On Sun, Dec 30, 2012 at 09:50:11PM -0000, pste...@apache.org wrote:
> Author: psteitz
> Date: Sun Dec 30 21:50:11 2012
> New Revision: 1426997
> 
> URL: http://svn.apache.org/viewvc?rev=1426997&view=rev
> Log:
> Documented commons.componentid ambiguiity and workaround.
> 
> Modified:
>     commons/proper/math/trunk/doc/release/release.howto.txt
>     commons/proper/math/trunk/pom.xml
> 
> Modified: commons/proper/math/trunk/doc/release/release.howto.txt
> URL: 
> http://svn.apache.org/viewvc/commons/proper/math/trunk/doc/release/release.howto.txt?rev=1426997&r1=1426996&r2=1426997&view=diff
> ==============================================================================
> --- commons/proper/math/trunk/doc/release/release.howto.txt (original)
> +++ commons/proper/math/trunk/doc/release/release.howto.txt Sun Dec 30 
> 21:50:11 2012
> @@ -70,7 +70,10 @@ to the SVN repository. Once the release 
>  The "download" page template is located at "src/site/xdoc/download_math.xml".
>  This file is updated automatically by running the command:
>  
> -  $ mvn commons:download-page
> +  $ mvn commons:download-page -Dcommons.componentid=math
> +
> +The command-line property override is necessary because the download plugin
> +expects the project name in the componentid property.

How neat that is! :)

On a related note: when creating the site, the generated Javadoc ends up in
a directory whose parent is "apidocs".  E.g. the Javadoc for release 3.1 is
in
  http://commons.apache.org/math/apidocs/index.html

Doc for older releases are in a directory whose parent is named after the
release identifier, e.g. doc for 3.0 is in
  http://commons.apache.org/math/api-3.0/index.html

I think that it would be more flexible that the "site" target creates a
directory with the appropriate version id.
That way, we would avoid the repetition of the problem raised in issue
  https://issues.apache.org/jira/browse/MATH-912

Could this be fixed with a similar trick?
[It would also require to change the links in the user guide: "apidocs"
appears explicitly there. There should probably be a variable instead, which
will be set as the document is processed to create the HTML page (?).]


Regards,
Gilles

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@commons.apache.org
For additional commands, e-mail: dev-h...@commons.apache.org

Reply via email to