Erik van der Poel <[EMAIL PROTECTED]> writes: > > > > CMap Character Set Encoding Notes > > ------------------ -------------------- --------------- ----------------- > > RKSJ-H/V JIS X 0208-1990 Shift-JIS HW JIS-Roman @ 1B > > According to Lunde's book, RKSJ stands for Roman, (half-width) Kana and > Shift-JIS. This means that it also has single-byte half-width Katakana > in it. The Roman and half-width Kana are from JIS X 0201-1976.
Well, presumably. The Adobe docs didn't mention this, but they also assumed that you know what RKSJ means... > > UniJIS-UCS2-H/V ISO 10646-1:2000 UCS2 P Latin in ASCII ra > > Is it really all of ISO 10646-1:2000? Or just the JIS X 020[18] > character sets, in UCS-2 encoding? Uhh... I can't tell. It looks like it covers the JIX X stuff. But without sitting down with a copy of all of Unicode 3.0 and grinding over the CMap I can't say for certain. It *looks* like it covers a lot, though. > > Hankaku - - HW Latin & hiragana & > > katakana @ 1B > > Are you sure there are Hiragana in there? Hey, that's what the Adobe docs said. I have no clue, personally. Haven't looked yet. > > Hiragana - - HW hiragana @ 1B > > Half-width Hiragana? Seems unlikely to me. Again, that's what Adobe claims. It's presumably possible that there's HW hiragana in some fonts, since these may be needed by publishers. But I think the mere human needs them not. > > CMap XLFD CHARSET_REGISTRY XLFD CHARSET_ENCODING > > ------------------ ----------------------- ----------------------- > > H/V JISX0208.1983 0 > > H/V JISX0208.1990 0 > > The JISX0208.19xx fonts in the X world are normally only used for > horizontal layout. You would probably need to use some other XLFD field > for the vertical fonts (or just skip those for now). That's right. > > RKSJ-H/V JISX0208.1983 0 > > I may be misunderstanding the machinery that you are going to use, but > if you are going to use the encoding as is, then RKSJ is definitely not > the same as JISX0208.1983. RKSJ is in Shift-JIS encoding. Or does the > machinery take care of transcoding from JISX0208.1983-0 to RKSJ? I don't know. I didn't see if this worked. So it probably doesn't. > > EUC-H/V JISX0208.1983 0 > > EUC has the 8th bit on both bytes set to one, so it is JISX0208.1983-1, > which is deprecated, so you might want to just skip this one. Yeah, I discovered that. > > UniJIS-UTF8-H/V ISO10646 1 > > UTF8 is not the same as ISO10646-1. Right, UCS-2 is what is wanted. That was a typo. > > The JIS X 0212-1990 > > standard does not include Latin glyphs. > > It does contain accented Latin characters, but I don't know whether the > Adobe fonts contain glyphs for those. Really? It does? The idea was for these CMaps to apply to *any* CID-keyed font, not just Adobe's. So if it's missing something then the CMap needs updating. What fonts contain and what *should* be mapped according to standards are two different things. I've gotten distracted trying to get XEmacs to do Japanese input. I'll be reapproaching the font problem when I get that going (which shouldn't take long). I need *something* to test with. 'james -- James A. Crippen <[EMAIL PROTECTED]> ,-./-. Anchorage, Alaska, Lambda Unlimited: Recursion 'R' Us | |/ | USA, 61.20939N, -149.767W Y = \f.(\x.f(xx)) (\x.f(xx)) | |\ | Earth, Sol System, Y(F) = F(Y(F)) \_,-_/ Milky Way. _______________________________________________ Fonts mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/fonts
