Re: [tlug] Re: font/char set question

["Jim Breen" <jimbreen@example.com>]

On 29/07/07, steven smith <sjs@example.com> wrote:
> Jim Breen wrote:
> > (BTW, font is not the issue here. It's all to dowith character
> > sets.)

> Not all fonts implement all the gliphs for all the
> characters in a character set, right?

Partly. Font files are invariably associated with a character set.
The ones for Japanese are associated with either JIS X 0208 (with or
without the JIS X 213 extensions) or Unicode/ISO 10646. Most such font
files implement all (in the case of JIS X 0208) or most (in the case of
Unicode) codepoints. (A glyph is associated with the shape of a particular
characater, e.g. whether the kusakanmuri radical in a kanji has 3 strokes
or four.)

> or am I just
> confusing myself here.

I think you are.

>   The dot or in this case the small
> box -- where does that come from?  Is it output from the OS
> or the application or just the character representation for
> an unimplemented character.  This is the character
> substitution you are talking about, right?

Yes. If you see a dot/lozenge/?/etc. in a field when you paste
some text, the substitution will have been done locally, e.g. by
the browser. If the dot/lozenge/?/etc. is in something sent out
by the server, it could be a result of either browser substitution
before the form contents were sent in by HTTP, or by the server.

I know this smacks of hypocracy, as my servers use EUC-JP (albeit
with UTF-8 available), but things would be much better if everything
changed over to Unicode/UTF-8 ASAP.

Jim

-- 
Jim Breen
Honorary Senior Research Fellow
Clayton School of Information Technology,
Monash University, VIC 3800, Australia
http://www.csse.monash.edu.au/~jwb/