- From: Daniel Veillard <daniel@veillard.com>
- Date: Thu, 9 Jun 2005 22:24:19 +0200
- To: Paul Grosso <pgrosso@arbortext.com>
- Cc: public-xml-core-wg@w3.org
On Thu, Jun 09, 2005 at 04:06:51PM -0400, Paul Grosso wrote: > > I just had an interest comment from someone: > > The XInclude spec is quite specific about where > to look for an XInclude object if the URI for it > is a relative path. The path is relative to the > location of the containing object. There is no > provision for any sort of search path to be used, > so doing something like this is, strictly speaking, > a violation of the XInclude spec. > > Admittedly, the comment was about adding an > application-level "base URI" sort of thing, but it > got me to thinking about XML Catalogs. > > Folks like Norm and me think that, in general, any > external identifier or URI reference could be > "redirected" by an XML Catalog if one is being used. Actually I have a grief with that in the case the URI-Reference is a directly accessible local path. http://bugzilla.gnome.org/show_bug.cgi?id=303290#c2 > I remember fiddling some words in the XML spec to > make sure we accounted for this in the case of > external identifiers, though we I go back to the > spec, the "weasel words" are pretty subtle. > > Assuming one wants to be able to use an XML Catalog > to allow for redirection of things the application > recognizes as external identifiers and URI references, > do we have to write something special in each of our > XML-related specs that say anything about URIs? > Does XInclude as written imply that use of a catalog > to redirect the href attribute would be a violation > of the XInclude spec? > > I'd be interested to hear what others think about this. That I dislike a local directly accessible resource being taken over a catalog. But for any remote, missing or public ID access I'm fine Catalogs taking over. Daniel -- Daniel Veillard | libxml Gnome XML XSLT toolkit http://xmlsoft.org/ daniel@veillard.com | Rpmfind RPM search engine http://rpmfind.net/ http://veillard.com/ |
Received on Thursday, 9 June 2005 20:24:20 UTC