• To: <xml-dev@l...>
• Subject: Re: A heavier-weight proposal for character entity definition
• From: "Rick Jelliffe" <ricko@a...>
• Date: Thu, 7 Feb 2002 18:35:32 +1100
• References: <342102887.1013070987@localhost>

From: "James Clark" <jjc@j...>
 
> Interesting.  Those are compelling use cases but this significantly 
> complicates things.  

Having an element or entity that just contains a combining character
looks dangerous.  This was discussed at the i18n SIG at W3C Boston 
last year. The behaviour of a combining character at the begining
of a file when displayed or edited is not defined AFAIK: so you are
at the whims of the system. Some systems, on seeing
 <!ENTITY tilde '~' > 
where ~ is the non-spacing (combining) form will just strip it out. 

So a combining character without a base is not "text", and therefore
not expressable with XML, at least as best practice.

Cheers
Rick    

• References:
  • Re: A heavier-weight proposal for character entitydefinition
    • From: James Clark <jjc@j...>

• Prev by Date: Re: Using <choice> to enforce rules
• Next by Date: RE: DTD validation without a DOCTYPE
• Previous by thread: Re: A heavier-weight proposal for character entity definition
• Next by thread: Re: A heavier-weight proposal for character entitydefinition
• Index(es):
  • Date
  • Thread