> ttf2tfm: ERROR: Invalid platform and/or encoding ID.
> Valid PID/EID pairs are:
> (1,0)
> (3,1)
> (3,10)
> #########################
> So I can say that this font have three cmaps, they are:
> Macintosh's Roman,(for Macintosh platform ID=1,and Roman's
> encoding ID is 0);
Ignore it.
> Microsoft's Unicode 2.0,(for Microsoft platform ID=3,and Unicode's
> encoding is 1);
Ignore it.
> Microsoft's UCS-4,(for Microsoft platform ID=3,and UCS-4's
> encoding is 10).
This is the right one (I forgot about it, sorry), since only 3,10 can
address character values > U+FFFF.
[Please insert a space after a comma or full stop; this makes it much
easier to read your mails.]
> 1. This font can be manipulate into subfonts only by your full
> unicode.sfd, because it have the PID/EID pair (3,10), is this true?
Not necessarily, but chances are high that there are values > U+FFFF.
> 2. In addition, if a font, say, just has the PID/EID pair (3,3),
> which is Microsoft's GB 2312, then how can I use it to generate
> subfonts so that I can use it in the Uniocode environment, is this
> the *virtual fonts* concept or not?
More or less, yes. For the example you mention, you need EUC.sfd to
generate subfonts which can be directly used with GB 2312 input
encoding. Then you can create virtual Unicode subfonts (which map to
those GB 2312 subfonts) with the sfd2uni.pl script.
> 3. In my tex source file, if I use the UTF8 environment, then must
> the tex source file itself be coded in UTF8?
Of course. Have you come to a different conclusion, or do you think
that the documentation implies something different?
> And is this thing just the same case for GBK and GB2312
> environments?
The `encoding' parameter of the CJK environment specifies the used
input encoding for the LaTeX file. In case the font encoding differ,
there is another optional parameter `fontencoding' (e.g., `HL' for
HLaTeX or `dnp' for Wadalab fonts). However, this optional parameter
can't be used to map, say, Unicode to GB 2312 or vice versa.
Werner
_______________________________________________
Cjk maillist - [email protected]
http://lists.ffii.org/mailman/listinfo/cjk
