Re: [tlug] japanese encoding question

[Katsuhiko Momoi <momoi@example.com>]

Brett Robson wrote:

>Hi, I thought I had this Japanese encoding stuff sorted out but my
>industrious Japanese colleagues seem to find new things to confuse me.
>
>I've got a problem with a character on one of our pages that is breaking
>some browsers. It is a roman numeral 2, ie it looks like II but one
>character.
>
>Its SJIS encoding is x8755 and according to MS IME has a kuten of 1322.
>However JIS X 0208:1997 does not have a row 13, it jumps from row 08 to
>16 (start of kanji).
>
>Does anyone know what's going on here? Where did this row 13 come from?
>Is row 13 a Microsoft "enhancement" or is their a latter JIS X 0208 that
>I can't find?
>
There is a collection of vendor specific characters in Shift_JIS and 
they are mapped to some empty rows of JIS X 0208 for convenience. The 
character you mention is one of these. If you use Windows, the Roman 
Numeral Two under question is 0x8755 but under Mac OS Japanese (Mac's 
Shift_JIS) it maps to 0x85A0. Since row 13 does not exist in the 
official JIS X 0208, if your platform uses JIS X 0208 as a mapping 
mechanism, e.g. Unix, the character isn't there and it won't be 
displayed. If you have a public web site visited by users of different 
platforms, you should avoid using these vendor-added characters under 
Shift_JIS.

- Kat

-- 
Katsuhiko Momoi
e-mail: katmomoi@example.com