RE: Newbie Startup Question
Adding that made things work perfectly. It descended into all three levels of modules and pulled everything in. -Original Message- From: Emmanuel Venisse [mailto:[EMAIL PROTECTED] Sent: Monday, September 26, 2005 11:24 To: continuum-users@maven.apache.org Subject: Re: Newbie Startup Question Can you try to add this line allowedScheme implementation=java.lang.Stringfile/allowedScheme after line 430 in apps/continuum/conf/application.xml? And after you try to use an url file:///path_to_your_pom Let me know the result. Emmanuel Yann Le Du a écrit : Hi Bob, Ran into the same issues :) Once, support of file:// was planned, but it was replaced by the 'Upload' feature. So file:// URLs are incorrect now. When you add a parent POM, all the modules are recursively added, but not the parent POM itself. I don't know whether this is on purpose. It's fine for me because the parent POM only contains inherited properties and is not proper to be built itself. Though, as you said, this mechanism fails with uploads, and there is an open issue : http://jira.codehaus.org/browse/CONTINUUM-196 So, for the moment, I just upload the sub-projects one by one. It's a little tedious, but once they're added, it's OK :) Regards, Yann --- Allison, Bob [EMAIL PROTECTED] a écrit : I am trying to get started using Continuum and am having trouble importing my first project. I am running Continuum 1.0-alpha-4 on the same machine as all of my sources and CVS server. My project has a top-level directory and three modules named 3rd-party, lib, and plugins. I have created a CVS repository on the machine to hold the sources. I do not have any CVS web viewer available on the machine. When I try to specify the POM as a file in Continuum's web page, it downloads the POM correctly, then fails to download the modules (I get the project builder downloading file:/tmp/summit-23448.tmp, file:/tmp/summit-2/3rd-party/pom.xml, file:/tmp/summit-2/lib/pom.xml, and file:/tmp/summit-2/plugins/pom.xml). Since the three module POMs were not copied to /tmp/summit-2, I get warnings that the file could not be downloaded and no projects are created. If I try to use a file:/ URL in Continuum's web page (file:/path/to/pom/pom.xml), I get back an error message requesting that I specify a valid URL and no attempt is made to create the project. If I comment out the modules and specify the POM as a file, it is processed without warnings and creates 0 projects and 1 project group. The project group does not appear in the Continuum web project list (that list still shows 0 projects) and I don't know where I can check to see if anything will actually build. Can anybody give me a hand? ___ Appel audio GRATUIT partout dans le monde avec le nouveau Yahoo! Messenger Téléchargez cette version sur http://fr.messenger.yahoo.com
[m1.1b1] IntelliJ IDEA plugin with Maven 1.1beta1
Hi, when generating the iml file with Maven 1.1beta1, I always get the following error when using that iml file: Cannot determine module type for the following module: mymodule. Assuming it is a JAVA module. But this assuming does not seem to work properly. The current workaround I have found is to manually edit the iml file as follows: Original: module version=4 relativePaths=true type= Add JAVA_MODULE as type: module version=4 relativePaths=true type=JAVA_MODULE please fix this, regards, Wim
Re: [m2] javadoc plugin and linkoffline
Hi, Trygve Laugstøl schrieb: I was thinking something like this: offlineLinks offlineLink urla/url packageListUrlb/packageListUrl -- is this really required? /offlineLink /offlineLinks Yeah, I would also appreciate keeping naming in sync with the javadoc-plugin propertly maven.javadoc.offlineLinks See http://maven.apache.org/reference/plugins/javadoc/properties.html Thanks -- Martin Skopp Riege Software International GmbH Support: mailto:[EMAIL PROTECTED], Information: http://www.riege.com This email is intended to be viewed with a nonproportional font. - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: [m2] javadoc plugin and linkoffline - online links in offline mode
Hi Trygve, Trygve Laugstøl schrieb: I was thinking something like this: offlineLinks offlineLink urla/url packageListUrlb/packageListUrl -- is this really required? /offlineLink /offlineLinks It's required to specify a *online URL* and a *offline local path* when you want to generates *online links* in offline mode, see http://maven.apache.org/reference/plugins/javadoc/faq.html#offlineLinks for and example with the maven.javadoc.offlineLinks property. To generate offline links in offline mode, the local path is sufficient, of course. Regards, -- Martin Skopp Riege Software International GmbH Support: mailto:[EMAIL PROTECTED], Information: http://www.riege.com This email is intended to be viewed with a nonproportional font. - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Nearly empty classpath while executing tests with maven2 beta 1
I am unable to figure out why when I run m2 with a test target the tests are executed with a nearly empty classpath. I have setup all my dependencies with the appropriate compile or test scope but these are are not being reflected in the classpath during execution. e.g. I added the following to my test case : System.out.println(); System.out.println(System.getProperty(java.class.path)); System.out.println(); Here's what I get in the output /usr/local/maven/core/boot/classworlds-1.1-alpha-2.jar My relevant plugin snippet in pom.xml is as follows : plugin groupIdorg.apache.maven.plugins/groupId artifactIdmaven-surefire-plugin/artifactId executions execution phasetest/phase /execution /executions goals goal idtest/id configuration !-- test**/Test*/test -- includes include implementation=java.lang.String**/Test*/include /includes /configuration /goal /goals /plugin Any suggestions what could be going wrong ? Thanks Dhananjay - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: Nearly empty classpath while executing tests with maven2 beta 1
On Mon, 26 Sep 2005, Dhananjay Nene wrote:
There's nothing wrong here. The system property java.class.path is
the classpath for the JVM, i.e. the CLASSPATH environment variable
when you start the jvm.
What you need/want is the 'classpath' of the classloader that runs
the tests: Thread.currentThread().getContextClassLoader(). That
classloader has a larger classpath, although it's hard to get the
classpath string from it. If you really need to, use this snippet
of code:
public List getRuntimeClasspath()
throws IOException
{
List classpathEntries = new ArrayList();
Enumeration n =
Thread.currentThread().getContextClassLoader().getResources( META-INF );
while ( n.hasMoreElements() )
{
URL url = (URL) n.nextElement();
String path = url.getPath();
int idx = path.indexOf( ! );
path = (idx 0) ? path.substring(0, idx) : path;
URL url2 = new URL(path);
path = url2.getFile();
classpathEntries.add( path );
}
return classpathEntries;
}
But normally you'd use ClassLoader.getResource().
-- Kenney
I am unable to figure out why when I run m2 with a test target the
tests are executed with a nearly empty classpath. I have setup all my
dependencies with the appropriate compile or test scope but these
are are not being reflected in the classpath during execution. e.g. I
added the following to my test case :
System.out.println();
System.out.println(System.getProperty(java.class.path));
System.out.println();
Here's what I get in the output
/usr/local/maven/core/boot/classworlds-1.1-alpha-2.jar
My relevant plugin snippet in pom.xml is as follows :
plugin
groupIdorg.apache.maven.plugins/groupId
artifactIdmaven-surefire-plugin/artifactId
executions
execution
phasetest/phase
/execution
/executions
goals
goal
idtest/id
configuration
!--
test**/Test*/test
--
includes
include
implementation=java.lang.String**/Test*/include
/includes
/configuration
/goal
/goals
/plugin
Any suggestions what could be going wrong ?
Thanks
Dhananjay
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
--
Kenney Westerhof
http://www.neonics.com
GPG public key: http://www.gods.nl/~forge/kenneyw.key
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
Re: Nearly empty classpath while executing tests with maven2 beta 1
Duh!! thats embarassing. The classpath is being set correctly. Thanks.
Kenney Westerhof wrote:
On Mon, 26 Sep 2005, Dhananjay Nene wrote:
There's nothing wrong here. The system property java.class.path is
the classpath for the JVM, i.e. the CLASSPATH environment variable
when you start the jvm.
What you need/want is the 'classpath' of the classloader that runs
the tests: Thread.currentThread().getContextClassLoader(). That
classloader has a larger classpath, although it's hard to get the
classpath string from it. If you really need to, use this snippet
of code:
public List getRuntimeClasspath()
throws IOException
{
List classpathEntries = new ArrayList();
Enumeration n = Thread.currentThread().getContextClassLoader().getResources(
META-INF );
while ( n.hasMoreElements() )
{
URL url = (URL) n.nextElement();
String path = url.getPath();
int idx = path.indexOf( ! );
path = (idx 0) ? path.substring(0, idx) : path;
URL url2 = new URL(path);
path = url2.getFile();
classpathEntries.add( path );
}
return classpathEntries;
}
But normally you'd use ClassLoader.getResource().
-- Kenney
I am unable to figure out why when I run m2 with a test target the
tests are executed with a nearly empty classpath. I have setup all my
dependencies with the appropriate compile or test scope but these
are are not being reflected in the classpath during execution. e.g. I
added the following to my test case :
System.out.println();
System.out.println(System.getProperty(java.class.path));
System.out.println();
Here's what I get in the output
/usr/local/maven/core/boot/classworlds-1.1-alpha-2.jar
My relevant plugin snippet in pom.xml is as follows :
plugin
groupIdorg.apache.maven.plugins/groupId
artifactIdmaven-surefire-plugin/artifactId
executions
execution
phasetest/phase
/execution
/executions
goals
goal
idtest/id
configuration
!--
test**/Test*/test
--
includes
include
implementation=java.lang.String**/Test*/include
/includes
/configuration
/goal
/goals
/plugin
Any suggestions what could be going wrong ?
