I have a multi-module web project with the following structure:
swear (pom)
|- fish-tissue (pom)
| |- ft-webapp (war)
| |- ft-ear (ear)
|- swear-core (pom)
| |- sc-model (jar)
| |- sc-webapp (war)
| |- sc-ear (ear)
|- swear-ear (ear)
sc-ear is an EAR for testing just the swear-core module and ft-ear is an
EAR for testing the fish-tissue module. swear-ear is the EAR that will
house the final deployment unit for the entire application.
Currently, both the swear-core and fish-tissue modules have a pom.xml
that builds all sub-projects, including the test EARs, along with a
pom-no-ear.xml that builds all sub-projects except for the EAR, and is
used when constructing the final swear-ear deployment unit. I achieve
this by placing the following modules section into the POM for swear:
<modules>
<module>swear-core/pom-no-ear.xml</module>
<module>fish-tissue/pom-no-ear.xml</module>
</modules>
This does not seem right to me since the pom.xml and the pom-no-ear.xml
are otherwise identical, with the exception of the one module (the EAR
module) that is conditionally included. This would mean that, depending
on when you pulled a POM artifact from the repository, you would get
different POMs.
I think that this could be remedied by using activation elements to
control the included modules, but activation only seems to be additive,
not subtractive. This is an issue since, by default, we want the
testing EARs to be generated, with the no-EAR mode to be only used for
generating a final deployment unit.
Does anyone know how to achieve this?
Thanks,
Chris Lieb
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
