On 5 Oct 2001, Lars Marius Garshol wrote: > I've received data encoded in ISO 2022-JP that I am unable to figure > out how to map to Unicode. The characters in question do not appear in > the old JIS0208.TXT, and I can't find them in UniHan.txt either
> These characters are the problem: > > (7A22, 7C22, 7964 and 7B64) Assuming what I wrote below is correct (and I didn't make any mistake in 'math'), they are: u+605D, u+91DE, u+5953, u+FA21 > Does anyone know what characters these are, and how to map them to > Unicode? Are they part of some vendor extension to JIS 0208? If so, I'm not sure, but it seems like they're a part of NEC Kanji character set (ref. Ken Lunde, CJKV Information Processing, p. 592). According to CJKV Information Processing NEC Kanji adds 360 Kanjis and 14 Non-Kanji from IBM Japanese character sets in rows 89-92. IBM Japanese is listed as 'kIBMJapan' in UniHan.txt. > does anyone know of a conversion table for that extension? Provided that what's above is the case, I guess you can rather easily construct a conversion table from UniHan.txt by comparing the table in p.585 of CJKV I.P. and the table in p. 594 of the same book. Jungshik Shin
