As far as I've found, currently it is not possible to specify a
dependency scope which will cause a dependency normally packaged
along with its compile scope dependencies in a final product like
EAR, but will not resolve its compile scope dependencies when
compiling the project which includes
Wed, 22 Jun 2011 15:07:02 -0700, /Manfred Moser/:
Use a dependency to the api with provided scope on jar or whatever code
projects you have.
And then use the implementation of the api as dependency on the ear.
Yeah, that's exactly what I'm currently doing, but it easily becomes
a nightmare
Could someone tell me what's wrong with the attached POM. It
should basically build an EAR with application.xml defining
jboss-seam as only EJB module. The problem is I'm getting:
$ mvn package
[INFO] Scanning for projects...
[INFO]
Sun, 26 Jun 2011 12:59:10 +0300, /Stanimir Stamenkov/:
I've really included jboss-seam as a dependency and I even
expect I don't need to define it as ejbModule in the EAR plugin
configuration, as it is already declared as
packagingejb/packaging, but without defining it as ejbModule
it ends
Sun, 26 Jun 2011 16:00:35 +0300, /Stanimir Stamenkov/:
O.k. I've tried with my own EJB module and I've found I have to
specify typeejb/type for the dependency to get automatically
included as EJB module in the EAR. So the attached pom.xml appears
to build as expected.
However I seem to get
Wed, 6 Jul 2011 13:00:00 +0800, /Ben Caradoc-Davies/:
Mercurial allegedly has some space-performance problems handling
renames of directories, which end up doubling the space used in the
repo. This might make the current implementation inefficient for
large projects.
Fri, 23 Sep 2011 14:12:00 -0400, /Gupta, Narendra/:
1. I have following directory for java source code
--lion
--com
test1 -- contains pom.xml and java source code with package
com.test1
test2 - contains pom.cml and java source code with pakage
com.test2
I want to configure the 'finalName' for a parent module and all its
child modules as:
build
finalName${artifactId}/fileName
/build
Is there a way to configure this as a default pattern in the parent
and avoid repeating it in the 'build' section for all child modules?
--
Thu, 13 Oct 2011 08:30:49 +0800, /Yuen-Chi Lian/:
Yes. Have you tried?
$ cat parent/pom.xml | grep finalName
finalName${project.artifactId}-${project.version}-b${buildNumber}/finalName
Ah, yes - thank you. I've already tried it before but for some
reason I had determined it doesn't work,
I'm trying to activate a build profile in a project, like:
profiles
profile
idunpack/id
activation
property
namedev.explode/name
valueyes/value
/property
/activation
build
pluginManagement
plugins
...
Fri, 14 Oct 2011 09:35:26 +0200, /Jörg Schaible/:
Any property used
to activate a profile must be given on the command line or be defined in the
settings.xml.
Thanks. That confirms my suspicions. The point I want properties
read from external file, but not settings.xml, is to maintain a
Sat, 15 Oct 2011 19:45:25 +0200, /Ansgar Konermann/:
Am 15.10.2011 18:35 schrieb Stanimir Stamenkov:
The point I want properties read
from external file, but not settings.xml, is to maintain a configuration
specific to a project. Using setting.xml for all projects could cause
conflicts if two
Wed, 12 Oct 2011 17:42:59 -0700, /Manfred Moser/:
11-10-12 05:30 PM, Yuen-Chi Lian wrote:
parent
groupIdy/groupId
artifactIdx/artifactId
versionv/version
relativePath../parent/relativePath
Dont use relative path
Tue, 1 May 2012 11:44:27 -0700, /Daivish Shah/:
I have following dependency declared in my UNIT Testing module project.
dependency
groupIdsample-project-ws/groupId
artifactIdsample-project-ws/artifactId
version${project.version}/version
typewar/type
scopetest/scope
Is it true attached artifacts, that is artifacts with a non-empty
classifier, inherit their dependencies from the main artifact?
Stated otherwise - is it possible to specify different set of
dependencies for an attached artifact such that when included in
another project, that different set of
Sat, 5 Jan 2013 14:38:38 -0800, /Manfred Moser/:
If the dependencies are not applicable for this artifact (e.g. the pom is
not valid for it because it is generated against Maven convention) then it
should not be an artifact in the same coordinates but rather be a
different module with different
Sat, 5 Jan 2013 23:52:50 +0100, /Anders Hammar/:
I always argue for having the API (e.g. ejb client) as a separate module
and then the impl (e.g. the ejb) as a separate module.
That's what we're using after all, but I've recently discovered this
ejb-client option and started wondering if
Sat, 5 Jan 2013 14:58:45 -0800, /Rajwinder Makkar/:
How can i pass property/value pair a part of maven build process on the
command line which insert things in manifest file ? Something like mvn
clean install --manifestdefineproperty property=value
Note, depending on the purpose of this
Sun, 06 Jan 2013 23:46:05 +0200, /Stanimir Stamenkov/:
Note, depending on the purpose of this property you might not be
doing something considered a good practice, but I can say from the
description you've given.