Thanks
Dhananjay
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
--
Kenney Westerhof
http://www.neonics.com
GPG public key: http://www.gods.nl/~forge/kenneyw.key
-
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]
Fw: Error While executing Maven2 deploy
Dear All, Can anyone please help in resolving the below issue in Maven2? Regards and Thanks, Deva Narayanan Devanathan/CHN/TCS 09/22/2005 06:01 PM To users@maven.apache.org cc Subject Error While executing Maven2 deploy Dear All, Attached is the exception encountered while executing the command m2 deploy using Beta-1 Version of Maven2. Kindly give me the possible solution to the below problem ASAP. Also I have attached the pom.xml and settings.xml for the reference. [ERROR] BUILD ERROR [INFO] [INFO] Diagnosis: Error deploying artifact [INFO] [ERROR] Cause: org.apache.maven.plugin.MojoExecutionException: Error deploying artifact at org.apache.maven.plugin.deploy.DeployMojo.execute(DeployMojo.java:154) at org.apache.maven.plugin.DefaultPluginManager.executeMojo(DefaultPluginManager.java:357) at org.apache.maven.lifecycle.DefaultLifecycleExecutor.executeGoals(DefaultLifecycleExecutor.java:479) at org.apache.maven.lifecycle.DefaultLifecycleExecutor.executeGoalWithLifecycle(DefaultLifecycleExecutor.java:452) at org.apache.maven.lifecycle.DefaultLifecycleExecutor.executeGoal(DefaultLifecycleExecutor.java:438) at org.apache.maven.lifecycle.DefaultLifecycleExecutor.executeTaskSegments(DefaultLifecycleExecutor.java:273) at org.apache.maven.lifecycle.DefaultLifecycleExecutor.execute(DefaultLifecycleExecutor.java:131) at org.apache.maven.DefaultMaven.execute(DefaultMaven.java:186) at org.apache.maven.cli.MavenCli.main(MavenCli.java:302) at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method) at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:39) at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:25) at java.lang.reflect.Method.invoke(Method.java:324) at org.codehaus.classworlds.Launcher.launchEnhanced(Launcher.java:315) at org.codehaus.classworlds.Launcher.launch(Launcher.java:255) at org.codehaus.classworlds.Launcher.mainWithExitCode(Launcher.java:430) at org.codehaus.classworlds.Launcher.main(Launcher.java:375) Caused by: org.apache.maven.artifact.deployer.ArtifactDeploymentException: Error deploying artifact: at org.apache.maven.artifact.deployer.DefaultArtifactDeployer.deploy(DefaultArtifactDeployer.java:89) at org.apache.maven.artifact.deployer.DefaultArtifactDeployer.deploy(DefaultArtifactDeployer.java:53) at org.apache.maven.plugin.deploy.DeployMojo.execute(DeployMojo.java:142) ... 16 more Caused by: org.apache.maven.artifact.metadata.ArtifactMetadataRetrievalException: Unable to retrieve metadata at org.apache.maven.artifact.repository.metadata.DefaultRepositoryMetadataManager.resolveAlways(DefaultRepositoryMetadataManager.java:212) at org.apache.maven.artifact.repository.metadata.DefaultRepositoryMetadataManager.resolveAlways(DefaultRepositoryMetadataManager.java:183) at org.apache.maven.artifact.transform.SnapshotTransformation.resolveLatestSnapshotBuildNumber(SnapshotTransformation.java:130) at org.apache.maven.artifact.transform.SnapshotTransformation.transformForDeployment(SnapshotTransformation.java:75) at org.apache.maven.artifact.transform.DefaultArtifactTransformationManager.transformForDeployment(DefaultArtifactTransformationManager.java:58) at org.apache.maven.artifact.deployer.DefaultArtifactDeployer.deploy(DefaultArtifactDeployer.java:62) ... 18 more Caused by: org.apache.maven.wagon.TransferFailedException: Invalid checksum file at org.apache.maven.artifact.manager.DefaultWagonManager.verifyChecksum(DefaultWagonManager.java:497) at org.apache.maven.artifact.manager.DefaultWagonManager.getRemoteFile(DefaultWagonManager.java:336) at org.apache.maven.artifact.manager.DefaultWagonManager.getArtifactMetadata(DefaultWagonManager.java:256) at org.apache.maven.artifact.repository.metadata.DefaultRepositoryMetadataManager.resolveAlways(DefaultRepositoryMetadataManager.java:198) ... 23 more Caused by: java.io.FileNotFoundException: /net/michiru/homes/userId/main/maven2/m2/User-apptest/local_repo/com/mycompany/app/user-apptest/1.0-SNAPSHOT/maven-metadata-User.xml.tmp.sha1 (No such file or directory) at java.io.FileInputStream.open(Native Method) at java.io.FileInputStream.init(FileInputStream.java:106) at org.codehaus.plexus.util.FileUtils.fileRead(FileUtils.java:258) at org.apache.maven.artifact.manager.DefaultWagonManager.verifyChecksum(DefaultWagonManager.java:468) ... 26 more [INFO] [INFO] Total time: 16 seconds [INFO] Finished at: Thu Sep 22 08:14:02 EDT
Fw :[m2-beta-1] bug for 'warSourceExcludes' property?
Maven devs: anybody has an idea about that? Should I open a bug report?
Best Regards / Cordialement,
Fabrice BELLINGARD
DINQ/DSIN/INSI/EATE/IDVS/AIDV
(+33) (01 61) 45 15 91 - [EMAIL PROTECTED]
- Réacheminé par FABRICE BELLINGARD - E222731/users/PSA le 26/09/2005
13:41 -
FABRICE
BELLINGARD -
E222731/users/PSAPour
Maven Users List
22/09/2005 10:58 users@maven.apache.org
cc
Objet
[m2-beta-1] bug for
'warSourceExcludes' property?
Hi all,
has anybody already used the 'warSourceExcludes' property of the War
plugin? (didn't find any thread about this)
Because I experience a strange behaviour with this property. When I grab
some Web projects from the CVS of my company, I usually get all the
WEB-INF/lib Jars that the developers put in the CVS:
- With Maven 1, I use the 'maven.war.src.excludes' property to exclude
those Jars ('maven.war.src.excludes=**/lib/*.*'), and it works well: the
Jars packaged in the War are only the ones defined in the pom.
- With m2, I tried to use the 'warSourceExcludes' property, which seemed to
me to be the equivalent property. However, when the War is generated, there
is no jar in the WEB-INF/lib folder. This is as if the plugin is excluding
files after having copied the needed dependencies in the WEB-INF/lib folder
(whereas in the Maven 1 version, it was excluding them before).
Is this a bug or am I missing something?
Thanks for your help! :o)
Best Regards / Cordialement,
Fabrice BELLINGARD
DINQ/DSIN/INSI/EATE/IDVS/AIDV
(+33) (01 61) 45 15 91 - [EMAIL PROTECTED]
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
RE: Problem with calling ant from site:site goal
Hi Achim, In goal, you must specify the goal of the plugin you are configuring, not the one of the phase you are using (if this ever makes sense). You are configuring maven-antrun-plugin, so you should use goalrun/goal. Look into this file (source code) : http://svn.apache.org/viewcvs.cgi/maven/components/trunk/maven-core/src/main/resources/META-INF/plexus/components.xml?view=markup In org.apache.maven.lifecycle.LifecycleExecutor component, you have the list of the existing phases, and their order of execution. As you guessed, the site phase doesn't exist. Then, each org.apache.maven.lifecycle.mapping.LifecycleMapping component defines, for a given artifact type, what plugins are called for each phase. You can see that the site plugin is never called. I don't exactly know what you are trying to achieve, but you can try to : * use another phase, if possible * edit components.xml (unguaranted, I never tried it myself) * edit the site plugin HTH, Yann --- Achim Hügen [EMAIL PROTECTED] a écrit : Hi, I'm trying to call ant during the site:site goal: build plugin groupIdorg.apache.maven.plugins/groupId artifactIdmaven-antrun-plugin/artifactId executions execution phasesite/phase configuration tasks echoTEST - /echo /tasks /configuration goals goalsite/goal /goals /execution /executions /plugin /plugins /build Unfortunately the echo task is never executed. I assume that there is no site phase. But I don't know and can't figure out in what phase the site:site goal is run. Achim - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] ___ Appel audio GRATUIT partout dans le monde avec le nouveau Yahoo! Messenger Téléchargez cette version sur http://fr.messenger.yahoo.com - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
maven cactus plugin and multiproject
Hi, it's not clear to me how cactus plugin supports multiproject. I have EJB in separate subproject, when I want to test EAR, cactus plugin uses cactus.build.goal.ejb (default ejb:ejb) to build the EJB artifact. But the EJB artifact is in separate subproject. I also created complex genapp application and the same problem is there too. The cactus:test-ear tries to build the EJB, but there's no EJB in EAR subproject. Jan - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: [m2] project descriptor tags
Hi Jesse, Yes, my trunk was older than that, that's why the XSD wasn't correct. I'm using beta-1 now. I had put an example of what I'm trying to achieve in the preceding mail - below my signature :) Here's it again : Thanks, Yann ~ project ~ parent ~ artifactIdmyparent/artifactId ~ groupIdcom.corp.myparent/groupId ~ version1.0.0/version ~ /parent ~ ~ modelVersion4.0.0/modelVersion ~ groupIdcom.corp.myparent /groupId ~ artifactIdmyproject/artifactId ~ nameMy project/name ~ version1.0.1-SNAPSHOT/version ~ ~ build ~ extensions ~ extension ~ groupIdant/groupId ~ artifactIdant-nodeps/artifactId ~ version1.6.5/version ~ /extension ~ /extensions ~ ~ [...] ~ ~ plugins ~ plugin ~ artifactIdmaven-antrun-plugin/artifactId ~ ~ executions ~ execution ~ phaseprocess-resources/phase ~ ~ goals ~ goalrun/goal ~ /goals ~ ~ inheritedtrue/inherited ~ ~ configuration ~ tasks ~ ant antfile=replace.xml/ ~ /tasks ~ /configuration ~ /execution ~ /executions ~ /plugin ~ ~ [...] ~ /plugins ~ /build ~ ~ dependencies ~ dependency ~ groupIdant/groupId ~ artifactIdant-nodeps/artifactId ~ version1.6.5/version ~ /dependency ~ ~ [...] ~ /dependencies ~ ~ [...] ~ /project --- Jesse McConnell [EMAIL PROTECTED] a écrit : well, that would depend on how _old_ that trunk version is...I think brett put it in just a month ago or so.. I'll be back in town soon and can help you out on monday maybe.. paste in some examples of what you are trying to do jesse On 9/23/05, Yann Le Du [EMAIL PROTECTED] wrote: Hi Jesse, I happen to try and use the extension mechanism for the antrun-plugin. The ant script I'm trying to run contains regexpmapper tags, so I need ant:ant-nodeps to be available to the plugin (if I add this dependency to antrun-plugin's POM, it's working). A few days ago, Kenney told me about this extension mechanism, but I never get to make it work - I admit now that I was using an ooold trunk version then :) Following your description, I just tried again, but to no avail. Do you see anything wrong in the way I use this mechanism below ? Also, I noticed this issue : http://jira.codehaus.org/browse/MNG-697 Does it mean this extension mechanism isn't operational at the present time ? Thanks, Yann [snip] --- Jesse McConnell [EMAIL PROTECTED] a écrit : sure: this is the pom.xml for the maven-jdbc-plugin...notice there are not dependencies for a database driver in the plugin pom itself. project modelVersion4.0.0/modelVersion parent artifactIdmojo/artifactId groupIdorg.apache.maven.plugins/groupId version1.0.1/version /parent artifactIdmaven-jdbc-plugin/artifactId version0.1-SNAPSHOT/version packagingmaven-plugin/packaging nameMaven JDBC Plugin/name inceptionYear2005/inceptionYear dependencies dependency groupIdorg.apache.maven/groupId artifactIdmaven-project/artifactId version2.0-beta-1-SNAPSHOT/version /dependency dependency groupIdplexus/groupId artifactIdplexus-utils/artifactId version1.0.1-SNAPSHOT/version /dependency /dependencies /project and here is the pom.xml for something that is using the plugin: project modelVersion4.0.0/modelVersion groupIdg/groupId artifactIdg-db/artifactId packagingjar/packaging version1.0/version nameg - db plugin playground/name parent groupIdg/groupId artifactIdg/artifactId version1.0/version /parent dependencies dependency groupIdoracle/groupId artifactIdoracle/artifactId version9201/version scopecompile/scope /dependency dependency groupIdoracle/groupId artifactIdoracle_nls_charset/artifactId version9201.12/version scopecompile/scope /dependency /dependencies build extensions extension artifactIdoracle/artifactId groupIdoracle/groupId version9201/version /extension extension artifactIdoracle_nls_charset/artifactId groupIdoracle/groupId version9201.12/version /extension /extensions plugins plugin artifactIdmaven-jdbc-plugin/artifactId version1.0-SNAPSHOT/version configuration driveroracle.jdbc.driver.OracleDriver/driver urljdbc:oracle:thin:@host:1790:DB/url usernameuser/username passwordpassword/password statementselect * from whatever/statement /configuration /plugin /plugins /build /project now, the dependency entries would in available to compile against for sources that might be in this subproject..but they would not be available to the
Re: [m2] Portable way of specifying JDKs in compiler plugin
For any avid readers..
