I don't recommend this approach, but you could add some POM configuration to install the 3rd-party dependency in a separate profile:
<profiles>
<profile>
<id>install-dependencies</id>
<build>
<plugins>
<plugin>
<artifactId>maven-install-plugin</artifactId>
<version>2.3.1</version>
<executions>
<execution>
<id>install-dropbox-sdk</id>
<phase>validate</phase>
<goals>
<goal>install-file</goal>
</goals>
<configuration>
<groupId>com.dropbox</groupId>
<artifactId>dropbox-sdk</artifactId>
<version>1.3.1</version>
<file>src/main/lib/dropbox-java-sdk-1.3.1.jar</file>
<packaging>jar</packaging>
</configuration>
</execution>
</executions>
</plugin>
</plugins>
</build>
</profile>
<profile>
<id>build</id>
<activation>
<activeByDefault>true</activeByDefault>
</activation>
<dependencies>
<dependency>
<groupId>com.dropbox</groupId>
<artifactId>dropbox-sdk</artifactId>
<version>1.3.1</version>
</dependency>
</dependencies>
</profile>
</profiles>
There are two profiles here: install-dependencies
and build
. The first installs the dropbox-sdk
dependency into your Maven repository and needs to be run once on every machine as follows:
mvn -Pinstall-dependencies validate
The second is enabled by default, and adds the Dropbox SDK as a dependency.
To be honest though, this isn't much better than running
mvn install:install-file -Dfile=src/main/lib/dropbox-java-sdk-1.3.1.jar -DgroupId=com.dropbox -DartifactId=dropbox-sdk -Dversion=1.3.1 -Dpackaging=jar
on every machine.
The other downside of this approach is that you'll have to add all dependencies of the dropbox-sdk
to your build as well- whereas if it is done properly by adding the JAR and a POM to a repository server, then Maven will calculate the transitive dependencies properly.