Skip to main content

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [List Home]
Re: [m2e-users] jaxb2 and XmlBeans Connectors -- testing with m2e 1.1

Ok, I see the problem now. I was not actually using the most recent m2e
version (oops).

Because of [1], m2e now creates source folder classpath entries for all
compile and test compile roots found in MavenProject, even for those
that do not exist in workspace.

[1] https://bugs.eclipse.org/bugs/show_bug.cgi?id=361549

--
Regards,
Igor

On 11-10-20 8:56 PM, Pino Silvaggio wrote:
forgot to mention that i test with indigo.

Let me create a small test that i can send you.

the error for antlr is simply the assert testing the presence of the generated source in the classpath, i got that too since the order seem to have changed.

Igor Fedorenko<igor@xxxxxxxxxxxxxx>  wrote:

How do you run antlr tests from m2e-extras repo? I just checked out
0.13.0.20110622-1538 tag, fixed version ranges in bundle manifests and
was able to successfully run antlr tests from my m2e 1.1 development
workspace.

--
Regards,
Igor

On 11-10-20 8:14 PM, Pino Silvaggio wrote:
i will investigate further but fyi m2e extras antlr tests fail,
granted with a different error but still...

Igor Fedorenko<igor@xxxxxxxxxxxxxx>   wrote:

This sounds like a bug, but I do not understand the problem enough
to
tell anything specific. Please provide a unit test (or a fix ;-) )
and
I'll have a look.

And just to confirm, project configurators that worked with m2e 1.0
and
did not use any .internal. m2e classes are expected to work with
1.1.

--
Regards,
Igor


On 11-10-20 6:22 PM, Pino Silvaggio wrote:
Hi,

I am currently testing my connectors with m2e 1.1 found on sonatype
forge site.

There seem to be a problem with the build where *build(int kind,
IProgressMonitor monitor)*
doesn't really build and my tests fail with the project having an
error
marker.

This is the same code working on m2e 1.0.

After analysis I see that *appliesToBuildKind* returns false when
it
should have returned true.

If I bypass the *super.build(...)* in my BuildParticipant with
*maven.execute(getSession(), getMojoExecution(), monitor);*
then everything works flawless...

Just wondering if this is a bug or not?

-

Pino




Back to the top