On 21/09/05, Mark Hobson [EMAIL PROTECTED] wrote:
[snip]
I'm not a big fan of (1) since system-specific information will be
held within the POM. Even if I use an expression like
${java.home}/../jdk1.4.2/bin/javac, this still assumes something
about the whereabouts of the JDK. For these scenarios, do you think
it would make sense to introduce a JDK version compiler config param
which cross-referenced a list of JDK definitions in settings.xml? The
resolved java.home could then be used in the plexus-compiler-*
projects when deriving the necessary rt.jar path.
See MNG-468, but hardcoding the JDK doesn't currently work due to MNG-862.
I prefer (2), but it appears there's no way of specifying the
bootclasspath in plexus-compiler-api. Ideally my project would have a
DOM Level 2 API dependency which I could flag as a bootclasspath jar
(yet another scope?) and that would take precendence over whatever JDK
I was using.
See MNG-973 and also MNG-1020 for this to work in eclipse.
I'd be interested in what options I have to achieve just compliation
in this setup using m2 beta1, and also how this problem could be
tackled more elegantly in future.
I couldn't find a way of doing this with beta1 due to the above issues :(
Mark
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
[beta-1-snasphot] trouble with bad entry in the cron's expression of a new schedule
Hi all, I have encountered a trouble by setting a bad cron expression for a new scheduler : 0 0 0,4,8,12,16,20 * * * instead of 0 0 0,4,8,12,16,20 * * ? The continuum application fails and stop. Then it's was impossible to restart the application (always failed during the startup). The only solution to delete and install again the application. I found this issue http://jira.codehaus.org/browse/CONTINUUM-304. It's probably difficult to write an good validator. But is it possible to correctly catch the exception ? Olivier This e-mail, any attachments and the information contained therein (this message) are confidential and intended solely for the use of the addressee(s). If you have received this message in error please send it back to the sender and delete it. Unauthorized publication, use, dissemination or disclosure of this message, either in whole or in part is strictly prohibited. ** Ce message electronique et tous les fichiers joints ainsi que les informations contenues dans ce message ( ci apres le message ), sont confidentiels et destines exclusivement a l'usage de la personne a laquelle ils sont adresses. Si vous avez reu ce message par erreur, merci de le renvoyer a son emetteur et de le detruire. Toutes diffusion, publication, totale ou partielle ou divulgation sous quelque forme que se soit non expressement autorisees de ce message, sont interdites. **
Plugin issues that span both m1 and m2
What's the official stance for plugin issues that span both m1 and m2? I opened MNG-1020 as a m2 version of MPECLIPSE-104 - is this good practice? I know ultimately m1 plugins will become wrappers around their m2 counterparts, but what about in the meantime? Cheers, Mark - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: Customn dependency scopes
Hi Tim, I have the same needs and wrote a plugin that merely copy all dependencies into a directory of your choice - it's largely inspired from war plugin, of course :) Let me know if you are interested. Yann --- Brett Porter [EMAIL PROTECTED] a écrit : Scope is a default for the packaging rule, not a packaging rule in itself. There is an open JIRA for the war plugin to allow this (specifically for applets and jnlp). It has not yet been implemented. You can work around it by writing your own small plugin (or use the antrun plugin), to copy the jar into place in an earlier phase. Cheers, Brett On 9/25/05, Tim Kettler [EMAIL PROTECTED] wrote: Hi, I've set up a simple web-app (war) project with Maven 2. Now I need to place some artifacts from my repository (a few jar's accessed by JNLP) in the root of the war. Just declaring them as dependencies puts them in the WEB-INF/lib directory. Is it possible to define a new kind of dependency scope (for example 'client') and then copy the artifacts in the 'package' goal to the right location? Thanks Tim - 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] ___ Appel audio GRATUIT partout dans le monde avec le nouveau Yahoo! Messenger Téléchargez cette version sur http://fr.messenger.yahoo.com - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: Customn dependency scopes
On 25/09/05, Brett Porter [EMAIL PROTECTED] wrote: There is an open JIRA for the war plugin to allow this (specifically for applets and jnlp). It has not yet been implemented. For those interested this is http://jira.codehaus.org/browse/MNG-896. Mark - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: Customn dependency scopes
On 25/09/05, Brett Porter [EMAIL PROTECTED] wrote: You can work around it by writing your own small plugin (or use the antrun plugin), to copy the jar into place in an earlier phase. BTW, I tried this with antrun using the maven ant tasks for transitive deps, but encountered MNG-1017. Admittedly it could be achieved with native ant tasks but that was just too evil for me to attempt. Mark - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[m2] getting error when pointing to new testResources
Hi, I'm trying to change the resources directory for tests and so I created the following build section in my POM: build sourceDirectory./sourceDirectory testSourceDirectory./testSourceDirectory resources resource directoryresources/directory /resource /resources testResources resource directoryresources/directory /resource /testResources /build As you can see the testResources section is identical to the resources section except for the name. However when I run m2 install I get the following error: [INFO] Reason: Failed to parse model from file '/Users/dev/projects/ source-tree/pom.xml'. Error: 'TEXT must be immediately followed by END_TAG and not START_TAG (position: START_TAG seen ...resource\n directory... @18:20) ' This error goes away if I delete the testResources section. As far as I can tell I'm using the syntax specified on the plugin descriptor page. Thanks - Ashley - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: [m2] getting error when pointing to new testResources
On 26/09/05, Ashley Williams [EMAIL PROTECTED] wrote: This error goes away if I delete the testResources section. As far as I can tell I'm using the syntax specified on the plugin descriptor page. Yep, the page is wrong :) Use testResource rather than resource within the testResources block. Mark - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[m2] Mojo, subproject and self-reference
I'm writting a mock generator and I need advice on how to organise my project. Amongst other module, I have : - The genrator itself - The runtime library used by the genrated code - The Mojo that will launch the genrator from + All the unit test. I wonder how to organise this into modules? I guess that the decomposition I have should be nice, but I have a problem for my unit tests. Indeed, I would like to be self-testing. I mean that I would like to use my mock generator to make my tests. One solution I have seen is to place the unit test into a separated module. But I don't really like it. An other one is to rely an a previous version of the mock generator being already deployed on the repository. Did you have already encounter similar problem ? What is the best practice ? Other question, maybe related : How did you make to test a plugin like sure-fire? Or how do you generate site docs for the site plugin ? _ Protégez votre boîte de réception: Phishing : comment l'identifier, le signaler et l'empêcher http://www.fr.msn.be/security/phishing/ - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: [m2] getting error when pointing to new testResources
Works now - you know I _nearly_ tried that as a last ditch attempt, honestly! Seems the project descriptor isn't auto-generated from the code like I thought it was though. On 26 Sep 2005, at 15:34, Mark Hobson wrote: On 26/09/05, Ashley Williams [EMAIL PROTECTED] wrote: This error goes away if I delete the testResources section. As far as I can tell I'm using the syntax specified on the plugin descriptor page. Yep, the page is wrong :) Use testResource rather than resource within the testResources block. Mark - 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: [m2] getting error when pointing to new testResources
On 26/09/05, Ashley Williams [EMAIL PROTECTED] wrote: Works now - you know I _nearly_ tried that as a last ditch attempt, honestly! Seems the project descriptor isn't auto-generated from the code like I thought it was though. Yeah it got me first time too - not too sure why the doc differs from the code since it is auto-generated? Mark - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: [m2] getting error when pointing to new testResources
On Mon, 2005-09-26 at 15:39 +0100, Ashley Williams wrote: Works now - you know I _nearly_ tried that as a last ditch attempt, honestly! Seems the project descriptor isn't auto-generated from the code like I thought it was though. It is generated but there is a bug in the generation. It's in JIRA. A problem with modello. -- jvz. Jason van Zyl jason at maven.org http://maven.apache.org - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: [m2] getting error when pointing to new testResources
On Mon, 2005-09-26 at 15:56 +0100, Mark Hobson wrote: On 26/09/05, Ashley Williams [EMAIL PROTECTED] wrote: Works now - you know I _nearly_ tried that as a last ditch attempt, honestly! Seems the project descriptor isn't auto-generated from the code like I thought it was though. Yeah it got me first time too - not too sure why the doc differs from the code since it is auto-generated? The issue[1] is that the TestResource class inherits the Resource class and there's a bug in the Modello plugin that renders the page. [1]: http://jira.codehaus.org/browse/MNG-523 -- Trygve - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: Newbie Startup Question
Can you try to add this line allowedScheme implementation=java.lang.Stringfile/allowedScheme after line 430 in apps/continuum/conf/application.xml? And after you try to use an url file:///path_to_your_pom Let me know the result. Emmanuel Yann Le Du a écrit : Hi Bob, Ran into the same issues :) Once, support of file:// was planned, but it was replaced by the 'Upload' feature. So file:// URLs are incorrect now. When you add a parent POM, all the modules are recursively added, but not the parent POM itself. I don't know whether this is on purpose. It's fine for me because the parent POM only contains inherited properties and is not proper to be built itself. Though, as you said, this mechanism fails with uploads, and there is an open issue : http://jira.codehaus.org/browse/CONTINUUM-196 So, for the moment, I just upload the sub-projects one by one. It's a little tedious, but once they're added, it's OK :) Regards, Yann --- Allison, Bob [EMAIL PROTECTED] a écrit : I am trying to get started using Continuum and am having trouble importing my first project. I am running Continuum 1.0-alpha-4 on the same machine as all of my sources and CVS server. My project has a top-level directory and three modules named 3rd-party, lib, and plugins. I have created a CVS repository on the machine to hold the sources. I do not have any CVS web viewer available on the machine. When I try to specify the POM as a file in Continuum's web page, it downloads the POM correctly, then fails to download the modules (I get the project builder downloading file:/tmp/summit-23448.tmp, file:/tmp/summit-2/3rd-party/pom.xml, file:/tmp/summit-2/lib/pom.xml, and file:/tmp/summit-2/plugins/pom.xml). Since the three module POMs were not copied to /tmp/summit-2, I get warnings that the file could not be downloaded and no projects are created. If I try to use a file:/ URL in Continuum's web page (file:/path/to/pom/pom.xml), I get back an error message requesting that I specify a valid URL and no attempt is made to create the project. If I comment out the modules and specify the POM as a file, it is processed without warnings and creates 0 projects and 1 project group. The project group does not appear in the Continuum web project list (that list still shows 0 projects) and I don't know where I can check to see if anything will actually build. Can anybody give me a hand? ___ Appel audio GRATUIT partout dans le monde avec le nouveau Yahoo! Messenger Téléchargez cette version sur http://fr.messenger.yahoo.com
