- From: Jirka Kosek <jirka@kosek.cz>
- Date: Tue, 17 Apr 2012 10:40:26 +0200
- To: veillard@redhat.com
- CC: Paul Grosso <paul@paulgrosso.name>, core <public-xml-core-wg@w3.org>
- Message-ID: <4F8D2C7A.9040209@kosek.cz>
>> On 2012 February 14, we published >> XInclude 1.1 Requirement and Use Cases >> http://www.w3.org/TR/xinclude-11-requirements/ >> summarizing the requirements and use cases for possible >> enhancements to XInclude addressing the issues: >> >> * @xpointer when parse="text" > > Sounds reasonable, though I wonder about implementations. > >> * copying attributes from the xinclude element to the root >> included element > > Internally libxml2 preserves the old xi:include element > so the attributes can be found (and as well as boundaries of inclusions > when people need to roll them back). But I'm surprized to suggest > the existence of a "root included element" especially as part of > our charter, to me that doesn't exist, it may exist but people > will include well balanced chunks or even pure text, and I'm not > sure you could even guess in advance if those attributes would be > preserved or not. I find that confusing. Indeed, but in my experience in 99% fragments with one root element are included so it makes sense to cater them specially. Of course we can transfer additional metatadata using PIs instead of attributes. This would work even for text nodes or well balanced chunks. Jirka -- ------------------------------------------------------------------ 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 ------------------------------------------------------------------
Received on Tuesday, 17 April 2012 08:41:01 UTC