[XML-DEV Mailing List Archive Home] [By Thread] [By Date] [Recent Entries] [Reply To This Message]

Re: Wellformed?

• To: lists@j...
• Subject: Re: Wellformed?
• From: David Carlisle <davidc@n...>
• Date: Tue, 25 May 2004 15:42:16 +0100
• Cc: xml-dev@l...
• In-reply-to: <02ee01c44263$ecfb1550$6403a8c0@ARIMATHEA> (lists@j...)
• References: <02a601c44214$0ddb9510$6403a8c0@ARIMATHEA> <003001c44217$0ea7de60$0300a8c0@karlglen188> <40B2E9C3.2070601@o...> <02ee01c44263$ecfb1550$6403a8c0@ARIMATHEA>

  The reason I ask is because I am trying to patch the AElfred parser to check
  for WFness of entity literals-- but I am not sure if I should be checking
  WFness upon reading the literal or upon inclusion (or both?)

WF on reading but namespace conformance after inclusion, I think.
Entities have to be WF "on their own" but the namespace rec doesn't have
any constraints at the entity level.

So it's perfectly OK to have namespace prefixes in an entity resolve to
different namespace URI at different parts of the document. (Although
msxml will error on that)


<!ENTITY x "<x:x/>">

...

<a xmlns:x="data:,a">&x;</a>
<a xmlns:x="data:,b">&x;</a>
...


David



________________________________________________________________________
This e-mail has been scanned for all viruses by Star Internet. The
service is powered by MessageLabs. For more information on a proactive
anti-virus service working around the clock, around the globe, visit:
http://www.star.net.uk
________________________________________________________________________

• Follow-Ups:
  • Re: Wellformed?
    • From: "Jeff Rafter" <lists@j...>

• References:
  • Wellformed?
    • From: "Jeff Rafter" <lists@j...>
  • Re: Wellformed?
    • From: "Karl Waclawek" <karl@w...>
  • Re: Wellformed?
    • From: Bob Foster <bob@o...>
  • Re: Wellformed?
    • From: "Jeff Rafter" <lists@j...>

• Prev by Date: SAX and Entity Literals containing content
• Next by Date: Re: Meta-Meta-Meta....
• Previous by thread: Re: Wellformed?
• Next by thread: Re: Wellformed?
• Index(es):
  • Date
  • Thread