Re: [m2] xsd updates
On 12/09/05, Brett Porter [EMAIL PROTECTED] wrote: It will next be updated when the next Maven release goes out. It will be locked down at 2.0 final, and after that changes to the POM will cause a model increment (4.0.1, etc). Anyone know if the 4.0.0 XSD has been updated for beta1? I'm still getting XSD validation errors on my beta1-specific pom.xml features. Mark - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: [beta-1-snasphot] trouble with bad entry in the cron's expression of a new schedule
Can you send the stacktrace? Emmanuel Olivier Lamy a écrit : Hi all, I have encountered a trouble by setting a bad cron expression for a new scheduler : 0 0 0,4,8,12,16,20 * * * instead of 0 0 0,4,8,12,16,20 * * ? The continuum application fails and stop. Then it's was impossible to restart the application (always failed during the startup). The only solution to delete and install again the application. I found this issue http://jira.codehaus.org/browse/CONTINUUM-304. It's probably difficult to write an good validator. But is it possible to correctly catch the exception ? Olivier This e-mail, any attachments and the information contained therein (this message) are confidential and intended solely for the use of the addressee(s). If you have received this message in error please send it back to the sender and delete it. Unauthorized publication, use, dissemination or disclosure of this message, either in whole or in part is strictly prohibited. ** Ce message electronique et tous les fichiers joints ainsi que les informations contenues dans ce message ( ci apres le message ), sont confidentiels et destines exclusivement a l'usage de la personne a laquelle ils sont adresses. Si vous avez reu ce message par erreur, merci de le renvoyer a son emetteur et de le detruire. Toutes diffusion, publication, totale ou partielle ou divulgation sous quelque forme que se soit non expressement autorisees de ce message, sont interdites. **
Re: [m2] trouble accessing internal repository with beta1
On 24.09.2005, at 02:31, Brett Porter wrote: Yep, sftp will work. ftp will work if wagon-ftp is added as an extension. So I switched my access method from scp to sftp with beta-1. Now I'm trying to deploy a project that does not yet exist in either the local or internal (company-wide) repository, and m2 fails like this: --- [INFO] [install:install] [INFO] [deploy:deploy] Uploading: sftp://internal.example.com/home/maven/maven2/blah/blah/ 0.1/blah-0.1.jar [INFO] Retrieving previous metadata from internal [INFO] [ERROR] BUILD ERROR [INFO] [INFO] Diagnosis: Error deploying artifact [INFO] [ERROR] Cause: org.apache.maven.plugin.MojoExecutionException: Error deploying artifact ... Caused by: org.apache.maven.artifact.deployer.ArtifactDeploymentException: Error deploying artifact: ... Caused by: org.apache.maven.artifact.metadata.ArtifactMetadataRetrievalException: U nable to retrieve metadata Caused by: org.apache.maven.wagon.TransferFailedException: Error occured while downloading from the remote repository:[internal] - sftp://internal.example.com/home/maven/maven2 ... Caused by: No such file at com.jcraft.jsch.ChannelSftp.throwStatusError(Unknown Source) at com.jcraft.jsch.ChannelSftp.stat(Unknown Source) at com.jcraft.jsch.ChannelSftp.get(Unknown Source) at com.jcraft.jsch.ChannelSftp.get(Unknown Source) at org.apache.maven.wagon.providers.ssh.SftpWagon.get (SftpWagon.java:216) --- In the internal repository, the jar, jar.md5 and jar.sha1 files have been created, but no pom. I'm wondering what file m2 is trying to retrieve? The No such file error message could be improved by adding the expected file name. The doco on this area is a little lean, but we're working on it now. Thanks Brett, your work on m2 and help on this list is highly appreciated! Cheers, -Ralph. On 9/23/05, Ralph Pöllath [EMAIL PROTECTED] wrote: Hi, since deployment via scp is broken for me in beta-1, is there a possible alternative, like sftp or ftp? Maybe I'm lacking sleep or caffeine, but I just cannot find any documentation on this. Cheers, -Ralph. On 22.09.2005, at 12:45, Ralph Pöllath wrote: On 22.09.2005, at 11:53, Brett Porter wrote: Not what I wanted to see :( Can you post this to JIRA for tracking, and I'll take a look at what has changed since alpha-3 to try and narrow it down. Done: http://jira.codehaus.org/browse/MNG-959 Do other jdks or servers succeed? I have no way of trying right now, but will comment on MNG-959 once I find out. Cheers, -Ralph. On 9/22/05, Ralph Pöllath [EMAIL PROTECTED] wrote: Connect thread maven.example.com session prio=5 tid=0x005651b0 nid=0x1e2 runnable [f0c89000..f0c89ac0] at java.net.SocketInputStream.socketRead0(Native Method) at java.net.SocketInputStream.read (SocketInputStream.java:129) at com.jcraft.jsch.IO.getByte(Unknown Source) at com.jcraft.jsch.Session.read(Unknown Source) at com.jcraft.jsch.Session.run(Unknown Source) at java.lang.Thread.run(Thread.java:552) main prio=5 tid=0x00500e30 nid=0x1804600 in Object.wait() [f07fe000..f08002c8] at java.lang.Object.wait(Native Method) - waiting on 0x650a5f08 (a com.jcraft.jsch.Channel $MyPipedInputStream) at java.io.PipedInputStream.read(PipedInputStream.java: 229) - locked 0x650a5f08 (a com.jcraft.jsch.Channel $MyPipedInputStream) at org.apache.maven.wagon.providers.ssh.ScpWagon.checkAck (ScpWagon.java:614) at org.apache.maven.wagon.providers.ssh.ScpWagon.get (ScpWagon.java:407) - 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] - 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]
Need comments on MNG-897, Ant support
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Hi, I know that many people are keen to have Ant supported as a plugin language in Maven 2.0. To this end, I'm working on adding native support for Ant, but I need some feedback on the approach I'm taking for binding the script into the M2 system. I've detailed my approach in the above issue, which can be found here: http://jira.codehaus.org/browse/MNG-897 Please feel free to comment on this. I'd like to get a feel for what others think might make binding build.xml files into M2 easier and more natural. Thanks in advance, john -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.6 (GNU/Linux) iD8DBQFDOBotK3h2CZwO/4URArPfAKCw4XZB6Peys69xIOgpBu/DgRZdPACeNxMK uaidM8rl1aGAI+su+yAnDFM= =AHED -END PGP SIGNATURE- - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Duplicate files and Doxia mojo
What the solution for dealing with supposedly duplicate files. We get the following error INFO - Caused by: org.apache.maven.reporting.MavenReportException: Some files are duplicates in the site directory or in the generated-site directory. INFO - Review the following files for the Default version: INFO - visioconvert/snddebugvm1web_files/snddebugvm1web_files/gif_1 INFO - resources/VisioConvert/SNDDebugVM1Web_files/SNDDebugVM1Web_files/gif_1.gif INFO - resources/VisioConvert/SNDDebugVM1Web_files/SNDDebugVM1Web_files/gif_1.js INFO - resources/VisioConvert/SNDDebugVM1Web_files/SNDDebugVM1Web_files/gif_1.htm INFO - visioconvert/snddebugvm1web_files/snddebugvm1web_files/vml_1 INFO - resources/VisioConvert/SNDDebugVM1Web_files/SNDDebugVM1Web_files/vml_1.js INFO - resources/VisioConvert/SNDDebugVM1Web_files/SNDDebugVM1Web_files/vml_1.htm INFO - resources/VisioConvert/SNDDebugVM1Web_files/SNDDebugVM1Web_files/vml_1.emz INFO - at org.apache.maven.doxia.DoxiaMojo.execute(DoxiaMojo.java:305) INFO - ... 16 more Now if the files were true duplicates I can understand, however when saving Visio and we have encountered it with or technology like flash, that the software will save all files with a common prefix. However the current site plugin doesn't seem to like that and it is tedious to say the least to have to keep changing the internal references for these files. Are we missing something big or what. We are primarily using APT, however these files are saved under the resources folder and I would expect all of those files to simple be copied to the output directory. Wb - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[m2] war:inplace
I didn't see this discussed anywhere. Are there plans to bring back this functionality on the Maven 2 plugin? If so, is there a proposed timeline? If not, why not? Should I put a feature request against the war plugin into JIRA? -- Stephen Duncan Jr www.stephenduncanjr.com - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: [m2] war:inplace
On 26/09/05, Stephen Duncan [EMAIL PROTECTED] wrote: I didn't see this discussed anywhere. Are there plans to bring back this functionality on the Maven 2 plugin? If so, is there a proposed timeline? If not, why not? Should I put a feature request against the war plugin into JIRA? Currently the m2 war plugin has a mode config param which currently supports 'war', 'exploded' or 'inplace'. The exploded mode has recently been deprecated since it breaks the install/deploy goals. Since inplace also breaks the install/deploy goals I expect this to go the same way. Does everyone think it's wise to remove these different modes of operation? I understand they break the install/deploy goals, but can't help thinking that removing them will break a lot more people's in-development build cycle. I know the build cycle for some of my projects will take considerably longer (currently an extra 8secs) zipping up and deploying a 15mb war into a container, which soon mounts up when you're doing it all day.. :( Mark - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: [m2] war:inplace
I can't see leaving them as modes that break the normal goals. As an additional goal of the plugin that you can call just to get your dependencies (and your compiled classes, though I rely on my IDE to take care of that) so that I can have my normal development process of running the webapp directly from my Eclipse workspace is a must. I think the M1 behavior is much preferred here. I guess I could always manually move the files over from the target directory to the src WEB-INF/lib and move on from there. But it'd be nice for it to be an automated part of the plugin... My only other issue is that I need to write a plugin to clean-up my Tomcat META-INF/context.xml file to fix the DocBase to match the war file name, change debugging settings, etc. and have that run during the WAR process. On 9/26/05, Mark Hobson [EMAIL PROTECTED] wrote: On 26/09/05, Stephen Duncan [EMAIL PROTECTED] wrote: I didn't see this discussed anywhere. Are there plans to bring back this functionality on the Maven 2 plugin? If so, is there a proposed timeline? If not, why not? Should I put a feature request against the war plugin into JIRA? Currently the m2 war plugin has a mode config param which currently supports 'war', 'exploded' or 'inplace'. The exploded mode has recently been deprecated since it breaks the install/deploy goals. Since inplace also breaks the install/deploy goals I expect this to go the same way. Does everyone think it's wise to remove these different modes of operation? I understand they break the install/deploy goals, but can't help thinking that removing them will break a lot more people's in-development build cycle. I know the build cycle for some of my projects will take considerably longer (currently an extra 8secs) zipping up and deploying a 15mb war into a container, which soon mounts up when you're doing it all day.. :( Mark - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] -- Stephen Duncan Jr www.stephenduncanjr.com - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: [m2] war:inplace
On 26/09/05, Stephen Duncan [EMAIL PROTECTED] wrote: My only other issue is that I need to write a plugin to clean-up my Tomcat META-INF/context.xml file to fix the DocBase to match the war file name, change debugging settings, etc. and have that run during the WAR process. Sounds like you need context.xml to be filtered - see MNG-791. This will allow you to specify tokens in your context.xml and their values in an associated filter.properties file. You can then supply different filter.properties files depending on which profiles are activated. Mark - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
RE: [m1 or 2] Odd project structure... how much pain will this be?
