Re: [docbook] Re: XInclude 1.1 Requirements

[David Cramer <david@thingbag.net>]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On 04/05/2012 03:36 PM, Jirka Kosek wrote:
> On 5.4.2012 14:14, Camille Bégnis wrote:
> 
>> I read it carefully and hope I understood it. It looks to be able
>> to solve many issues indeed, those we fight daily. If I
>> understand correctly solving the duplicate IDs issue for example
>> would be delegated to the processor though. How much likely
>> processor developers will be to support all of this? Hearing 
>> their feedback would be particularly interesting.
> 
> Hi,
> 
> in XSLT 2.0 it is fairly easy to handle duplicates IDs (once you
> decide how to handle them). There is prototype implementation
> preceding this XInclude proposal, but processing will be very
> similar.
> 
> http://docbook.org/docs/transclusion/#d6e362
> 
> If I recall it correctly this mechanism is transparently supported
> in XSLT 2.0 stylesheets.

Even with 1.0, there are ways to deal with duplicate ids. The thing
you lose now if you xinclude in duplicate ids is validation of the
wrapper document. The editor reports the duplicate ids while authoring
or your sanity validation of the doc before building fails. So the
solution has to allow for validating the document in a way that uses
the desired id fixup. Norm's blog post implies that this would be done
via a pipeline.

David
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iQEcBAEBAgAGBQJPfhDhAAoJEMHeSXG7afUhfLcH/1BaILlWCyW+1SEgNhKe5t52
v5Db2YeBRqdj5aMKiyqNu2DimEQe8ZkO2i4HO06p/utG94pTo5A8H+1YHcmVgXxS
mFCxG26eUHIqfvnhg9noh6So1Jjsf7PV42nBHM54sGhAcVX+Meoo83g9DYuqiPOh
biwwa83bShQYidSz57/qUH7+wIVRWTdvcijJJ0FL1lioLSc1r/xpn9vQP2iOptPC
uyqzlJFREdq/wvWW7zNNpd0GX6dKlOjpaPWG/1Afp61seF2/E+DPbOAlNfk8yOyQ
zupvILKV0yAplbWZBuaO/IpeC8w6bQv1RnJ81rHO7i/qNwiyGILCpZvgPLQeNfw=
=GFX1
-----END PGP SIGNATURE-----