Re: [docbook-apps] Use of xincludes vs. entities

[Ron Catterall <ron@catterall.net>]

My biggest problem with Xinclude is the duplicate ID if a piece of text 
is included twice etc.  Would it be possible to reset an ID of an 
Xincluded test on the Xinclude statement with an attribute, e.g.

<xi:include href="venn.xml" ***newid="qwerty"*** 
xmlns:xi="http://www.w3.org/2001/XInclude";>
   <xi:fallback><para><emphasis> MISSING INCLUDE FILE - venn.xml 
</emphasis></para></xi:fallback>
</xi:include>

Would this be a major enhancement?  Is it a reasonable suggestion?

On the indexing comment, I use things like:
<!ENTITY  Alf 
'<indexterm><primary>Alferes-M&oacute;r</primary></indexterm>Alferes-M&oacute;r'>

in a file of entities.  This does generate a first, crude pass at an 
index.  It also avoids continual re-typing of things like &oacute;

Ron

On 9/27/10 7:09 AM, Sam Fischmann wrote:
> I agree with Ron's simple summary: use XInclude for large chunks of
> text and entities for small. The basic problem is that there's no
> elegant way to change bits of the XIncluded text based on the context
> of its inclusion:
>
> * XML files + multiple catalog sets are awkward and not supported well
> in editors; you basically have to configure the editor each time you
> open a book.
>
> * Profiling + entities is sometimes inappropriate and forces you to
> group variables by name instead of the context in which they are used,
> which causes maintenance problems.
>
> Jirka, your proposal is valuable because it would allow me to define
> small text variables for each of several contexts and store them in
> separate definition files. I could then use profiling to conditionally
> include definition files at the top level of a book and not have to
> look at miles of logical cruft every time something is wrong with a
> text variable in my documentation. It does not require multiple
> catalog sets, would be easy for editors to support, and unlike pure
> entity-based inclusion, there's an elegant system for overrides. The
> only unknown here for me is how validation would work. I can see
> myself putting definitionfile tags at the top of every single file via
> a global entity so that XIncluded files validate correctly, but that's
> inelegant again.
>
> Hope these thoughts are informative or helpful...
>
> -Sam
>
> On Sun, Sep 26, 2010 at 12:37 PM, Jirka Kosek<jirka@kosek.cz>  wrote:
>> Sam Fischmann wrote:
>>> Sorry for mailing so much on this topic, but thanks for the link. The
>>> information here is great to see, but I'm curious about the current
>>> status of any work related to this proposal.
>>
>> It is work in progress which can be thrown away in any point or changed
>> dramatically. However user feedback is more then welcome in driving
>> further development of transclusion proposal and in deciding whether and
>> how add DocBook specific transclusion support.
>>
>> --
>> ------------------------------------------------------------------
>>   Jirka Kosek      e-mail: jirka@kosek.cz      http://xmlguru.cz
>> ------------------------------------------------------------------
>>        Professional XML consulting and training services
>>   DocBook customization, custom XSLT/XSL-FO document processing
>> ------------------------------------------------------------------
>>   OASIS DocBook TC member, W3C Invited Expert, ISO JTC1/SC34 member
>> ------------------------------------------------------------------
>>
>>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: docbook-apps-unsubscribe@lists.oasis-open.org
> For additional commands, e-mail: docbook-apps-help@lists.oasis-open.org
>
>
>

-- 
Ron Catterall Ph.D. D.Sc.
ron@catterall.net
http://catterall.net

S/MIME Cryptographic Signature