It sounds to me like two types of documentation are needed. One is the/an
ideal Maven Way of setting up a consultant's multi-client project,
assuming you are starting fresh and know you will be using Maven. I
personally think that Maven is ideal for exactly that kind of environment,
but then I'm already among the converted. :-)
The second task is providing migration examples, such as those sought in the
original query on this thread. This is substantially harder, because the
starting points can vary rather dramatically. The question comes up often on
this list, so perhaps the best thing to do is provide a centralized FAQ-type
page in the documentation with examples and solutions drawn directly from
the list.
The simplest first approach might be just a list of links to the appropriate
threads in the list archives. Add it to the Migrating from Ant section of
the existing documentation, perhaps. Fleshing those out and cleaning them up
would of course be great, but people are busy. Adding a link in the
documentation each time a new variant of the migration question comes up
would be relatively quick and might be sufficient.
I notice Strategy for migrating from Ant builds is one of the items listed
as needed documentation for M2. I did a quick search on migrate ant in the
archives and turned up a reasonable starting set, most of which are for M1.
I haven't moved to M2 yet myself, but if there are no strong objections,
I'll take a shot at assembling and organizing the links for the gurus to
look over and decide whether they are helpful as-is or in need of cleaning.
The best thing would be for the people who asked for the help originally to
write up and submit what it was they finally did that worked for them...
assuming they did go with Maven in the end. This is probably too much to
expect, but if we had a centralized set of links for the migration problem
it would make it easier to send follow-up emails to the originator later and
perhaps find out what happened. Exit interviews are quite useful when trying
to improve marketing.
Jay
-Original Message-
From: Andy Glick [mailto:[EMAIL PROTECTED]
Sent: Sunday, September 25, 2005 4:16 PM
To: users@maven.apache.org
Subject: Re: [m1 or 2] Odd project structure... how much pain will this be?
Brett Porter wrote:
Is it worthwhile publishing a few documents that show how various
project types would be set up for m2, like this?
- Brett
On 9/24/05, Kenney Westerhof [EMAIL PROTECTED] wrote:
On Fri, 23 Sep 2005, Eric Biesterfeld wrote:
You're pretty much set up for maven use, except for the 'overlay' system.
I'll demonstrate using maven2.
I see:
/base/pom.xml - base project
/base/client/pom.xml - grouping project for all clients, has parent
../pom.xml /base/client/X/pom.xml - project for client X, has parent
../pom.xml
As you compile the client/a/java/* to the base/target/ I expect those
classes don't 'overlay' existing ones, because your build will break -
the base classes are always newer than the client/a sourcefiles.
So now you get 1 + count(clients) jars. For each client, that's 2
jars. If you can live with that, then so far it's easy. If not, you
can add an 'assembly' goal to each client/a/pom.xml (ofcourse
configuration specified in client/pom.xml) to merge the two jars.
Btw, each client/a/pom.xml has a dependency on base/pom.xml.
For the property/html overlay: the property part (which is probably
going to end up in a jar /META-INF/ somewhere?) the assembly plugin
and/or the resources plugin could take care of that.
You need to use the unpack goal to unpack the base dependency into
client/a/target/classes/ (${project.build.outputDirectory} (or another
location if they shouldn't end up in the jar).
This goal is bound to the 'generate-resources' lifecycle phase.
Then maven2 will copy your client a's resources to the same location,
overwriting the other ones.
Or you could just specify a resources section in the client a pom
that has '../../html' in it. Be sure to specify that section before
the resources section that defines the client a resources.
That order will have the effect of client a's resources overwriting
the existing ones.
But it would ofcourse be best to have disjunct sets of resources for
base and the client projects, so that the client projects just 'add'
to the base project, not change it's behavior.
Maybe you can factor out the common features in the client projects,
and make projects for each of those features.
Then each client project just depends on the features that are
appropriate for that client. But it's probable that this won't work in
your particular case.
So, here are some ideas. Hope it helps!
I've worked at a number of shops that followed development practices similar
to these, where they had base project contents and by customer
customizations.
So, yes, I think that publishing documents, or even full or partial project
mockups that demonstrated various strategies for the
RE: [m1.1b1] IntelliJ IDEA plugin with Maven 1.1beta1
Can you open an issue on Jira please, if this one doesn't yet exist. Arnaud Hi, when generating the iml file with Maven 1.1beta1, I always get the following error when using that iml file: Cannot determine module type for the following module: mymodule. Assuming it is a JAVA module. But this assuming does not seem to work properly. The current workaround I have found is to manually edit the iml file as follows: Original: module version=4 relativePaths=true type= Add JAVA_MODULE as type: module version=4 relativePaths=true type=JAVA_MODULE please fix this, regards, Wim - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
FW: [m2b1] tests not seeing classes?
I found the problem. It had nothing to do with maven and everything to do with the VM's incompetent error reporting. The problem was that the DateLiteralBean class below linked to classes in jarA which linked to classes in jarB. JarB was not in the classpath. -Original Message- From: Mike Perham Sent: Saturday, September 24, 2005 11:07 AM To: Maven Users List Subject: [m2b1] tests not seeing classes? I have a project which is compiling some Hibernate beans into target/classes and then running an integration test with Derby to ensure that the database creation and simple queries work, etc. 2750 [main] ERROR net.sf.hibernate.proxy.LazyInitializer - CGLIB Enhancement failed java.lang.NoClassDefFoundError: com.webify.framework.triples.beans.DateLiteralBean at java.lang.Class.getDeclaredMethods0(Native Method) at java.lang.Class.privateGetDeclaredMethods(Class.java:1655) at java.lang.Class.getDeclaredMethod(Class.java:1262) at net.sf.cglib.proxy.Enhancer.getCallbacksSetter(Enhancer.java:558) But when I look in target/classes, the classes are right where they are expected. Directory of ...\target\classes\com\webify\framework\triples\beans 09/23/2005 11:41 AM 1,208 DateLiteralBean.class 09/23/2005 11:41 AM 1,336 DateLiteralBean.hbm.xml I suspect Maven2 is the problem only because this project compiles and the tests pass perfectly when I use our current ant-based build system. I tried -X to get more debugging but it didn't print out the test classpath. Is there a way to see the surefire classpath when the tests are run? Any advice on how to debug this issue would be appreciated. mike - 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: [m2] war:inplace
I'm not really sure of the best way. Right now, in src/main/webapps/META-INF/context.xml I have the docBase hardcoded to the full path to my local workspace where I'm developing from. In my previous build systems (Ant and Maven 1), as part of the war packaging system, I used Ant's XSLT task and an XSLT file targetting the context.xml file to replace those hardcoded values to the desired values (name of the war without .war for the docbase). I plan to implement this functionality as a Maven 2 plugin once I figure out how to develop a Maven 2 plugin. This approach allows me to have just one context.xml file that I actually touch. Filtering would require me to have two files: the one with the filter tokens, and the one with those tokens replaced by my local settings to use to deploy to Tomcat for development. It would, however, prevent me from CVS'ing my local settings... -Stephen On 9/26/05, Mark Hobson [EMAIL PROTECTED] wrote: On 26/09/05, Stephen Duncan [EMAIL PROTECTED] wrote: My only other issue is that I need to write a plugin to clean-up my Tomcat META-INF/context.xml file to fix the DocBase to match the war file name, change debugging settings, etc. and have that run during the WAR process. Sounds like you need context.xml to be filtered - see MNG-791. This will allow you to specify tokens in your context.xml and their values in an associated filter.properties file. You can then supply different filter.properties files depending on which profiles are activated. Mark -- Stephen Duncan Jr www.stephenduncanjr.com - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[m2] Xmlbeans v1 plugin?
Jason, your plugin matrix just has a XML Beans plugin listed but there are two flavors, just like with Xdoclet, v1 and v2. The plugin you listed is for XMLBeans v2, do you know if there is a v1 plugin available? http://docs.codehaus.org/display/MAVEN/Maven+Plugin+Matrix mike - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: [m2] Xmlbeans v1 plugin?
On Mon, 2005-09-26 at 13:27 -0500, Mike Perham wrote: Jason, your plugin matrix just has a XML Beans plugin listed but there are two flavors, just like with Xdoclet, v1 and v2. The plugin you listed is for XMLBeans v2, do you know if there is a v1 plugin available? http://docs.codehaus.org/display/MAVEN/Maven+Plugin+Matrix I'm working with Brett, John and everyone else to catalog the plug-ins but as far as the XMLBeans plug-in goes I don't believe there is a v1 for m2. But David Jencks or Brett can correct me on that. But I'm looking inside the mojo repository and all I see is an xmlbeans plugin that uses XMLBeans 2.0 so I'm pretty sure there isn't a version that uses XMLBeans 1.x. mike - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] -- jvz. Jason van Zyl jason at maven.org http://maven.apache.org - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
RE: RE : CruiseControl issues trying to install and run from maven.Pleasehelp.
I don't understand this, but all I did was create a batch file that called
the original maven.bat, and used that in my config.xml, and it started
working.
Thank You
Mick Knutson
Sr. Java/J2EE Consultant
BASE logic, inc.
(415) 648-1804 (S.F., CA)
http://www.BASELogic.com
HP Consulting Services (Walnut Creek, CA)
From: David Jackman [EMAIL PROTECTED]
To: Mick Knutson [EMAIL PROTECTED]
Subject: RE: RE : CruiseControl issues trying to install and run from
maven.Pleasehelp.
Date: Mon, 26 Sep 2005 13:35:20 -0600
No idea on that one. I've never used log4j.xml
-Original Message-
From: Mick Knutson [mailto:[EMAIL PROTECTED]
Sent: Monday, September 26, 2005 1:33 PM
To: David Jackman
Subject: RE: RE : CruiseControl issues trying to install and run from
maven.Pleasehelp.
where do I put the logger XML properties if I am using log4j.xml for
cruiseControl?
Thank You
Mick Knutson
Sr. Java/J2EE Consultant
BASE logic, inc.
(415) 648-1804 (S.F., CA)
http://www.BASELogic.com
HP Consulting Services (Walnut Creek, CA)
From: David Jackman [EMAIL PROTECTED]
To: Mick Knutson [EMAIL PROTECTED]
Subject: RE: RE : CruiseControl issues trying to install and run from
maven.Pleasehelp.
