RE: [docbook-apps] interested in feedback on my current strategy for writing manuals

["Robert P. J. Day" <rpjday@mindspring.com>]

On Mon, 19 May 2003, Jeff Beal wrote:

> My only suggestion is to try to remove steps (1) and (3).  I don't see any
> reason why you can't go from 
>  <il>
>    <e>....</e>
>    <e>....</e>
>  </il>
> 
> directly to
> 
> <itemizedlist>
>   <listitem><para>....</para></listitem>
>   <listitem><para>....</para></listitem>
> </itemizedlist>

i thought about that, but i was thinking that i might be able
to define a number of elements that are just different kinds of 
lists, and <e> was just a simple was of writing "list element",
since i might need to do it more than once.

certainly, it depends on how often i'll need that shortcut,
so i'll give it some thought.

> 
> Similarly:
>     <xsl:template match="chapter/section">
>      <xsl:element name="beginpage"/>
>      <xsl:element name="section">
>       <xsl:apply-templates/>
>      </xsl:element>
>     </xsl:template>
> 
> can probably be re-written as
>     <xsl:template match="chapter/section">
>      <fo:block break-after="page">&#160;</fo:block>
>      <xsl:apply-imports/>
>     </xsl:template>
> 
> to be able to go directly from DocBook to FO.

i actually broke that into two steps since i wanted to localize
ugliness like explicit page breaks.  again, it can certainly be 
done in one step, but if you look at the above two steps, the
first rewrites docbook in terms of *only* other docbook elements,
while the second introduces fo: elements, which is why i wanted
to keep them separate.

> Once you've gotten that far, there's a Saxon extension (if you're using
> Saxon, otherwise disregard) called 'next-in-chain' that allows you to chain
> stylesheets such that the output of one stylesheet (your Pidgen -> DocBook
> stylesheet) becomes the input of another stylesheet (Docbook -> FO).  This
> would effectively collapse out another step.  If you then configure FOP to
> use Saxon as the XSLT processor and pass source XML and XSLT to FOP instead
> of FO, you could possibly cut this all the way down to one step.

hmmm ... i'll have to check that out, thanks.

rday