• To: "'Henry S. Thompson'" <ht@i...>
• Subject: RE: Substitution Groups
• From: "Michael Kay" <mike@s...>
• Date: Sat, 19 Feb 2005 11:36:00 -0000
• Cc: "'Paul Spencer'" <xml-dev@b...>,"'Xml-Dev'" <xml-dev@l...>
• In-reply-to: <f5bfyzszu22.fsf@e...>
• Thread-index: AcUWc0x5EBuF+U+aQwO9gRRFbcp4FwAAnU9A

> 
> You gloss is not accurate.  The REC also says, wrt the semantics of
> <import> [1], 
> 

We had a discussion on this on xmlschema-dev in October and I thought the
conclusion was that 3.15.3 was indeed flawed. 

If document A includes document B (or if C includes A and B), the components
defined in A are never added to the "appropriate property of the schema
which corresponds to the document B", and under 3.15.3 they are therefore
not available to discharge QName references in B.

We're dealing with <include> here, not <import>.

Michael Kay
http://www.saxonica.com/

• Follow-Ups:
  • Re: Substitution Groups
    • From: ht@i... (Henry S. Thompson)

• References:
  • Re: Substitution Groups
    • From: ht@i... (Henry S. Thompson)

• Prev by Date: RE: What should TrAX look like? (Was: Re: Article on JAXP 1.3 "Fast and Easy XML Processing")
• Next by Date: RE: Transmitting XML between different applications
• Previous by thread: Re: Substitution Groups
• Next by thread: Re: Substitution Groups
• Index(es):
  • Date
  • Thread