I've really meant to say: ... but I *can't* say from the
description you've
Mon, 8 Apr 2013 20:15:15 +1000, /Olivier Lamy/:
Release Notes - Apache Maven 2.x Compiler Plugin - Version 3.1
** Bug
* [MCOMPILER-187] - incremental stuff detect changes even if
nothing has changed means too much compilation
I'm not sure it is the same, but it seems I'm still observing
Thu, 11 Apr 2013 23:38:17 -0500, /Wayne Fay/:
I'm not sure it could be a clue but it happens for some (I haven't
verified all) WAR modules.
What code are you compiling in a WAR module? Best practice says to move the
Java code to a Jar module. Are you precompiling jsps with jspc or ...?
No
Thu, 11 Apr 2013 23:38:17 -0500, /Wayne Fay/:
I'm not sure it could be a clue but it happens for some (I haven't
verified all) WAR modules.
[...]
Is this a project you can share via Github? Or can you make a sample that
(consistently) exhibits the behavior, and share just that part?
All
Sat, 13 Apr 2013 20:55:17 +0300, /Stanimir Stamenkov/:
Thu, 11 Apr 2013 23:38:17 -0500, /Wayne Fay/:
I'm not sure it could be a clue but it happens for some (I haven't
verified all) WAR modules.
[...]
Is this a project you can share via Github? Or can you make a
sample that
(consistently
Mon, 15 Apr 2013 22:10:19 +1000, /Olivier Lamy/:
+ excludes
+exclude**/package-info.java/exclude
+ /excludes
[...]
As all package-info.java files never go to
${project.build.outputDirectory} a compilation is triggered.
This is really
I want to copy an artifact with its transitive dependencies to a
specified location. The artifact is not a dependency of the project
the given build step would be part of.
I've tried the copy goal of the maven-dependency-plugin [1], but
it doesn't have a transitive option. I've also tried
Fri, 31 May 2013 17:14:31 -0400, /Evan Ward/:
I'm trying to configure the javadoc report plugin so that I get the same
output with `mvn javadoc:jar` and `mvn site`. I've read the usage page
[1] which suggests that configuring the plugin in the reporting
section will apply the same configuration
where this feature would be located: either
by undeprecating destination param some way , either removing the
requirement for a maven pom for copy-deps or by creating a brand new goal.
Thanks a lot.
Le 2 juin 2013 17:13, Stanimir Stamenkov s7a...@netscape.net a écrit :
I want to copy an artifact
Mon, 3 Jun 2013 07:04:58 -0700 (PDT), /Evan/:
http://maven.apache.org/plugins/maven-javadoc-plugin/usage.html#Javadoc_Configuration
That section warns me that the behaviour is different, but not *how* it
differs. In the example at the bottom of that page the javadoc:javadoc uses
elements from
[See my reply below the quote.]
Wed, 24 Jul 2013 20:20:49 +0200, /Grégory Joseph/:
I can't seem to find an accurate trace of this in the release notes,
so I thought I'd just ping the list - Changes in how exclusions are
applied transitively between Maven 2.2.1 and 3.1 ?
Here's a situation: A
Sat, 5 Apr 2014 22:45:59 -0700 (PDT), /ghostwolf59/:
Just wonder if I could expect issues to crop up by upgrading maven from
v3.0.4 to the most recent stable version available (v3.2.1 while writing
this)
Note sure if this could affect you:
https://jira.codehaus.org/browse/MNG-5591
(fixed
I'm having a POM like:
org.liquibase
liquibase-maven-plugin
3.8.3
...
I don't
19 à 18:04, Stanimir Stamenkov a écrit :
I'm having a POM like:
org.liquibase
liquibase-maven-plugin
(possibly 5-10 of them). All in all, it doesn't appear
feasible unless I'm missing something with your suggestion?
Stanimir Stamenkov wrote on 2019-12-22 09:04 (GMT -08:00):
I'm having a POM like:
org.liquibase
Wed, 01 Jan 2020 03:14:25 +0100, /Steinar Bang/:
Karl Heinz Marbaise:
(The purpose of the javadoc aggregate is to get a javadoc jar
attached to the parent pom artifact, that I can link to with a javadoc
banner in the README.)
The question is also why running javadoc everytime?
Hm... I
Mon, 23 Dec 2019 21:43:36 +0100 (CET), /Manfred Moser/:
well... with using the profile you can add more plugins and all sorts of other
stuff and the command will stay that long
I must admit I'm completely puzzled by your suggestion – why I would be
adding all sorts of stuff gaining nothing
Tue, 24 Dec 2019 09:34:05 -0600, /Jeff Jensen/:
Perhaps your specific example requires no further profile configuration,
however, since a profile enables configuring many things and calling one
goal/mojo is only one, Manfred generally suggests that it's no more effort
to run a profile with
Tue, 24 Dec 2019 17:27:18 +0100 (CET), /Manfred Moser/:
Typically a separate invocation like needing to update the data schema involves
more than one plugin.
