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

Re: Blueberry is not "closed" (was: Closing Blueberry)

• From: David Brownell <david-b@p...>
• To: Joel Rees <rees@m...>, xml-dev@l...
• Date: Tue, 24 Jul 2001 09:36:30 -0700

> A parser designed for XML 1.0 will do one of several things:

Your list was incomplete.  For example, there are at least "give a warning
and proceed" and "report a validity error and proceeed" options.  I'd suggest
either of those could be appropriate transition strategies.


> The preferred response is that it should stop at the XML declaration and
> say, "I don't know how to handle this." 

"Fatal error" option.

> A less desirable response would be to ignore the declaration and proceed. 

"ignore" option ... seems unwise to me, users will be mystified when
a blueberry-related error ("blueberror"? :) breaks something later.

 
> The least desirable response would be that the parser would only reject text
> known to be necessarily illegal in the combination of XML 1.0/UNICODE 2.0.

That seems to be a core conformance requirement for a parser that's
claiming to conform to the XML 1.0 specification; a parser that doesn't
do that has a bug.

- Dave

• References:
  • Re: Blueberry is not "closed" (was: Closing Blueberry)
    • From: Joel Rees <rees@s...>

• Prev by Date: XML referencing (was Re: Blueberry is not "closed")
• Next by Date: Re: Blueberry is not "closed" (was: Closing Blueberry)
• Previous by thread: Re: XML referencing (was Re: Blueberry is not "closed")
• Next by thread: reminder: OASIS ebXML IIC TC participation
• Index(es):
  • Date
  • Thread