Description of problem: With new javapackages-tools and xmvn too many packages end up with source JARs installed. I'm not sure if it's good or bad thing, but I'm rather against that. If we don't want source JARs to be installed by default then perhaps we should add a default XMvn rule to javapackages-tools that would skip installation of sources by default. This rule could be overridden by packages that do want to install sources. Stano, do we want source JARs installed by default? Version-Release number of selected component (if applicable): 3.0.2-2
*** Bug 1006721 has been marked as a duplicate of this bug. ***
Yeah, most likely disabling source jar installation is best solution at this time. Later we *might* want to create a rule in guidelines that will install them in some agreed-upon directory different from javadir.
There are also packages which generate not only -sources jar, but also -javadoc and -tests jars (e.g.: httpcomponents-client). Tests might be useful, but javadoc artifacts are IMO not needed.
Maybe we should disable automatic installation of attached artifacts (unless explicitly enabled with %mvn_package)?
(In reply to Mikolaj Izdebski from comment #4) > Maybe we should disable automatic installation of attached artifacts (unless > explicitly enabled with %mvn_package)? Since we were not installing them in the first place before XMvn 1.0 this might be the best solution. No unexpected changes to existing packages... Or am I missing something?
If by default we skip installation of any artifact with non-empty classifier then existing packages would not need any changes to keep behavior of XMvn < 1.0.0. Packages that want to install some or all attached artifacts will have to enable them explicitly with %mvn_package.
Fixed upstream in 2b481d9b I added the following rule: <rule> <optional>true</optional> <artifactGlob> <classifier>*?</classifier> </artifactGlob> <targetPackage>__noinstall</targetPackage> </rule>
Fixed in javapackages-tools-3.3.1-1
I believe that this bug is fixed in javapackages-tools-3.3.1-1, which is available in Fedora Rawhide, so I am closing this bug now. The build containing the fix can be found at Koji: http://koji.fedoraproject.org/koji/buildinfo?buildID=469008