AbstractJavadocMojo#getClasspath(..) should use subProject's managedVersionMap
------------------------------------------------------------------------------

                 Key: MJAVADOC-198
                 URL: http://jira.codehaus.org/browse/MJAVADOC-198
             Project: Maven 2.x Javadoc Plugin
          Issue Type: Bug
    Affects Versions: 2.4
            Reporter: Detelin Yordanov
         Attachments: AbstractJavadocMojo.patch

Hi,
   We had a problem using Eclipse artifacts that contain version qualifiers, 
e.g. artifact foo version 3.3.0-SomeQualifier is not resolved
when the dependency version definition uses a version range e.g.:
<dependency>
<artifactId>foo<artifactId>
<version>[3.3.0,4.0.0)</version>
<groupId>some Group...</groupId>
<dependency>

We found a workaround for this described here: 
http://jira.codehaus.org/browse/MECLIPSE-405.
The workaround is to use maven 2.0.9+ and define concrete versions for the 
eclipse artifacts in a <dependencyManagement> section of our project, thus 
overriding the range version definitions in some of the eclipse poms.

e.g.:

<dependencyManagement>
    <dependencies>
        <dependency>
            <groupId>org.eclipse.equinox</groupId>
            <artifactId>common</artifactId>
            <version>3.3.0-v20070426</version>
        </dependency>
     ....
    </dependencies>
<dependencyManagement>

This helped us to build our project without getting version range issues, 
however when we ran javadoc:javadoc we found out that
the javadoc dependency resolution does not take into account the 
<dependencyManagement> section and we still get
the error:

An error has occurred in JavaDocs report generation: Couldn't find a version in 
[3.2.0-v20060603, 3.3.0-v20070426] to match range [3.3.0,4.0.0)
  org.eclipse.equinox:common:jar:null

When we examined the getClasspath(..) method of AbstractJavadocMojo we found 
out that it uses the ArtifactResolver#resolveTransitively(..)
method that lacks the "managedVersions" Map parameter.
We made an according patch to use the method that specifies it, and our problem 
was solved.

So the question is whether the usage of the #resolveTransitively(..) that lacks 
"managedVersions" parameter is intentional or not. 
If there is no problem with it, we would be very happy if you could change 
this, so that we can successfully use the javadoc plugin in our project.

Kind Regards,
   Detelin
                


-- 
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators: 
http://jira.codehaus.org/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira

        

Reply via email to