This is SiteExporter utility, used to export (publish) the Tapestry Confluence
wiki content (pages and their attachments) to static web content directories and
files. To run this program manually, you need to have a copy of the Tapestry
static website checked out to a local directory:

  svn checkout https://svn.apache.org/repos/infra/websites/production/tapestry/content

To build the SiteExporter Java (command-line) program itself:

  mvn cxf-wadl2java:wadl2java
  mvn install
  
The first step above builds some generated WADL code that is referenced in
SiteExporter.java, so you'll need to do that step (and maybe a maven update
project) before compiler errors will go away in your IDE. 

To run the SiteExporter program, you will need to pass credentials for a valid
Confluence user, and specify the directory where the static content resides.
This can be done in one of two ways:

1) Add a profile to your .m2/settings.xml file, with your user name, password,
and content directory:

       <profile>
            <id>confluence</id>
            <properties>
                <confluence.user>user</confluence.user>
                <confluence.password>passwd</confluence.password>
                <cxf.site.output>/path/to/content</cxf.site.output>
            </properties>
        </profile>

2) Alternatively, and less conveniently, you can specify the user name, password
and content directory as system properties on mvn command line:

   -Dconfluence.user=username -Dconfluence.password=password -Dcxf.site.output=/path/to/content

To export only the changed pages in the wiki space (this is normally only
run from BuildBot):

  mvn -Pconfluence exec:java
  
How to regenerate all pages of a given Confluence space
=======================================================

By default, only modified pages (as indicated by RSS feed) are regenerated.
However, sometime it is necessary to regenerate all pages of a given
Confluence space. For example, after changing the export template or modifying
the navigation pages. To force regeneration of all pages of a given Confluence
space delete the main.pageCache file from your copy of the
following SVN directory:

https://svn.apache.org/repos/infra/websites/production/tapestry/content/cache/

At that point you can either commit the deletion and wait for BuildBot to notice the change
(via an hourly cron), or you can use Maven to re-export all Confluence files to your local
computer:

  mvn -Pconfluence,force exec:java

after which you'll need to commit any changes in the content directory via a Subversion
commit. The changes should be "live" in just a few seconds, due to the svnpubsub process.

*Most* of the site content is managed through Confluence. However, some content
is (and should be) checked into directly into the svn:

 - 5.3.7/ (and any other 5.x.x versions)
 - dtd/
 - images/
 - resources/
 - schema/
 - styles/
 - tapestry3/
 - tapestry4.1/

The Confluence spaces are exported and checked into the svn using a BuildBot 
process:

 - http://ci.apache.org/builders/tapestry-site-production