Re: [tlug] oasys -> linux

[Edward Middleton <edwardmiddleton@example.com>]

On Sat, 2003-07-26 at 21:10, Tony Laszlo wrote:
> ISSHO just received some floppies from a 
> person who wishes to donate some interesting 
> lexicological data. Mostly East Asian characters. 
> It was produced on a machine running OASYS. 
> files look like this: 
> 
> 823081 Jul 24 21:24 ???6??120.OA2
> 369541 Jul 24 21:20 bf0ussvu.usa
> 640 Jul 24 21:24 oasbind.mng
> 4 Jul 24 19:33 oasys2.pat
> 
> It is rumored that doccat and perhaps 
> one other package from dehenken.co.jp 
> can convert this stuff, but I couldn't 
> find a doccat trial version that would 
> work. 
> 
> Two questions. 
> Is there anything out there that can 
> handle this conversion?
> Is there any way for this person to 
> generate the data in a format that is 
> more convertible (knowing only that 
> he is working with some kind of 
> word processing software on oasys 
> that can handle mutiple languages, 
> e.g., cjkv ?
> 
> Thanks!
> 
> Tony Laszlo
> http://www.issho.org/modules.php?op=modload&name=phpWiki&file=index&pagename=LaszloBlog
> (going for the record: blog with the longest URL)

I don't have a solution but this might get you started.

According to "CJKV Information Processing" by Ken Lunde OASYS word
processors uses the Fujitsu Japanese processing Extended Feature JEF
character set.

It includes
JIS C 6226-1978		6,802
JEF Extended kanji	4,039
JEF Extended non-kanji	1,010
User-defined characters	3,102

and the encoding for the JIS C 6226-1978 character set is identical to
EUC-JP.

According to this email
http://oss.software.ibm.com/pipermail/icu4c-support/2002-April/000201.html
It doesn't map very easily to Unicode.

Edward Middleton