love to know what people are building and the size of things, I think our 
 project is larger than most (I'm not stepping into a pissing contest, simply 
 stating the facts).  We're working on a massive refactoring both in codeline 
 management and actual code refactoring.

 Even with the refactoring done, that will only help to a certain degree.  
 With testing, we're talking about 1 - 1.5 hours.
 The resulting assembly for our main application is 195 - 200 mb (full of 
 jsps, images, classes, jars, etc).  I wish I could better show you the size 
 of the application (we have three ears and 4 - 5 wars depending on what 
 you're building).  There's no real honest to goodness code in the jsps (as it 
 should be), all business logic is handled else where.  As a release engineer, 
 it's not my business to make sweeping changes like that.  I can make 
 suggestions, but in the end, it's up to development to decide what projects 
 have merit/profit/etc associated with them.

 It'd be nice if you could work at the lowest directory all the time (the one 
 responsible for building the ear file), but when your change effects a module 
 that the ear you want depends on (or MULTIPLE subprojects), most developers 
 here build from the top.  Also, when profiles.xml lives at the root of the 
 overall project and contains 90% the configuration for a developers box, 
 sitting at a lower level, aren't these properties unavailable (doesn't maven 
 only look in the working directory for profiles.xml)?

 If maven wouldn't try to rebuild every jar and then every war and then every 
 ear, life would be much better.

 -Original Message-
 From: Wayne Fay [mailto:[EMAIL PROTECTED]

 Sent: Friday, November 09, 2007 8:10 AM
 To: Maven Users List
 Subject: Re: maven constantly rebuilding everything

 Sounds like your projects are entirely too big -- break 'em up a bit.
 Even 10min builds sounds rather long -- is this compilation only or
 including testing? And what portion of that time is consumed by your
 tests -- at this level (compilation), you should be running very fast
 unit tests, generally not integration tests.

 Artifacts/modules are cheap in Maven. War modules in particular
 should not have a lot of code associated with them -- move the code
 into Jars and extract the code out of your JSPs etc into classes that
 go into those Jars.

 Also when you're working, you don't clean install from the top
 parent, but instead run it from the module you're working on. At
 least, this is what we do...

 Wayne

 On 11/9/07, EJ Ciramella [EMAIL PROTECTED] wrote:
  That's incredible, doing that takes 30 - 40 minutes here.

  Most developers here are used to a 10 build time.

[ANN] Maven Source Plugin 2.0.4 Released

2007-11-12 Thread Dennis Lundberg


The Maven team is pleased to announce the release of Maven Source
Plugin, version 2.0.4.

http://maven.apache.org/plugins/maven-source-plugin/

Release Notes - Maven 2.x Source Plugin - Version 2.0.4


** Bug
* [MSOURCES-26] - source:jar ignoring remote-resources
* [MSOURCES-27] - Specifying excludes in the configuration removes
the default excludes

** Improvement
* [MSOURCES-18] - Honour targetPath
* [MSOURCES-22] - configuration option for source:jar to exclude
resources


Enjoy!

- The Maven Team

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Re: Running a Test Conditionally -- Including/Excluding Tests with Profiles

2007-11-12 Thread Dan Kigelman

Mm..  Is there any way to do what I'm trying though?  Execute one or
more of the tests if and only if?  Maybe by running on different
executions of surefire?

I tried to attach a .zip of my test code, but the message was rejected
as spam.  Here is the sample program as an in-text attachment.  It was
generated with archetype:create and modified to show the problem.

Thank you again for your help,

-- Dan
.
|-- pom.xml
`-- src
   |-- main
   |   `-- java
   |   `-- sample
   |   `-- test
   |   `-- App.java
   `-- test
   `-- java
   `-- sample
   `-- test
   |-- ExecuteAlwaysTest.java
   `-- ExecuteSometimesTest.java

pom.xml:
-
project xmlns=http://maven.apache.org/POM/4.0.0;
xmlns:xsi=http://www.w3.org/2001/XMLSchema-instance;
  xsi:schemaLocation=http://maven.apache.org/POM/4.0.0
http://maven.apache.org/maven-v4_0_0.xsd;
  modelVersion4.0.0/modelVersion
  groupIdsample.test/groupId
  artifactIdexecute-test-conditionally/artifactId
  packagingjar/packaging
  version1.0-SNAPSHOT/version
  nameexecute-test-conditionally/name
  urlhttp://maven.apache.org/url
  dependencies
dependency
  groupIdjunit/groupId
  artifactIdjunit/artifactId
  version3.8.1/version
  scopetest/scope
/dependency
  /dependencies

  build
plugins
  plugin
groupIdorg.apache.maven.plugins/groupId
artifactIdmaven-surefire-plugin/artifactId
configuration
  excludes
exclude**/ExecuteSometimesTest.java/exclude
  /excludes
/configuration
  /plugin
/plugins
  /build

  profiles
profile
  idconditionProfile/id
  activation
property
  namesome_condition/name
  valuetrue/value
/property
  /activation
  build
plugins
  plugin
groupIdorg.apache.maven.plugins/groupId
artifactIdmaven-surefire-plugin/artifactId
executions
  execution
idrareTest/id
goals
  goaltest/goal
/goals
phasetest/phase
configuration
  includes
include**/ExecuteSometimesTest.java/include
  /includes
  excludes /
/configuration
  /execution
/executions
  /plugin
/plugins
  /build
/profile
  /profiles
/project

App.java: (ignore me)
--
package sample.test;

/**
 * Hello world!
 *
 */
public class App
{
public static void main( String[] args )
{
System.out.println( Hello World! );
}
}


ExecuteAlwaysTest.java:


package sample.test;

import junit.framework.Test;
import junit.framework.TestCase;
import junit.framework.TestSuite;

/**
 * Unit test for simple App.
 */
public class ExecuteAlwaysTest
extends TestCase
{
/**
 * Create the test case
 *
 * @param testName name of the test case
 */
public ExecuteAlwaysTest( String testName )
{
super( testName );
}

/**
 * Rigourous Test :-)
 */
public void testApp()
{
assertTrue( true );
}
}

ExecuteSometimesTest.java:
--

package sample.test;

import junit.framework.Test;
import junit.framework.TestCase;
import junit.framework.TestSuite;

/**
 * Unit test for simple App.
 */
public class ExecuteSometimesTest
extends TestCase
{
/**
 * Create the test case
 *
 * @param testName name of the test case
 */
public ExecuteSometimesTest( String testName )
{
super( testName );
}

/**
 * Rigourous Test :-)
 */
public void testApp()
{
fail();
}
}


On Nov 12, 2007 11:07 AM, Wayne Fay [EMAIL PROTECTED] wrote:
 Without looking at the code I can't be sure, but I would assume that
 you can't include -and- exclude the same file. Probably the exclude
 list is processed last' in the code which means it won't run.


 Wayne

 On 11/11/07, Dan Kigelman [EMAIL PROTECTED] wrote:
  Thank you Wayne,
 
  I will be careful about that.  It seems it was working on my linux laptop,
  but I'm sure you've averted some of my future headaches for executing this
  on all kinds of shells.  I renamed the profile and activation to the
  following:
 
  profile
idconditionProfile/id
activation
  property
namesome_condition/name
valuetrue/value
  /property
/activation
build
 
  I execute now with mvn test -Dsome_condition=true but still with the same
  result:
 
  ---
   T E S T S
  ---
  Running sample.test.ExecuteAlwaysTest
  Tests run: 1, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 0.04 sec

Re: Running a Test Conditionally -- Including/Excluding Tests with Profiles