The best and most common example is a release profile. It often includes things
like javadoc, signing, obfustcating, creating
Tue, 17 Oct 2023 11:56:12 -0700, /David Karr/:
[...]
cvc-complex-type.3.2.2: Attribute 'combine.self' is not allowed to
appear in element 'configuration'.
I, along with multiple other people, assumed that this indicates a bug in
m2e, because it clearly works in the command-line build.
Tue, 17 Oct 2023 22:11:29 +0300, /Stanimir Stamenkov/:
I'm not familiar with this attribute but looking at
<https://maven.apache.org/xsd/maven-4.0.0.xsd>:
You can control how child POMs inherit configuration from parent POMs
by adding combine.children or combine.self
attr
Sun, 24 Sep 2023 15:45:49 +0200, /Delany/:
[...]
Is there a way to specify these in a properties file and have each argument
on its own line?
Have you tried using @file (a-la javadoc options):
https://docs.oracle.com/en/java/javase/17/docs/specs/man/javac.html#command-line-argument-files
[With some more context from the original message.]
Sun, 24 Sep 2023 15:45:49 +0200, /Stanimir Stamenkov/:
Sun, 24 Sep 2023 15:45:49 +0200, /Delany/:
I can also quickly run a build to catch on issue
that I'm trying to eradicate by adding
-Dep="-XepDisableAllC
Sun, 24 Sep 2023 18:04:51 +0200, /Delany/:
Thank you. Its not happening though. I also tried it from the pom, so
-Xplugin:ErrorProne
-XepExcludedPaths:.*/target/.* @${build.root}/ep.config
@files are Java 9+ feature for tools other than javadoc – just making
sure you're not using Java 8
Sun, 24 Sep 2023 20:01:14 +0200, /Thomas Broyer/:
ErrorProne itself doesn't support @file arguments, so I think the question
remains unanswered: can Maven load a property's value from a file?
(possibly through a plugin; I have looked at build-helper-maven-plugin and
apparently it doesn't
Wed, 4 Oct 2023 15:52:14 +, /Joseph Kesselman/:
My only other real gripe with Maven at the moment is that I haven't
yet found a way to tell the javadoc plugin "Yes, I know there's lots
that can be improved, but for now please consider that a Warning
rather than an Error."
If I'm not
the additional
"requires")
On 7/2/22, 4:46 PM, "Stanimir Stamenkov" wrote:
I can't figure out what's the best practice or just proper way for
setting up a JPMS Maven project with test sources that have additional
Java SE module dependencies to those
I can't figure out what's the best practice or just proper way for
setting up a JPMS Maven project with test sources that have additional
Java SE module dependencies to those of the main sources.
I have the following basic project structure:
pom.xml
src/
├─ main/
│ └─ java/
I want to produce a Java 8 compatible library that provides some Java 9+
classes, packaged as a Multi-Release JAR:
* https://maven.apache.org/plugins/maven-compiler-plugin/multirelease.html
I've taken a slimmed down _Multi-Release Parent_
Sun, 31 Jul 2022 14:18:52 +0300, /Stanimir Stamenkov/:
Sat, 23 Jul 2022 19:28:50 +0300, /Stanimir Stamenkov/:
I want to produce a Java 8 compatible library that provides some Java
9+ classes, packaged as a Multi-Release JAR:
*
https://maven.apache.org/plugins/maven-compiler-plugin
Sat, 23 Jul 2022 19:28:50 +0300, /Stanimir Stamenkov/:
I want to produce a Java 8 compatible library that provides some Java 9+
classes, packaged as a Multi-Release JAR:
* https://maven.apache.org/plugins/maven-compiler-plugin/multirelease.html
O.k. From the very same page:
*Challenges
Fri, 20 Jan 2023 13:38:32 +0200, /Delany/:
Hi Tom,
You can turn off filtering if you don't need it. The docs suggest having
separate folders for plaintext/binary resources
https://maven.apache.org/plugins/maven-resources-plugin/examples/filter.html
The docs also mention a configuration to
Thu, 1 Jun 2023 13:07:01 +, /Geoffry Roberts/:
[ERROR] Internal error: java.nio.file.InvalidPathException: Illegal char <:> at
index 3:
jar:file:\C:\Program%20Files\apache-maven-3.9.2\lib\maven-core-3.9.2.jar!\org\apache\maven\project
-> [Help 1]
I see the error has to do with a ":".
