Re: Encoding name problems when running Jigsaw with ORP plus Classpat h

[Chris Gray]

"Aaron M. Renn" wrote:

> Wu, Gansha (address@hidden) wrote:
> > This is the third of the series.
> >
> > The encoding name convention is different between Classpath and standards. 
> > For instance, ISO-8859-1 appears in Classpath like "8859_1", I wonder why 
> > Classpath does this way, because "8859_1" even didn't appear in IANA's 
> > official names of character set 
> > (http://www.iana.org/assignments/character-sets) and J2SE documentations 
> > (e.g, 
> > http://java.sun.com/j2se/1.3/docs/api/java/lang/package-summary.html). 
> > While Classpath uses these names to locate decoder classes, a mapping 
> > should be done somewhere. Classpath provides a alias mechanism to use 
> > System.properties to do the mapping. But who will insert these properties?
> > Any guys have ideas about this?
>
> Good question.  IIRC, I choose 8859_1 because the Japhar or someone else used
> it.  There are so many variants of character set names people might specify
> that I put in this alias mechanism.  It is also necessary to handle
> encoding class names that have -'s.  You are correct that no work has been
> done to set this up however.

Sun's printed documentation for Java2 mentions a whole load of funky charset 
names when
describing the String class.  8859_1 is there, together with MacCroatian.

The great thing about Java standards is , there's so many to choose from. :)

Chris Gray

VM Architect, ACUNIA