On Saturday 13 September 2008 11:43:03 am Benson Margulies wrote:
> Mapping either of these to a restricted xsd:string runs them straight
> into the 'valid XML 1.0 Unicode subset' issue. More fun yet, a Java
> 'character' will hold a non-BMP Unicode character, and those range
> from 0 to 2**32-1, sort of. I think that mapping them to restrictions
> on positive integers makes more sense.

Well, I'm not so sure.    A Java String COULD be mapped to a list of Integers 
for the same reason, but I don't think anyone would argue that's the best 
option.   :-)

Seriously, is there anyway to make it configurable.    The main reason is that 
I think on some platforms/toolkits, if they see a String with maxLength 1, 
they'll map it to a char.   Thus, it may make a little more sense for them.

Also, if you get an "integer", it's up to the caller to then realize that it 
represents a unicode value and they would need to handle the codepage 
conversion and such.   That's easy in Java, but not so easy in things like 
C/C++.  If it's a String, it's kind of assumed the underlying toolkit has 
done whatever is needed to convert it to whatever is needed on that platform.

Anyway, if it could be at all configurable, I think that would be great.   I 
think the restricted string would work for 95% of the usecases.  

-- 
Daniel Kulp
[EMAIL PROTECTED]
http://www.dankulp.com/blog

Reply via email to