Mailing-List: contact dev-help@forrest.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@forrest.apache.org
Received-SPF: neutral (hermes.apache.org: local policy)
Message-ID: <428B1BA8.8060405@apache.org>
Date: Wed, 18 May 2005 11:40:40 +0100
From: Ross Gardler <rgardler@apache.org>
User-Agent: Mozilla Thunderbird 1.0 (Windows/20041206)
MIME-Version: 1.0
To: dev@forrest.apache.org
Subject: Re: Solving FOR-470 and getting rid of xdocs now?
References: <1331654960.20050517120618@soethe.net>
 <1116343272.13322.0.camel@localhost.localdomain>
 <428A3245.7040202@apache.org>
In-Reply-To: <428A3245.7040202@apache.org>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit

Juan Jose Pablos wrote:
>>On Tue, 2005-05-17 at 12:06 +0200, Ferdinand Soethe wrote:
> 
> 
>>Looking at the details, I keep wondering if it wouldn't save us much
>>work to get rid of x-docs and fix FOR-470 in one step.
> 
> 
> In order to get rid of x-docs, we need to create the stylesheet for
> xdocs --> xhtml so, I recond that this needs to be done first...


I think (hope) ferdinand was talking about the xdocs directory rather
than the DTD. The Xdocs directory now contains all processes data
regardless of it's format (sxw, html, doc etc.) so the name is misleading.

Changing the internal format to XHTML is much more than just writing a
stylesheet from xdocs -> XHTML. We also need to change all the skins
which currently work with XDoc. The existing pipelines would all have to
change as well, we would need to add the new transformation into it (if
we use this approach). We need to plan this move carefully.

However, I am -1, even on getting rid of the xdocs directory. It is a
big job. We should have all our docs conform to the "correct" way of
doing things, that means changing all the documentation, all the
plugins, site-author, fresh-site etc.

Then there are the names of the propoeries we use in the sitemaps, they
would no longert make sense (project:content.xdocs} vs
{project:content). And we would have to work out how to handle the
situation where we want raw, unprocessed data from files that would
normally be processed, such as html and xml.

Lets not underestimate the work involved with these two tasks.

-1 on even thinking about it before 0.7.

Ross