Date: Mon, 26 Sep 2005 12:28:27 -0600
1. CC uses Log4j, so your log4j.properties file can indicate the
logging
level. My log4j.properties looks like this:
log4j.rootCategory=INFO,FILE,HUB
# Log errors to the console (for the CruiseControl service, the
console output is sent to service_out.log and service_err.log)
log4j.appender.Console=org.apache.log4j.ConsoleAppender
log4j.appender.Console.Threshold=ERROR
log4j.appender.Console.layout=org.apache.log4j.PatternLayout
log4j.appender.Console.layout.ConversionPattern=[cc]%d{MMM-dd
HH:mm:ss} %-14.14c{1}- %m%n
# FILE is file logger with rotation
log4j.appender.FILE=org.apache.log4j.RollingFileAppender
log4j.appender.FILE.Threshold=INFO
log4j.appender.FILE.layout=org.apache.log4j.PatternLayout
log4j.appender.FILE.layout.ConversionPattern=%d [%-9t] %-5p
%-16c{1} %x- %m%n
log4j.appender.FILE.File=cruisecontrol.log
log4j.appender.FILE.MaxFileSize=5000KB
log4j.appender.FILE.MaxBackupIndex=4
# HUB is a socket hub appender
log4j.appender.HUB=org.apache.log4j.net.SocketHubAppender
log4j.appender.HUB.Threshold=DEBUG
log4j.appender.HUB.layout=org.apache.log4j.PatternLayout
log4j.appender.HUB.layout.ConversionPattern=[cc]%d{MMM-dd
HH:mm:ss} %-14.14c{1}- %m%n
log4j.appender.HUB.port=4445
Change the log4j.appender.*.Threshold to DEBUG to get debug logging.
Yours doesn't have to have 3 separate appenders. The Console one is
enough to see what's going on when it happens.
2. Yes. Create a Maven.bat file that looks something like this:
echo C:\Maven_1.0.2\bin\maven.bat %*
call C:\Maven_1.0.2\bin\maven.bat %*
Then change the mavenscript attribute in your config.xml to point to
the
new Maven.bat instead.
Hope this helps
..David..
-Original Message-
From: Mick Knutson [mailto:[EMAIL PROTECTED]
Sent: Monday, September 26, 2005 12:04 PM
To: David Jackman
Subject: RE: RE : CruiseControl issues trying to install and run from
maven.Pleasehelp.
This seems to be only for running Maven by itself.
1. How do I change the log level of CC?
2. Can I turn on maven debug in a BAT files similiar to what you just
gave, but running through CC?
Thank You
Mick Knutson
Sr. Java/J2EE Consultant
BASE logic, inc.
(415) 648-1804 (S.F., CA)
http://www.BASELogic.com
HP Consulting Services (Walnut Creek, CA)
From: David Jackman [EMAIL PROTECTED]
To: [EMAIL PROTECTED]
Subject: RE: RE : CruiseControl issues trying to install and run from
maven.Pleasehelp.
Date: Fri, 23 Sep 2005 08:29:34 -0600
Well, I can't see anything here that jumps out, either. What I
really
want to look at now is the command line CC uses to start maven. I
thought it would be in the log, but I guess not. I created my own
Maven
script that looks like this:
@echo off
echo C:\Maven_1.0.2\bin\maven.bat %*
call C:\Maven_1.0.2\bin\maven.bat %*
so the command line will be sent to stdout. You will also get a copy
of
the command line in the log if you up the logging level to debug
instead
of info (you'll get a load of other stuff, too, which may be helpful
here).
The command line for Maven that I get looks something like this:
C:\Maven_1.0.2\bin\maven.bat
-Dfoundation.src.changed=true
-Dcclastbuildtimestamp=20050921230327
-Dlabel=220
-Dcclastgoodbuildtimestamp=20050921230327
-Dlastbuildsuccessful=true
-Dcvstimestamp=2005-09-22 13:26:13 GMT
-Dcctimestamp=20050922072613
-b -p project.xml multiproject:deploy multiproject:site
site:deploy
Based on the error you're getting, I think that either the
cclastbuildtimestamp or cclastgoodbuildtimestamp properties aren't
getting set correctly. There's probably a space before the value in
the
[m2] Newbie Questions
So, I'm trying to get started with Maven 2, and I've been finding that even with a skeletal project, a lot of the plugins just don't seem to work. I'm particularly curious about these: [INFO] [INFO] Building System Core [INFO]task-segment: [javadoc:javadoc] [INFO] [INFO] [javadoc:javadoc] [INFO] C:\Program Files\Java\jdk1.5.0_04\jre\..\bin\javadoc.exe -package -sourcePath C:\checkouts\system-core\src -classpath C:\checkouts\system-core\target\classes;C:\repository\junit\junit\3.8.\j unit-3.8.1.jar;C:\repository\log4j\log4j\1.2.8\log4j-1.2.8.jar -author -bottom Copyright ... All Rights Reserved. -charset ISO-8859-1 -d C:\checkouts\system-core\target\javadoc\apidocs -doctitle System Core 1.3.1-DEV API -stylesheetfile C:\checkouts\system-core\target\javadoc\apidocs\stylesheet.css -use -version -windowtitle System Core 1.3.1-DEV API @files 'C:\Program' is not recognized as an internal or external command, operable program or batch file. [INFO] [ERROR] BUILD ERROR [INFO] [INFO] Diagnosis: An error has occurred in JavaDocs report generation. [INFO] [INFO] [INFO] Building System Core [INFO]task-segment: [clover:check] [INFO] [INFO] [clover:instrument] Clover Version 1.3.9, built on July 06 2005 loaded from: C:\repository\clover\clover\1.3.9\clover-1.3.9.jar 30 day Evaluation Version distributed via the Maven Jar Repository. Clover is not free. You have 30 days to evaluate it. After this, please visit http://www.cenqua.com to obtain a licensed version of Clover Updating database at 'C:\checkouts\system-core\target/clover/clover.db' Instrumented 53 source files. [INFO] [resources:resources] [INFO] [compiler:compile] Compiling 53 source files to C:\checkouts\system-core\target\clover\classes [INFO] [INFO] BUILD FAILURE [INFO] [INFO] Reason: Compilation failure [INFO] [INFO] C:\checkouts\system-core\target\clover\src\com\cerner\system\util\Thread UUID.java:[22,52] package com_cenqua_clover does not exist ...repeat ad nauseam... Can anyone give me some pointers on why these might be happening? Matthew Beermann | Software Engineer | Cerner Corporation | 816.201.7026 | [EMAIL PROTECTED] | www.cerner.com CONFIDENTIALITY NOTICE This message and any included attachments are from Cerner Corporation and are intended only for the addressee. The information contained in this message is confidential and may constitute inside or non-public information under international, federal, or state securities laws. Unauthorized forwarding, printing, copying, distribution, or use of such information is strictly prohibited and may be unlawful. If you are not the addressee, please promptly delete this message and notify the sender of the delivery error by e-mail or you may call Cerner's corporate offices in Kansas City, Missouri, U.S.A at (+1) (816)221-1024. --
Re: [m1 or 2] Odd project structure... how much pain will this be?
We're slated to talk about it tomorrow, and we're planning from then.
I had originally discounted Maven 1 because the multiproject system
looked so messy (and I really like transitive dependencies) but maven
2 beta looks like it's sufficent for our needs.
I'm still wary about the lack of documentation, but if we do go with
Maven I will write up what we've done as a case study.
On 9/26/05, Jay H. Hartley [EMAIL PROTECTED] wrote:
It sounds to me like two types of documentation are needed. One is the/an
ideal Maven Way of setting up a consultant's multi-client project,
assuming you are starting fresh and know you will be using Maven. I
personally think that Maven is ideal for exactly that kind of environment,
but then I'm already among the converted. :-)
The second task is providing migration examples, such as those sought in the
original query on this thread. This is substantially harder, because the
starting points can vary rather dramatically. The question comes up often on
this list, so perhaps the best thing to do is provide a centralized FAQ-type
page in the documentation with examples and solutions drawn directly from
the list.
The simplest first approach might be just a list of links to the appropriate
threads in the list archives. Add it to the Migrating from Ant section of
the existing documentation, perhaps. Fleshing those out and cleaning them up
would of course be great, but people are busy. Adding a link in the
documentation each time a new variant of the migration question comes up
would be relatively quick and might be sufficient.
I notice Strategy for migrating from Ant builds is one of the items listed
as needed documentation for M2. I did a quick search on migrate ant in the
archives and turned up a reasonable starting set, most of which are for M1.
I haven't moved to M2 yet myself, but if there are no strong objections,
I'll take a shot at assembling and organizing the links for the gurus to
look over and decide whether they are helpful as-is or in need of cleaning.
The best thing would be for the people who asked for the help originally to
write up and submit what it was they finally did that worked for them...
assuming they did go with Maven in the end. This is probably too much to
expect, but if we had a centralized set of links for the migration problem
it would make it easier to send follow-up emails to the originator later and
perhaps find out what happened. Exit interviews are quite useful when trying
to improve marketing.
Jay
-Original Message-
From: Andy Glick [mailto:[EMAIL PROTECTED]
Sent: Sunday, September 25, 2005 4:16 PM
To: users@maven.apache.org
Subject: Re: [m1 or 2] Odd project structure... how much pain will this be?
Brett Porter wrote:
Is it worthwhile publishing a few documents that show how various
project types would be set up for m2, like this?
- Brett
On 9/24/05, Kenney Westerhof [EMAIL PROTECTED] wrote:
On Fri, 23 Sep 2005, Eric Biesterfeld wrote:
You're pretty much set up for maven use, except for the 'overlay' system.
I'll demonstrate using maven2.
I see:
/base/pom.xml - base project
/base/client/pom.xml - grouping project for all clients, has parent
../pom.xml /base/client/X/pom.xml - project for client X, has parent
../pom.xml
As you compile the client/a/java/* to the base/target/ I expect those
classes don't 'overlay' existing ones, because your build will break -
the base classes are always newer than the client/a sourcefiles.
So now you get 1 + count(clients) jars. For each client, that's 2
jars. If you can live with that, then so far it's easy. If not, you
can add an 'assembly' goal to each client/a/pom.xml (ofcourse
configuration specified in client/pom.xml) to merge the two jars.
Btw, each client/a/pom.xml has a dependency on base/pom.xml.
For the property/html overlay: the property part (which is probably
going to end up in a jar /META-INF/ somewhere?) the assembly plugin
and/or the resources plugin could take care of that.
You need to use the unpack goal to unpack the base dependency into
client/a/target/classes/ (${project.build.outputDirectory} (or another
location if they shouldn't end up in the jar).
This goal is bound to the 'generate-resources' lifecycle phase.
Then maven2 will copy your client a's resources to the same location,
overwriting the other ones.
Or you could just specify a resources section in the client a pom
that has '../../html' in it. Be sure to specify that section before
the resources section that defines the client a resources.
That order will have the effect of client a's resources overwriting
the existing ones.
But it would ofcourse be best to have disjunct sets of resources for
base and the client projects, so that the client projects just 'add'
to the base project, not change it's behavior.
Maybe you can factor out the common features in the client projects,
and make projects for each of those
how to specify where my ear file is for JBoss to deploy?
I have a sub-project-ear that my maven script creates my final ear, and I want to know how the jboss plugin knows that it is in a sub-project, not my main project? Thank You Mick Knutson Sr. Java/J2EE Consultant BASE logic, inc. (415) 648-1804 (S.F., CA) http://www.BASELogic.com HP Consulting Services (Walnut Creek, CA) - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: FW: [m2b1] tests not seeing classes?
