[rfc-i] xi:includes in references

Nico Williams nico at cryptonector.com
Tue Jan 20 08:50:58 PST 2015

On Tue, Jan 20, 2015 at 08:44:07AM -0800, Paul Hoffman wrote:
> On Jan 20, 2015, at 5:02 AM, Julian Reschke <julian.reschke at gmx.de> wrote:
> > XInclude processors are allowed to put this in (and it makes a lot
> > of sense to do so). If we want to validate post-Xinclude results,
> > we'll either need a preprocessing step (filtering out the
> > attribute), or allow it either "anywhere" or at least on those
> > elements where it's likely to turn up.
> I would vote for a pre-processing step, unless there is a semantic
> description of what xml:base "anywhere" means.

If the base attribute records the source for some element, then leaving
it in seems both, useful (for debugging) and useless (semantically).
The semantic uselessness [to processors of this schema] of a debugging
facility doesn't bother me.

It (the xml:base attribute) might prove useful for documentation
purposes if pre-processed .xml is published ("oh, that element was
XIncluded from ...").


More information about the rfc-interest mailing list