OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.

 


Help: OASIS Mailing Lists Help | MarkMail Help

docbook-apps message

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]


Subject: Re: [docbook-apps] indexterm with zone attribute


Thomas Schraitle wrote:
> :-) To be a bit more concrete, I was thinking of something like this:
> 
>  <sect1 xml:id="intro">
>     <title>...</title>
>     <indexterm zone="intro foo">
>       <primary> ... </primary>
>     </indexterm>
>     <sect2 xml:id="bla"> ... </sect2>
>     <sect2 xml:id="foo"> ... </sect2>
>  </sect1>
> 
> Ok? I'm not sure how accurate it will be in the final PDF regarding the 
> page numbers. I never tried it. 


The idea that sort of contradicts this is break-before ='page'
on sect2!

perhaps better to stick a blank
<para xml:id='end'/> prior to the major block?


  For this case, I generally use in my books the class attribute
> with "startofrange" and "endofrange". The above code can be rewritten 
> like this:
> 
>  <sect1 xml:id="intro">
>     <title>...</title>
>     <indexterm xml:id="idx.intro" class="startofrange"> 
>       <primary> ... </primary>
>     </indexterm>
>     <sect2 xml:id="bla"> ... </sect2>
>     <sect2 xml:id="foo">
>       <title>...</title>
>       <!-- Some paras, figures, etc. -->
>       <indexterm class="endofrange" startref="idx.intro"/>
>     <sect2>
>  </sect1>
> 
> So actually you have 2 methods creating index ranges. As always, both have 
> advantages and disadvantages. The second method is probably a bit more 
> verbose but I think it is also more accurate. You have to think a bit, 
> where to set the 2nd indexterm. For example, if you have an additional 
> sect3, you have to move it *into* the sect3 section as the content modell 
> doesn't allow it inbetween sections.

Note the fo constraint above? Using indexterm, a para in the end of
the first sect would be better?

This does look more robust though.

> 
> Another disadvantage is profiling: If the last section ("foo" in this 
> case) is removed from the profiling step, the end indexterm will be 
> removed too. This leads to an accidently missing end point and the range 
> can not be created. 

<grin/> I bet Bobs stylesheets shout at you when you do that!



> I'm searching for a XSLT method to insert the last indexterm 
> automatically, but that's a different story.
> 
> Just some weird thoughts about this topic. :-)

Indexing is a weird topic anyway.
certainly a pretty rare skill IMHO



regards

-- 
Dave Pawson
XSLT XSL-FO FAQ.
http://www.dpawson.co.uk


[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]