Mike Perham wrote: I found the problem. It had nothing to do with maven and everything to do with the VM's incompetent error reporting. The problem was that the DateLiteralBean class below linked to classes in jarA which linked to classes in jarB. JarB was not in the classpath. 2750 [main] ERROR net.sf.hibernate.proxy.LazyInitializer - CGLIB Enhancement failed java.lang.NoClassDefFoundError: com.webify.framework.triples.beans.DateLiteralBean at java.lang.Class.getDeclaredMethods0(Native Method) at java.lang.Class.privateGetDeclaredMethods(Class.java:1655) at java.lang.Class.getDeclaredMethod(Class.java:1262) at You might want to reconsider your evaluation of the VM's error reporting. If it were merely a question of not finding com.webify.framework.triples.beans.DateLiteralBean, then you would have seen java.lang.ClassNotFoundException. Instead the VM produced java.lang.NoClassDefFoundError. In my experience the error report and the Javadoc are somewhat misleading, but each provides information that should allow you to diagnose the problem, if you look hard enough, which you did. From the Javadoc: Thrown if the Java Virtual Machine or a ClassLoader instance tries to load in the definition of a class (as part of a normal method call or as part of creating a new instance using the new expression) and no definition of the class could be found. The searched-for class definition existed when the currently executing class was compiled, but the definition can no longer be found. The misleading aspect of the error message is as follows: when a classloader goes to load a class that is visible to it, you will see the NoClassDefFoundError under but not limited to the following circumstances: 1) the named class or some other class being loaded as a consequence of loading the named class is visible to the classloader and 1 or more methods or fields referenced in that class do not exist in the the byte code being loaded 2) some other class or possibly a library or resource referenced in the newly loaded class's implicit class graph cannot be found and loaded. When the error message or the Javadoc state the class they are being overly specific, any downstream loader failure during the load of the named class will cause NoClassDefFoundError to be thrown. If you interpret this error message to mean java.lang.ClassNotFoundException, then yes the message error reporting is incompetent, but if you interpret the message to mean, that the classLoader attempted to load the class named com.webify.framework.triples.beans.DateLiteralBean and at some point during the class' init or instance's ctor initialization an unspecified loader error was detected, then its not exactly incompetent, its just not very helpful. - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: Delivering mock objects for a public API [was Re: [m2] custom compiler mojo]
On 9/25/05, Trygve Laugstøl [EMAIL PROTECTED] wrote: On Sat, 2005-09-24 at 04:31 +, John Fallows wrote: On 9/23/05, Trygve Laugstøl [EMAIL PROTECTED] wrote: On Fri, 2005-09-23 at 08:03 +, John Fallows wrote: I've created a custom compiler mojo that extends AbstractCompilerMojo from the maven-compiler-plugin. Is there a special reason for this? We where hoping that a new Compiler implementation would suffice. The Compiler interface is defined in plexus-compiler-api and you can select your own implementation by setting the compilerId flag. [snip] In the snipped out part I understand what your issue is and agree that it is a real issue. Perhaps there is a better approach? The issue is not specific to mock objects, we could use the same thing for our TCK type tests that we have for Maven SCM. See [1] for a issue to follow (and vote for). I'm still not entirely sure how you would separate out the tests for the artifacts itself and the reusable tests. Anyway, if you are mocking a Compiler implementation I'd be interested to put that back into the Plexus repository. We'd need to decide on the correct lifecycle phase binding for the mock compiler. Strictly speaking, the input to the test-compile phase should include the compiled mock classes as a dependency. Presumably the default directories for source and target would be src/mock/java and target/mock-classes. We'd also want to take care of mock resources, making sure they are copied from src/mock/resources to target/mock-classes. This is sounding more like an extension of the current m2 lifecycle. validate generate-sources process-sources generate-resources process-resources compile process-classes generate-mock-sources process-mock-sources generate-mock-resources process-mock-resources mock-compile generate-test-sources process-test-sources generate-test-resources process-test-resources test-compile test package integration-test verify install deploy [1]: http://jira.codehaus.org/browse/MNG-932 Yes, I understand that it would be generally useful to deliver a subset of the unit test code as a JAR. The above proposal still stands though, although the name mock can be treated as a placeholder for that subset until we have a better name. Alternatively, is there any mileage in the idea of having completely separate sub-modules for main, mock, and test, all inside the api-module? Kind Regards, John Fallows Kind Regards, John Fallows.
Re: [m2] Newbie Questions
On Mon, 2005-09-26 at 16:01 -0500, Beermann,Matthew wrote: So, I'm trying to get started with Maven 2, and I've been finding that even with a skeletal project, a lot of the plugins just don't seem to work. I'm particularly curious about these: Did you use the suggested method of getting started in the getting started guide? Just wondering what your project looks like. -- jvz. Jason van Zyl jason at maven.org http://maven.apache.org We know what we are, but know not what we may be. -- Shakespeare - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: [m2] Newbie Questions
The first bug is already open (try setting java_home to the short path without spaces). It's only on Windows when Java is in program files. The second is fixed with the upcoming beta-2 release. If you find a bug that is not in JIRA already, please let us know! - Brett On 9/27/05, Beermann,Matthew [EMAIL PROTECTED] wrote: So, I'm trying to get started with Maven 2, and I've been finding that even with a skeletal project, a lot of the plugins just don't seem to work. I'm particularly curious about these: [INFO] [INFO] Building System Core [INFO]task-segment: [javadoc:javadoc] [INFO] [INFO] [javadoc:javadoc] [INFO] C:\Program Files\Java\jdk1.5.0_04\jre\..\bin\javadoc.exe -package -sourcePath C:\checkouts\system-core\src -classpath C:\checkouts\system-core\target\classes;C:\repository\junit\junit\3.8.\j unit-3.8.1.jar;C:\repository\log4j\log4j\1.2.8\log4j-1.2.8.jar -author -bottom Copyright ... All Rights Reserved. -charset ISO-8859-1 -d C:\checkouts\system-core\target\javadoc\apidocs -doctitle System Core 1.3.1-DEV API -stylesheetfile C:\checkouts\system-core\target\javadoc\apidocs\stylesheet.css -use -version -windowtitle System Core 1.3.1-DEV API @files 'C:\Program' is not recognized as an internal or external command, operable program or batch file. [INFO] [ERROR] BUILD ERROR [INFO] [INFO] Diagnosis: An error has occurred in JavaDocs report generation. [INFO] [INFO] [INFO] Building System Core [INFO]task-segment: [clover:check] [INFO] [INFO] [clover:instrument] Clover Version 1.3.9, built on July 06 2005 loaded from: C:\repository\clover\clover\1.3.9\clover-1.3.9.jar 30 day Evaluation Version distributed via the Maven Jar Repository. Clover is not free. You have 30 days to evaluate it. After this, please visit http://www.cenqua.com to obtain a licensed version of Clover Updating database at 'C:\checkouts\system-core\target/clover/clover.db' Instrumented 53 source files. [INFO] [resources:resources] [INFO] [compiler:compile] Compiling 53 source files to C:\checkouts\system-core\target\clover\classes [INFO] [INFO] BUILD FAILURE [INFO] [INFO] Reason: Compilation failure [INFO] [INFO] C:\checkouts\system-core\target\clover\src\com\cerner\system\util\Thread UUID.java:[22,52] package com_cenqua_clover does not exist ...repeat ad nauseam... Can anyone give me some pointers on why these might be happening? Matthew Beermann | Software Engineer | Cerner Corporation | 816.201.7026 | [EMAIL PROTECTED] | www.cerner.com CONFIDENTIALITY NOTICE This message and any included attachments are from Cerner Corporation and are intended only for the addressee. The information contained in this message is confidential and may constitute inside or non-public information under international, federal, or state securities laws. Unauthorized forwarding, printing, copying, distribution, or use of such information is strictly prohibited and may be unlawful. If you are not the addressee, please promptly delete this message and notify the sender of the delivery error by e-mail or you may call Cerner's corporate offices in Kansas City, Missouri, U.S.A at (+1) (816)221-1024. -- - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: [m2] properties
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
this should work now (committed it this afternoon). However, I should
note that in Chris' email he was using model properties (POM-level
properties) to set the version of the POM...to me, this has a bad smell,
since the project version may be influenced by active profiles, etc. way
too easily, which in turn may cause a reproducability problem.
With model properties and profile properties, you should be really
careful to avoid letting users override the sort of core project
information (groupId, artifactId, version, packaging, etc.) by simply
providing a profile in their settings.xml...at least, IMHO.
Cheers,
John
Brett Porter wrote:
| I guess a lot of people are going to want to use dotted properties, so
| we should support it.
|
| It probably is attempting to do reflection on m2.
|
| - Brett
|
|
| On 9/26/05, Chris Berry [EMAIL PROTECTED] wrote:
|
|Hi John,
|Don't know if it's a bug but
|properties
|m2.version2.0-beta-2-SNAPSHOT/m2.version
|/properties
|.
|version${m2.version}/version
|
|does NOT work.
|
|But, without the . in the property name, it does work. I.e.
|properties
|m2version2.0-beta-2-SNAPSHOT/m2version
|/properties
|.
|version${m2version}/version
|This works.
|Cheers,
|-- Chris
|
|-
|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 v1.2.6 (GNU/Linux)
iD8DBQFDOJALK3h2CZwO/4URAvnbAJ44JReEYzlyF4xXj7yXkYZQygRQkQCggPlt
4Xwqr6brD0Ite8MQAdXE0yM=
=NauG
-END PGP SIGNATURE-
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
RE: using maven.test.skip
Hi, Andy,
Thanks a lot for the prompt reply! Using your technique, I have encountered
two issues during testing,
1) the variable maven.test.skip.value seems to be not presisting between
goals. E.g. after I set this variable in preGoal of war:install, then went
on to use a j:if test=empty(...)} to test it in preGoal of test:test, it
is always tested true.
I don't think I understand how to define a property for a maven plugin. Any
pointer on this subject is appreciated.
2) To proceed, I hardcoded the value in the maven:set/ tag.
It is something like:
maven:set plugin=maven-test-plugin
property=maven.test.skip
value=true/
Then there is an runtime error:
You must define an attribute called 'value' for this tag.
This is the way I defined the namespace for my maven plugin:
project
xmlns:ant=jelly:ant
xmlns:define=jelly:define
xmlns:j=jelly:core
xmlns:doc=doc
xmlns:maven=jelly:maven
I wonder if you may have any idea on why this error? Basically if I can
solve Q2 here, I can manage to roll my first ever custom-made maven plugin
:-)
BTW, I am using maven1.0.2
Cheers,
AK
-Original Message-
From: news [mailto:[EMAIL PROTECTED] On Behalf Of Andy Glick
Sent: Monday, 26 September 2005 11:38 AM
To: users@maven.apache.org
Subject: Re: using maven.test.skip
Anthony Kong wrote:
Hi, all,
I have written a custom plugin for a in-house project which will
produce a ear file at the end.
There is a number of goals defined in this plugin. One of these is:
goal name=projecct:ear-build description=Invokes ejb:install and
war:install goal
j:set var=maven.test.skip value=false/
ant:echo{maven.test.skip} is set to ${maven.test.skip}/ant:echo
attainGoal name=ejb:install/
j:if test=${maven.test.failure}
fail message=There were test failures!/
/j:if
j:set var=maven.test.skip value=true/
ant:echo{maven.test.skip} is set to ${maven.test.skip}/ant:echo
attainGoal name=war:install/
j:set var=maven.test.skip value=false/
/goal
Basically i want to skip the junit test in war:install (ejb:install
will invoke the junit test). However the trick does not work. The
junit test in war:install is not skipped, even though from ant:echo I
can see it is set to true. Any suggestion to how to debug/solve this
problem?
