2008/7/4 Stuart McCulloch <[EMAIL PROTECTED]>:
> 2008/7/3 Craig Walls <[EMAIL PROTECTED]>:
>
>> It's my understanding from previous discussion on this list, that I can
>> use a little bit of <Embedded-Dependency> magic to create a manifest for a
>> WAR bundle that includes a Bundle-ClassPath: header for the various
>> dependencies. But, so far, I can't get Bundle-ClassPath into the generated
>> MANIFEST.MF.
>>
>> Here's what the pertinent parts of my pom.xml file look like:
>>
>> <plugin>
>> <groupId>org.apache.felix</groupId>
>> <artifactId>maven-bundle-plugin</artifactId>
>> <version>1.4.1</version>
>> <extensions>true</extensions>
>> <configuration>
>> <supportedProjectTypes>
>> <param>jar</param>
>> <param>bundle</param>
>> <param>war</param>
>> </supportedProjectTypes>
>> <instructions>
>>
>> <Embed-Dependency>*;scope=compile|runtime;inline=false</Embed-Dependency>
>> <Embed-Directory>WEB-INF/lib</Embed-Directory>
>> <Embed-Transitive>true</Embed-Transitive>
>> <Private-Package>com.semantra.*</Private-Package>
>> </instructions>
>> </configuration>
>> <executions>
>> <execution>
>> <id>bundle-manifest</id>
>> <phase>process-classes</phase>
>> <goals> <goal>manifest</goal>
>> </goals> </execution>
>> </executions> </plugin>
>>
>> When I do the build, I get a MANIFEST.MF in target/classes/META-INF, but
>> it doesn't have a Bundle-ClassPath header.
>>
>> I've also tried to add the following instruction:
>>
>>
>> <Bundle-ClassPath>.,WEB-INF/classes,{maven-dependencies}</Bundle-ClassPath>
>>
>>
>> This results in a Bundle-ClassPath header that contains exactly what is
>> specified above...that is, {maven-dependencies} is not expanded.
>>
>> What am I doing/understanding incorrectly?
>>
>
> this looks similar to Sahoo's issue where the manifest goal behaves
> differently to the bundle goal:
>
> http://issues.apache.org/jira/browse/FELIX-620
>
> both the manifest and bundle goals share a lot of the same code to invoke
> Bnd, so there must be
> something else going on that's different in the manifest goal - btw, I
> won't get to this until next week
> because I'm spending most of Friday on a plane and sleeping off jet lag for
> much of the weekend...
>
FYI, this issue has now been fixed in trunk - I'll deploy a 1.4.2-SNAPSHOT
tomorrow
--
Cheers, Stuart
