Sean Schofield wrote:
> I was able to follow the instructions fairly well except for the major
> part of the .ihtml not converting.  Here is the hack that I did to get
> everything working:

It was converting. As i said, it was just that it
was being generated into a different directory.

> Instead of source.ihtml and binary.ihtml I just wrote source.xml and
> binary.xml in the normal xdocs fashion (using the mirrors.ihtml stuff
> as a template for my combo box.)

Well i don't understand how you would have got past
xml validation for that. Switch it off i suppose.
That was the point of using *.html as the source format.

> Then in my site.xml I had:
> 
>     <project-information href="" label="Project" tab="project">
>         <index_pm href="index.html" label="Welcome"/>
>         <license label="License" href="license.html" />
>         <downloads label="Downloads">
>             <bindownload label="Binary"
> href="http://myfaces.apache.org/binary.cgi"/>
>             <srcdownload label="Source"
> href="http://myfaces.apache.org/source.cgi"/>
>             <bindownload_html href="binary.html"/> <!-- these dummy
> links are necessary -->
>             <srcdownload_html href="source.html"/> <!-- these dummy
> links are necessary -->
>         </downloads>
>      </project-information>
> 
> ...
> 
>      <!--
>      NOTE: foo tab doesn't really exist (its label is empty String.) 
> This allows us to create bogus
>      links for the binary.xml and source.xml files.  These bogus links
> are necessary to trick
>      Forrest into converting them into html even though they are not
> linked anywhere.  The real
>      links are the cgi scripts tha will eventually make use of these
> html templates.
>      -->
>      <foo href="" label="" tab="foo">
>          <bindownload_html label="" href="binary.html"/>
>          <srcdownload_html label="" href="source.html"/>
>      </foo>
> 
> So the foo tab links forced the generation of the HTML page but you
> never see them because the tab is empty string.  The links in the
> download section were necessary b/c otherwise when the cgi redirected
> you to the page you would see the empty tab with empty entries.  You
> can check out the end results at http://myfaces.apache.org.

That is an interesting workaround.

--David 

Reply via email to