On Thu, Sep 18, 2003 at 02:10:03PM -0700, David Remy wrote: > I am getting close to having a revised XMLBeans website using Forrest > ready for publication. I have a few questions that I am hoping someone > will know about. > > 1) I am thinking of having the website as a top level directory under > xml-xmlbeans in cvs (a peer to v1 and soon v2). It looks like other > projects have their website in their source tree, is that common > practice?
Yes, most projects have the 'source' for their documentation (XML) in the source tree, and then the actual rendered HTML/PDF documentation in the separate 'xml-site' CVS module. > 2) Given the above, I am able to generate the website using forrest and > it creates the build directory containing the published website on my > machine. What is the strategy for getting the website generated into > xml.apache.org/xmlbeans? That's where the xml-site module comes in. The contents of build/site/ should be imported into xml-site/targets/xmlbeans/. Then login to www.apache.org and 'cvs checkout' those files to /www/xml.apache.org/xmlbeans/ > Do I just copy the build directory somewhere or is there a more elegant > way? Is there an automated way to have the website generated from the > Forrest source? That is what ForrestBot is for, right? I'm not clear > on how to set that up. The website needs to be periodically updated from CVS (xml-site module). I think Berin Lautenbach has established a cron job to update XML sites, so all you need to do in future is render the docs and update xml-site/targets/xmlbeans. As xml-site contains the whole xml.apache.org website, it can be a real PITA to update for people on modem connections. Forrestbot is a service which automates the render and commit part of the process. I can set it up for xmlbeans once the XML is checked into CVS. > 3) There is some nice html formatted documentation that would be great > to have integrated with the site. Does anyone know if there is a way > to have Forrest integrate html pages into the Forrest site build? Or > at least a good way for them to co-exist? If HTML content is put in src/documentation/ (not src/documentation/xdocs), it will be copied through unmodified. Alternatively, if you want to have the regular Forrest site menus grafted onto the HTML, you could give them a .ehtml extension and drop them somewhere in src/documentation/content/xdocs/. Only contents below the <body> tag will be used. Also, if your HTML uses CSS, a bit of custom tweaking may be needed. There are examples of both these techniques in the Forrest 'seed' site (type 'forrest seed site' in a clean directory to demonstrate). Cheers, --Jeff (lurking Forrest committer and forrestbot perpetrator;) > thanks for any help here! > rem > --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