Thu, 1 Jun 2023 18:49:59 +0300, /Stanimir Stamenkov/:
Thu, 1 Jun 2023 13:07:01 +, /Geoffry Roberts/:
[ERROR] Internal error: java.nio.file.InvalidPathException: Illegal
char <:> at index 3:
jar:file:\C:\Program%20Files\apache-maven-3.9.2\lib\maven-core-3.9.2.jar!\org\apache\maven\p
Following "Maven Compiler Plugin 3.12.0 Released" [1] I'm trying it out
in a small project of mine [2]. After updating the
maven-compiler-plugin 3.11.0 -> 3.12.0 I've noticed the following error
occurring in my dirty(!) working copy:
> mvn -V package
Apache Maven 3.9.6
Thu, 21 Dec 2023 09:10:44 +0100, /Slawomir Jaranowski/:
Release under vote ... you can check candidate version:
https://lists.apache.org/thread/scb8qw1or396l4hqgf5h78z5h0km60z4
Tried maven-compiler-plugin 3.12.1 from the staging repository – no
error during the build.
--
Stanimir
Mon, 8 Jan 2024 22:05:42 +0200, /Stanimir Stamenkov/:
Mon, 8 Jan 2024 13:49:41 -0600, /Neil Aggarwal/:
call mvn exec:exec -Dexec.executable="java"
-Dexec.args="-Djava.library.path=C:\OneDrive\Dev\Jacob -cp %%classpath
com.fiscalassets.tax.PrintForm1098s application.properti
Mon, 8 Jan 2024 13:49:41 -0600, /Neil Aggarwal/:
I am trying to put my Maven command in a batch file so I can click
on the icon to run it.
[...]
If I try putting call before the mvn command:
call mvn exec:exec -Dexec.executable="java"
-Dexec.args="-Djava.library.path=C:\OneDrive\Dev\Jacob -cp
Sat, 6 Jan 2024 00:29:57 -0600, /Neil Aggarwal/:
OK, so I changed to try to use the exec:exec goal.
I updated the goal in my pom.xml (Attached) and tried this command line:
mvn exec:exec -Dexec.executable="java" -Dexec.args="-cp %classpath
com.fiscalassets.tax.PrintForm1098s
Mon, 5 Feb 2024, /sebb/:
Under Windows,
mvn test -Dtest=org.apache.commons.compress.harmony.unpack200.tests.ArchiveTest
fails with
Unknown lifecycle phase
".apache.commons.compress.harmony.pack2000.tests.ArchiveTest".
but the following works:
mvn test
Mon, 5 Feb 2024, /sebb/:
That looks like a Windows quoting issue; Windows does not like '.' in
parameter values
Try
mvn clean test -D"test=org.apache.commons.compress.harmony.unpack200.tests.*"
Dot (.) doesn't appear a standard delimiter on Windows:
*
Fri, 1 Mar 2024, /Olivier Lamy/:
users will not be able to rebuild from sources.
is it a requirement? No real idea as already said in this thread, you
need to ask Sonatype.
But it breaks the concept of opensource as you cannot build from the sources :)
The sources uploaded to Maven Central
Thu, 29 Feb 2024, /Tamás Cservenák/:
Resolver 1.9.18 uses the temp file technique you describe:
copies to (random) temp file located in the same directory where target
file is, and then atomically moves the file to its place
(unless OS is windows, for obvious reasons).
Doesn't atomic file
specific to atomic moves. It could happen one process has opened a file
for reading, preventing another from overwriting it immediately. I
guess such updates need a retry mechanism, at least for the atomic move
part.
On Thu, Feb 29, 2024 at 1:40 PM Stanimir Stamenkov wrote:
Thu, 29 Feb 20
See my reply below the quote...
Fri, 22 Mar 2024, /Debraj Manna/:
[...]
I tried the below
org.springframework.boot
spring-boot-maven-plugin
true
org.apache.maven.plugins
maven-jar-plugin
Sat, 30 Mar 2024, /Debraj Manna/:
As suggested by Stanimir if I remove the test-jar type and also the
includes like below then mvn package --projects=:agent -T 2 --also-make
works
[...]
But if I add back the below inclusion it again starts throwing the
NoClassDefFoundError
[...]
At least
Thu, 28 Mar 2024, /Debraj Manna/:
But I am still facing issues in the below case.
[...]
com.spotnana
synapse
0.0.1-SNAPSHOT
tests
test-jar
This doesn't seem right. I'm not aware of test-jar –
should be just jar (or just omit it as it is implied).
test
Now if I am
-216
On Thu, Mar 28, 2024 at 3:11 PM Stanimir Stamenkov wrote:
Thu, 28 Mar 2024, /Debraj Manna/:
com.spotnana
synapse
0.0.1-SNAPSHOT
tests
test-jar
This doesn't seem right. I'm not aware of test-jar –
should be just jar (or just omit it as it is implied
66 matches
Mail list logo