Re: [xml-dev] Re: Question external entities and relative URis

[David Carlisle <davidc@nag.co.uk>]

On 19/07/2010 15:23, Mike Sokolov wrote:
> Just an opinion, but it seems to me that in this matter of including
> files, a guiding principle is modularity.  So if file B includes C using
> a relative reference, the much more usual case is to expect the link
> from B->C to be relative to B's location, and not the location of some
> as-yet-to-be named file (A) that might include B.  If it were the other
> way, then the B->C nexus wouldn't really be modular: the validity of
> that reference would depend on the scope in which B was embedded, which
> usually is not what you want.
>
> Full disclosure: I don't really know what the XML specs say, just
> reasoning from basic faith in rationality :)

but the xml document doesn't directly refer to the file with a relative 
reference so in the xml entity case it's not at all clear what your A b 
and C are.

the xml document has a reference & foo ; to an entity named foo  that 
might be in this file, or another. What the specs 9more or less) say is 
if that entity is declared using a SYSTEM identifier with a local URI, 
then that local URI should be resolved relative to the base URI of the 
entity in which it appears (ie, the URI of the dtd file, not teh URI of 
the original xml document, if they are different)

David

________________________________________________________________________
The Numerical Algorithms Group Ltd is a company registered in England
and Wales with company number 1249803. The registered office is:
Wilkinson House, Jordan Hill Road, Oxford OX2 8DR, United Kingdom.

This e-mail has been scanned for all viruses by Star. The service is
powered by MessageLabs. 
________________________________________________________________________