You might want to try the following:
preGoal name=ejb:install
j:set var=maven.test.skip.value value=false/ /preGoal
preGoal name=war:install
j:set var=maven.test.skip.value value=true/ /preGoal
preGoal name=test:test
maven:set plugin=maven-test-plugin var=maven.test.skip
value=${maven.test.skip.value}/
/preGoal
The reason that I would recommend this strategy is as follows:
1) you cannot rely upon a single setting of maven.test.skip.
2) because your effort to set the value of maven.test.skip before executing
war:install isn't working, I think that you'll have better results if you
inject the value of maven.test.skip into the maven-test-plugin
3) to do that maven:set is the preferred method, and the way that this set
of code ought to work is that you should be able to explicitly set the value
of maven.test.skip in the test plugin as you execute it
If this doesn't work, please try the following modification to the j:set
var tags:
j:set var=maven.test.skip.value value={appropriate value}
scope=parent/
and if that doesn't work, feel free to report the fact and we can try to
debug whatever it is that is going on.
Hope that this helps.
-
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: [m1.1b1] IntelliJ IDEA plugin with Maven 1.1beta1
Please try 1.1 beta-2 first. The IDEA plugin has been improved. - Brett On 9/27/05, Arnaud HERITIER [EMAIL PROTECTED] wrote: Can you open an issue on Jira please, if this one doesn't yet exist. Arnaud Hi, when generating the iml file with Maven 1.1beta1, I always get the following error when using that iml file: Cannot determine module type for the following module: mymodule. Assuming it is a JAVA module. But this assuming does not seem to work properly. The current workaround I have found is to manually edit the iml file as follows: Original: module version=4 relativePaths=true type= Add JAVA_MODULE as type: module version=4 relativePaths=true type=JAVA_MODULE please fix this, regards, Wim - 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: Duplicate files and Doxia mojo
This is a bug - it's already open on the site plugin. - Brett On 9/27/05, Wendell Beckwith [EMAIL PROTECTED] wrote: What the solution for dealing with supposedly duplicate files. We get the following error INFO - Caused by: org.apache.maven.reporting.MavenReportException: Some files are duplicates in the site directory or in the generated-site directory. INFO - Review the following files for the Default version: INFO - visioconvert/snddebugvm1web_files/snddebugvm1web_files/gif_1 INFO - resources/VisioConvert/SNDDebugVM1Web_files/SNDDebugVM1Web_files/gif_1.gif INFO - resources/VisioConvert/SNDDebugVM1Web_files/SNDDebugVM1Web_files/gif_1.js INFO - resources/VisioConvert/SNDDebugVM1Web_files/SNDDebugVM1Web_files/gif_1.htm INFO - visioconvert/snddebugvm1web_files/snddebugvm1web_files/vml_1 INFO - resources/VisioConvert/SNDDebugVM1Web_files/SNDDebugVM1Web_files/vml_1.js INFO - resources/VisioConvert/SNDDebugVM1Web_files/SNDDebugVM1Web_files/vml_1.htm INFO - resources/VisioConvert/SNDDebugVM1Web_files/SNDDebugVM1Web_files/vml_1.emz INFO - at org.apache.maven.doxia.DoxiaMojo.execute(DoxiaMojo.java:305) INFO - ... 16 more Now if the files were true duplicates I can understand, however when saving Visio and we have encountered it with or technology like flash, that the software will save all files with a common prefix. However the current site plugin doesn't seem to like that and it is tedious to say the least to have to keep changing the internal references for these files. Are we missing something big or what. We are primarily using APT, however these files are saved under the resources folder and I would expect all of those files to simple be copied to the output directory. Wb - 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: [m2] properties
Actually it was just a snippet ;-) The version was really meant to
be in a plugin definition (I just got lazy). Although I could see
one interested in, say, having a single inherited version when
building multiple projects -- so I could see it's legitimate use.
In the end, it's hard to build a powerful tool that isn't also
dangerous. You can cut your finger off w/ a skill saw, but that
doesn't make it any less useful and necessary -- it is assumed that
you will be careful.
Cheers,
-- Chris
On 9/26/05, John Casey [EMAIL PROTECTED] wrote:
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
this should work now (committed it this afternoon). However, I should
note that in Chris' email he was using model properties (POM-level
properties) to set the version of the POM...to me, this has a bad smell,
since the project version may be influenced by active profiles, etc. way
too easily, which in turn may cause a reproducability problem.
With model properties and profile properties, you should be really
careful to avoid letting users override the sort of core project
information (groupId, artifactId, version, packaging, etc.) by simply
providing a profile in their settings.xml...at least, IMHO.
Cheers,
John
Brett Porter wrote:
| I guess a lot of people are going to want to use dotted properties, so
| we should support it.
|
| It probably is attempting to do reflection on m2.
|
| - Brett
|
|
| On 9/26/05, Chris Berry [EMAIL PROTECTED] wrote:
|
|Hi John,
|Don't know if it's a bug but
|properties
|m2.version2.0-beta-2-SNAPSHOT/m2.version
|/properties
|.
|version${m2.version}/version
|
|does NOT work.
|
|But, without the . in the property name, it does work. I.e.
|properties
|m2version2.0-beta-2-SNAPSHOT/m2version
|/properties
|.
|version${m2version}/version
|This works.
|Cheers,
|-- Chris
|
|-
|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 v1.2.6 (GNU/Linux)
iD8DBQFDOJALK3h2CZwO/4URAvnbAJ44JReEYzlyF4xXj7yXkYZQygRQkQCggPlt
4Xwqr6brD0Ite8MQAdXE0yM=
=NauG
-END PGP SIGNATURE-
-
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]
[m2] Archetype Plugin parameters and templates
Hi, The M1 Genapp plugin offers the possibility to define custom parameters and templates... So here my questions about the M2 Archetype plugin: (1) Does it offer the same possibilities? (2) Is it possible to prompt the user for the parameter's value? (3) Where should I start to build my own template? (4) Finally, how to add a specific script/behaviour to a template in M2? Thanks, Eric
RE: using maven.test.skip
At 08:32 PM 9/26/2005, Anthony Kong wrote:
2) To proceed, I hardcoded the value in the maven:set/ tag.
It is something like:
maven:set plugin=maven-test-plugin
property=maven.test.skip
value=true/
Then there is an runtime error:
You must define an attribute called 'value' for this tag.
Anthony,
You might want to try adding the artifact namespace to your project tag. As
in xmlns:artifact=artifact. I have found that with M1 1.0.2 and 1.1b1
that the artifact namespace was necessary in order to set values in plugins.
I realized that I did have an existing project on which I could experiment.
I've included a fragment of a maven.xml file, but I'm using Maven 1.1b2 so
your mileage may vary.
The coverage goal will execute the jblanket, jcoverage, and emma coverage
tools. Coverage tools instrument the code under test and then run the
testcases and report the % of the CUT that is covered. In this case, using
preGoals from the coverage tools I first checked to see if
maven.test.skip.value was not empty and removed it if it was and then set
the value of maven.test.skip.value to true. In the postGoal for test:test I
changed the value of maven.test.skip.value to false. So in this example I
am varying the value of the Jelly variable serving as a sentinel and
setting the a plugin variable's value. If you are interested, I've got a
trace of the execution of this maven.xml file which I'm willing to mail to you.
Hope that this helps.
project default=coverage
xmlns:j=jelly:core
xmlns:license=license
xmlns:maven=jelly:maven
xmlns:util=jelly:util
xmlns:ant=jelly:ant
goal name=coverage
attainGoal name=jblanket:coverage/
attainGoal name=jcoverage/
attainGoal name=emma:report/
/goal
preGoal name=jblanket:coverage
echo message=value of maven.test.skip.value is
${maven.test.skip.value}/
j:if test=${!empty(maven.test.skip.value)}
j:remove var=maven.test.skip.value/
echo message=executing j:remove of maven.test.skip.value/
/j:if
j:set var=maven.test.skip.value value=true/
/preGoal
preGoal name=jcoverage:on
echo message=value of maven.test.skip.value is
${maven.test.skip.value}/
j:if test=${!empty(maven.test.skip.value)}
j:remove var=maven.test.skip.value/
echo message=executing j:remove of maven.test.skip.value/
/j:if
j:set var=maven.test.skip.value value=true/
/preGoal
preGoal name=emma:init
echo message=value of maven.test.skip.value is
${maven.test.skip.value}/
j:if test=${!empty(maven.test.skip.value)}
j:remove var=maven.test.skip.value/
echo message=executing j:remove of maven.test.skip.value/
/j:if
j:set var=maven.test.skip.value value=true/
/preGoal
preGoal name=test:test
echo message=value of maven.test.skip.value is
${maven.test.skip.value}/
maven:set plugin=maven-test-plugin property=maven.test.skip
value=${maven.test.skip.value}/
maven:get plugin=maven-test-plugin property=maven.test.skip
var=reported.maven.test.skip/
echo message=value of reported.maven.test.skip is
${reported.maven.test.skip}/
/preGoal
postGoal name=test:test
j:if test=${!empty(maven.test.skip.value)}
echo message=maven.test.skip.value is ${maven.test.skip.value}/
j:set var=maven.test.skip.value value=false/
/j:if
/postGoal
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
Re: [m2] Archetype Plugin parameters and templates
On Mon, 2005-09-26 at 21:49 -0400, Eric Jacob wrote: Hi, The M1 Genapp plugin offers the possibility to define custom parameters and templates... So here my questions about the M2 Archetype plugin: (1) Does it offer the same possibilities? Yes, the templates for a archetype are actually velocity templates so more flexible then what's in genapp. (2) Is it possible to prompt the user for the parameter's value? Archetypes were designed to do that. The parameters can come from anywhere as long they end up in a Map. The Map is used when rendering the velocity templates to parameterize the output. I would like to do more work on the archetypes in order to provide some metadata about the parameters for a particular archetype so that better UI tools can be made. But even with that being the case there is reason why you couldn't make a UI for the generation of a project from an archetype. (3) Where should I start to build my own template? You can look at the existing archetypes which are simply JARs that contain velocity templates and some metadata about what is actually in the archetype. I will add some doco to the getting started guide about archetypes as I think there is little known about them but they are going to be highly useful. If you follow the current getting started guide and create a project: m2 archetype:create -DgroupId=com.mycompany.app -DartifactId=my-app This will actually download the archetype to your local maven repository and you can take a look at it. It will be in: $repo/org/apache/maven/archetypes/maven-archetype-quickstart/1.0-alpha-3 (4) Finally, how to add a specific script/behaviour to a template in M2? The templates are velocity templates so you get the directives that velocity provides and you can make your own velocity macros or tools. A tool is simply a Java object that you drop into a velocity context so you can really do whatever you want. I would think velocity itself would give you enough flexibility. Thanks, Eric -- jvz. Jason van Zyl jason at maven.org http://maven.apache.org In short, man creates for himself a new religion of a rational and technical order to justify his work and to be justified in it. -- Jacques Ellul, The Technological Society - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
